Kant Yearbook Volume 12 (2020) Issue 1 (Sep 2020) Kant and Neo-Kantianism

Table of contents :
[18684602 - Kant Yearbook] Titelseiten
[18684602 - Kant Yearbook] On Method_ The Fact of Science and the Distinction between Natural Science and the Humanities
[18684602 - Kant Yearbook] Durkheim’s French Neo-Kantian Social Thought_ Epistemology, Sociology of Knowledge, and Morality in The Elementary Forms of Religious Life
[18684602 - Kant Yearbook] Eduard von Hartmann, die Postulatenlehre und die Genese des Neukantianismus im Kontext spiritistischer Debatten um 1900
[18684602 - Kant Yearbook] The Problem of Schematism in Kant and its Transformation in Southwest Neo-Kantianism
[18684602 - Kant Yearbook] Cassirer and Kant on the Unity of Space and the Role of Imagination
[18684602 - Kant Yearbook] Cohen’s Logik der reinen Erkenntnis and Cassirer’s Substanzbegriff und Funktionsbegriff
[18684602 - Kant Yearbook] List of Contributors
[18684602 - Kant Yearbook] Topics of the Kant Yearbook 2021, 2022 and 2023

Citation preview

Kant Yearbook 12/2020

Kant Yearbook 12/2020 Kant and Neo-Kantianism

Edited by Dietmar H. Heidemann (University of Luxembourg) Editorial Assistant: Sven Seidenthal (University of Luxembourg) Editorial Board: Henry E. Allison (University of California, San Diego), Karl Ameriks (Notre Dame), Gordon Brittan (Montana State University), Daniel O. Dahlstrom (Boston University), Klaus Düsing (Universität zu Köln), Corey Dyck (University of Western Ontario), Kristina Engelhard (Universität Trier), Brigitte Falkenburg (TU Dortmund), Hannah Ginsborg (University of California at Berkeley), Michelle Grier (University of San Diego), Thomas Grundmann (Universität zu Köln), Paul Guyer (Brown University), Robert Hanna (Independent Philosopher), Markus Kohl (Chapel Hill), Lothar Kreimendahl (Universität Mannheim), Guido Kreis (Aarhus University), Georg Mohr (Universität Bremen), Angelica Nuzzo (Brooklyn College/CUNY), Dieter Sturma (Universität Bonn), Robert Theis (University of Luxembourg), Jens Timmermann (University of St Andrews), Ken Westphal (Boğaziçi Üniversitesi), Marcus Willaschek (Universität Frankfurt/Main)

The Kant Yearbook is an international journal that publishes articles on the philosophy of Immanuel Kant. Each issue is dedicated to a specific topic. Each annual topic will be announced by way of a call for papers. The Editorial Board of the Kant Yearbook is com­posed of renowned international experts, and selects papers for publication through a double blind peer review process. Online access for subscribers: http://dx.doi.org/10.1515/kantyb

ISSN 1868-4599 (Print) ISSN 1868-4602 (Online) Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at: http://dnb.dnb.de. © 2020 Walter de Gruyter GmbH, Berlin/Boston Cover image: Martin Zech, Bremen Printing and binding: CPI books GmbH, Leck www.degruyter.com

Brigitte Falkenburg

On Method: The Fact of Science and the Distinction between Natural Science and the Humanities Abstract: This article examines Cohen’s “transcendental method”, Windelband’s “critical method”, the neo-Kantian distinctions between natural science and the humanities (i. e., human or cultural sciences), and Weber’s account of ideal-typical explanations. The Marburg and the Southwest Schools of neo-Kantianism have in common that their respective philosophies of science focused on method, but they substantially differ in their approaches. Cohen advanced the “transcendental method”, which was taken up and transformed by Natorp and Cassirer; later, it became influential in neo-Kantian approaches to 20th century physics. Windelband distinguished between facts and values, linking the former to the “genetic” method of history and the latter to the “critical” method of philosophy; and between the “nomothetic” and “idiographic” methods of the empirical sciences, a distinction further elaborated by Rickert. The distinction does not give rise to a sharp discrimination but is rather what Weber would later call an ideal type. All these approaches contribute in different ways to understanding the structure of scientific knowledge, focusing on different aspects of the general path of the empirical sciences between rationalism and empiricism.

Introduction The founders of the Marburg and the Southwest schools of Neo-Kantianism, Hermann Cohen and Wilhelm Windelband, had to a certain extent common philosophical grounds. Both opposed the psychologistic interpretation, which had shaped the reception of Kant’s transcendental philosophy after 1800. Early 19th century philosophy was marked by the contrast between positivism and idealism. However, both traditions interpreted Kant’s epistemology in psychological terms. Hegel contributed to this view in post-Kantian German idealism; whereas Fries and his followers carried psychologism on in a first influential movement back to Kant. On the side of positivism, following Comte, many philosophers appealed to the triumphs and advances of the empirical sciences. One of the relevant developments from the middle of the 19th century onwards was the emergence of physiology and scientific psychology, and the associated program of naturalizing the human mind. Helmholtz was the main contributor to the natuhttps://doi.org/10.1515/kantyb-2020-0001

2

Brigitte Falkenburg

ralization of psychology by doing research on its physiological foundations. In particular, he interpreted Kant’s principle of causality as a structure of consciousness that is based on physiological processes and can be investigated by means of empirical psychology. Cohen and Windelband sharply criticized Helmholtz’ program. Both advocated a “critical idealism” which focused on legitimizing the objective validity of Kant’s principles (Beiser 2014, 492– 493). Despite the affinities of their interpretation of Kant, their versions of critical idealism and their approaches to philosophy and the sciences differed substantially. Cohen introduced the “transcendental method” and advanced its role for the foundations of the exact sciences, in The Principle of the Infinitesimal Method and its History (1883) and the second edition of Kant’s Theory of Experience (1885). His Logic of Pure Cognition (1902) transformed the transcendental method into a logic of origin (Logik des Ursprungs, Cohen 1902, 28 ff.). His contributions to the foundations of the humanities (the human or cultural sciences)¹ were not so clear, even though the second and third part of his philosophical system covered ethics and aesthetics (1904; 1912) and his system explicitly aimed at a philosophy of culture. Windelband, in contrast, worked out a “critical method”, according to which the principles of philosophy are grounded in universal values. In addition, he emphasized the methodological distinction between the “nomothetic” and the “idiographic” methods of the empirical sciences (1894) and accused Cohen of neglecting the human sciences. Regardless of these differences and because of their common opposition to psychologism, Cohen and Windelband both relied on a rationalistic account of cognition, focusing on the methods of the natural and human sciences and the corresponding conditions a priori of possible experience. In addition, both neo-Kantians conceded that scientific cognition changes with time. Their views about the rational foundations and the historical dimension of scientific cognition were in a certain tension. In the end, both attempted at resolving this tension through opposing approaches. (1.) The neo-Kantians of the Marburg school tended towards logicism and constructivism. (1.1) Cohen’s transformation of Kant’s critical method reflects the above mentioned tension in the relation between his “transcendental method” and the “fact of science”. (1.2) In order to make the fact and the method compatible, he explained the constitution of scientific facts in terms of the “infinites-

 In the literature on Dilthey and neo-Kantianism, the expressions “human science” (Geisteswissenschaft) and “cultural science” (Kulturwissenschaft) have become established for the humanities. Therefore I use them in the following.

On Method

3

imal method”. (1.3) He advanced a constructivist account of science, which Natorp and Cassirer shared and which became influential in neo-Kantian approaches to 20th century physics. (2.) The neo-Kantians of the Southwest School focused on differentiating between historical and rational methods and could not avoid the predominance of the historical elements of cognition. Windelband marked a sharp distinction between contingent historical facts and universally valid values. On this basis, he distinguished (2.1) the “critical” method of philosophy from the “genetic” method of the empirical sciences, and (2.2) the “nomothetic” method of natural science from the “idiographic” method of the human sciences, a distinction criticized by Dilthey and further elaborated by Rickert. (2.3) Max Weber finally made the balance between facts, values, and the rational and historical elements of scientific cognition, in his approach to social science. His method of ideal-typical explanations suggests a going back and forth between constructive modelling and empirical corrections of the models. (3.) My general conclusion is that all of these neo-Kantian approaches contribute important insights to understanding the structure of scientific knowledge, given that they focus on different aspects of the general path of the sciences between rationalism and empiricism.

1 The “Transcendental Method” and the “Fact of Science” Cohen interpreted Kant’s critical method as a transcendental method which aims at demonstrating the conditions of the possibility of objective experience. He identified the objects of experience with the “fact of science”, that is, the structure of Newtonian physics. It is well known that Kant himself never used the term “transcendental method”. Kant distinguished three methods of philosophy: the traditional analytic-synthetic method of early modern science and philosophy, the skeptical method and the critical method. Cohen reinterprets the critical method in terms of the “regressive” step of the analytic method, an issue which is in need of explanation. In his pre-critical philosophy, Kant employed traditional analytic and synthetic methods. In early modern science, both parts of the method had been considered to belong together as a two-step regressive-progressive procedure of first developing the principles of a theory (in the regressive, analytic step) and then justifying it (in the progressive, synthetic step). The method traced back to ancient geometry, to Pappus’ commentary on Euclid’s geometry (Pappus 1589). Galileo and Newton transformed it into what became known as the resolutive-com-

4

Brigitte Falkenburg

positive or analytic-synthetic method of early modern science. Newton described it in the Opticks (Newton 1730, 404 – 405), Roger Cotes in his preface to the second edition of Newton’s Principia (Newton 1713, 386). In post-Cartesian philosophy, both parts of the method were split and distinguished as an inductive analytic method, or “method of discovery”, on the one hand, and a deductive, synthetic method, or “method of instruction”, on the other, as explained in the influential Logic of Port-Royal (Arnauld and Nicole 1685, 233), the latter being the method “more geometrico” of Spinoza’s Ethics. Kant, in his famous Prize Essay of 1764, considered it the appropriate method of mathematics, in contradistinction to the “analytic” method of philosophy. However, in his defence of the analytic method for the purposes of philosophy he emphasized that, in analogy to Newtonian science, both steps of the method should be employed in order to derive reliable metaphysical concepts (UD, AA 02: 286). He repeated this methodological stance in the preface to the second edition of the Critique of Pure Reason, in the passage on the “experiment of pure reason” (CPR B XVIII–XXI).² After his critical turn, he no longer considered analysis and synthesis in the traditional sense as an appropriate method of deriving and justifying metaphysical concepts with objective validity, but only as a heuristic tool for presenting a doctrine. Making use of this, he then reversed the method, according to which analysis should always precede synthesis (Newton 1730, 404). The first Critique presents the doctrine of transcendental idealism following the synthetic method; whereas the Prolegomena explains the principles of transcendental idealism following the analytic method, in order to make this doctrine more accessible to the readers (Prol, AA 04: 263; 274– 279). In contrast to the traditional analytic-synthetic methods, Kant’s critical method aims at proving the principles of pure reason as synthetic judgments a priori with apodictic certainty and objective validity. Kant claimed that the strength of his critical method lay in what we today call transcendental arguments (Stern 2019). The same goes for Cohen and his “transcendental method”, but a closer look at Cohen’s version of this method however reveals two features that are at odds with one another.

 See Falkenburg (2018a, 17– 18; 2018b). Kant’s pre-critical analytic method and its later transformations are discussed in detail in Falkenburg (2020a), Chapters 2 and 6. For the traditional analytic-synthetic method see Engfer (1982) and Beaney (2018). – The following sections 1.1– 1.3 are partially based on Falkenburg (2020b).

On Method

5

1.1 The transcendental method Cohen explains that he follows the regressive or analytic method of the Prolegomena rather than the progressive or synthetic method of the first Critique (Cohen 1877, 24– 25). But he also claims that the transcendental method aims at concepts and principles a priori which are sufficient and necessary to establish and reinforce the fact of science.³

Accordingly, Geert Edel emphasizes: On closer examination, however, it becomes apparent that the ‘transcendental method’ is not merely a variant of Kant’s theoretical program, but rather an explicit stipulation of the analytic mode of presentation against Kant, against the synthetic approach of the ‘critique of reason’ and thus a radical departure from the original Kantian theory conception, with which Cohen simultaneously carries out the […] transformation of the concept of epistemology itself. (Edel 2010, 95)

This “radical departure” from Kant’s transcendental philosophy goes together with transforming epistemology, the “theory of cognition” (Erkenntnistheorie), into a “critique of cognition” (Erkenntniskritik), as Cohen explains in the Infinitesimal Method (1883, 103 – 104). This transformation of epistemology is closely related to his view that experience in Kant’s sense is identical with the objects and principles of mathematical physics. For Cohen, Kant’s concepts and laws a priori are the mathematical concepts and laws of Newton’s physics. By suggesting to take cognition […] as a fact, which came about in science and continues to complete itself (Cohen 1883, 103),

he identifies the objects of Kant’s Critique of Pure Reason with the “fact of science”, i. e., with the results of the successful natural science of his time, and in particular, with the objects and principles of mathematical physics. In the Infinitesimal Method, he clearly expresses the crucial features of his reinterpretation of Kant’s critical method and its relation to the “fact of science”, without explicitly calling his critique of cognition then the “transcendental method”:

 “[…] welche hinreichend und nothwendig sind, das Factum der Wissenschaft zu begründen und zu festigen.” Cohen (1885, 77; my translation). For the relation between Cohen’s transcendental method Kant’s analytic method of the Prolegomena, see also Edel (1987, 20 ff.; 2010, 69 – 70; 94– 95); Luft (2015a, 51); Pringe (2017, 116 – 117).

6

Brigitte Falkenburg

The critique of cognition […] resolves science into the presuppositions and foundations assumed in and for its propositions. Cognitive-critical idealism takes as its objects not so much things and events, nor even consciousness as such, but scientific facts. […] then cognitive-critical idealism is the scientific form of idealism, [a form] which is attained by the concept of the transcendental. For transcendental refers to the possibility of a cognition that receives the value of a priori or scientific validity (Geltung). Hence cognitive-critical idealism is synonymous with transcendental logic, for its task is to discover the fundamental synthetic propositions, or the grounding (Grundlegung) of cognition on which science builds itself, and on whose validity the latter depends. (Cohen 1883, 104)

Compared to idealism in general on the one hand and epistemology on the other hand, his “critique of cognition”, or the transcendental method, has four distinguishing marks: 1. It proceeds from the “propositions of science” to their “presuppositions and foundations”, following the regressive or analytic method. 2. It does not start from “things and events”, but from the “scientific facts”, in particular, the objects and principles of mathematical physics. 3. It is “transcendental” in referring to the “possibility of a cognition” which has objective, “scientific validity” and is a priori. 4. It is identical or “synonymous with transcendental logic” and aims at discovering the “synthetic propositions” which are the foundations a priori of scientific cognition. To employ (1.) the analytic method with (2.) the facts of science as a starting point suggests an inductive procedure, which is supposed to trace from the historical stage of natural science back to the underlying principles, and which is not expected to be stronger than an inference to the best explanation. To consider (3.) the method as capable of demonstrating the objective validity and apriority of these principles and to identify it (4.) with transcendental logic, in contrast, raises a much stronger claim. It claims that the transcendental method is as strong as Kant’s transcendental deduction of the categories, that is to say, it is capable of demonstrating that the principles obtained in this way are necessary and sufficient to establish the facts of science (Cohen 1877, 77). If Cohen’s appeal to the analytic method is to be taken seriously, both claims seem incompatible. Cohen’s critical idealism and the transcendental method are due to his antipsychologistic interpretation of Kant, which also led him to reject the doctrine of space and time as pure forms of intuition, starting from the second edition of Kant’s theory of experience (Cohen 1885). His critical idealism was directed against the positivistic zeitgeist, according to which objective knowledge is based on scientific experience, understood as the cognition of empirical facts

On Method

7

and things, and according to which cognition refers to the contents of consciousness. However, as is so often the case, this critically opposed attitude was decisively influenced by the zeitgeist. Cohen’s “critique of cognition” reduces Kant’s transcendental philosophy to a metatheory of the exact sciences. But his view that philosophical cognition must meet, or even exceed, the standards of objectivity of the exact sciences not only corresponded to the positivist paradigm of the late nineteenth century. In addition, it was also present in the Cartesian tradition of rationalism. German idealism after Kant, and above all Hegel, had rejected Descartes’ mathematical ideal of cognition. By assuming that the objects of objective knowledge were not empirically given, but provided by natural science, and in particular mathematical physics, Cohen thus stood in opposition to post-Kantian German idealism as well as the positivism of his time. His background for this assumption was ultimately Plato’s view that mathematics plays a decisive role in the ascent to cognition of ideas. Cohen’s Neo-Kantianism was thus a mathematically founded idealism, characterized on the one hand by a rejection of Hegel’s philosophy and on the other by sharp opposition to empiricist views. To rely on the “fact of science” and to insist on its time-dependence introduces an empirical element in Cohen’s transcendental method, which is obviously in a certain tension with his strong claim of providing the necessary and sufficient logical foundations of this fact by means of the transcendental method. In an instructive study of this tension, Ursula Renz argues that such a combination of rationalistic and empiricist elements may give rise to a consistent epistemology, if rationalism is understood as the position that assumes the intelligibility of the world, on the one hand, and that nonetheless, in its concrete claims concerning scientific cognition, depends on the historical stage of science, on the other (Renz 2018, 10). Indeed, Cohen was not alone with the ambiguity of his transcendental method between rationalism and empiricism. It reflected the methods of Newtonian science itself, and in addition, also of Kant’s own transcendental philosophy.⁴ The followers of neo-Kantianism later attempted to resolve the ambiguity by interpreting the foundations of physics in terms of

 DiSalle (2013) discusses the ambiguity of Newton’s scientific methodology between rationalism and empiricism. The many facets of the path of physics between rationalism and empiricism in 19th and 20th century are investigated by Scheibe (2001; 2007). That even Kant’s critical theory of nature, and in particular, the Metaphysical Foundations (AA 04), contain substantial empirical, or inductive, elements has been investigated in our research project on inductive metaphysics (DFG FOR 249, https://indmet.weebly.com/#); see Engelhard et al. (2020), Engelhard (forthcoming), Seide (2020, Chapter 4.2), Seide (forthcoming).

8

Brigitte Falkenburg

a relative a priori, from Reichenbach (1920) to Heisenberg (1959) or Mittelstaedt (1964; 2013). To summarize, Cohen’s (re‐)interpretation of Kant is based on the following four assumptions, which are closely related to his account of the transcendental method: (i) The object of Kant’s general theory of nature is the “fact of science”, and in particular, the successful theories of mathematical physics. (ii) Experience, for Kant, means scientific cognition, or: the objects of the Critique of Pure Reason are the scientific phenomena and principles of Newtonian science. (iii) Kant’s critical method is identified with the “transcendental” method, which aims primarily at providing the foundations of the exact sciences. (iv) Kant’s account of intuition as a cognitive faculty that contributes to the foundations of scientific cognition has to be rejected. The first three assumptions characterize the Kant interpretation of the Marburg school, from Cohen to Natorp and Cassirer. They bring Kant’s Critique close to his Metaphysical Foundations of Natural Science (1786). The fourth assumption stems from the 19th-century tradition of understanding Kant’s theory of intuition as a psychological theory and Cohen’s anti-psychologism. Cohen’s own systematic philosophy is characterized by turning away from a critique of cognition in Kant’s sense towards a new transcendental logic of cognition, which he presented in the Logic of Pure Cognition (Cohen 1902). The Infinitesimal Method (Cohen 1883) made the turn from Kant to developing this logic of cognition. Cohen’s mathematical claims were harshly criticized by Russell (1903, 338 ff.) and others, but recently defended by Mormann (2018) in view of nonstandard analysis. However, Cohen’s intention of 1883 was not so much to contribute to the philosophy of mathematics as to explain which conception in his system is to replace Kant’s theory of intuition.

1.2 The Constitution of Scientific Facts The Infinitesimal Method (Cohen 1883) is intended to show that the contents of an empirical science, which according to Kant are given in intuition, fall under the category of reality. Hence they can be expressed in terms of intensive quantities and thus are to be constructed according to the infinitesimal method, i. e., the differential calculus. The underlying idea is that Kant’s category of reality captures the sensory qualities (or Qualia) of empirical objects in terms of graduated concepts. The category of reality makes it possible to grasp the de-

On Method

9

grees of qualities in mathematical terms and to quantify them within a scale. For example, the degree of heat of a body is expressed in terms of temperature, which is a measurable intensive quantity. Other physical properties of bodies can also be expressed by intensive quantities, such as their hardness or elasticity; or their colour, which corresponds to a wavelength according to the physics of radiation (which was investigated in Cohen’s days by Kirchhoff, Planck and others). In this way, the empirical properties of things can made the subject of mathematical physics. For the mechanical properties of bodies, Kant developed the respective concepts a priori in the Metaphysical Foundations of 1786. Cohen generalizes this approach to the objects of scientific cognition in general and suggests to trace their cognition back to the mathematical concepts of a physical dynamics, such as Newton’s theory of gravitation. Physics is the theory of motions. Kant’s Metaphysical Foundations were based on the concept of motion, or of something moving in space. Mathematical physics describes the planetary motions and other physical processes in terms of differential equations that describe the interactions of bodies and fields. When Cohen wrote the Infinitesimal Method, several other theories existed in addition to Newton’s mechanics: Maxwell’s electrodynamics, thermodynamics, the foundations of hydrodynamics, and Boltzmann’s kinetic theory which explains heat by the motions of molecules. Since Newton (and Leibniz), the primary mathematical tool of any physical dynamics was the calculus. Therefore, Cohen used the infinitesimal concept for his project of conceptualizing or logifying Kant’s modal category of empirical existence by reducing it to the category of reality and its degrees. Cohen (1883, 27– 28) explained this issue as follows, by referring to Kant’s principles of pure reason and linking the Anticipations of Perception and the Postulates of Empirical Thought to the Analogies of Experience: 32. Reality and existence. – Let us finally try to envision the independent achievement of reality against the postulate of modality. The principle of existence refers to the cognitive value that is inherent in the facts of perception […]. But of those facts of perception A and B are presupposed just as much as in the analogies. If A and B are established, then I call that relation among them fact, which for the connection of A and B is based on perception. Perception itself, however, denotes a methodical tool of experience, similar to delineating a circle in geometry. […] If perception contained the sufficient criterion of objectivity, there would be no need for any further critical equipment: we would then know where we could grasp the infinite nature. Rather, perception and the reality corresponding to it denote only that very relation between the conceived elements of the equation; existence itself is the fundamental concept, the postulate of reality. Rather, perception and the corresponding reality only denote that relation defined by existence between the conceptual elements of the equation; existence itself is the fundamental concept, the postulate of reality […]. But this principle also presupposes reality […]. One could say that the principle of

10

Brigitte Falkenburg

the anticipations of perception contains the problem of the critique of cognition in a nutshell. […] so the completion of reality […] becomes necessary, which cuts through the knot of intuition: overcoming all extension […].⁵

This intricate passage provides insight to the following principles of the Infinitesimal Method: 1. Cohen reduces ‘existence’, the category of modality, to ‘reality’, the category of quality. Accordingly, he reinterprets the Anticipations of Perception and the Postulates of Empirical Thought in terms of the Analogies of Experience. 2. For him, perception is not a sufficient condition of objective cognition. In order to make its contents objective, he suggests to express the “facts of perceptions A and B” in terms of relational concepts. 3. Therefore, for him the facts of cognition are not the isolated contents A or B of perception, but only their relation: “If A and B are established, then I call that relation among them fact”. So far Cohen’s “critique of cognition”, which suggests to express the given contents or facts of perception in relational terms, based on the category of reality. For him, the objects of cognition are not given as isolated contents of perception, but constituted as relational scientific facts. In a further step, Cohen brings the relation of the Critique to the Metaphysical Foundations into play. The Critique describes the properties of things in terms of intensive magnitudes, according to the category of reality; or, in Cohen’s reinterpretation, as “intensive reality”. The Metaphysical Foundations describe (or reconstruct) the empirical properties of bodies by concepts a priori that can be subject to mathematical constructions.

 “32. Realität und Dasein. – Versuchen wir endlich, die selbständige Leistung der Realität gegenüber dem Postulate der Modalität zu vergegenwärtigen. Der Grundsatz des Daseins bezieht sich auf den Erkenntniswert, welcher den Tatsachen der Wahrnehmung beiwohnt […]. Aber von jenen Tatsachen der Wahrnehmung wird ebensosehr wie in den Analogien A und B vorausgesetzt. Wenn A und B gesetzt sind, so nenne ich dasjenige Verhältnis unter ihnen Tatsache, welches für den Zusammenhang von A und B auf Wahrnehmung beruht. Wahrnehmung selbst aber bezeichnet ein methodisches Mittel der Erfahrung, ähnlich wie das Kreisbeschreiben für die Geometrie. […] Wenn Wahrnehmung das zureichende Kriterium der Objectivität enthielte, so würde es aller weiteren kritischen Zurüstung nicht bedürfen: wir wüssten alsdann, wo wir sie fassen können, die unendliche Natur. Vielmehr bezeichnet die Wahrnehmung und die ihr entsprechende Wirklichkeit eben nur jenes durch Dasein definirte Verhältniss unter den gedachten Elementen der Gleichung; das Dasein selbst ist der Grundbegriff, das Postulat der Wirklichkeit […]. Aber auch dieser Grundsatz setzt die Realität voraus […]. Man kann sagen, der Grundsatz der Anticipationen der Wahrnehmung enthalte das Problem der Erkenntnisskritik zusammengefasst. […] so wird die Ergänzung der Realität […] nothwendig, welche den Knoten der Anschauung durchhaut: alle Extension überwindet […].” (Cohen 1883, 27– 28; my translation).

On Method

11

Identifying the objects of the Critique and the Metaphysical Foundations then gives rise to express Kant’s intensive magnitudes of the Anticipations of Perception, or “intensive reality”, in terms of the calculus (Cohen 1883, 28): 33. Intensive reality. […] Only in this way does the given attain that apparent solidity and independence which it by no means possesses within intuition. […] This presupposition of intensive reality is latent in all principles, and must therefore be made independent. This presupposition is the meaning of reality and the secret of the conception of the differential; the logical secret revealed by the critique of cognition. ⁶

This passage may be understood as supporting the following line of reasoning: 4. The mathematical concept of the infinitesimal is the foundation of the calculus, which serves to express the dynamic properties of physical phenomena. 5. The calculus describes how finite magnitudes are generated from infinitesimals, or, in Kant’s terms, how intensive magnitudes are constituted. 6. Therefore, for Cohen the calculus serves to express the (relational) “facts” of science, that is, the connections of the “given” contents of perception, and constitutes “intensive reality”. Therefore, Cohen considers the infinitesimal method as the key to understanding the constitution of scientific facts and the objects of scientific cognition. Of course, he does not want to claim that empirical things or their qualities can be generated from nothing by means of the infinitesimal method of mathematics, as the phrase “the meaning of reality and the secret of the conception of the differential” (§ 33) may suggest when read with misintent. He rather wants to emphasize that the dynamic character of reality can only be understood in an adequate philosophical way by employing the principle of the infinitesimal method. For him, to understand reality adequately means to dispense with Kant’s theory of intuition. Here, the phrase concerning the “completion of reality […], which cuts through the knot of intuition: overcoming all extension” (§ 32) is revealing. It marks a decisive contrast to Kant’s theory of nature. Kant’s project was to show, with the first Critique and the Metaphysical Foundations, how the empirical or dynamic properties of matter can be constructed in intuition. We

 33. Intensive Realität. […] Dadurch erst erlangt das Gegebene jene anscheinende Solidität und Selbstverständlichkeit, die es innerhalb der Anschauung keineswegs besitzt. […] Diese Voraussetzung der intensiven Realität ist in allen Grundsätzen latent, sie muss daher selbständig gemacht werden. Diese Voraussetzung ist der Sinn der Realität und das Geheimniss des Differentialbegriffs; das logische Geheimnis, das die Erkenntniskritik enthüllt.” (Cohen 1883, 28; my translation).

12

Brigitte Falkenburg

may understand Cohen here as follows. In order to adequately understand the objects of scientific cognition, we must replace the extensive magnitudes, which according to Kant can be constructed in intuition, by a mathematical conception of the intensive magnitudes. This spells the end for Cohen’s project. The main idea underlying it, which then also enters into his Logic of Pure Cognition of 1902, should be repeated here: The facts of science or objects of scientific cognition are not given, but constructed or constituted; and they do not consist in isolated facts, but in relational properties. Reinterpreting Kant’s Critique, Cohen suggests that the objects of scientific cognition are constituted, or objectified, as physical objects by means of mathematical methods. Their physical quantities then become subject to a physical dynamics which grasps the phenomena in mathematical terms of physical magnitudes and differential equations; from falling bodies and the planetary motions to electric currents, electromagnetic waves and thermodynamic processes; and in 20th century physics, far beyond. Cohen considers the objects of scientific cognition in terms of intensive magnitudes which are generated according to the principle of the infinitesimal method; not according to existence, but according to their relational structure. In this regard, Cohen’s approach to scientific cognition is strikingly modern. Concerning the relational structure of physical properties, it was taken up by Natorp (1910) and Cassirer (1910). Furthermore, concerning the constitution and objectification of physical objects, he had prominent followers in the philosophy of 20th century physics (Heisenberg 1959; Mittelstaedt 1998; 2009; Von Weizsäcker 1985).

1.3 A Constructivist Philosophy of Science However, Cohen’s approach also poses a problem. He replaces Kant’s category of (empirical) existence by the (logical) conception of intensive reality; he understands this conception in such a way that scientific facts are constituted according to the principle of the infinitesimal method. This is a kind of logicism that gives rise to a constructivist theory of science. Constructivism in a most general sense means the view that the objects of cognition are not given, but constructed; that is, made by human beings or depending on human thought and action in nature. Concerning this view, Cohen joins a long line of philosophers, from Hegel to Cassirer, Goodmann, Sellars, and Davidson. These philosophers represent very different philosophical doctrines; Hegel’s speculative idealism (Hegel 1807), Cassirer’s philosophy of symbolic forms (Cassirer 1923 – 1927), Goodman’s pluralism (Goodman 1978) which

On Method

13

takes up Cassirer’s symbolic idealism, Sellars’ naturalism (Sellars 1997), and Davidson’s criticism of the dualism of a conceptual scheme and its given conceptual content (Davidson 1973). What all these thinkers have in common is a strong criticism of the empiricist “myth of the given” (Heidemann 2002). Cohen enters here as the founder of constructivism concerning scientific cognition. For him, the objects of science are constructed or constituted in mathematical terms. He expresses this attitude very clearly in his Introduction, with Critical Remarks, to the Ninth Edition of Lange’s “History of Materialism” (Cohen 1896, 1914), in which he emphasizes that he regards Kant’s concept of the matter of sensation (CPR A 20/B 34) as an empiricist residual in Kant’s theory of nature: Such a remnant, not quite dispensed with, is the description of sensation as matter; although […] it is thought, alone and exclusively, can satisfy the demands of sensation; for the real, which is defined as the object of sensation, was recognized as a category, in which he grounded reality. (Cohen 1914, 121)

As a consequence, Cohen plays down the importance of empirical data for physics. This prompts him to adopt a constructivist interpretation of empirical science, in sharp contrast to the empiricism of the most influential physicist-philosophers of his time. Here, next to Helmholtz, Ernst Mach has to be mentioned. Cohen insists on the theoretical preconditions of scientific experience, which Kant put forward in preface B to the Critique of Pure Reason (B XII–XIII). Furthermore, in recent philosophy of science, it was Thomas S. Kuhn who argued that scientific experience is “theory laden” and who interpreted theory change in terms of paradigm shifts (Kuhn 1962). Cohen, whose constructivism concerning the facts of science was never adequately received, refers to Hertz’ Principles of Mechanics for support. Hertz writes about theory change in this field: […] to many physicists it appears simply inconceivable that any further experience whatever should find anything to alter in the firm foundations of mechanics. Nevertheless, that which originates in experience can again be annulled by experience. That all too common definition of the basic laws can evidently only arise because in it the elements of experience are somewhat concealed in them and mixed (verschmolzen) with the unchangeable elements necessary in thought (denknotwendig). (Hertz 1894, 9)⁷

In the history of classical mechanics, theory change only concerned the sequence of different “pictures” (Bilder) of theoretical mechanics. Hertz’s “pic Translation amended as by Patton in Cohen (2014, 126). Literally, “verschmolzen” means “amalgamated”.

14

Brigitte Falkenburg

tures”, representations or depictions of mechanics also received little attention for a long time. Today, they are understood in the sense of the models of a theory and Cohen discusses in detail their significance. In doing so he indeed made an important contribution to the contemporary debates on the atomistic and competing dynamic explanations of the structure of matter, as Patton (2004) shows in detail. Here, however, primarily the constructivist features of Cohen’s approach to Hertz’ mechanics deserve attention. Cohen discusses Hertz’ criteria of the permissibility (i. e., logical consistency), correctness (i. e., empirical adequacy) and purposiveness (i. e., structural clarity and simplicity) of physical theories, or their depictions (Cohen 1914, 125 – 127). With regard to correctness or adequacy, he comments on the above passage of Hertz’ Principles of Mechanics as follows: If such a mixture (Verschmelzung) of the logical elements with those of experience is possible, this demonstrates a connection of thought-elements whose intimacy is not captured adequately (hinreichend) by the expression ‘concealed’. If the latency is revealed, the effective, positive character of this connection must emerge. (Cohen 1914, 126)

As a matter of course he here subsumes the “logical elements” and “those of experience” under “thought-elements”; and then he insists on the “positive effect” of amalgamating the logical and the empirical elements of science, an effect which he relates to the possible surplus meaning of the fundamental laws of the theory in both directions. From here, Cohen makes the bridge from the question of the correctness or adequacy of a theory to its “fitness to the purpose”, i. e., the clarity and simplicity of its logical structure; and he emphasizes that this question really deals with the transcendental method, namely with thinking the concepts into the things to be constructed by means of them (Hineindenken der Begriffe in die mittelst ihrer zu konstruierenden Dinge) […]. (Cohen 1914, 127)⁸

In this way, Cohen establishes the tendency of the Marburg school of neo-Kantianism to interpret the theory dependence of experimental experience in such a way that even the values of physical quantities may be considered constructs of our thinking. Cohen’s constructivism was directed against the sensualistic conception of facts advanced by the empiricists of his day. This constructivist ten My translation. Patton translates: “[…], namely with the attempt to understand concepts in the things to be constructed using them” (Luft 2015b, 127). This is not completely off the point, but does not capture the constructivist aspect of Cohen’s transcendental method, which he clearly expresses here.

On Method

15

dency is later also found in Natorp’s and Cassirer’s philosophy of physics (Natorp 1910; Cassirer 1910; 1921; 1937). With this view, however, the neo-Kantians of the Marburg school are at danger of throwing out the “baby” of their critical theory of cognition, i. e., the insight into the theory dependence of all observations, experiments and data of empirical science, with the constructivist “bath water”, by no longer admitting independent scientific facts. Natorp (1910) further elaborated Cohen’s constructivist conception of science. To a certain extent he already reflected on the development of physics which began to take shape with the advent of quantum and relativity theory after 1900. Natorp considered scientific facts and the objects of the exact sciences not as given (gegeben), but rather as “posed as a task” (aufgegeben). In Natorp’s view, Cohen’s “fact of science” is not established once and for all, rather it is evolving. Cohen had conceded the historical dimension of the facts of science, but not explained how the “transcendental method” should take it into account. For Natorp, in contrast, the objects of scientific cognition are “fieri” (with the ambiguous meaning of becoming, or being created) rather than “fact” (Natorp 1910, 11 ff.). Their cognition or constitution was, for him, an “infinite task” (Natorp 1910, 16 ff.). In this way, he took up another element of Kant’s transcendental philosophy, namely the insight that in any empirical science the objects of cognition are constructed according to regulative principles, with the horizon of scientific cognition being continuously expanded. Even Albert Einstein in his later years referred to this insight, attributing it however to Kant himself rather than Natorp: I […] came to understand the truly valuable which is to be found in his doctrine […] only quite late. It is contained in the sentence, ‘The real is not given to us (gegeben) but put to us (aufgegeben) (by way of a riddle). (Einstein 1949, 680)

2 Facts, Values, and Ideal Types Concerning the tension between the rational and the historical elements of scientific cognition, the Southwest school of neo-Kantianism is closer to the other end of the scale which ranges from logicism (as a follower of rationalism) to historicism (as a follower of empiricism). Windelband never framed a philosophical system. His major philosophical work was on the history of philosophy. His systematic philosophical thoughts only entered into the repeatedly expanded collection of his lectures and essays, which range from the 1870s to the first decade of the 20th century, the Preludes (Windelband 1915a; 1915b). Like Cohen, he understood himself in the tradition of Kant’s transcendental philosophy but saw the

16

Brigitte Falkenburg

need of transforming it. In the preface to the first edition (1883) of the Preludes he claimed: “Understanding Kant means transcending him.”⁹ Windelband’s transformation also aimed at validating the principles of Kant’s philosophy in a renewed version. In contrast to Cohen, however, he did not focus on the fact of science, but on the validation of Kant’s principles in terms of values. His systematic philosophical thought was addressed to the distinction between history and philosophy, and their respective objects, that is, facts and values. For Windelband, the universal validity of the principles of Kant’s philosophy rests on values. The principles of pure reason correspond to the “True”, the principles of morality to the “Good”, and the principles of aesthetic judgments to the “Beautiful” (Windelband 1915a, 35). This approach to the ideal of philosophy and its concretization in the three main areas of philosophy became decisive for Windelband’s account of the “critical method”, his classification of the empirical sciences as “nomothetic” and “idiographic” disciplines; as well as Rickert’s transformation of this classification, and Weber’s approach to social science.

2.1 “Genetic” and “Critical” Method In the programmatic article What is Philosophy? (1882), Windelband elaborates on the distinction between contingent historical facts and universally valid principles with regard to the history of philosophy and the search for a universal conception of philosophy. There, he argues that in view of the multitude of historical doctrines of philosophy and their diversity it is impossible to define a general concept of philosophy; and furthermore, that the very attempt to define philosophy as a science also falls short of its specific character: Indeed, the historical facts demand that we refrain from such an unconditional subordination of philosophy to the concept of science […]. The historian’s open eye will rather have to see in it a multi-branched, proteus-like cultural phenomenon that cannot simply be schematized or categorized.¹⁰

 “Kant verstehen, heißt über ihn hinausgehen.” (Windelband 1915a, IV); translation by Pollok (2010, 348).  “In der Tat verlangen die historischen Tatsachen, von einer so unbedingten Unterordnung der Philosophie unter den Begriff der Wissenschaft […] Abstand zu nehmen. Der offene Blick des Historikers wird vielmehr in ihr eine vielverzweigte, proteusartige Kulturerscheinung sehen müssen, die sich nicht einfach schematisieren oder rubrizieren läßt.” (Windelband 1915a, 6; my translation).

On Method

17

He adds that in view of the many historical faces of philosophy it is also far from being clear how philosophy as a science might be distinguished from other scientific disciplines: Also to this second question there is no generally valid answer given by history […]. The sciences can be distinguished partly according to their objects, partly according to their methods: but in neither of these two respects can a characteristic be identified that is the same for all historical phenomena of philosophy.¹¹

Such a tension between the historical doctrines or systems of philosophy and the search for a general conception of philosophy had already been stated by Kant, in the Critique of Pure Reason (all quotations after Kant 2008): Historical cognition is cognitio ex datis, a rational cognition, however, cognitio ex principiis. […]. Hence he who has properly learned a system of philosophy, e. g., the Wolffian system, […] still has nothing other than a complete historical cognition of the Wolffian philosophy. (CPR A 836/B 864).

Kant strictly distinguishes such a historical system of philosophy (which corresponds to a scholastic concept of philosophy) from the cosmopolitan concept of philosophy, which is “personified and represented as an archetype in the ideal of the philosopher” (CPR A 838 – 9/B 866 – 7), and he claims: In this way philosophy is a mere idea of a possible science, which is nowhere given in concreto, but which one seeks to approach in various ways until the only footpath […] is discovered, and the hitherto unsuccessful ectype […] is made equal to the archetype. (CPR A 838/B 866)

Windelband follows Kant in claiming that the general concept of philosophy is “a mere idea”, claiming that the “universality that is at issue here is not a factual but an ideal one […]”.¹² He refers to the normative character which he attributes to the ideal of philosophy as a mark of distinction between the objects and methods of philosophy and those of the sciences, pointing out that:

 “Auch auf diese zweite Frage gibt die Geschichte […] keine allgemeingiltige Antwort. Unterscheiden kann man die Wissenschaften teils nach ihren Gegenständen, teils nach ihren Methoden: aber in keiner von beiden Hinsichten läßt sich ein für alle historischen Erscheinungen der Philosophie gleichbleibendes Merkmal feststellen.” Ibid.; my translation.  “Die Allgemeingiltigkeit, um die es sich hier handelt, ist keine tatsächliche, sondern eine ideale […].” (Windelband 1915a, 42; my translation).

18

Brigitte Falkenburg

philosophy in the systematic (not in the historical) sense, [is] nothing else but the critical science of universally valid values. The science of universal values: this refers to the objects; the critical science: this refers to the method of philosophy. (Windelband 1915a, 29)¹³

To consider philosophy a science of universal values probably indicates the sharpest difference between the Southwest and the Marburg schools of neo-Kantianism. Windelband here again follows Kant, however, who at the end of the architectonics chapter of the Critique of Pure Reason related the cosmopolitan concept of philosophy to the “essential ends” of humanity: Thus the metaphysics of nature as well as morals, but above all the preparatory (propaedeutic) critique of reason that dares to fly with its own wings, alone constitute that which we can call philosophy in a genuine sense. This relates everything to wisdom […] Mathematics, natural science, even the empirical knowledge of humankind, have a high value as means, for the most part to contingent but yet ultimately to necessary and essential ends of humanity, but only through the mediation of a rational cognition from mere concepts, which, call it what one will, is really nothing but metaphysics. (CPR A 850/B 878)

Windelband neither supports Kant’s claim that all sciences are related to the essential ends of humankind, nor the identification of systematic philosophy with metaphysics. Nevertheless, he shares the teleological aspect of the idea of philosophy expressed here. Indeed he stresses the teleological character of the principles of philosophy in his Critical or Genetic Method? of 1883. There, he calls Kant’s synthetic judgments a priori of all three Critiques “axioms”, expanding the common usage of the term from logic and theoretical philosophy, to ethics and aesthetics. For him philosophy does not differ in logical structure from the sciences, but with regard to the validity of its principles or “axioms”: With an expansion of the customary parlance, one can thus also speak of ethical and aesthetic axioms, and the task of all philosophical investigations can then also be formulated thus: the problem of philosophy is the validity (Geltung) of axioms. (Windelband 1883, 108; Luft 2015b, 274)¹⁴

 “[…] wenn ich unter Philosophie im systematischen (nicht im historischen) Sinne nichts anderes verstehe, als die kritische Wissenschaft von den allgemeingiltigen Werten. Die Wissenschaft von den allgemeingiltigen Werten: das bezeichnet die Gegenstände; die kritische Wissenschaft: das bezeichnet die Methode der Philosophie.” (my translation). For the normative aspects of philosophy according to Windelband see also Beiser (2014, 295 – 298).  Translation by Alan Duncan. In his short introduction to the article, Luft (2015b, 267) however correlates “validity” with “Wert” (value) instead of “Geltung” (validity).

On Method

19

The task of philosophy is therefore to demonstrate the validity of “axioms” such as: the principle of causality, the categorical imperative, or the principle of disinterested satisfaction of an aesthetic judgment. Here finally comes Windelband’s “critical method” into play. He distinguishes two kinds on necessity, which may be attributed to the validity of the “axioms”: “factual validity” or “teleological necessity”; he associates the demonstration of the former with the “genetic method”, and the demonstration of the latter with the “critical method”: There is no logical necessity with which the axioms’ validity could be proven. Hence, two possibilities remain: either one shows the factual validity, one seeks to demonstrate that […] these principles are accepted and recognized as valid in the empirical reality of psychic life (Seelenleben) – or one shows that another sort of necessity inheres in them, namely the teleological necessity that their validity must be recognized as unconditioned, if certain ends are to be achieved. This is the point at which the genetic and the critical conception of philosophy part ways. For the genetic method, the axioms are actual modes of conception that have formed and asserted themselves in the development of human ideas, feelings and wilful decisions; for the critical method these axioms are norms […] that should be valid under the precondition that thought would achieve the end of being true, willing – the end of being good, feeling – the end of perceiving beauty, in a way that is to be universally accepted and recognized. (Windelband 1883, 109; Luft 2015b, 275)

Hence, the “genetic” method is based on contingent historical facts, whereas the “critical” method aims at demonstrating that thought, volition and taste ultimately underlie the universally valid norms of truth, being good, and beauty. He then argues that the teleological structure of these norms does not come along with any “metaphysical hypostasizing of the concept of the end” and is by no means in conflict with scientific explanations which are not teleological (Windelband 1883, 109; Luft 2015b, 275)). But Windelband has another problem which his approach to philosophy. He explains the “critical method” in terms of Plato’s normative ideas of the “True”, the “Good”, and the “Beautiful”. One may understand these ideas as regulative principles in Kant’s sense, which underlie any theoretical cognition, moral decisions and aesthetic judgments. But Windelband does not explain how the “critical method” works. He gives no methodological rules on how to follow or get closer to the normative ideas of what is true, good or beautiful. History cannot teach it, as he emphasizes in Critical or Genetic Method?, in opposition to Hegel’s view that reason will realize itself in the course of history. According to Windelband, the contingent course of history always has to be compared with the universal normative ideas of philosophy, which are the

20

Brigitte Falkenburg

ends that constitute the universal guiding clue for the critical method. (Windelband 1883, 133; Luft 2015b, 284).

With his distinction between the “genetic” and the “critical method”, he insists on the gap between contingent facts and universal values, nature and norms, historical knowledge and rational principles, rejecting relativism and historicism. Beiser (2014, 510) emphasizes that Windelband attempted without success to respond to late 19th century historicism and relativism, finally approaching Hegel’s views. In addition, he struggled with the question of how the values of philosophy and the actions of free will may be compatible with deterministic laws of nature (Beiser 2014, 511– 516), a question which obviously arises from Kant’s third cosmological antinomy without Kant’s distinction between phenomena and noumena.

2.2 The “Nomothetic” and the “Idiographic” Method Windelband bound the three main areas of philosophy to norms which correspond to the ideas of the True, the Good, and the Beautiful. The empirical sciences belong for him to the realm of truth. That is, they depend on cognitive values indebted to the normative ideal of true cognition, but not on ethical or aesthetic norms. This view is reflected in his distinction between the “idiographic” and the “nomothetic” method of the empirical sciences, which he explains in his presidential address of 1894, History and Natural Science. The distinction belongs to the philosophy of science, he considers it “a theme from logic, especially from methodology, from the theory of science” (Windelband 1894, 138)¹⁵. For him logic is an applied discipline employed in the practice of the sciences, which range from the “rational” disciplines philosophy and mathematics to the “empirical sciences” (Windelband 1894, 141). Windelband emphasizes that the empirical sciences should not be distinguished according to their objects ‘nature’ and ‘mind’, as the usual distinction between natural science (Naturwissenschaft) and human science (Geisteswissenschaft) indicates. Psychology in particular falls short of this distinction, as a science that has the mind as its object but investigates it with the methods of natural science (Windelband 1894, 142). Instead, he suggests to classify the sciences according to their methods, distinguishing the  My translation. The English translation of History and Natural Science by Guy Oakes in Luft (2015b, 287– 298) is very instructive, but not sufficiently precise in detail with regard to Windelband’s philosophical terminology. Here and in the next quotation I suggest a translation that is as literal as possible.

On Method

21

“nomothetic” from the “idiographic” method. The “idiographic” method is historical and focuses on the description of individual facts (hence it comes close to what he called the “genetic” method, in 1883); whereas the “nomothetic” method is nomological, it aims at establishing general laws: Here we now have before us a purely methodological classification of the empirical sciences, which is to be based on reliable logical terms. The principle of classification is the formal character of their cognitive goals. Some of them search for general laws, others for specific historical facts. […]. Thus we may say: The empirical sciences search in the cognition of reality either for the universal in the form of natural law or for the individual in the historically specified formation (Gestalt); in part they consider the always invariable form; in part the individual, in itself specified content of the actual events. The former are sciences of laws, the latter are sciences of events; the former teach what always is, the latter what once was. Scientific thinking is – if one is allowed to introduce new artificial expressions – in the one case nomothetic, in the other idiographic. If we want to keep to the customary expressions, we may further speak in this sense of the contrast between natural sciences and historical disciplines. (Windelband 1894, 144– 145; my translation).

Windelband distinguishes between both methods according to their cognitive goals of capturing the logical subjects of either universal judgments, that is, the general and invariable laws of nature; or of singular judgments, that is, individual historical facts or events. He emphasizes that the respective distinction between natural sciences and historical disciplines is not strict, as the example of psychology shows, and that there are scientific disciplines that combine both methods, in particular, evolutionary biology. One year before Windelband made this distinction, the first volume of Dilthey’s Introduction into the Human Sciences (Einführung in die Geisteswissenschaften) had appeared. Based on his account of the understanding in the human sciences and on the way to developing hermeneutics, Dilthey strongly opposed Windelband’s distinction. He had three kinds of objections: to the distinction as such, given that there are natural sciences with idiographic elements and human sciences with nomothetic goals; to the claim that psychology belongs to the nomothetic disciplines; and finally, to the view that singular historical facts may be understood as such, without embedding them in any general framework of regularities (see Makkreel 2016, Sect. 2.2). Of these objections the first one is somewhat unjust, given that Windelband himself admitted that the distinction is not sharp. The second makes a substantial point, namely that Windelband, with all his emphasis on the autonomy of historical methods, in relation to psychology was not free from contemporary positivism (just like Cohen in relation to mathematical physics as the predominant “fact of science”). However,

22

Brigitte Falkenburg

most important is the third objection, to which above all Max Weber reacted with his conception of ideal-typical explanations. Hermann Rickert presented a refined classification of the empirical sciences in Science and History (1926) and The Limits of Concept Formation in Natural Science (1929; for details, see Staiti 2018). He modified Windelband’s distinction of nomothetic and idiographic methods and disciplines in the following ways. He emphasizes that the objects of investigation should not be omitted in an adequate classification of the empirical sciences and adds them to Windelband’s methodological distinction. With regard to the objects, he proposes to distinguish the natural sciences from the cultural, rather than human, sciences and to define the objects of the cultural sciences in terms of values, relying on Windelband’s “genetic” and “critical” method in the field of moral philosophy. In addition, Rickert suggests avoiding the expression “idiographic”, and distinguishing the historical method from the nomothetic; he also stressed that his distinctions do not give rise to a sharp discrimination. For him, the empirical sciences are located in a continuum between the extremes of classical mechanics as the prototype of mathematical physics, and individual history as the prototype of a discipline focusing on facts only; and these extremes meet in astronomy (Wagner and Härpfner 2015). One may say that his distinction between the sciences in terms of methods and objects is what Weber called an ideal-type. In a correction of Windelband’s less differentiated ideal-type of the empirical sciences, Rickert had already adapted this ideal-type to the “fact of science” of his day (to take up Cohen’s term), i. e., to the variety of the empirical disciplines existing around 1900.

2.3 Ideal-Typical Explanations Rickert’s criticism and further elaboration of Windelband’s distinction had indeed substantial impact on Weber’s account of social science (Oakes 1990). In philosophy and sociology, Weber is most known for his postulate of value neutrality and the corresponding sociological debate on value judgments (Albert and Topisch 1971). His theory of ideal-typical explanations, in contrast, is a neglected topic. It has long been criticized for its lack of applicability (Watkins 1952) but recently it has once again drawn some attention in the philosophy of social science (Aronovitch 2012; Swedberg 2017). Weber’s conception of ideal-types serves to make a bridge in terms of causal processes from Rickert’s historical method to “nomothetic” explanations. Weber agreed with Rickert that values belong to the objects of cultural and social science, even though their scientific investigation should be value neutral. In addi-

On Method

23

tion, he agreed with Dilthey that historical facts and events cannot be understood without embedding them in an interpretative framework of regularities. Therefore, Weber insisted on explaining historical constellations and their formation in causal terms. A famous example is his explanation of the rise of capitalism under the conditions of protestant ethic (Weber 1904 – 05). In his essay “Objectivity” in Social Science and Social Policy (1904), he explains his conceptions of ideal-types and ideal-typical explanations and the way in which they apply to concepts of economics, religion, political science, and to the history of social phenomena. To avoid misunderstandings, it is worth to the crucial passage of his text in detail here: We have in abstract economic theory an illustration of those synthetic constructs which have been designated as “ideas” of historical phenomena. It offers us an ideal picture of events on the commodity-market under conditions of a society organized on the principles of an exchange economy, free competition and rigorously rational conduct. This conceptual development brings together certain relationships and events of historical life into a complex, which is conceived as an internally consistent system. (Weber 1904, 89 – 90)

Here, Weber considers the market as a historical phenomenon of which economics has an idealized typified conception, the ideal type. The ideal-type of the market is an “ideal picture”, or model, of social actions under certain social conditions. This model is a “synthetic construct” of the dynamics of social life as an “internally consistent system” of social relations, i. e. an idealized model of the dynamics which occurs in a market under certain conditions, such as “the principles of an exchange economy, free competition and rigorously rational conduct”. He continues: Substantively, this construct in itself is like a utopia which has been arrived at by the analytical (gedankliche) accentuation of certain elements of reality. Its relationship to the empirical data consists solely in the fact that where market-conditioned relationships of the type referred to by the abstract construct are discovered or suspected to exist in reality to some extent, we can make the characteristic features of this relationship pragmatically clear and understandable by reference to an ideal-type. (Weber 1904, 90)

The idealized model of the market is obtained by theoretical “accentuation”, picking out certain elements of reality and neglecting others, just as the physicists do in their models of classical point mechanics or of an ideal gas. The relation of such a model to empirical reality is the assumption that the model relations, or “characteristic features” of the “ideal-type”, refer “to some extent” to the relationships which are “discovered or suspected to exist” in empirical reality. The model or “ideal-type” is a heuristic tool for developing hypotheses:

24

Brigitte Falkenburg

This procedure can be indispensable for heuristic as well as expository purposes. The ideal typical concept will help to develop our skill in interpretation in research: it is no ‘hypothesis’ but it offers guidance to the construction of hypotheses. It is not a description of reality but it aims to give unambiguous means of expression to such a description. (Weber 1904, 90)

Weber here anticipates the insight of recent philosophy of science that models do not describe empirical reality but are rather instruments to investigate it (Morgan and Morrison 1999). Models mediate between the phenomena and abstract theories, making it possible to go back and forth between the poles of the phenomena (empirical data) and rational cognition (theory), in order to improve the models and their theoretical foundations. Weber then passes from the example of exchange economy in modern society to another case of an ideal-type, the economic model of a medieval city: It is thus the ‘idea’ of the historically given modern society, based on an exchange economy which is developed for us by quite the same logical principles as are used in constructing the idea of the medieval ‘city economy’ as a ‘genetic’ concept. When we do this, we construct the concept ‘city economy’ not as an average of the economic structures actually existing in all the cities observed but as an ideal-type. (Weber 1904, 90)

Here he emphasizes that such a model does not refer to the average of empirical data but only to certain ideal features of its object. Hence, the ideal type does not aim at generating a statistical model of social phenomena. To stress this point, Weber repeats that the model is an idealization. It takes up many individual phenomena and condenses them into an ideal, abstract picture of a cognitive object which as such does not exist in empirical reality. Like the mass points of classical mechanics or the ideal gas of thermodynamics, the market and the rational economic man of neoclassical economics or the medieval city economy never existed in empirical reality. The ideal-type is a “utopia”, in the literal sense of that which exists nowhere: An ideal type is formed by the one-sided accentuation of one or more points of view and by the synthesis of a great many diffuse, discrete, more or less present and occasionally absent concrete individual phenomena, which are arranged according to those one-sidedly emphasized viewpoints into a unified analytical construct (Gedankenbild). In its conceptual purity, this mental construct (Gedankenbild) cannot be found empirically anywhere in reality. It is a utopia. (Weber 1904, 90)

After having explained in this way the constructive features of modelling an ideal-type, Weber finally sketches the relation between model and reality. This relation is dynamic. The model has to be compared with the empirical data,

On Method

25

i. e., the individual historical phenomena to which it refers. The comparison works in two directions, going back and forth between the individual phenomena themselves and the idealized assumptions of the model; in which the model helps to select the phenomena to which it applies, and the phenomena in turn help to improve the model assumptions. In this way, the historical data serve to modify the ideal-type in order to capture the structure of empirical reality more adequately: Historical research faces the task of determining in each individual case, the extent to which this ideal-construct approximates to or diverges from reality, to what extent for example, the economic structure of a certain city is to be classified as a ‘city-economy.’ When carefully applied, those concepts are particularly useful in research and exposition (Veranschaulichung). (Weber 1904, 90)

Weber furthermore adds a causal aspect to his conception of an ideal-type (1904, 93 – 110). To understand the historical dimension of social phenomena, he introduces genetic concepts. In particular, he discusses the way in which the predominant ideas, thoughts, or ideals of an epoch may have causal influence on the evolution of social phenomena, such as the church, the state, etc. He also points out that such ideas themselves are ideal types of the thoughts of individual human beings. An ideal-typical explanation then combines the ideal type of a historical constellation with a causal explanation. It reconstructs the causal process in which a specific social structure may have emerged under certain social conditions, such as the rise of capitalism in Western Europe under the condition of protestant ethics, to take up his major example (Weber 1904 – 05). Weber’s ideal-typical explanations have been criticized for raising substantial problems, given that they cannot explain collective social phenomena from the behaviour of the individual social agents (Watkins 1952). But explanations of this kind, which today are subject to the physics of social systems, were not Weber’s objective, at least not in the Objectivity essay of 1904 (for Weber’s later account in his Economy and Society, see Swedberg 2017). His ideal types aim at idealization, not at the average behaviour of individual agents nor at the average features of individual social facts. Therefore, ideal-typical explanations do not cover individual historical phenomena by statistical laws. They apply the general law of causality to individual social phenomena and their evolution under specific social conditions, to which the values associated with moral and religious beliefs belong. It has been observed that in this way they merge interpretative understanding of the motivations of social agents with causal explanation at an individual level, that is (in contrast to natural science), without making causal generalizations (Aronovitch 2012). Weber’s ideal-typical explanations combine elements of the “nomothetic” and the “idiographic” or “historical” method in

26

Brigitte Falkenburg

such a way that the causally relevant factors for the formation of specific social phenomena and historical constellations are captured by an idealized model, which has repeatedly to be compared with social reality and modified by adapting it to new data.

3 The Legacy of Neo-Kantian Philosophy of Science As diverse as the approaches to philosophy and the sciences are in the Marburg School, from Cohen to Natorp and Cassirer, and the Southwest School, from Windelband to Rickert and his influence on Weber, they have one feature in common. All of them struggle with finding the balance between the rational and the empirical or historical elements of scientific cognition, between general laws or principles and individual facts or events, between Kant’s principles a priori reflecting the structure of Newtonian science and the historical stage of the empirical sciences. All of them contribute different important insights into the structure of scientific knowledge, focusing in Kant’s tradition on different aspects of the general path of the sciences between rationalism and empiricism. Cohen’s significance in this field lies in the fact that he has contributed a constructivist approach to the philosophy of the exact sciences, which offers profound insights into the theory dependence of scientific experience and avoids the misunderstandings of scientific cognition typical of later versions of constructivism (for the latter, see Scheibe 1997). Cohen’s philosophy of science is the first, neglected version of a scientific constructivism, which still calls for an adequate reception. Natorp and Cassirer have further developed this approach, but its reception has so far been largely limited to Cassirer’s work on the philosophy of physics and of culture. Cohen’s approach is still topical with regard to his double criticism not only of empiricism and its “myth of the given”, but also of a metaphysical realism that wants to codify the “fact of science”, i. e., the characteristics of our scientific world view, in naturalistic terms. Windelband and Rickert are much more on the side of history and the limitations of cognition a priori. Their significance for the philosophy of the human and cultural sciences I cannot even begin to assess here. Regarding the relation between the rational and the historical elements of scientific knowledge, Windelband made a heroic attempt to save the rationality of philosophical principles in terms of universal values against relativism and historicism, and he created a preliminary, rough ideal-type of the distinction between natural science and the historical disciplines in terms of the “nomothetic” and the “idiographic” meth-

On Method

27

od. Rickert refined this distinction, as a distinction of both the methods and contents of natural science vs. cultural science. Weber’s conception of ideal-typical explanations finally made the bridge from modelling individual social phenomena to the causal reconstruction of their genesis under specific historical conditions and under the influence of specific values shared by the social agents. Weber’s ideal types are close to models as mediators in the sense of recent philosophy of science. They make it possible to go back and forth between data and theories in order to develop more differentiated models. His ideal-typical explanations aim at an idealized reconstruction of historical phenomena and the way in which they arise, rather than at a naturalization of social phenomena in terms of the statistical behaviour of social agents. Like Cohen’s philosophy of physics, but for very different reasons, Weber’s ideal-typical explanations deserve more attention in the philosophy of science.

Acknowledgements Dietmar Heidemann invited me to contribute this article to the Kant Yearbook and I am very grateful to him for giving me this opportunity. In addition I would like to thank Kieran Salt for language corrections.

Bibliography Albert, Hans and Ernst Topitsch (eds.) (1971): Werturteilsstreit. Darmstadt: Wissenschaftliche Buchgesellschaft. Arnauld, Antoine and Nicole, Pierre (1685): La logique ou l’art de penser. Quoted after: Logic or the Art of Thinking. Transl. and ed. by Jill Vance Buroker. Cambridge: Cambridge University Press, 1996. Aronovitch, Hilliard (2012): Interpreting Weber’s Ideal-Types. in: Philosophy of the Social Sciences 42, 356 – 369. Beaney, Michael: Analysis. in: The Stanford Encyclopedia of Philosophy (Summer 2018 Edition), Edward N. Zalta (ed.), URL = . Beiser, Frederick C. (2014): The Genesis of Neo-Kantianism (1796 – 1880). Oxford: Oxford University Press. Cassirer, Ernst (1910): Substanzbegriff und Funktionsbegriff. Reprint: Gesammelte Werke, ed. by Birgit Recki. Vol. 6. Hamburg: Meiner. Cassirer, Ernst (1921): Zur Relativitätstheorie Albert Einsteins. Reprint: Gesammelte Werke, ed. by Birgit Recki. Vol. 10. Hamburg: Meiner.

28

Brigitte Falkenburg

Cassirer, Ernst (1923 – 27): Philosophie der symbolischen Formen, vol. I-III. Reprint: Gesammelte Werke, ed. by Birgit Recki. Vol. 11 – 13. Hamburg: Meiner. Cassirer, Ernst (1937): Determinismus und Indeterminismus in der modernen Physik. Reprint: Gesammelte Werke, ed. by Birgit Recki. Vol.19. Hamburg: Meiner. Cohen, Hermann (1871, 18852): Kant’s Theorie der Erfahrung. Berlin: Dümmler. Repr. of the 3rd ed. (1918) in: Werke, Bd. 1.1. Hildesheim: Olms. Cohen, Hermann (1877): Kants Begründung der Ethik. Berlin: Dümmler. Reprint of the 2nd ed. (1910) in: Werke, Bd. 2 Hildesheim: Olms, 2011. Cohen, Hermann (1883) Das Prinzip der Infinitesimalmethode und seine Geschichte. Ein Kapitel zur Grundlegung der Erkenntniskritik. Berlin: Dümmler. Reprint of the 4th ed. in: Werke, Bd. 5.I. Hildesheim: Olms 1977. Quoted after: “Introduction” to The Principle of the Infinitesimal Method and its History. Engl. transl. by David Hyder and Lydia Patton of Part I, pp. 1 – 11, in: Luft (2015b), pp. 101 – 106. Cohen, Hermann (1896) Einleitung mit kritischem Nachtrag zu Fr. Alb. Lange’s Geschichte des Materialismus in fünfter Auflage. Leipzig: Brandstetter. Cohen, Hermann (1914) Das Verhältnis der Logik zur Physik. In: Einleitung mit kritischem Nachtrag zur 9. Auflage der Geschichte des Materialismus von Friedrich Albert Lange in dritter, erweiterter Auflage. Leipzig: Brandstetter, 58 – 94. Reprint in: Werke, Band 5.II. Hildesheim: Olms 1977. Quoted after: The Relation of Logic to Physics from the Introduction, with Critical Remarks, to the Ninth Edition of Lange’s “History of Materialism”. Engl. transl. by Lydia Patton, in: Luft (2015b), pp. 117 – 136. Cohen, Hermann (1902, ²1914): System der Philosophie. Erster Teil: Logik der reinen Erkenntnis. Berlin: Cassirer. Nachdruck d. 2., verb. Aufl. (1914) in: Werke, Band 6. Hildesheim: Olms, 1977. Cohen, Hermann (1904, ²1907) System der Philosophie. Zweiter Teil: Ethik des reinen Willens. Berlin: Cassirer. Nachdruck d. 2. Aufl. (1907) in: Werke, Band 7. Hildesheim: Olms, 1977. Cohen, Hermann (1912): System der Philosophie. Dritter Teil: Ästhetik des reinen Gefühls. Berlin: Cassirer. Nachdruck in: Werke, Band 8 – 9. Hildesheim: Olms, 2005. Damböck, Christian (2018): Philosophie und Wissenschaft bei Hermann Cohen / Philosophy and Science in Hermann Cohen. Cham: Springer. Davidson, Donald (1973): On the Very Idea of a Conceptual Scheme. In: Proceedings and Addresses of the American Philosophical Association 47, pp. 5 – 20. DiSalle, Robert (2013): The transcendental method from Newton to Kant. In: Studies in History and Philosophy of Science 44, 448 – 456. Edel, Geert (1987): Einleitung, in: Helmut Holzhey (Hrsg.): Hermann Cohen, Werke, Bd. 1, Teil 1, Kants Theorie der Erfahrung. Hildesheim: Olms, pp. 8*-59*. Edel, Geert (2010): Von der Vernunftkritik zur Erkenntnislogik. Die Entwicklung der theoretischen Philosophie Hermann Cohens. Alber 1988. Quoted after the 2nd, completely revised ed.: Waldkirch, Edition Gorz 2010. Einstein, Albert (1949): Remarks to the Essays Appearing in this Co-operative Volume. In: Paul Arthur Schilpp (ed.), Albert Einstein. Philosopher – Scientist, Evanston, Ill.: Libr. of Living Philosophers, 1949, pp. 663 – 688. Engelhard, Kristina (forthcoming): Kant’s non-aprioristic practice of metaphysics in his Metaphysical Foundations of Natural Science. In manuscript.

On Method

29

Engelhard, Kristina, Gebharter, Alexander, Feldbacher-Escamilla,Cristian and Seide, Ansgar (eds.) (2020):Traditional and Inductive Metaphysics. Special Issue of Grazer Philosophische Studien (to appear). Falkenburg, Brigitte (2018): Kant and the Scope of the Analytic Method. In: Studies in History and Philosophy of Science 71 (2018), pp. 13 – 23. Falkenburg, Brigitte (2020a): Kant’s Cosmology. From the Pre-Critical System to the Antinomy of Pure Reason. Cham: Springer Nature (to appear). Falkenburg, Brigitte (2020b): Hermann Cohens Bedeutung für die Philosophie. In: Görge K. Hasselhoff (ed.), “Diese Einheit von Erzeugen und Erzeugniss fordert der Begriff des reinen Denkens.” – Vorträge zu Erkenntnistheorie und Religion im Denken Hermann Cohens (Schriften aus dem Nachlass von Dieter Adelmann, 4), Potsdam: Universitätsverlag Potsdam (to appear). Goodman, Nelson (1978): Ways of Worldmaking. Indianapolis: Hackett. Dt.: Weisen der Welterzeugung. Frankfurt am Main: Suhrkamp, 1984. Hegel, Georg Wilhelm Friedrich (1807): Die Phänomenologie des Geistes. Bamberg/Würzburg: Goebhardt. Heidemann, Dietmar H. (2002): Hegel, Sellars und der Mythos des Gegebenen. in: Hegel-Jahrbuch 2002/1, pp. 362 – 368. Heisenberg, Werner (1959): Physik und Philosophie. Stuttgart: Hirzel. Engl.: Physics and Philosophy: The Revolution in Modern Science. London: George Allen and Unwin, Ltd., 1959. Hertz, Heinrich (1894): Die Prinzipien der Mechanik in neuem Zusammenhange dargestellt. Leipzig: Barth. Quoted after the Engl. Transl. by T. E. Jones and J. T. Walley: The Principles of Mechanics in a New Form. London/New York: MacMillan & Co., 1899. Kant, Immanuel (2008): Critique of Pure Reason, ed. by Paul Guyer and Allen W. Wood, Cambridge University Press. Kuhn, Thomas S. (1962): The Structure of Scientific Revolutions. Chicago: Chicago University Press. Luft, Sebastian (2015a) The Space of Culture: Towards a Neo-Kantian Philosophy of Culture. Oxford University Press. Luft, Sebastian (ed.) (2015b): The Neo-Kantian Reader. London/New York: Routledge. Makkreel, Rudolf (2016): “Wilhelm Dilthey”. In: The Stanford Encyclopedia of Philosophy (Fall 2016 Edition), Edward N. Zalta (ed.), URL = . Mittelstaedt, Peter (1964): Philosophische Probleme der modernen Physik. Mannheim: Bibliographisches Institut 1964; 4th ed.: 1972; 7th ed.: 1989. Engl. Translation of the 4th ed.: Philosophical Problems of Modern Physics. Reidel: Dordrecht, 1975. Mittelstaedt, Peter (1998): The Constitution of Objects in Kant’s Philosophy and in Modern Physics. In: Elena Castellani (ed.), Interpreting Bodies. Princeton University Press 1998, pp. 168 – 180. Mittelstaedt, Peter (2009): The Constitution of Objects in Classical Physics and in Quantum Physics. In: Michel Bitbol, Pierre Kerszberg and Jean Petitot (eds.), Constituting Objectivity. Dordrecht: Springer, pp. 169 – 181. Mittelstaedt, Peter (2013): Rational Reconstructions of Modern Physics. 2nd ed., Dordrecht: Springer.

30

Brigitte Falkenburg

Morgan, Mary S., and Margaret Morrison (1999): Models as Mediators: Perspectives on Natural and Social Science. Cambridge University Press. Mormann, Thomas (2018): Zur Mathematischen Wissenschaftsphilosophie des Marburger Neukantianismus. In: Damböck (2018), pp. 101 – 133. Natorp, Paul (1910, ²1921): Logische Grundlagen der exakten Wissenschaften. Leipzig/Berlin: Teubner. Newton, Isaac ([1713]: The Principia. Mathematical Principles of Natural Philosophy. A new translation by I. Bernhard Cohen and Anne Whitman. Berkeley: Univ. of California Press, 1799. Newton, Isaac (1730): Opticks. Based on the Fourth Edition London, 1730. Ed. by I. Bernhard Cohen, New York: Dover 1979. Oakes, Guy (1990): Weber and Rickert: Concept Formation in the Cultural Sciences. Cambridge, Mass.: MIT Press. German ed.: Die Grenzen kulturwissenschaftlicher Begriffsbildung : Heidelberger Max-Weber-Vorlesungen 1982. Frankfurt am Main: Suhrkamp (1990). Pappus of Alexandria (1589), Pappi Alexandrini Mathematicae Collectiones. Venedig. Engl. Transl.: Book 7 of the Collection. Part 1. Edited with Translation and Commentary by Alexander Jones. New York 1986, 82. Patton, Lydia (2004): Hermann Cohen’s history and philosophy of science. Dissertation, McGill University (2004). https://philpapers.org/archive/PATHCH.pdf (access: April 11, 2020). Pollok, Konstantin (2010): The “Transcendental Method”. On the Reception of the Critique of Pure. Reason in Neo-Kantianism. in: Paul Guyer (ed.), Cambridge Companion to Kant’s Critique of Pure Reason, Cambridge University Press, 346 – 379. Pringe, Hernán (2017): Cohen and Natorp on Transcendental and Concrete Subjectivity. In: CR: The New Centennial Review, Vol. 17, pp. 115 – 134. Reichenbach, Hans (1920). Relativitätstheorie und Erkenntnis Apriori. Berlin: Springer. Transl. by Maria Reichenbach: The Theory of Relativity and A Priori Knowledge. Berkeley: University of California Press, 1965. Renz, Ursula (2018): Zwischen erkenntnistheoretischem Rationalismus und wissenschaftsphilosophischem Empirismus. Zu Cohens Philosophiebegriff. In: Damböck (2018), pp. 1 – 12. Rickert, Heinrich (1926): Kulturwissenschaft und Naturwissenschaft. 6th and 7th expanded editions, Tübingen: Mohr Siebeck. Engl. transl.: Science and history: A critique of positivist epistemology, Princeton: Van Nostrand 1962. Rickert, Heinrich (1929): Die Grenzen der naturwissenschaftlichen Begriffsbildung. Eine logische Einleitung in die historischen Wissenschaften, 6th improved edition. Tübingen: Mohr Siebeck. Abridged Engl. transl.: The Limits of Concept Formation in Natural Science, Cambridge University Press, 1986. Russell, Bertrand (1903): The Principles of Mathematics. London: Allen & Unwin, 19372. Scheibe, Erhard (1997): Missverstandene Naturwissenschaft. In: R. Enskat (Hrsg.), Wissenschaft und Aufklärung. Leske und Budrich: Opladen, pp.9 – 29. Scheibe, Erhard (2001): Between Rationalism and Empiricism. Selected Papers in the Philosophy of Physics. Ed. by Brigitte Falkenburg. Berlin: Springer. Scheibe, Erhard (2007): Die Philosophie der Physiker. München: C.H.Beck (revised paperback edition).

On Method

31

Seide, Ansgar (2020): Die Notwendigkeit empirischer Naturgesetze bei Kant. Reihe: Quellen und Studien zur Philosophie, 144. Berlin/Boston: de Gruyter. Seide, Ansgar (forthcoming): The Relation between Empirical and A priori Elements in Kant’s Special Metaphysics of Nature. In manuscript. Sellars, Wilfried (1997): Empiricism and the Philosophy of Mind. Cambridge/Mass.: Harvard University Press. Dt.: Der Empirismus und die Philosophie des Geistes. Paderborn: Mentis, 1999. Staiti, Andrea (2018): “Heinrich Rickert”, The Stanford Encyclopedia of Philosophy (Winter 2018 Edition), Edward N. Zalta (ed.), URL = . Stern, Robert A. (2019): “Transcendental Arguments”. In: The Stanford Encyclopedia of Philosophy (Summer 2019 Edition), Edward N. Zalta (ed.), URL = . Swedberg, Richard (2017): How to use Max Weber’s ideal type in sociological analysis. In: Journal of Classical Sociology 18, pp. 181 – 196. Von Weizsäcker, Carl Friedrich (1985): Aufbau der Physik. München: Hanser. Engl. transl.: The Structure of Physics. Edited, revised and enlarged by Thomas Görnitz and Holger Lyre. Dordrecht: Springer 2006. Wagner, Gerhard, and Claudius Höpfner (2015): Neo-Kantianism and the social sciences: from Rickert to Weber. In: Nicolas de Warren and Andrea Staiti (eds.), New Approaches to Neo-Kantianism. Cambridge University Press. Watkins, J. W. N. (1952): Ideal Types and Historical Explanation. In: The British Journal for the Philosophy of Science 3, pp. 22 – 43. Weber, Max (1904): Die “Objektivität” sozialwissenschaftlicher und sozialpolitischer Erkenntnis. In: Johannes Winckelmann (ed.), Max Weber: Gesammelte Aufsätze zur Wissenschaftslehre. 3rd, extended ed. Tübingen: Mohr Siebeck 1968, pp.146 – 214. Quoted after the Engl. transl.: “Objectivity” in Social Science and Social Policy. in: E. A. Shils and H. A. Finch (eds.), Max Weber: The Methodology of the Social Sciences. New York: Free Press, 1949, pp. 50 – 112. Weber, Max (1904 – 05): Die protestantische Ethik und der “Geist” des Kapitalismus. In: Archiv für Sozialwissenschaft und Sozialpolitik 20 (1904), pp. 1 – 54, and 21 (1905), pp. 1 – 110. Engl. Transl. by T. Parsons: The Protestant Ethic and the Spirit of Capitalism. London: Routledge, 1992. Windelband, Wilhelm (1882): Was ist Philosophie? (Über Begriff und Geschichte der Philosophie) In: Windelband (1915a), pp. 1 – 54. Windelband, Wilhelm (1883): Kritische oder genetische Methode? In: Windelband (1915b), 99 – 135. Quoted after: Critical or Genetic Method? Engl. Transl. by Alan Duncan, in: Luft (2015b, 271 – 286). Windelband, Wilhelm (1894): Geschichte und Naturwissenschaft. (Straßburger Rektoratsrede) in: Windelband (1915b), pp.136 – 160. Quotations: my translation. Engl. transl. by Guy Oakes: “History and Natural Science” (Presidential Address Strasbourg), in: Luft (2015b, 287 – 298). Windelband, Wilhelm (1915a): Präludien. Aufsätze und Reden zur Philosophie und ihrer Geschichte. Fünfte, erweiterte Auflage. Erster Band. Tübingen: Mohr Siebeck. Windelband, Wilhelm (1915b): Präludien. Aufsätze und Reden zur Philosophie und ihrer Geschichte. Fünfte, erweiterte Auflage. Zweiter Band. Tübingen: Mohr Siebeck.

Dustin Garlitz

Durkheim’s French Neo-Kantian Social Thought: Epistemology, Sociology of Knowledge, and Morality in The Elementary Forms of Religious Life Abstract: This article presents Durkheim as a Neo-Kantian social thinker and a source of the theory of emotional contagion. The Elementary Forms of Religious Life is examined as Durkheim’s paradigm case of Neo-Kantianism. He is first considered among the intellectual context of French Neo-Kantianism and its figures Charles Renouvier, Émile Boutroux, and Octave Hamelin, all whom were influential in his formative years. Durkheim’s Neo-Kantianism in The Elementary Forms of Religious Life is then juxtaposed to the Neo-Kantian legal philosophy of Emil Lask and Hans Kelsen. Agued is that Durkheim’s notions of distortion and emotional contagion are his leading contributions to Neo-Kantianism.

1 Introduction “To understand Kant, is to go beyond him.” – Wilhelm Windelband, Präludien, Tübingen 1884, IV.

Émile Durkheim’s Social Thought has been interpreted in a variety of intellectual contexts since the early twentieth century. Anthony Giddens equates certain writings of Durkheim’s to “sociological Kantianism” (Giddens 1979, 91). In particular, a more recent commentator like Susan Stedman Jones believes that the young Durkheim of The Rules of Sociological Method (1895) is a positivist, while the mature Durkheim of The Elementary Forms of Religious Life (1965 [1912]) is an idealist (Stedman Jones 2001, 43). This historical treatment of Durkheim’s Social Thought is done among the intellectual context of French Neo-Kantianism, and it focuses on the more idealist Durkheim of the Elementary Forms. It is in the Forms that one is able to identify the emergence of Durkheim’s French Neo-Kantianism with most clarity. Durkheim’s development of a distinct epistemology and sociology of knowledge in this study serves as important criteria for judging the trajectory of Neo-Kantianism, namely since the dates of its authorship and publication marked the conventional endings of the intellectual movement in academic institutions on the European Continent. https://doi.org/10.1515/kantyb-2020-0002

34

Dustin Garlitz

One should note that Alasdair MacIntyre finds that Durkheim’s overall legacy is his self-establishment as the true heir to Auguste Comte’s positivism in the French Social Sciences, despite Stedman Jones former description of him as a positivist early in his academic career. MacIntyre believes that as time progressed in Durkheim’s academic career (with his founding of the first European university sociology department at Bordeaux in 1895), not only did Durkheim realize he was the true heir to Comte, but more importantly, other academics and colleagues surrounding him in his academic field acknowledged that he was too (Donagan et al. 1986, 87). There are certain limits and boundaries to Durkheim’s idealism in the Elementary Forms, even with the high degree of Neo-Kantian thinking evident in the study. Nonetheless, I will establish in this historical treatment of Durkheim and his thought that he in fact was a true Neo-Kantian, and that his social philosophy was greatly informed by other key figures in the history of French Neo-Kantianism. These figures include Charles Renouvier, Émile Boutroux, and Octave Hamelin. Despite such early influences of Durkheim’s, I will, in turn, present his Neo-Kantian social thought as distinct from their philosophies. His explicit arguments are reconstructed and contrasted to the themes of two other prominent Neo-Kantian Social Thinkers who made significant contributions to legal philosophy: Emil Lask (Substratum¹) and Hans Kelsen (Purification of Concepts²), and considered among the Social Neo-Kantian context of ideas in Georg Simmel (Social Forms³) and Ernst Cassirer (Symbolic Forms⁴).

2 Neo-Kantian Context and Intellectual Background The Kantianism that Durkheim was exposed to in his formative years was one that was filtered through Renouvier. Charles Renouvier saw himself as providing “neo criticism” of Kant (Logue 1993, 5) that did not extend the Kantian tradition but, rather, critically rewrote Kantian epistemic themes into an intelligible body

 ‘Substratum’ is a term used by Neo-Kantian philosopher Emil Lask in his Legal Philosophy (1905) and that is also used by Durkheim in his social thought. See Lask (1950).  The ‘Purification of Concepts’ is a central theme employed by jurist Hans Kelsen in his Pure Theory of Law. See Kelsen (1992).  See Simmel (1971).  See Cassirer (1965).

Durkheim’s French Neo-Kantian Social Thought

35

of knowledge applicable to the social world.⁵ Following Renouvier’s lead, Terry Godlove finds that “Durkheim transplanted these Kantian conceptions [of the First Critique] from their native first person context into the third person ‘human’ sciences, including those concerned with the interpretation of religion” (Godlove 1989, 4). Barry Schwartz situates a radical transplantation of intellectual thought by commenting that, “more than a century after the appearance of the [First] Critique, rationalism met its first sociology challenge” with Durkheim’s intelligible body of knowledge (Schwartz 1981, 38), despite Kant not being a rationalist. Moreover, Schwartz stresses that the French sociology of knowledge⁶ that Durkheim pioneered at the turn of the twentieth century was developed out of a confrontation with Kantian idealism (Schwartz 1981, 12), and was a repudiation of “the autonomy of the mind by showing how the most fundamental categories of understanding correlate with the divisions of society”; this was “an attempt to save empiricism by standing Kant on his head” (Schwartz 1981, 12). Further, in neglecting orthodox Neo-Kantianism and rather following in the Renouvierian French tradition of a new critical rewriting of Kantian theoretical philosophy, Kant’s crucial mistake, says Durkheim, was his assumption that time and space are homogenous categories. They are always classified in terms of the organization and recurrent activities of the group. This correspondence betrays their social origin. Although Durkheim shares with Kant the conviction that the mind is a positive organizer of experience, he maintains that its ordering categories are socially produced. Because he locates the sources of Kant’s a priori in society rather than the individual, Durkheim’s “synthesis” of rationalism and empiricism turns out to be no compromise at all. Durkheim’s theory of knowledge would make Locke himself blush (Schwartz 1981, 2– 3).

The sense in which Durkheim’s theory of knowledge would make Locke himself blush is that it destroys any a priori categories by deriving organization solely from contingent social forms. Ken Morrison and Steven Lukes, among others, have documented that after graduating with his Ph.D. in Philosophy from the École Normale Supérieure in Paris, Durkheim spent the 1885 – 1886 academic year on a fellowship studying empiricism in Marburg and Leipzig. In addition, he analyzed the relation of Ethics to the broader study of Social Science in Berlin under Wilhelm Wundt. In the latter studies he learned that “the true object of

 To even more of an extreme, commentator Donald Nielsen writes that, “Renouvier is hardly a Kantian epigone. His system represents a genuine philosophical synthesis of its own, despite its partially Kantian lineage” (Nielsen 1999, 41).  Warren Schmaus has also focused on Durkheim’s sociology of knowledge, as developed in The Elementary Forms of Religious Life and ‘Primitive Classification’. See Schmaus (1994; 2004) and Schmaus and Pickering, W.S.F. (eds.) (2001).

36

Dustin Garlitz

morality is to make man feel that he is not the whole, but part of the whole – and how insignificant he is by reference to the plurality of contexts which surround him” (Durkheim 1972, 92; Nielsen 1999, 120; Poggi 2000, 2; LaCapra 1972, 33; Lukes 1972, 85). Both Morrison and Lukes argue that such courses of study (especially the empiricism) made more sense to the 28 year old Durkheim than his studies in traditional French Neo-Kantianism with his dissertation advisor Émile Boutroux (who had actually, in turn, studied with the orthodox German NeoKantian Hermann von Helmholtz in Heidelberg a decade prior to Durkheim’s graduate studies) at the ENS from 1879 – 1885. By ‘making sense’, at least these two commentators’ refer that Durkheim found the German courses of study more relevant and applicable to the presentation and advancement of Sociology as a rigorous subject matter embracing the precision of the scientific method and centered upon practical and objective social research. One can also see how Wundt’s moral theme of an anti-individualistic nature informs the later Durkheim’s aggregate notion of the collectivity marking social reality. In such a social reality, he would develop a micro-notion of ‘semi-autonomous’ moral realities existing within the broader context of the initial social reality, in the sense of its inchoate, preformed stage, that very social reality perpetuating the livelihood of the collectivity and its distinct individual and collective representations connected as layering of webs. We are imagining a form of social organization whose categorical forms are somehow given without prior development. For Durkheim, mythical thought and moral beliefs express a reality different than individual reality. However, Durkheim had learned from Renouvier’s Neo-Kantianism that ‘the whole is greater than the sum of its parts’. Such studies in France may have only contributed to and raised the importance of Wundt’s ethical theory in Germany. This relates to him generating an intelligible body of anti-individualist social thought entailing these themes when he arrived back in France. Along with Ken Morrison and Steven Lukes, Durkheimian commentators Terry Godlove and Barry Schwartz find Durkheim’s break with the Kantian tradition clearly visible by his first three major studies (The Division of Labor in Society [1892], the aforementioned Rules [1895], and Suicide [1897]), whereas Giddens and Stedman Jones argue that Durkheim extended French Neo-Kantianism by pursing a groundbreaking sociology of knowledge in Primitive Classification (1903), and the monumentally ambitious Elementary Forms.

3 The Forms The Elementary Forms is a paradigm case of Durkheim’s relation to Neo-Kantianism and positivism. In the Elementary Forms, Durkheim immediately presents

Durkheim’s French Neo-Kantian Social Thought

37

the thesis that “there are no religions that are false” (Durkheim 1912, 15). For Durkheim, justifications by religious followers can be negatively judged by the collective of a social group or society, but beyond these judgments, all religions at ground level express the conditions of the social reality that generates them (Durkheim 1912, 15). As far as a theory of knowledge is concerned, Durkheim is attempting to find the origins of the basic categories of the faculty of understanding in social experience itself, hence it representing a social version of a Kantian project. Back in his study Primitive Classification, co-authored by his nephew Marcel Mauss, he argued that the very notion of category or class does indeed derive from society, and that his sociology of knowledge is centered upon the idea that “the unity of this first logical system merely reproduces the unity of society” (Durkheim and Mauss 1967). The first logical categories had to be purely social in nature. We are imagining human subjects acquiring logical categories through socially mediated experience, and this problem arises because Durkheim is trying to take the Kantian categories out of an a priori context. Space is a concept⁷ for Durkheim (as it was for his Octave Hamelin, his colleague at Bordeaux); instead of it serving as a Kantian sensible intuition characterized by its receptivity, it has the discursivity of a traditional concept. This carrying forward of the late nineteenth century Neo-Kantian elimination of intuition results in Durkheim’s notion of the discursive concept of space being particular not to the individual but to the social group; the concept of space is shared by the members of a social group and is derivative of the everyday life of the collective of society. In arguing that our complex system of classification and knowledge stems from the fact that we do in fact live in such social groups, Durkheim implies it is contingent that group categories of thought precede the formations of intellectual categories. Durkheim believes that society, as a type of ‘substratum’⁸, did not simply work as a model in which classificatory thought followed; it was actually its own divisions which served as divisions for the system of classification itself (Thompson 1985, 135). The Neo-Kantian implication of such an organizational truth is that the first logical categories had to be ones of a purely social nature. The first classes of ‘things’ were those of classes of individuals, into which such ‘things’ were integrated. It is exactly because individuals were indeed grouped, and thought of themselves in the form of groups, that in their ideas they grouped other things. Furthermore, because these two types of group-

 See Godlove (1996, 441– 455).  See Lask (1950).

38

Dustin Garlitz

ing occurred at such a foundational level of the substratum⁹ they emerged very early in society as distinct form one another (Thompson 1985, 135) and as simply amalgamated, generalized social conditions for the possibility of experience of collective life and its representations. There is no typical Neo-Kantian circularity in this argument of Durkheim’s, in that he believes that discursive practices do not change because of collective representations. Commentators Godlove and Schwartz really emphasize the rupture of Durkheimian sociology of knowledge from the critical Kant’s theoretical philosophy simply because Durkheim thinks the essential categories of thought are constructed products of society and do not spring from individual minds. If one looks at Durkheim as rewriting (or inverting) Kant in such a fashion, it is difficult to find some degree of validity in Giddens and Stedman Jones’s NeoKantian readings of his later social thought. As far as him generating a theory of knowledge in the Elementary Forms, these commentaries find that two aspects of a guiding thesis are therefore evident. First, returning to his thought in Primitive Classification, intellectual categories leading to complex systems of thought, such as science, logic, and philosophy derive their classificatory framework from the fact that human beings tend to live in groups and thereby group their ideas. Second, once again, intellectual categories are derivative of group categories. This second aspect contradicts Kant’s discussion of apprehension in the First Critique’s Transcendental Deduction (A), where the Kantian contention is that internal mental categories are primary in the apprehension of the external world. In the critical form, this requires forms of intuition, categories of understanding, and the content of sensibility. One must remember that for Kant, the A Deduction terminology of apprehension is reorganized and rearticulated into the theory of a ‘transcendental unity’ of apprehension in the B Deduction, with Durkheim superimposing Kantian terminology by rejecting this degree of epistemic universality (the universality of Kantian categories and forms of intuition) found innate in the individual’s mind. He is rather opting for a pooling, sharing, or grouping of knowledge as a founding characteristic of the ‘construction of social reality’¹⁰ and the dominant collective that define and govern its phenomenal content as the Kantian manifold of group intellectual thought. There are no noumena for Durkheim; the Kantian ‘thing-in-itself’ has to be knowable since all knowledge can be reduced to the social. In the observation of social facts, Durkheim believes such social facts have to be treated like things, in that we do not know a priori, or rather, innately what ideas form their basis. The social scientist

 Departing from Lask (1950).  Departing from John Searle’s Construction of Social Reality. See Searle (1995).

Durkheim’s French Neo-Kantian Social Thought

39

is encouraged to examine the various currents of social life and go back to their source as the vehicle of individual and collective representations in order to arrive at knowledge as a source distinct from their true origins. All facts of social reality are knowable for Durkheim, but he believes that certain facts can certainly exist without serving any purpose at all and can be useless in an attempt to construct a full explanation of social phenomena as categories of group intellectual thought. However, to argue for Giddens and Stedman Jones’s Neo-Kantian readings of Durkheim Elementary Forms, one cannot deny the very epistemic and metaphysical nature of the questions being asked by Durkheim throughout the study. Two such questions that come to mind, through the pro-Neo-Kantian scholarship of Pickering, is “Who is God?” (Pickering 1984, 487) and ‘What is Reality?’. Pickering believes that Durkheim’s thought in the Elementary Forms should be compared to Kant’s argument for the existence of God: we must hypothesize God’s existence, since without that the notion of morality can have no sense. Yet Kant nowhere doubts the motivational power of respect for the moral law. Second, to push the parallel between Durkheim’s society and Kant’s God, then he/she should take some account of the recent consensus that Kant’s allegiance to the practical postulates weakens or perhaps lapses entirely in the later writings. As far as the Kantian theorem is concerned for Durkheim’s, in place of God, he puts society (Pickering 1984, 487; Allen et al. 1998, 75). During the Enlightenment, the individual had emerged as a type or form of God to himself, and Pickering rephrases the ‘Who is God?’ question to read, ‘What is there in the beginning?’: ‘Religion or Society?’. The aprioristic influence of Kant is undeniable through this reading of Durkheim. Durkheim, of course, answers such a revised fundamental question by declaring that society comes before all, it is the Kantian a priori, in that it is an intelligible epistemic category, as contingent forms of Durkheim’s society necessary and universal, and that all religion (no matter how primitive or advanced the religion) is nonetheless derived from society. However, one can find a Kantian method of inquiry employed by the thought that in the beginning of society religion was the ‘matrix’ of all that is social and that in the absence of God, man thinks of himself as a self-legitimizing being (Poggi 2000, 69). Kant has been appropriated here in Durkheim’s social thought; there still is a ‘matrix’ involved, the mapping now, however, is no longer of cognition but of society. In terms of thinking of society as God once again brings up the undeniable Renouvierian influence in Durkheim’s French Social Neo-Kantianism, for both Renouvier and Boutroux believed that they were pioneering a Neo-Kantian ‘science of morality’. This Neo-Kantian strengthening of philosophy by way of the scientific method all goes back to Preface B of the First Critique, with the Kantian intention in the re-

40

Dustin Garlitz

write made clear: Metaphysics needs to be grounded in the precision and “secure path of science” to achieve the stature in society it deserves (CPR B VII). Durkheim interprets Kantian categories in social terms. Durkheim is a Neo-Kantian in that he applies Kantian categories to a distinct model of social reality, representations, and symbols.

3.1 Social Reality As far as the other philosophical question at hand in the Elementary Forms, that of, ‘What is reality?’, Durkheim believes that reality is nothing more than a system of symbols (Rawls 2004, 181). A contrast must be drawn with the Kantian view that the world is composed of enduring, causally interacting, spatio-temporal substances. The challenge for a Neo-Kantian reading of Durkheim is to show that these views are not so far apart as to make them impossible to locate on the same map. Earlier in his academic career, a young Durkheim (Durkheim 1972) proposed that that no religion whatsoever could ever be regarded merely as fantasy and that every religion that has ever existed has found its fundamental basis in social reality. He acknowledges that these religions have done very little in terms of providing an explanation of reality, admitting that they do not express the ‘things’ of the physical world as they are, per se. However, it is his belief that they “interpret, in a symbolic form, social needs and collective interests. They represent the various connections maintained by society with the individuals who go to make it up as well as the things forming part of its substance” (Durkheim 1972, 50). This is a very Neo-Kantian style of analysis, in that Durkheim acknowledges that there is a ‘substratum’ in the form of society and its social conditions, and (departing from the Laskian substratum), that society is in part comprised of ‘pre-conceptualized’ content’¹¹. Pre-conceptual on grounds it lacks Kantian discursivity. This is so because Durkheimian individuals are already bringing their own individual representations to a collective notion of social reality. In its construction process full of the generation of collective representations, each Durkheimian individual has already very early made sense of the world by attaching and ascribing certain meaning to abstract symbols. The Neo-Kantian circularity in this argument is that society as a priori has already informed the individuals’ individual representations and their normative inter-

 The language of this argument incorporates a phrase from Robert Hanna and Dietmar Heidemann’s conception of Kant and Non-Conceptual Content. See Hanna, Robert (2005, 247– 290) and Heidemann (ed.) (2013).

Durkheim’s French Neo-Kantian Social Thought

41

pretation of a certain segment of social reality’s economy¹² of symbols. Interpreting society as the Kantian ‘manifold’ may lead one astray; Durkheim finds that society as a collective totality contains ‘pre-formed’¹³ segments, which are conceived by him to represent social things, and as a collective structure society as such has already been impregnated with conceptual content¹⁴. This is a direct relationship between the pre-formed and the a priori on these grounds, in terms of what is defined as discursive and conceptual in the critical Kant’s theoretical philosophy. Durkheim’s discussion is Neo-Kantian is the sense that it is pre-conceptual rather than straightforwardly empirical and categorical in construction. On these terms, pre-conceptual material can be set off from the empirical whilst avoiding the a priori. The Neo-Kantian conceptual content was inserted¹⁵ earlier rather than later in the multiple, continuous phases (time is reorganized in Durkheim’s French Neo-Kantianism as a discursive concept) which constitute the making of the totality of society and both its individual and collective representations. In a social reality of a closed-society there alone are mutual exchanges of symbols communicating meaning and value as well as representations characterizing certain more abstract realities. This all makes for a substrate¹⁶ constantly shedding layers of generalized forms. For Kant, ‘apperception’ is the reflexive awareness of representations, but Durkheim’s construction of social reality is a purely collective notion. This implies that there is one single coherent, yet extremely complex social reality that is realized and constructed in multiple phases (providing for the aforementioned variances in the ascription of the meaning of symbols) over the course of a continuous progression of time. Once again, even time is a concept (of social nature) for Durkheim, since it finds the “rhythm of social life” as its basis (Durkheim 1973, 215). He finds that the conceptual content is already evident in the early phases of social progression, (religion being the most primitive of all social phenomena), more than a Kelsenian type of pure theory of law ‘insertion’ of conceptual content being administered during some later point. The conceptual content has always been with us and is contingent in the narratives of society and its

 In the Derridean sense as developed with signs in Of Grammatology and Structure, Sign, and Play in the Discourse of the Human Sciences. See Derrida (1978, 278 – 293; 1998).  Pre-formed in the Laskian sense that was so important in the early intellectual development of the young Martin Heidegger. In this regard, see Kisiel (1995, 197– 240).  As departing from the advances in scholarship made in the earlier conceptions of Kant and Non-Conceptual Content.  In this reading of Kelsen’s Pure Theory of Law and its purification of concepts, conceptual content is capable of being inserted at a certain stage.  A Neo-Kantian substrate in the tradition of Lask’s Legal Philosophy.

42

Dustin Garlitz

categories of totality. Therefore, as distinctively Neo-Kantian (rather than materialist), Durkheim is arguing for pre-conceptualized segments or fragments of reality that are subject to an arduous epistemic ‘retrieval’ by the individual subject, rather than an insertion of conceptual content early in the reality constructing process in order to relay meaning to the collectivity. He labels the study of the substratum as ‘morphology’ in his Neo-Kantian informed sociology, and grounds it beneath the normative sphere encompassing ‘institutions’ and the aforementioned symbolic sphere comprised of “collective representations” (Thompson 1985, 16). There is definitely a Neo-Kantian presence felt in Durkheim’s argument that there is pre-conceptualization in the fragmentary construction of social reality and its grounding substrate, rather than an insertion of conceptual content late in a construction process of the symbolic sphere. However, he is opposing what another Neo-Kantian philosopher (Hans Kelsen) is doing, by inserting (or at least acknowledging) the conceptual content (Durkheim: religion; Kelsen: law) so early in the construction of social reality and its substrate framework, which is supposed to ascribe a more aggregate sense of meaning and purpose. Both Kelsen and Durkheim are extremely interested in the ‘purification’ of concepts, but the ‘purification’ of the concepts of society and its ‘bedrock’¹⁷ of religions works differently in practice than a pure theory of something as socially normative as law and jurisprudence. Durkheim departs from Kelsen’s Pure Theory of Law methodology and its Grundnorm ¹⁸ in that he is not working top-down in the manner that his individual cannot start with a direct accessing of the Grundnorm of society, an universalizable collective representation to work down from, one that he would be able to make sense of in its lower hierarchies of further constructed collective and individual representations. That is just plain wrong for Durkheim and one should be cautious against taking society in the sense of the Grundnorm. The tension here rests on the Kantian antinomies of pure thought that underscore such an intellectual concept, as well as its intersubjective dimension. The problem with Kelsen’s notion of ‘purity’ for Durkheim is that social reality is entirely a collective entity and this makes a Grundnorm of society inapplicable to the Durkheimian individual subject. Whereas the Kelsenian jurist would use such a notion as a guide to coherently administer jurisprudence, Durkheim’s social reality is impregnated with a barrage of individual and collective representation, on top of what is a multitude of symbols¹⁹, each

 A term from Ludwig Wittgenstein and his concept of Normativity. See Wittgenstein (2001).  As explicated by Kelsen throughout his Pure Theory of Law.  In the postmodern semiotics sense.

Durkheim’s French Neo-Kantian Social Thought

43

of which has the capacity to be arbitrarily interpreted by certain individuals in different ways. Although Durkheim’s purest notion of society as substratum is already impregnated with such developed conceptual content, Durkheim finds that it is this very attempt at connection-making in the symbolic sphere paired with the actual composition of the individual subject’s interests and proclivities that provide for a high degree of realness. In the Forms, Durkheim embodies this degree of realness accessible in its social construction and workings even further by proposing the idea that “Religions are the primitive way in which societies becomes conscious of themselves and their history, and that they are to the social order what sensations are to the individual” (Durkheim 1972, 250). He imagines a skeptic who argues that there is a fundamental and primary ‘distortion’ of the ‘things’ represented in social reality by these religions, a ‘distortion’ of the same magnitude as in the case of the ‘distortion’ in the processes in the imagery of the religious custom practitioner. This discussion employing the notion of ‘distortion’ is once again extremely Neo-Kantian, this time even bearing more of a resemblance to Kelsen’s methodology than Lask’s, with whose chief aim of the discussion to further advanced a theory of the ‘purification’ of concepts. This advance surfaces in Durkheim’s study as emotional contagion. A significant theory of Durkheim’s in this Neo-Kantian light is that in the a priori of society, there exists an inevitable degree of emotional contagion. This emotional contagion must be ‘purified,’ or, distilled, to identify the true conceptual contents of which society is founded on. That preserved society is already a collective notion for Durkheim; in his Neo-Kantianism all individual and collective representations are identified as conceptual, despite one of Kant’s signature doctrines is his insistence that he has identified the set of pure concepts that makes experience possible. Further, he believes that in order to even ‘think’ conceptually (being since society is the a priori), one must “isolate” certain social conditions and ‘constraints’ (Durkheim 1973, 214) of the substratum (remember that in the Rules Durkheim finds that even the moral beliefs of the individual tend to constrain the appearance and identification of ‘social facts’ as generalized ‘things’ in reality). Society is the a priori on these grounds because it is conceptualized as an innate epistemic category. The consequence of such presuppositions is that Durkheim’s notion of emotional contagion can never be fully fleshed out and conceptual purity of the structures and categories of totality will always be unattainable. Durkheim’s theories of categories welter of empiricism. They are unattainable because individuals are complex, and the social reality they live in is even more complicated. Think about all of the possible groupings, customs, and practices initiated by even the simplest of societies. Furthermore, when the division of labor process becomes more specialized

44

Dustin Garlitz

and societies become further advanced, ‘social facts’ as ‘things’ can really become less intelligible (a multitude of symbols²⁰ only adding to the haze) and the prospect of a ‘purification’ of concepts starts to become unattainable. On these terms, Durkheim emerges as a true Neo-Kantian social thinker. Once again, simply focusing on the social irrationalism and emotional contagion that the individual can generate in ‘reality’ and its economy²¹ of symbols, Durkheim writes in the Forms that, […] it is a well-known law that the sentiments aroused in us by something spontaneously attach themselves to the symbol which represents it. For us, black is a sign of mourning; it also suggests sad feelings and ideas. This transference of sentiments comes simply from the fact that the idea of a thing and the idea of its symbol are closely linked in our minds; the result is that the emotions provoked by the one extend contagiously to the other. But this contagion, which always takes place in some degree, is much more extensive and marked when the symbol is of something simple, definite and easy to represent, while the thing itself, owing to its dimensions, the number of its parts and the complexity of their arrangement, is difficult to conceive of. For we are unable to treat an abstract entity, which we can represent with difficulty, and in a confused manner, as the source of strong sentiments which we feel (Durkheim 1972, 256).

In is in such context that Durkheim derives his social theory of emotional contagion. Social reality is purportedly complex, and the individuals which partially comprise its collective framework (representations tend to take on a life of their own) have always been acknowledged by Durkheim to be erratic and unpredictable. None of the former is good news for any type of traditional epistemic process entailing a Neo-Kantian ‘purification’ of concepts, especially one done while entangled in the interconnectedness²² of the social realm full of these emotionfilled individuals. Such presuppositions entailing a decline in subjectivity that had already been irrational to begin with, in part, is what makes Durkheim a distinct social theorist of cultural modernity. The wave of relativism that transpired, (despite the multitude of non-relativists from the period including Kant himself, Hegel and Fichte), over the course of nearly a century-and-a-half after the publication of Kant’s First Critique penetrates its deepest point in Durkheim’s new cultural relativism. He counters the claim by the aforementioned relativist with a rejoinder: such ‘distortions’ in the processes of the religious-observers’ imagery are of the same type as when sensations distort the physical ‘things’

 In the sense of Derrida’s economy of signs and the logic of postmodernism.  In the Derridean sense.  Evoking the Hegelian usage of the term.

Durkheim’s French Neo-Kantian Social Thought

45

they are supposed to convey to the individual subject. Furthermore, he finds that, Sound, colour, and temperature do not exist in the world any more than gods, demons or spirits do. By the fact alone that the representation presupposes a subject who thinks – (in one case, an individual, in the other, a collective subject) – the nature of the subject is a factor in the representation and distorts the thing represented. The individual, in picturing by means of sensation the relations he has with the world about him, puts into these images something that is not there, some qualities that come from himself. Society does the same in picturing its structure by means of religion (Durkheim 1972, 250).

The former is Durkheim’s articulation of the necessity for both individual and collective representations and a complex social reality that has the capacity to generate and absorb the realities each is intended to capture and picture. Because humans (the Durkheimian individual) are filled with the very emotions spoken about, the ‘distortion’ in both the relaying and reception of meaning in those abstractions of reality (the individual and the collective representations) occurs entirely among the verisimilitudes of modern subjectivity, in terms of society being full of emotional contagion. Even though Durkheim has theorized about a high degree of chaos occurring among the social reality of ‘things’, the fundamental ‘distortion’ in collectively pictured realities (representations) is in fact the products of individuals’ performance in and interpretations of simple functional roles. This sense of functioning is intended for presentation within the context of certain social customs and practices basic or foundational to any typical source of a Durkheimian substrate. When the simplicity of the social practice or custom (is this case, of a religious nature) is overlooked or misinterpreted by the individual is when there is the birth of new types of representations (individual and collective). In such a capacity, these representations tend to complicate social reality for the collective rather than clarify, and what emerges is a form of social Neo-Kantianism greatly distinct from other social thinkers and intellectual movements of the time period.

3.2 Representations: Individual and Collective The perpetuation of the notion of the symbolic²³ is advanced only further in the Elementary Forms (Thompson 1985, beginning on 121). Here a complex construc-

 The symbolic and their relation to the mythical were lively topics to other Neo-Kantians as well, in particular, Ernst Cassirer in the Second Volume of The Philosophy of Symbolic Forms, as

46

Dustin Garlitz

tion of social reality is proposed that incorporates Durkheim’s concept of ‘representation’, which, as we have seen throughout the former section, had always been contingent in such conceptualization. ‘Representations’ relate to the concrete, everyday world of experience, one that is based on physical objects, physical necessities, human nature, and psychological drives. As is evident from the passages cited above, Durkheim, as early as 1898 (in Leçons de sociolgie, and Individual and Collective Representations, from Revue de Métaphysique et de Morale) developed his notion of ‘representation’ with the dual capacity for both individual and collective representations. He argues that society does not depend upon the nature of the individual personality and in the latter publication theorizes that, Society has for its substratum the mass of associated individuals. The system of which they form by uniting it together, and which varies according to their geographical disposition and the nature and number of their channels of communication, is the base from which social life is raised. The representations which form the network of social life arise from the relations between the individuals thus combined or the secondary groups that are between the individuals and the total society. If there is nothing extraordinary in the fact that individual representations, produced by action and reaction between neural elements, there is nothing surprising in the fact that collective representations, produced by the action and reaction between individual minds that form the society, do not derive directly from the latter and consequently surpass them. The conception of the relationship which unites the substratum and the social life is at every point analogous to that which undeniably exists between the physiological substratum and the psychic life of individuals (Durkheim 1912, 24– 25).

Even though Durkheim believes that it is through a type of aggregate substratum at the level of the intersubjective that collective representations are exterior to individual minds to the degree that they do not derive from them but are associated with them (Durkheim 1912, 26). He theorizes about a dual sense of dependence and distinction, and that simply since it is the collective substratum that generates reality’s social phenomena, such phenomena must bear the marks of this origin (Durkheim 1912, 30). However, in terms of a purely collective social consciousness, it may be initially structured on the nature of the substratum (i. e., the quantity of social elements and conditions in a closed system, and the structure in which they are grouped and distributed), but once a preliminary set of representations has been generated by such a substratum, the representa-

well as his Symbol, Myth, and Culture, The Myth of the State, and Language and Myth. As much as the Neo-Kantian Cassirer wants to ground culture in reality, he acknowledges that you cannot do away with the mythical. See Cassirer (1946; 1953; 1965; 1981).

Durkheim’s French Neo-Kantian Social Thought

47

tions, in part, become autonomous “realities”, each with its “own way of life” (Durkheim 1912, 30). These semi-autonomous “realities” engage in a type of ‘free play’²⁴ where they have the power and freedom to attract and repel each other and to move towards various syntheses of a more Neo-Kantian ‘self-regulative’ manner. Durkheim finds that since such syntheses are determined by the natural affinities of the ‘mini-realities’ and not by the conditions for the possibility of their experience (i. e., the social matrix which marked their origin), a new set of representations is born out of these ‘self-regulative’ syntheses. These newest ‘representations’ reflect even less of the initial substratum and its social structure, and much more of the ‘quasi-spontaneously’ formed collective representations that, in turn, were generated by such substratum’s social conditions (Durkheim 1912, 31). Where his nephew Marcel Mauss theorizes about a ‘gift’²⁵ economy of customs, Durkheim conceives of a social reality of symbols, one that is that is multi-layered and loosely interdependent rather than purely causal. This social reality’s outer layers of broadly construed, self-regulative ‘representations’ tend to take on a life (or form) of their own as ‘mini-realities’. Nonetheless, there is still very strong evidence for interconnectedness, comprehensiveness, and Kantian universality to each such ‘mini-reality’ acknowledged by Durkheim in the Forms, where he believes that, “religion, far from ignoring society and making an abstraction of it, is in its image; it reflects all its aspects, even the most vulgar and the most repulsive” (Durkheim 1973, 193 – 194). Representations can therefore be interpreted as ‘self-sustaining’ systems of reality that found their initial codification in the very substrate of society and its groups, institutions, customs, and practices. Religion, being the model of representation for the most primitive societies, has in fact endured modernity and proven to be a social phenomenon of reiterative magnitude, one such that characterizes how complex and chaotic the mythical component of social reality can be for Durkheim.

3.3 Symbols and Representations As far as ‘reality’ being defined by Durkheim as simply a ‘system of symbols’, one of his presuppositions is that there are no unknowable symbols of any kind. He writes in the Forms that, “one must know how to go underneath the

 In the Derridean sense, as developed in Structure, Sign, and Play in the Discourse of the Human Sciences and Of Grammatology.  See Mauss (2002).

48

Dustin Garlitz

symbol to the reality it represents and which gives it meaning. The most barbarous and the most fantastic rites and the strangest myths translate some human need, some aspect of life, either individual or social” (Durkheim 1912, 14). Even the young Durkheim believed that divinity was nothing more than the symbolic expression of “transfigured” society and the circularity of his argument entailed that the role of religion in such a society was “transfiguring” in itself (Nielsen 1999, 206). In the Elementary Forms, Durkheim is attempting to perform a Neo-Kantian ‘retrieval’ of conceptual content and meaning by returning to the sources and foundations of society (the categories of totality). In his Neo-Kantian ‘isolation’ of the social conditions for the possibility of religious experience, he administers what I would like to call a ‘hermeneutics of retrieval’²⁶ by proposing that “the first systems of representation with which men have pictured to themselves the world and themselves were of religious origin” (Durkheim 1912, 21). Stedman Jones finds that if reality is simply a system of symbols, and all symbols are knowable, then representations, accordingly, are simply ‘ways of seeing’ the already knowable (Stedman Jones 2001, 66). That is, they are forms of the very consciousness which stem from the activity of the understanding in association with the data we receive through our senses. Collective representations, therefore, are the ‘shared’ ways of seeing that are central to a culture or a society. Nonetheless, Durkheim does not deny that there is meaning encoded and pictured in all representations, and this once again brings about the strongly anti-individualist nature of Durkheim’s social philosophy and its representations. Kant uses the term ‘representation’ throughout the First Critique in a particular way; the significance of the way Durkheim uses it implies, of course, that it does not indicate the private, the subjective, or an illusion. All representations refer to content. Even though individuals are regarded to be highly complex because, in part, of the economy of symbols they generate and employ in their practice of everyday life, it is certainly evident that the collective facts which inform collective representations are considered by Durkheim to be much more advanced and complicated than such singular meanings that emotion-filled individuals attached to their symbols used in the practice of everyday life.

 This may be similar in spirit, possibility, to Paul Ricoeur’s notion of a ‘hermeneutics of recovery’ in his narrative approach to selfhood, philosophical anthropology, and the human sciences.

Durkheim’s French Neo-Kantian Social Thought

49

3.4 The ‘thing-in-itself’ Stedman Jones finds that it is precisely to underscore the reality of representations that he rejects their identification with things-in-themselves (Stedman Jones 2001, 6). Instead of rendering this a complete rejection of Kant’s original thought, in is possible to interpret Durkheim a product of the intellectual context of French Neo-Kantianism. This is because it was his early hero Charles Renouvier who was one of the first and most prominent figures in the history of nineteenth century French philosophy to deny the unknowable nature of the ‘thing-in-itself’. Instead of representing a methodology bordering on French Neo-Kantian social constructivism, Durkheim’s social realism²⁷ differs from a German Neo-Kantian figure such as Emil Lask, in that there is not a ‘real’ substratum mirroring the ‘real’ thing-in-itself (i. e., you can not look directly at it). In place of a constructivist outlook, Durkheim’s social realism holds that society as substratum can never be truly out of sight, and because of such clear visibility it can be objectively studied to the point where its facts become intelligible to the social scientist. In only one instance has Durkheim hinted that there may be a ‘second’ substratum, a type of “organic substratum [that serves] as the basis of all psychic life.” (Thompson 1985, 31). This was introduced by Durkheim in the context of a purely hypothetical example used to counter Antonio Labriola’s Essays on the Materialist Conception of History in an 1897 edition of Revue philosophique. Regardless of its actual presence, the purpose of a second, organic substratum, in theory, is to generate the very framework for ‘psychic life’²⁸ which Durkheim  As explicated by Robert Alun Jones in his conception of Durkheim’s classical social realism. See Jones (1999).  Georg Simmel, another one of the prominent Neo-Kantian social theorists, focused much of his analysis on ‘psychic life’ (especially the ‘psychic life’ of a ‘social space’ he calls “the metropolis”), and even though Durkheim had a tremendous amount of respect for his method of analysis, he saw himself critiquing this very method in his Social Inquiry of Sociology and its Scientific Field (1900). His general problem with Simmel’s Neo-Kantian Social Thought revolves about his fundamental question of, “what are the meanings of the expression of ‘social forms’ and ‘forms of association in general’? If one wanted to speak only of the manner in which individuals are placed in contact with one another in association, of the dimensions of association, of its density – in a word, of its external and morphological aspect – the notion would be definite; but it would be too restricted to constitute, by itself alone, the subject manner of a science. For it would be equivalent to reducing sociology to the exclusive investigation of the substratum on which social life rests. As a matter of fact, however, our author attributes to the term ‘social forms’ a much more extended significance. By it he understands not only the modes of grouping, the static condition of association, but also the most general forms of social relations. The term refers to the largest forms of relations of every kind that mesh in society and to the nature

50

Dustin Garlitz

believes to be within the boundaries of sociology’s subject matter. Thinking like a Neo-Kantian, once this conceptual framework²⁹ has been set up to the point where one knows what the actual boundaries of the particular science’s subject matters are, then the social phenomena can be studied objectively by the social researcher to the point where they are identified as intelligible facts of this particular discipline, a discipline which happens to ground itself in social reality more than another prior to it (Durkheim famously wrote that “individualistic sociology is only applying the old principles of materialist metaphysics to social life” (Durkheim 1912, 29).

3.5 ‘Space’ as a Concept: The Social Implications The French Neo-Kantian denial of the ‘thing-in-itself’ goes hand-in-hand with the shifting of ‘space’ from a pure intuition characterized by its receptivity (in Kant’s Transcendental Aesthetic) to a discursive concept in Durkheim’s epistemology and sociology of knowledge. Where Durkheim picked up from his French Neo-Kantian hero Charles Renouvier the denial that the ‘thing-in-itself’ is completely unknowable, he borrowed from his French Neo-Kantian colleague at Bordeaux, Octave Hamelin, that space truly is a discursive concept³⁰ and not merely a sensible intuition. In the First Critique, Kant refers to space and time as “organs of perception”, the terms used as forms of intuition, and from this point of view, recent commentator Ken Morrison emphasizes that it is not a stretch for Durkheim to posit that “the first framework for understanding the world and classifying other things in relation to spaces is the model of spatial relationship” (Morrison 1995, 202). Morrison continues by emphasizing that “understanding the concept of direction, for instance, depends upon spatial relationships and upon this depends the understanding of the outside world” (Morrison 1995, 202). Durkheim’s conception of the substratum is of the central Spencerian ‘organ’, (as Spencer was so important in his social thought very early in his academic career dating from his work The Division of Labor in Society), to which society depends upon in its livelihood. It is supposed to serve as the conditions for the possibility of social experience, as far as the generation and perpetuation of a layering of multiplicities³¹ (i. e., sheets of ‘mini-realities’) that become intelligi-

of phenomena with which we are presented as being directly pertinent to sociology” (Durkheim 1960, 357– 358).  See Godlove (1989).  See Godlove (1996, 441– 455).  In what would become the Bourdieusian sense of the term.

Durkheim’s French Neo-Kantian Social Thought

51

ble to certain sets of individuals. Such a collective notion of subjectivity, (or possibly an inchoate, early twentieth century notion of intersubjectivity that stretches in conception from Edmund Husserl back to the early nineteenth century thought of Hegel), is further articulated by Durkheim in his writing that, Social life has various manifestations, the nature of which we shall indicate presently. All of them, however, have this in common: They emanate from a group, simple or complex; the group is their substratum. Obviously, the study of sociology is the study of the social substratum […] The composition of society consists in certain combinations of people and things which by necessity are connected in space. The explanatory analysis of this substratum, however, should not be confused with that of the social life which builds on it. The way in which society emerges fully formed is one thing; the manner in which it acts is another. These are realities of two kinds […] Man modifies the social substratum in a thousand ways, and the resultant differences have great sociological significance because of both the causes which they depend and the effects that they produce […] The social stratum must, above all, be determined in its external form. This external form is chiefly defined by (1) the size of its territory; (2) the space which the society occupies, that is, its peripheral or central position in regards to ‘continents,’ and the way it is enclosed by other societies, and so on; and (3) the form of its frontiers […] Yet the substratum of collective life is not the only thing of a social character that exists in nature; the life that flows from it or sustained by it necessarily has the same character and belongs to the in the province of the same science. Besides the social ways of being, there are social modes of doing (Durkheim 1960, 360).

From the former application of Kant’s philosophical thought in the First Critique’s Transcendental Aesthetic to an emergent early twentieth century cultural sociology, it is evident that space has to be held as a prototypical Neo-Kantian concept in Durkheim’s social thought and not a traditional Kantian intuition. The distinctness of Durkheim’s Neo-Kantianism, however, comes in his justification and reasoning, “that the concept of space is not a general abstraction but rather has its origins in relation to the social group from which the individual perception of space develops” (Morrison 1995, 202). The construction of a collective sense of spatial orientation³² in Durkheim’s social reality cannot be overlooked. There were strong anti-individualist themes in Durkheim’s social thought from very early in his academic career (think: the ‘collective conscience’ of The Division of Labor in Society), and here too these themes (the ‘collective conscience’ resurfaced as ‘collective consciousness’ (Thompson 1985, 131)) reemerge and validate themselves in his belief that it is impossible to conceptualize the world spatially unless there is some sort of commonly “shared standard in terms of which spatial relations can be judged” (Morrison 1995, 202). This re-

 Invoking Maurice Merleau-Ponty and his Phenomenology of Perception. See Merleau-Ponty (2012).

52

Dustin Garlitz

inforces Stedman Jones’s interpretation of collective representations as ‘shared ways’ of seeing. There are further implications of such a notion of ‘sharing’ in the Durkheimian construction of orientation. Morrison situates Durkheim as theorizing about the need for a “center point of space” for the collective, from which all other “spaces” are able to “radiate” (Morrison 1995, 202). As far as a social group or primitive tribe’s social camp in the Forms, Morrison’s reading is that its social organization would serve as the ‘common’ standard and this social organization of the tribal camp carries over to the use of “the concept of space as a model for the mental organization of space” (Morrison 1995, 202). It is here, in this “mental organization mirrored and modeled on the social organization” of such a tribal camp, that we find the ‘focal’ point of spatial direction for the collective of such a primitive tribe, social group, or society in the Elementary Forms (Morrison 1995, 202– 203). This is a distinct sense of modernity in terms of cultural theorizing, since pivotal figures in twentieth century French Structuralism ranging from Claude LéviStrauss³³ to Pierre Bourdieu³⁴ run with passages from the Elementary Forms documenting this particular spatial outlook that stretches and transposes Kant’s original framework in the Transcendental Aesthetic to its very functional and conceivable limit, As Hamelin has shown, space is not the vague and indetermined medium which Kant imagined; if purely and absolutely homogenous, it would be of no use, and could not be grasped by the mind. Spatial representation consists essentially in a primary co-ordination of the data of sensuous experience. But this co-ordination would be impossible if the parts of space were qualitatively equivalent and if they were interchangeable. To dispose things spatially there must be a possibility of placing them differently, of putting some at the right, others at the left, these above, those below, at the north of or at the south of, east or west of, etc., etc., just as to dispose states of consciousness temporally there must be a possibility of localizing them at determinate dates. That is to say that space could not be what it is if it were not, like time, divided and differentiated. But whence come these divisions which are so essential? By themselves there are neither right nor left, up nor down, north nor south, etc. All these distinctions evidently come from the fact that different sympathetic values have been attributed to various regions. Since all the men of a single civilization represent space in the same way, it is clearly necessary that these sympathetic values, and the distinctions which depend upon them, should be equally universal, and that almost necessarily implies that they be of social origin (Durkheim 1912, 23 – 24).

A segmented account of social space results in a fragmentation in the administration of social customs and practices of the primitive society. Where the Neo-

 See Lévi-Strauss (1955; 1966; 1969; 1974; 1983).  See Bourdieu (1977; 1985; 1989; 1990).

Durkheim’s French Neo-Kantian Social Thought

53

Kantian Frenchman Gaston Bachelard develops The Poetics of Space ³⁵, poetics mean little to Durkheim on this matter, as he is more concerned about the implications and the social conditions for the possibility of the production of space. He differs from Henri Lefebvre’s social account of The Production of Space ³⁶ in the fact that he is still a Neo-Kantian, and is therefore attempting to map social space the same way that Kant of the First Critique attempts to map cognition or take an inventory of the mind. In such a mapping, if social space can be relativized and differentiated like other products of the collective substrate (i. e., representations and symbols), fertile spaces can be identified, which provides the social conditions for the possibility of the substrate to enlarge. Once Durkheim’s entire social system has been fully relativized, all its ‘things’ (facts, symbols, representations) tend to take on a life of their own. The point is that there is in fact ‘life’ prolonged and further generated to the point that Durkheim’s substrate (which has the primary function is his system to provide the conditions for the possibility to study this social life) becomes a Neo-Kantian entity that will self-sustain itself to spawn countless generations of social realities.

4 Conclusion In this historical study of his social thought, I have established Durkheim’s significant place in the trajectory of Neo-Kantianism by analyzing and reconstructing his explicit arguments in the Forms (and among other places) that find themselves centered upon five interconnected themes: (3.1.) reality; (3.2.) representations; (3.3) symbols; (3.4) the ‘thing-in-itself’; (3.5) space. By contrasting his notion of a social substratum to the Neo-Kantianism substrate of Lask, I have shown that primitive social phenomena such as religion are much more grounded in the ‘bedrock’ of the substrate than the normative conception of law. By considering Durkheim’s attempt at a ‘purification’ of concepts and social conditions in light of Kelsen’s pure theory of law, I have only added to his distinctness as a French Neo-Kantian. This is because I identified his belief that the social substrate³⁷ is already constructed of pre-conceptualized segments, and this, in turn, further feeds the belief that Kelsen’s ‘insertion’ of ‘conceptual content’ later in the construction of the societal categories of the faculty of understanding is both irrelevant and unattainable in Durkheimian thought (since it

 See Bachelard (1994).  See Lefebvre (1991).  Departing from Lask once again.

54

Dustin Garlitz

does not work from the top-down). I have only added to Durkheim’s distinctness as a Neo-Kantian social thinker by presenting, analyzing, and reconstructing his explicit arguments entailing notions of ‘distortion’ and emotional “contagion”, which I find are the leading themes that have historically secured his prominent place in the movement’s trajectory.

Bibliography Allen, N.J., Pickering, W.S.F. and Watts Miller, W. (eds.) (1998): On Durkheim’s Elementary Forms of the Religious Life, London. Bachelard, Gaston (1994): The Poetics of Space. The Classic Look at How We Experience Intimate Places, Boston/MA: Beacon Press. Bourdieu, Pierre (1977): Outline of a Theory of Practice, translated by Richard Nice, Cambridge/New York, Cambridge University Press. Bourdieu, Pierre (1990): The Logic of Practice, translated by Richard Nice, Stanford, CA, Stanford University Press. Bourdieu, Pierre (1989): Social Space and Symbolic Power, in: Sociological Theory 7 (1), pp. 14 – 25. Bourdieu, Pierre (1985): The Social Space and the Genesis of Groups, in: Theory and Society 14 (6), pp. 723 – 744. Cassirer, Ernst (1965): The Philosophy of Symbolic Forms, Three Volumes, translated by Ralph Mannheim, New Haven: Yale University Press. Cassirer, Ernst (1981): Symbol, Myth, and Culture. Essays and Lectures of Ernst Cassirer. 1935 – 1945, edited by Donald Philip Verene. New Haven: Yale University Press. Cassirer, Ernst (1953): Language and Myth, translated by Susan K. Langer, New York, Dover Publications. Cassirer, Ernst (1946): The Myth of the State, New Haven: Yale University Press. Derrida, Jacques (1978): Structure, Sign, and Play in the Discourse of the Human Sciences, in: Writing and Difference, translated by Alan Bass, Chicago: The University of Chicago Press, pp. 278 – 293. Derrida, Jacques (1998): Of Grammatology (Corrected Edition), translated by Gayatri Chakravorty Spivak, Baltimore, MD/London: The Johns Hopkins University Press Durkheim, Émile (1965 [1912]): The Elementary Forms of the Religious Life, translated by Joseph Ward Swain, New York. Durkheim, Émile (1953 [1924]): Sociology and Philosophy, Illinois. Durkheim, Émile (1960): Émile Durkheim. 1858 – 1917, edited by Kurt H. Wolff, Ohio State. Durkheim, Émile (1972): Selected Writings, edited and translated by Anthony Giddens, Cambridge. Durkheim, Émile (1973): On Morality and Society, edited by Robert N. Bellah, Chicago. Durkheim, Émile and Mauss, Marcel (1967): Primitive Classification, translated by Rodney Needham, Chicago. Donagan, Alan, Perovich, Jr., Anthony N. and Wedin, Michael V. (eds.) (1986): Human Nature and Natural Knowledge: Essays Presented to Marjorie Grene on the Occasion of Her Seventy-Fifth Birthday, Dordrecht.

Durkheim’s French Neo-Kantian Social Thought

55

Giddens, Anthony (1979): Émile Durkheim, New York. Godlove, Terry (1989): Religion, Interpretation, and Diversity of Belief: The Framework Model from Kant to Durkheim to Davidson, Cambridge. Godlove. Terry (1996): Is “Space” a Concept?: Kant, Durkheim, and French Neo-Kantianism, in: Journal of the History of the Behavioral Sciences 32 (4), pp. 441 – 455. Hanna, Robert (2005): Kant and Nonconceptual Content, in: European Journal of Philosophy 13 (2), pp. 247 – 290. Heidemann, Dietmar H. (ed.) (2013): Kant and Non-Conceptual Content, London: Routledge. Jones, Robert Alun (1999): The Development of Durkheim’s Social Realism, Cambridge: Cambridge University Press. Kant, Immanuel (2003): Critique of Pure Reason, translated by Norman Kemp Smith, New York. Kelsen, Hans (1992): Introduction to the Problems of Legal Theory, translated by Bonnie Litschewski Paulson and Stanley L. Paulson, Oxford: Claredon Press. Kisiel, Theodor (1995): Why Students of Heidegger will have to Read Emil Lask, in: Man and World 28, pp. 197 – 240. LaCapra, Dominic (1972): Émile Durkheim: Sociologist and Philosopher, London. Lask, Emil (1950): Legal Philosophy, in: The Legal Philosophies of Lask, Radbruch, and Dabin, translated by Kurt Wilk, MA: Harvard University Press. Lefebvre, Henri (1991): The Production of Space, Malden, MA/Oxford: Blackwell. Lévi-Strauss, Claude (1966): The Savage Mind, Chicago/London: The University of Chicago Press. Lévi-Strauss, Claude (1969): The Elementary Structures of Kinship, Boston: Beacon Press. Lévi-Strauss, Claude (1955): The Structural Study of Myth, in: The Journal of American Folklore 68 (270): Myth: A Symposium, pp. 428 – 444. Lévi-Strauss, Claude (1974): Structural Anthropology, Vol. 1, translated by Claire Jacobson and Brooke Grundfest Schoepf, New York: Basic Books. Lévi-Strauss, Claude (1983): Structural Anthropology, Vol. 2, translated by Monique Layton, Chicago and London: The University of Chicago Press. Logue, William (1993): Charles Renouvier: Philosopher of Liberty, London. Lukes, Steven (1972): Émile Durkheim: His Life and Work- A Historical and Critical Study, New York. Mauss, Marcel (2002): The Gift, New York/London: Routledge. Merleau-Ponty, Maurice (2012): Phenomenology of Perception, translated by Donald A. Landes, London and New York: Routledge. Morrison, Ken (1995): Marx, Durkheim, Weber: Formations of Modern Social Thought, London. Nielsen, Donald A. (1999): Three Faces of God: Society, Religion, and the Categories of Totality in the Philosophy of Émile Durkheim, New York. Pickering, W.S.F. (1984): Durkheim’s Sociology of Religion: Themes and Theories, London. Pickering, W.S.F. (ed.) (2000): Durkheim and Representations, London. Poggi, Gianfranco (2000): Durkheim, Oxford. Rawls, Anne Warfield (2004): Epistemology and Practice: Durkheim’s The Elementary Forms of Religious Life, Cambridge. Simmel, Georg (1971): On Individuality and Social Forms, edited by Donald N. Levine, Chicago/London: The University of Chicago Press.

56

Dustin Garlitz

Schmaus, Warren (1994): Durkheim’s Philosophy of Science and the Sociology of Knowledge: Creating an Intellectual Niche, Chicago/London: The University of Chicago Press. Schmaus, Warren and Pickering, W.S.F (eds.) (2001): Emile Durkheim: Critical Assessments, London: Routledge. Schmaus, Warren (2004): Rethinking Durkheim and his Tradition, Cambridge: Cambridge University Press. Schwartz, Barry (1981): Vertical Classification: A Study in Structuralism and the Sociology of Knowledge, Chicago. Searle, John R. (1995): The Construction of Social Reality, New York: The Free Press. Stedman Jones, Susan (2001): Durkheim Reconsidered, Cambridge. Thompson, Kenneth (ed.) (1985): Readings from Émile Durkheim, London. Wittgenstein, Ludwig (2001): Remark 217, in: Philosophical Investigations, translated by G.E.M. Anscombe, Malden, MA/Oxford: Blackwell.

Hauke Heidenreich

Eduard von Hartmann, die Postulatenlehre und die Genese des Neukantianismus im Kontext spiritistischer Debatten um 1900 Abstract: Most scholars dealing with neo-Kantianism emphasize the existence of an explicit origin that characterizes neo-Kantianism as a clearly defined ‘movement’ with certain objects. In spite of the fact that differences between various neo-Kantian authors are mentioned in the current discourse (e. g. ‘Südwestdeutsche’ and ‘Marburger Schule’) scholars claim a singular unity of all neo-Kantian debates represented in the problem of ‘Geltung’. This is used to construct a certain relevance of neo-Kantianism in today’s philosophy. My contribution will pursue a historical approach following Klaus Christian Köhnke’s and Ulrich Sieg’s demands. Instead of sketching systematic topics I pay further attention to the context in which an essence of neo-Kantianism was claimed or rejected. In doing so I will examine the discussions on the doctrine of postulates in Kant’s philosophy mainly led by Hans Vaihinger’s ‘Philosophy of As If’ and the impact of Eduard von Hartmann on the genesis of neo-Kantianism. I conclude that the definition of a consistent neo-Kantian ‘movement’ was an effect of rejecting spiritualist interpretations of the doctrine of postulates and in the following a rejection of the whole doctrine as ‘mysticism’ in canonical Kant interpretations.

1 Einheit als systematische Schulgemeinschaft? Simply defined, neo-Kantianism, in a historical sense, was the movement in 19th-century Germany to rehabilitate Kant’s philosophy. Neo-Kantianism was the predominant philosophical movement in Germany in the final decades of the 19th century, and its influence spread far and wide, to Italy, France, England and Russia (Beiser 2017, 1).

Mit diesem programmatischen Satz eröffnete Frederick C. Beiser 2017 seinen Text zur Entstehung des Neukantianismus. Der Neukantianismus sei nicht nur eine klar abgrenzbare Bewegung mit einem gemeinsamen Anliegen – „to rehabilitate Kant’s philosophy“ – sondern als solche auch „predominant“ im Deutschland des 19. Jahrhunderts gewesen. Beiser zufolge lässt sich das ‚Wesen‘ oder die Essenz dessen, was er als Neukantianismus beschreibt, also klar festlegen: als eine Be-

https://doi.org/10.1515/kantyb-2020-0003

58

Hauke Heidenreich

wegung, deren Ursprünge sozusagen bereits zu Kants Lebzeiten liegen, z. B. bei Fries, Herbart oder Beneke (Beiser 2017, 3). Den Neukantianismus als eine einheitliche Bewegung oder Schulgemeinschaft zu definieren, ist, trotz der vielerorts geäußerten Kritik an einer vereinheitlichenden Darstellung, weitestgehend Konsens in der heutigen Forschung zum Neukantianismus (Zeidler 2014, 5; Krijnen 2012, 69; Flach 2012, 12– 13; Gabriel 2007, 92; Orth 1994, 28 – 29). Meist wird diese Schulgemeinschaft grob in zwei Hauptstränge unterteilt und mit bestimmten Namen versehen, Hermann Cohen und Paul Natorp für den Marburger Neukantianismus sowie Wilhelm Windelband und Heinrich Rickert für den Südwestdeutschen Neukantianismus (Pascher 1997). Diese dichotomische Einteilung wurde mit dem Erscheinen des von Christian Krijnen und Andrzej J. Noras herausgegebenen Sammelbandes Marburg versus Südwestdeutsch (Krijnen/Noras 2012) nochmals systematisch verstärkt. Vor diesem Hintergrund ist gerade die Relevanz des Neukantianismus für heutige philosophische Debatten von mehreren Autorinnen und Autoren betont worden (Renz 2002, 44; Flach 2012, 11; Höffe 2012, 212; Stolzenberg 1995, 3). Einen gemeinsamen Grund aller philosophischen Debatten des Neukantianismus erkennen zu können scheint also gerade die zentrale Voraussetzung zu sein, die Relevanz desselben für die heutige Philosophie betonen zu können. Dieses Beharren auf einer Relevanz, die sich umgekehrt auch nur aus einer grundsätzlichen systematischen Einheit ergeben könne, wird vor allem auch als Argument gegen historische Forschungen gesetzt. Werner Flach etwa vertritt die These, dass historische Arbeiten zum Neukantianismus nicht „die Bastionen der Ignoranz zu schleifen“ imstande seien, die sich in „Empirismus, Phänomenologie, Analytischer Philosophie, Philosophischer Hermeneutik, Strukturalismus und Neostrukturalismus sowie Postmoderne“ finden ließen (Flach 2012, 10). Daher müsse jede historische Arbeit den Primat des „sachlichen Anliegen[s] des Neukantianismus“, nämlich seine Essenz als „Lehre von der Kultur“ und der „Geltungswertigkeit“ (Flach 2012, 12) – akzeptieren. Die systematische Einheit einer Debattenlage wird also als Voraussetzung der geschichtlichen Forschung installiert, was uns auf Beisers These eingangs zurückverweist, dessen Ausgangspunkt einer historischen Arbeit eben gerade in der Position bestand, dass die Einheit des Neukantianismus objektiv erwiesen sei und zudem auf Kants Lebzeiten zurückgehe. Auffällig ist in diesem Kontext, gegen wen Flach die Einheit des Neukantianismus konkret erhebt. Abgesehen nämlich von seiner Front gegen verschiedene von ihm gesehene Denkformen der Gegenwart (Empirismus, Phänomenologie, Neustrukturalismus) ist der einzige explizit benannte Gegenspieler der Ideenhistoriker Klaus Christian Köhnke. Köhnke hatte mit seinem Buch Entstehung und Aufstieg des Neukantianismus seinerzeit ein Defizit historischer Forschungen zum

Eduard von Hartmann und die Genese des Neukantianismus

59

Neukantianismus ausgemacht und gerade den positiven Effekt historischer Annäherungen betont. Historische Arbeit könne in erster Linie die Verwobenheiten philosophischer Konzepte in den zeitgenössischen Kontext herausarbeiten und damit „eine bloße Bestätigung des gegenwärtigen Denkens anhand eines ihm entsprechend ausgewählten geschichtlichen Materials“ vermeiden (Köhnke 1993, 9 – 10). Köhnke sieht zwei Hauptprobleme im Umgang mit dem Neukantianismus: einerseits die Tendenz, den Neukantianismus als reine Vorgeschichte heutiger Philosophie zu betrachten, und andererseits, aber damit zusammenhängend, die Gefahr, mit diesem Vorgehen der „neukantianischen Glorifizierung der eigenen Geschichte“ unkritisch anheimzufallen (Köhnke 1993, 11). Demgegenüber müsse man zu den Quellen zurückgehen und erneut die Frage nach der Faktizität stellen. (Köhnke 1993, 14). Denn, und offensichtlich zum Unbehagen Flachs, betont Köhnke nun, dass der Neukantianismus weniger durch eine bestimmte philosophische Programmatik hervorgebracht oder bestimmt worden sei, sondern durch politische Konfliktlagen Ende des 19. Jahrhunderts, namentlich die Nachmärzzeit, der Materialismusstreit, der Kulturkampf und die „Krise des Liberalismus um 1878/79“ (Köhnke 1993, 15). Gerade diese Diagnosen brächten es mit sich, dass besonders die Vorgeschichte und Entstehung des Neukantianismus neu erzählt werden müsse. Auffällig ist, dass Köhnke den Ursprung des Neukantianismus weiterhin nicht in einer besonderen philosophischen Tat erkennt, sondern als „Etikett“, das man einem philosophischen Gegner anhängte (Köhnke 1993, 213 – 214; Luft 2015, XX – XXI; Sieg 2016, 33), in einer Zeit, die von nationalistischen Vereinnahmungen der philosophischen Vergangenheit stark geprägt war (Köhnke 1993, 179 – 190). Es ist daher nicht verwunderlich, dass der Wert historischer Forschung zum Neukantianismus in der Folge von Köhnkes Text von einzelnen Autoren betont wurde (Sieg 1994; 2016). Denn die Einheit des Neukantianismus wird offensichtlich auch aus einem gegenwärtigen Interesse heraus beschworen. Gerade in diesem Kontext lohnt daher der von Köhnke geforderte Rückgang zu den Quellen. Im Kontext der Diagnose mehrerer Autoren, dass der historische Ursprung des Neukantianismus mit einer Etikettierung zusammenhänge, stellt sich daher die Frage, wie genau die Entstehung eines philosophischen Systems mit historischen Frontstellungen zusammenhängt, oder anders ausgedrückt: durch welche konkreten Praktiken des Ausschlusses und des Erhebens von Wahrheitsansprüchen wurde aus einem „Umschlagplatz“ für philosophische Interessen (Orth 1994, 16) eine „philosophische Schulgemeinschaft“ (Sieg 1994)? Zur konkreten Vorgehensweise: Ich werde zuerst auf die vermeintlichen Ursprünge neukantianischer Debatten und deren diskursive Genese eingehen. Hier wird zu zeigen sein, dass bereits diese Ursprünge Teil von Aushandlungs- und Abgrenzungsprozessen im historischen Kontext sind. Dann werde ich die Re-

60

Hauke Heidenreich

zeption dieser Debatten durch Eduard von Hartmann näher untersuchen und die Kontroverse mit Vaihinger Ende der 1870er Jahre beschreiben. Die bereits in diesem Zusammenhang thematisierte Kritik am höchsten Gut wird im Weiteren in ihrer Beziehung zu spiritistischen Debatten um 1900 analysiert. Die zentrale Bedeutung spiritistischer Deutungen der kantischen Morallehre im Kontext, der um 1900 sich konstituierende Kantforschung und deren Interpretation der Postulatenlehre und des höchsten Gutes wird dabei genauer zu zeigen sein. Dabei ist in einem weiteren Schritt auf die wichtige Rolle der Texte Ernst Haeckels und Eduard von Hartmanns in der Umdeutung des Wortes ‚Neukantianismus‘ um 1900 von einem fremdbeschreibenden Kampfbegriff zu einer abgrenzenden Selbstbeschreibung einzugehen. Haeckels und Hartmanns Positionen bezüglich des höchsten Gutes bildeten bedeutende Segmente in der (Neu‐)Verortung einer neukantianischen ‚Einheit‘ im Diskurs, deren Effekte das heutige Verständnis des Neukantianismus massiv prägen. In einem Fazit werden die Kontexte und ihre Rezeption noch einmal abschließend dargestellt und theoretische Schlussfolgerungen gezogen.

2 Spuren der Differenz im historischen Kontext Auf Basis welcher Ausschlüsse wird also die Einheit erkannt? Und was sind die historischen Entstehungsbedingungen dieser erkannten Einheit? Der von Köhnke geforderte Rückgang zu den Quellen, der zur Beantwortung dieser Fragen dienen kann, hat daher gerade die Spuren der Differenz in der heutigen Forschung und den zeitgenössischen Texten zu verfolgen, die sich im Zusammenhang mit dem Anspruch auf eine Einheit des Diskurses finden. Ernst Wolfgang Orth hatte bereits vor Jahren einen literarischen Ort ausgemacht, den er als „institutionelle Manifestation“ des Neukantianismus bezeichnete, der innerhalb der von Orth diagnostizierten offensichtlichen Heterogenität der zeitgenössischen Debatten in gewisser Weise den neukantianischen Konsens herstellte: die Kant-Studien (Orth 1994, 13). Gegründet wurde diese bis heute zentrale Zeitschrift der Kantforschung von dem halleschen Philosophieprofessor Hans Vaihinger. Der Pfarrerssohn Vaihinger hatte am Tübinger Stift studiert (Simon 2014, 23), war ab 1883 außerordentlicher Professor in Straßburg und später in Halle, wo er 1894 auch zum Professor ernannt wurde. Zum Zeitpunkt der Gründung hatte sich Vaihinger bereits den Ruf eines herausragenden Kantforschers erworben. Er hatte 1881 und 1892 die zwei ersten Bände seines groß angelegten Kommentars zur Kritik der reinen Vernunft (Vaihinger 1881; 1892) publiziert, die Vaihinger als maßgeblichen Kantforscher etablierten. Zudem war Vaihinger ein überzeugter Parteigänger eines Autors, der immer wieder als wichtige Grün-

Eduard von Hartmann und die Genese des Neukantianismus

61

dungsfigur des Neukantianismus erkannt wurde: Friedrich Albert Lange (Beiser 2017, 2; Bergunder 2016, 93; Pascher 1997, 50; Sieg 1994, 119). Schon zeitgenössisch als „durchschlagendsten Sieg der neukantischen Bewegung“ (Vorländer 1908, 420) bewertet, markiert Langes Geschichte des Materialismus in der Sicht des Cohen-Schülers Karl Vorländer sozusagen den Ursprung des Neukantianismus. Lange hatte sich in seinem Hauptwerk ausgiebig mit Kants Erkenntnislehre beschäftigt und kam in diesem Kontext auch zu einer scharfen Kritik der Moralphilosophie. Denn immerhin sei es doch der zentrale Irrtum der bisherigen Kantforschung gewesen, dass Kant die Ideen Seele, Freiheit und Gott so sicher festgestellt habe, „wie die Gesetze des Sternenlaufs“. Eine solche Deutung sei dem „Geist der Vernunftkritik“ völlig fremd, da ja gerade die praktische Philosophie „der wandelbare und vergängliche Theil der Kant’schen Philosophie“ gewesen sei (Lange 1877, 18). Statt also auf Basis der Ideen Gott, Freiheit und Unsterblichkeit „unsre Erkenntnis zu erweitern“ (Lange 1877, 81), wie viele „ungeschickte“ Gegner des Materialismus es täten, solle man das Naturgesetz selbst als „nur Erscheinung“ auffassen, während die „Naturanlage“ unserer Vernunft mit „Nothwendigkeit“ dazu führe, „neben der Welt, die wir mit unseren Sinnen wahrnehmen, noch eine eingebildete Welt anzunehmen“ (Lange 1877, 83). Dezidiert lehnt Lange die von Kant installierte objektive Realität der Freiheit (CprR, AA 05: 03 – 04; 44; 115; 132– 134) ab: „Kant wollte nicht einsehen, […] dass die ‚intelligible Welt‘ eine Welt der Dichtung ist und dass gerade hierauf ihr Werth und ihre Würde beruht. Denn Dichtung in dem hohen und umfassenden Sinne, wie sie hier zu nehmen ist, kann nicht als ein Spiel talentvoller Willkür zur Unterhaltung mit leeren Empfindungen betrachtet werden, sondern sie ist eine notwendige und aus den innersten Lebenswurzeln der Gattung hervorbrechende Geburt des Geistes […]“ (Lange 1877, 88). Langes scharfer Angriff auf jegliche „metaphysische“ Deutungen der kantischen Morallehre, besonders auf die objektive Realität der Postulate, war durchaus keine singuläre Diskursposition in der Zeit. Bereits 1865 – und damit ein Jahr vor der Erstausgabe von Langes Geschichte – hatte Otto Liebmann die Existenz des Dings an sich in Kants Philosophie als „transscendente Vogelscheuche“ bezeichnet, die als „fremder Tropfen Bluts im Kriticismus“ als ein „Wegweiser in’s irrationale Jenseits“ diene und daher abgeschafft werden müsse (Liebmann 1865, 205). Der Trendelenburg-Schüler Gustav Teichmüller ließ am Ende seines aus der Egoperspektive Kants abgefassten satirischen Berichts von Kants postmortaler Reise durch das Jenseits, wo dieser keinen Lohn für seine Glückwürdigkeit erhalten habe, Kant in die Welt der Erscheinungen zurückkehren und mit einer offenbar dezidiert nicht ironisch gemeinten Apologie zur Exegese der kantischen Morallehre schließen: „Meine treuen Anhänger mögen sich jeder Täuschung und schwärmerischen Illusion in Betreff der himmlischen Glückseligkeit, allwo ein

62

Hauke Heidenreich

Einklang zwischen Freiheits- und Natur-Gesetzen herrsche, gänzlich entschlagen und, wie ich, aus Achtung vor dem Sittengesetz, d. h. um der Glückseligkeit würdig zu werden, immer hier in der Welt der Erscheinungen bleiben“ (Teichmüller 1877, 45). In diesen Diskurs hatte sich bereits ein Jahr vor Erscheinen der dritten Auflage von Langes Hauptwerk Hans Vaihinger eingeschaltet. Der zu diesem Zeitpunkt erst Anfang 20-jährige Vaihinger verfasste eine Schrift, in der er Langes Philosophie als den wahren Weg bezeichnete, Kants System nun endlich „ganz und voll“ zu verstehen (Vaihinger 1876, 4). Auffällig aber ist die Front, in der Vaihinger diese Diagnose verortet. Es geht in diesem Kontext um den damals aufgrund seiner Philosophie des Unbewussten weithin bekannten Philosophen Eduard von Hartmann, dessen oberste Prinzipien, im Gegensatz zu Lange, nur aus einer mystischen „Eingebung des Unbewußten“ kämen (Vaihinger 1876, 11), dessen „mystische Conceptionen nur Phantasieen sind“ (Vaihinger 1876, 12).

3 Hartmann, Vaihinger und eine Debatte um den Neukantianismus Hartmann hatte auf den Mystikvorwurf Vaihingers mit einem eigenen Text geantwortet, der schon im Titel die programmatische Tragweite des Disputs anzeigt: Neukantianismus, Schopenhauerianismus und Hegelianismus in ihrer Stellung zu den philosophischen Aufgaben der Gegenwart. Hierin warf Hartmann Lange vor, mit der Geschichte des Materialismus eine „durch geschichtliche Studien angeschwollene Tendenzschrift“ verfasst zu haben (Hartmann 1877, 1). Vaihingers Verteidigungsschrift für Lange zeige in ihrer „Hätschelung des Materialismus“ und der „Antipathie gegen Idealismus, Metaphysik und Speculation“ gerade das wahre unwissenschaftliche und unmoralische Wesen von Langes Schrift (Hartmann 1877, 6). Der eigentliche Grund für Langes Bezugnahme auf Kant bestehe darin, dass auch Kant es immer abgelehnt habe, Metaphysiker zu sein, während er in Hartmanns Diktion eigentlich ein „unbewusster Metaphysiker“ gewesen sei, das aber in der Forschung meist als „nebensächliche Auswüchse“ oder „senile Abirrungen“ entschuldigt werde. Hartmann zufolge verschleiere Lange also absichtlich seine Quellen, um den Anschein zu erzeugen, eine völlig antimetaphysische Philosophie präsentiert zu haben, deren Quellen selbst aber nichts Anderes als eben Metaphysik seien. Lange inszeniere sich als „Kantianer, und Vaihinger feiert ihn als das Haupt und den Führer des im letzten Jahrzehnt entstandenen Neukantianismus“, was aus dem Grund aber eigentlich folgerichtig sei, als auch Kants Philosophie selber eine „wunderlich verschlungen[e]“ Mischung aus

Eduard von Hartmann und die Genese des Neukantianismus

63

„subjectivem Idealismus und transcendentalem Realismus“ sei (Hartmann 1877, 19). Lange und Vaihinger würden sich, in dem sie alles Übersinnliche für Dichtung erklärten, zudem absichtlich von der kantischen Freiheitslehre entfernen (Hartmann 1877, 20). Nach Hartmann mangele es nun im Anschluss an den Vaihingerschen Skeptizismus an einer Philosophie, die weder in den Nihilismus, noch in eine erneute Hegelsche Spekulation umschlägt und synthetisches Denken ohne Anspruch auf „apodictische Gewissheit“ möglich macht. Wenn man es nämlich wie Vaihinger und Lange mit dem Skeptizismus übertreibe, wie vor allem „junge Leute ohne speculative Anlagen“ es manchmal täten, lande man letztlich beim Materialismus der Sozialdemokratie. Vaihinger und Lange hätten Kants Kritizismus zum Skeptizismus verzerrt und unter dem Deckmantel der „Kantvergötterung“ ausgegeben (Hartmann 1877, 25 – 26). Der „Geist“ der kantischen Lehre, nicht der „Buchstabe“, solle daher nun wiedererweckt werden [Hartmann spricht in diesem Zusammenhang von „Wiedererweckung“], „nur der Geist ist es, der lebendig macht“ (Hartmann 1877, 29). Im Jahr 2006 hatte der Schweizer Philosoph Jean-Claude Wolf eine Monografie über Hartmann vorgelegt.Wolf wies ausdrücklich auf den Widerspruch hin, dass Hartmann sich mit seiner Morallehre als Gegner des Neukantianismus inszeniere, gleichzeitig aber Kants Ethik radikalisiere, sie sozusagen „kantischer als Kant“ mache, „sofern er auch die Aussicht auf einen ‚transzendenten Optimismus‘ (oder einen Lohn im Himmel) verwarf“ (Wolf 2006, 14). Diese Ablehnung jeglichen Eudämonismus falle mit Hartmanns Ablehnung des zeitgenössischen Katholizismus und der Sozialdemokratie zusammen (Wolf 2006, 14; 18), gegen letztere sieht Wolf dezidiert einen Zusammenhang mit Hartmanns scharfer Abgrenzung gegen jedweden Versuch einer Harmonisierung von Glück und Freiheit, die er eben der Sozialdemokratie anlaste. Wenn man die philosophischen Errungenschaften Langes daher zusammenfassen wolle, so erkennt es in der Folge der Philosoph und Theologe Adolf Lasson in einer Rezension des Hartmannschen Textes, sei dies: (1) „Philosophie als Wissenschaft ist unmöglich“ und (2) die Forderung einer „Metaphysik als speculative Begriffsdichtung“ sei nur „eine Reihe von ‚Hirngespinnsten‘“. Zudem, für Lasson das Schlimmste, „dieser vollkommene Widersinn“ firmiere auch noch unter dem Namen Kants: „Einmal muss es doch gesagt werden: diese Herren NeoKantianer von der Lange’schen oder einer verwandten Art nennen sich nach Kant etwa mit demselben Rechte, wie die Neo-Lutheraner sich nach Luther, oder die Jesuiten sich nach Jesus nennen“ (Lasson 1877, 226). Nach einer weiteren Rezension Carl Gerhards sei, die Langesche und Vaihingersche Philosophie „für ‚das eigentliche Resultat der kantischen Erkenntnistheorie‘ anzusehen“, nur dann möglich, „wenn man eben die realistischen Stellen in Kant’s Schriften ignorirt“, in denen für Kant die Existenz des Dings an sich „feststand, obschon er seine Be-

64

Hauke Heidenreich

schaffenheit als unerkennbar bezeichnet“. Für den ‚Neukantianismus‘ sei Philosophie letztlich nur eine Art bewusster Selbsttäuschung (Gerhard 1878, 101– 102). Hartmann wirft offensichtlich den Terminus Neukantianismus als Kampfbegriff in den Diskurs, um einer gegen ihn gerichtete Polemik zu begegnen, genauer: um den Mystikverdacht zu entkräften, den Vaihinger gegen ihn ausgesprochen hatte. Neukantianer seien, in Hartmanns Sicht, Autoren wie eben Lange und Vaihinger, die, unter Verschleierung der eigenen metaphysischen Wurzeln, jegliche Spekulation unter dem Verdikt der Mystik in den Nihilismus verkehrten und durch ihre falsche Kantauslegung letztlich der Sozialdemokratie in die Hände spielten; ein Jahr vor Inkrafttreten der Sozialistengesetze auch ein im philosophischen Gewand vorgetragenes Bekenntnis zur politischen Macht im Kaiserreich, d. h. zur Bismarck-Regierung. Auffallend ist die schnelle Akzeptanz des Hartmannschen ‚Neukantianismus‘-Begriffs in der direkten Rezeption der Kontroverse, die offensichtlich ebenfalls mit politischen Positionen zusammenhing; hier ist auf Lassons Erwähnung der Jesuiten hinzuweisen, die ja seit den Kulturkampfgesetzen in Deutschland des Landes verwiesen waren (Clark 2006, 648). Der Neukantianismus wird also in diesem Kontext eben nicht mit einem ‚Zurück zu Kant‘, sondern mit einer Entfernung von dem „Geist“ von dessen Philosophie zusammengebracht; geradezu als Signum für eine Gruppe Philosophen, welche den Namen Kant nur vorschieben, um ihre eigenen dogmatischen oder quasimetaphysischen Wurzeln zu verdecken und durch diese Taktik eigentlich dem historischen Materialismus, wenn nicht sogar dem Katholizismus das Wort zu reden. Ist es also eine Reaktion auf die von Hartmann, Lasson u. a. vorgebrachten Vorwürfe, dass Vaihinger in seinem Commentar von 1881 wiederum seine Methode zur Analyse der ersten Kritik nun so beschrieb, dass Deutungen der kantischen Lehre aus dem „Geist“ und aus dem „Buchstaben“ sich ergänzen müssen (Vaihinger 1881, VI)? Wollte sich Vaihinger von dem Verdacht befreien, ein heimlicher Sozialdemokrat zu sein? Hartmann hatte jedenfalls in seiner Rezension von Vaihingers zweitem Kantkommentar ausdrücklich eine allmähliche Annäherung Vaihingers an seine eigene Position erkannt (Hartmann 1893, 344). Weiterhin ist auffällig, dass Vaihinger das Projekt Langes bemerkenswerter Weise nicht als ‚Neukantianismus‘ bezeichnete, sondern als ‚Kritizismus‘. Die beiden Termini sind in ihrer diskursiven Setzung im historischen Kontext einander entgegengesetzt und eben zu diesem Zeitpunkt noch nicht identisch, wie es die heutige Forschung teilweise nahezulegen scheint (Zeidler 2009, 126 – 128). Als allerdings Hartmann in seiner Kritik der „Kathederphilosophie“ des „erkenntnistheoretischen Neukantianismus“, die durch ihre Ignoranz gegenüber der zweiten Kritik nur einer „egoistischen Pseudomoral fröhnen“, sein Konzept des Pessimismus in Kants Lehre eintragen wollte (Hartmann 1880, XII), kam es zu

Eduard von Hartmann und die Genese des Neukantianismus

65

Widerstand gegen Hartmanns Thesen. Hartmann hatte nämlich vorgeschlagen, das kantische höchste Gut und die Postulate durch den Pessimismus als „ethisches Postulat“ zu ersetzen (Hartmann 1880, 1; 8; 17– 19). Das höchste Gut bei Kant wollten die Rezensenten dann doch nicht aufgeben (Schaarschmidt 1881; Sommer 1880; Volkelt 1881). Zumindest noch nicht.

4 Spiritismus und Postulatenlehre 1888 kam es zur Publikation eines Textes, der in der Folge massiven Einfluss auf die Rezeption der Postulatenlehre und des höchsten Gutes haben sollte. Es handelte sich um die Wiederveröffentlichung der Mitschriften von Kants Vorlesung L1 über Metaphysik durch den Münchner Philosophen und Okkultisten Carl du Prel (Prel 1964).¹ Du Prel behauptete in seinem eigens verfassten Vorwort: „Man muß sehr viel überflüssige Zeit haben, um die Frage zu untersuchen, ob Kant Spiritist war. Wenn aber die Frage gestellt wäre, ob er heute Spiritist sein würde, so müßte ich diese Frage bejahen“ (Prel 1964, 57). Als Beleg dafür führte du Prel nicht nur die von ihm gekürzt herausgegebenen Vorlesungsmitschriften und die Träume eines Geistersehers an, sondern auch die Kritik der praktischen Vernunft, in der Kant „mystische“ Ideen vertreten habe, besonders in Bezug auf die menschliche Freiheit (Prel 1964, 44– 45). Kant habe daher die Präexistenz und Unsterblichkeit der Seele, das materielle Dasein als Ausnahme sowie das transzendentale hingegen als Regel gelehrt und die Notwendigkeit von du Prels eigener Forschung zum Seelenbeweis antizipiert (Prel 1964, 63). Vaihinger, über zwanzig Jahre in Briefkontakt mit du Prel und damit über dessen mystische Kantdeutung genauestens informiert (Kaiser 2008, 51), setzte sich zweimal öffentlich mit du Prels Text auseinander (Vaihinger 1891; 1892, 512– 514). Der spätere Herausgeber des Standardwerks Ueberwegs Grundriss der Geschichte der Philo-

 Zur Darstellung des Einflusses von du Prel auf die Kantforschung um 1900 vgl. Stengel (2011, 698 – 700); Kaiser (2008); Heidenreich (2019) und Prunea-Bretonnet (2018). Nicht zutreffend ist die Diagnose von Prunea-Bretonnet (2018, 243), du Prels Vorlesungen über Psychologie seien 1889 erschienen. Ihr Erscheinen wurde bereits in den spiritistischen Psychischen Studien, (Band XVI, 1888, Heft 9, 452) angezeigt. Offensichtlich hatte du Prel den Text ein Jahr vordatiert. Vgl. dazu auch Kaiser (2008, 67). Zudem setzt Prunea-Bretonnet (2018, 243) einen scharfen Bruch zwischen Kant und Swedenborg voraus, der du Prels Position von vornherein als „irrational“ einstuft und den historischen Kontext du Prels an dieser Stelle ignoriert. Zur Kritik dieses scharfen Bruchs vgl. Stengel (2011, 639 – 671; 2015). Erstere Arbeit von Stengel zitiert auch Prunea-Bretonnet, ohne allerdings Stengels Kritik an einem scharfen Bruch zu erwähnen.

66

Hauke Heidenreich

sophie Max Heinze veröffentlichte ein ganzes Buch (Heinze 1894), in dem er sich unter Bezug auf Vaihinger mit du Prel auseinandersetzte. Du Prels Thesen zur mystischen Kantdeutung – Mystik verstand du Prel im Gegensatz zu Vaihinger nicht als Polemik, sondern als positive Selbstbeschreibung (Prel 1964, 17– 19; 24) – waren daher weitgehend im Diskurs bekannt², als 1898 der Berliner Philosoph und Pädagoge Friedrich Paulsen mit seinem Immanuel Kant. Sein Leben und seine Lehre an die Öffentlichkeit trat. Das Auffällige an Paulsens Buch war, dass er unter Bezug auf genau dieselben drei Kanttexte wie du Prel den „ethisch-metaphysischen Idealismus“ Kants (Paulsen 1899, X) betonen wollte. Eben durch Kants Zwei-Welten-Lehre und seinen Primat der praktischen Vernunft könne der Verstand erkennen, „dass es jenseits der sinnlichen Welt eine absolute Wirklichkeit gibt“ (Paulsen 1899, 6). Gegen die Sicht von Vaihinger sei auch die erste Kritik nicht nur negativ, sondern vielmehr „eine positive Metaphysik, nämlich eine idealistische Weltanschauung“, „nicht die Niederreissung der übersinnlichen Welt, sondern vielmehr die definitive Befestigung des Glaubens an sie und unserer Zugehörigkeit zu ihr“ (Paulsen 1899, 120). Genau mit diesen Gedanken zur Morallehre befinde sich Kant „an den Grenzen der Mystik“ (Paulsen 1899, 67). Trotz Kants „nicht ganz ohne Ernst“ gemeinter Zustimmung zu Swedenborgs Sicht des mundus intelligibilis habe Kant aber die „spiritistischen Erscheinungen“ nicht ernst genommen (Paulsen 1899, 87) und man müsse Kant daher vor allen Autoren in Schutz nehmen, die ihn „zum Spiritisten machen wollen“ (Paulsen 1899, 105). Paulsens Kritikpunkt richtet sich dem folgend auf die Postulatenlehre und die Lehre vom höchsten Gut. Denn damit führe Kant, gleichsam durch die Hintertür seiner Philosophie, den Eudämonismus wieder in die Moral ein, den Kant vorher aber eigentlich schon als moralisch nicht tauglich erkannt habe. Wie solle denn, so Paulsens Frage, ein zeitlicher Fortschritt im Jenseits postuliert werden, wo es doch keinerlei sinnliche Anschauungen in der Zeit mehr gebe, es sei denn, Kant wolle „auf das indische System der Wiedergeburt, der Seelenwanderung“ zurückkommen (Paulsen 1899, 323 – 324). Als Otfried Höffe 2012 seinen Kommentar zur zweiten Kritik publizierte, sah er offensichtlich genau dieselben Gefahren in Kants Lehre vom höchsten Gut: Diese und die Postulate Unsterblichkeit und Dasein Gottes seien ein widerrechtlich in Kants Morallehre verbliebener „Rest-Eudaimonismus“ (Höffe 2012, 167; 173), den man aus strategischen Gründen verwerfen müsse, um Kants wahre Lehre erhalten

 Noch 1912 charakterisierte Ernst Troeltsch (1912, 967), die Mystik habe „Wahlverwandtschaft zur Autonomie der Wissenschaft und bildet das Asyl für die Religiosität wissenschaftlich gebildeter Schichten“. Dass er sich damit offensichtlich auf du Prel bezog, zeigt seine 1897 publizierte Rezension von Danckelmans Kant als Mystiker?, in der Troeltsch seine Kenntnis von du Prels Text und dessen Selbstbeschreibung als Mystiker anzeigt (Graf/Brandt 2007, 366).

Eduard von Hartmann und die Genese des Neukantianismus

67

zu können³, sowie eine Theorie der Seelenwanderung „aus anderen Regionen und im Abendland von Pythagoras und den Orphikern“ vertreten, wonach „die Seele nach dem Tod in ein anderes Geschöpf übergeht“ (Höffe 2012, 184). Höffes Kritik an den Postulaten zielte letztlich gegen die von ihm sog. „Postmoderne“. Nicht diese, sondern die kantische Lehre (ohne Postulate) habe die alte Metaphysik von Gott, Freiheit und Unsterblichkeit überwunden (Höffe 2012, 33). Auch Paulsen ging es darum, die von ihm bei Kant gesehene Metaphysik klar von der alten „dogmatischen“ zu unterscheiden (Paulsen 1899, 89 – 91). Doch wichtiger noch war es offensichtlich, die Metaphysik Kants von denen abzugrenzen, die bereits, unter Bezug auf dieselben Kanttexte wie Paulsen, Kants Metaphysik als Mystik in Anspruch genommen hatten. Es ist nicht sicher, inwieweit Paulsen du Prels Text kannte. Sicher ist aber, dass Paulsen den o.g. Text von Heinze (1894) kannte (Paulsen 1899, 244), wo dieser explizit darauf einging, dass du Prel mit seiner Wiederveröffentlichung der L1 behauptet habe, „dass Kant die heutige mystische Philosophie vorausgegriffen, ja in seiner Psychologie ein ganzes System der Mystik entworfen habe“ (Heinze 1894, 483). Wichtig ist, hier darauf hinzuweisen, dass 1899 ein Schüler von du Prel, Walter Bormann, einen Text veröffentlichte, in dem die Lehre vom höchsten Gut und die intelligible Welt erstmals explizit als das „Okkulte“ bezeichnet wurde (Bormann 1899, 114). Dieser Text wurde in den Kant-Studien Vaihingers rezensiert. Der Rezensent Fritz Medicus, ein Schüler Vaihingers und ein Lehrer des Theologen Paul Tillich (Graf/Christophersen 2004)⁴, behauptete daraufhin, dass Kant durchaus eine „leise Neigung zum Geisterglauben“ gehabt habe, diese aber eine nur „willkürliche Ausmalung des leeren Begriffs der intelligiblen Welt, ohne wissenschaftlichen Wert“ gewesen sei, genauso wie „Bormanns Identifizierung von Kants intelligibler Welt und dem Reich der Zwecke mit den ‚wissenschaftlichen‘ Erfahrungen des Occultismus“ (Medicus 1900, 335). Eine weitere Schrift Bormanns, in welcher er sich ausdrücklich als „Okkultist“ auf die Seite von „Professor Paulsen“ stellte (Bormann 1900, 84), wurde in den Kant-Studien ebenfalls angezeigt.⁵ Gerade in dieser aufgeladenen Gemengelage äußerte sich Vaihinger zu Paulsens Buch und stellte eine folgenreiche Diagnose: Es seien „in der letzten Zeit mehrere Versuche gemacht worden, gegen den Kritiker Kant den Metaphysiker Kant

 „Die Moralphilosophie kann und muß auch ohne die Postulatenlehre überzeugen, was es dem säkularen religionsskeptischen Zeitgenossen erleichtert, in der philosophischen Ethik Kantianer zu sein“ (Höffe 2012, 178).  Bemerkenswerterweise promovierte Tillich bei Medicus mit einer Arbeit unter dem Titel Mystik und Schuldbewußtsein in Schellings philosophischer Entwicklung (Graf/Christophersen 2004, 53).  Bormanns Text wurde als Sonderabdruck publiziert, was in den Kant-Studien (6) (1901, 348) unter der Rubrik „Neu Eingegangenes“ kurz erwähnt wurde.

68

Hauke Heidenreich

auszuspielen“ und der negativen ersten Kritik eine „eine wirkliche und positive Metaphysik der Vernunft“ entgegen zu stellen. Diese Gegenbewegung sieht Vaihinger in einem aufkeimenden Interesse von „Freunde[n] der Swedenborg’schen Theosophie“ an „Kant’s ‚Vorlesungen über Metaphysik‘“ begründet und „in Paulsens Kantbuch an den Tag […] treten“ (Vaihinger 1900a, 135).Vaihinger zufolge habe Kant durchaus auch in der kritischen Periode, besonders im Hinblick auf das Reich der Zwecke, noch an metaphysische Gegenstände geglaubt. Was aber Kant nur „unter tausend Verklausulierungen“ und einem „kritischen Schleier“ verstecke, stelle Paulsen nun „in das hellste Tageslicht“ (Vaihinger 1900a, 140). Entgegen Paulsens Vorgehen, Kants „Erdichtungen“ (Vaihinger 1900a, 147), die er auch noch in „peinliche Verklausulierungen“ hülle, als das Zentrale der Morallehre auszugeben, müsse man erkennen, dass die Postulate nur „ein ‚Als ob‘ jene Ideen wirklich wären“ seien (Vaihinger 1900a, 154). Dieses wiedererwachte Interesse an Kants Vorlesungen hatte Vaihinger bereits Jahre zuvor explizit mit du Prel in Verbindung gebracht (Vaihinger 1895, 420). Vaihingers Strategie produziert im Kontext der Rezension Medicus’ den Eindruck, Paulsen würde durch seine Metaphysik, die sich zu sehr auf die Lehre vom höchsten Gut und die Postulate konzentriere, selber Spiritismus betreiben, obwohl Paulsen sich sowohl von dem einen als auch dem anderen ausdrücklich distanziert hatte. Vaihingers spätere Philosophie des Als Ob (Vaihinger 1913) wurde also in diesem historischen Kontext explizit als Antwort auf spiritistische Vereinnahmungen der kantischen Postulatenlehre in den Diskurs gesetzt; offensichtlich übrigens unter Bezug auf Langes „Welt der Dichtung“, welche nun offenbar Vaihinger mit seiner Fiktionstheorie gleichsetzt und als Deutung des Übersinnlichen anbietet, die gerade der spiritistischen Mystik entgegengesetzt wird. Die Postulatenlehre galt nun, vor allem unterstützt von Vaihingers Autorität als Professor und Herausgeber der Kant-Studien ⁶, als ein Einfallstor des Spiritismus in Kants Lehre. Und diese Frontstellung, die objektive Realität der Postulate durch eine Fiktionalität zu ersetzen oder jene gleich ganz abzuschaffen, wurde in der Folge von wichtigen Autoren der Zeit erkannt und tradiert (Heinze 1901, 283; Cohen 1910, 369; Cohen 1907, 413 – 415; Bauch 1904a, 110). Es bildete sich ein Kanon, die Postulatenlehre als konstitutives Außen jeder sich als wissenschaftlich behauptender Kantforschung zu behandeln.Vaihingers Fiktionshypothese bildete dabei einen äußerst wichtigen Diskursgegenstand zur Deutung bzw. Verwerfung der Postulate, der bis heute als kanonisch zitiert wird (Gardner 2015, 1820 – 1821).

 Vaihinger popularisierte die Front gegen den Spiritismus als konstitutives Außen jeder wissenschaftlichen Kantforschung in dieser Zeit massiv in den Kant-Studien: Vaihinger (1899; 1900b; 1901a; 1901b; 1902).

Eduard von Hartmann und die Genese des Neukantianismus

69

5 Materialismus, Postulate und Neukantianismus Wir haben bisher gesehen, dass das Wort ‚Neukantianismus‘, das heute zumeist als Kennzeichen oder Beschreibung eines einheitlichen Kanons einer bestimmten philosophischen Richtung gesehen wird, in seiner zeitgenössischen Verwendung zuerst von Hartmann, einem sich als Gegner des Neukantianismus ausgebenden Autors, als Kampfbegriff benutzt wurde, der polemisch eine sich nahe am Materialismus haltende und darum politisch anrüchige Kantforschung performativ beschreiben sollte. Wir haben zudem bemerkt, dass Hartmanns Ablehnung des höchsten Gutes noch auf Widerstand in der philosophischen Forschung stieß, vor allem im Kontext von Hartmanns beabsichtigter Ersetzung der Postulate durch seinen eigenen Pessimismus. In der Diskussion um du Prel, Vaihinger und Paulsen wurde im Diskurs ein Kanon gebildet, der, zeitgleich zur Entstehung der Akademie-Ausgabe, auf der Verwerfung oder Fiktionalisierung der Postulatenlehre begründet wurde, genau als Antwort auf spiritistische Deutungen der Morallehre Kants. Gerade der Vorwurf Vaihingers an Paulsen, aufgrund einer metaphysischen Interpretation angeblich sich selbst dem Spiritismus anzunähern – eine Sicht, die vom Okkultisten Walter Bormann ebenfalls vertreten wurde – führte in gewisser Weise vor Augen, welche Themen, Deutungen und Strategien den Ausschließungsprozeduren der akademischen „Fachphilosophie“ (Ziche 2000, 62) ausgesetzt werden müssen. Das Auffällige besteht nun darin, dass alle diese Debatten um die Postulatenlehre geführt wurden. Die Postulatenlehre wurde paradoxerweise zum zentral diskutierten Segment, das aber aus dem Grund heraus diskutiert werden musste, um es aus dem Diskurs der Wissenschaft auszuschließen. Gerade diese Ausschließungsprozesse aber suchen den Diskurs um Kant um 1900 massiv heim, sie bestimmen die Debatte um die Morallehre und um den Spiritismus, letztlich auch um die Entstehung einer einheitlichen Kantforschung. Und der Kampfbegriff ‚Neukantianismus‘ spielt nun in diesen Debatten eine zentrale Rolle zur Verortung dieser durch akademische Macht konstituierten ‚Fachphilosophie‘. Denn 1899 mischte sich ein weiterer Forscher in die Debatte ein: der bekannte Darwinist Ernst Haeckel, seit 1863 Zoologie-Professor in Jena. Haeckel hatte in diesem Jahr gerade erst seine Welträthsel (Haeckel 1899) veröffentlicht, die sich bis zum Weltkrieg zu einem Bestseller entwickelten. Gerade auch in akademischen Debatten wurde das Buch intensiv diskutiert und dies hatte einen Grund: Direkt im Vorfeld der Diskussion um Paulsen hatte auch Haeckel eine Deutung der Postulatenlehre anzubieten. Dass Haeckels Text geradezu eine Provokation für die sich konstituierende „Fachphilosophie“ darstellte (Ziche 2000, 62), lag daran, dass Haeckel Kant unterstellte, mit dem Abfassen der

70

Hauke Heidenreich

zweiten Kritik sich eines Rückfalls in dualistische und christliche, in Haeckels Sicht also letztlich voraufklärerische Denkmuster schuldig gemacht zu haben. Indem Kant die Lehre vom höchsten Gut formuliert habe, deren Postulate Haeckel als „Großmächte des Mystizismus“, die „drei Haupt-Gespenster“ bezeichnete, sei Kant dem Aberglauben und der Mystik aufgesessen (Haeckel 1899, 107). Äußerst bemerkenswert ist nun Haeckels Beschreibung der Wirkung dieses kantischen Rückfalls: abergläubisch seien dadurch z. B. die „Neokantianer“, die den Rückgang auf Kant predigten und dadurch gerade die mystischen Versatzstücke der kantischen Morallehre in den heutigen Diskurs überführten (Haeckel 1899, 108). Haeckel unterstellt den Neukantianern eine religiös motivierte Sicht, die sich an Kants Postulatenlehre orientiere und letztlich das „höchste Gebiet des Aberglaubens“, den Unsterblichkeits-Glauben, befestige (Haeckel 1899, 219). Zwar ist nach Haeckel jeder religiöse Glaube prinzipiell ein Aberglaube, weil er durch „falsche Phantasie-Dichtungen“ hervorgerufene übernatürliche Erscheinungen annehme (Haeckel 1899, 348), doch gibt es in diesem Zusammenhang einige Abstufungen. Denn die „zielbewußten und rücksichtslosen Angriffe der ultramontanen Kirche auf die Wissenschaft, gestützt auf die Trägheit und Dummheit der Volksmassen“ (Haeckel 1899, 359) seien ungleich schlimmer als andere Formen des Aberglaubens. Dagegen müsse mit wissenschaftlichem und politischem „Kulturkampf“ vorgegangen werden bis zum Sieg von „freie[r] Wissenschaft und freie[r] Lehre“ sowie des „Kultur-Staates“ (Haeckel 1899, 388). Gerade das, was Haeckel in Bezug auf die Postulate Mystik nennt, ist auch hier ein direkter Bezug auf Carl du Prel, dessen „Gespenster-Glauben“ (Haeckel 1899, 353) bzw. Mystik Haeckel, anders als Vaihinger, nicht als Triebkraft einer falschen Kantauslegung, sondern als Effekt einer gewissermaßen konsequenten Auslegung des widersprüchlichen und letztlich abergläubischen Systems von Kants Morallehre in Szene setzt (Haeckel 1899, 353 – 354). Als wiederum Friedrich Paulsen 1900 eine Rezension zu Haeckels Text verfasste, zeigt er sich verwundert, wen Haeckel denn mit der „Schule der Neokantianer“ überhaupt im Auge habe. Ihm selbst sei eine derartige, die sich zentral auf Kants Postulate beziehe, „nicht bekannt; auf die aber, die man vielleicht da und dort so nennt, paßt die Charakteristik wieder nicht: denn grade sie legen viel weniger Gewicht auf die Rettung von Gott, Freiheit und Unsterblichkeit, als auf die Kritik der dogmatischen Metaphysik und die neue Theorie der Erfahrung“ (Paulsen 1900, 59). Die Definition ‚Neukantianismus‘ wird von Haeckel in den Diskurs geworfen, um performativ eine Gruppe Autoren zu beschreiben, die sich einer falschen philosophischen Methode bedienen würden, nämlich einer zu starken Fokussierung auf die Postulatenlehre. Paulsens Buch wurde kurz vor dieser Kontroverse wiederum von Hermann Cohen rezensiert, der Paulsen vorwarf, keine Kenntnis davon haben, was bei Kant mit dem Wort „kritisch“ gemeint

Eduard von Hartmann und die Genese des Neukantianismus

71

sei (Cohen 1899, 610). Paulsen, der Cohen zufolge „eine Schwäche für KantSchriften aus den 1760er Jahren“ habe (Cohen 1899, 611), sei es schlicht entgangen, dass die Essenz der kantischen Philosophie „in der Untersuchung und Entdeckung der synthetischen Grundsätze der mathematischen Naturwissenschaft“ bestehe (Cohen 1899, 612). Trotz Paulsens Abgrenzung gegen das höchste Gut behauptet auch Cohen, dass Paulsen sich zu sehr daran festhalte, denn die Ersetzung von Gott durch die Vernunft, die Kant in Cohens Augen durchgeführt habe, sei eben nicht nur formal, sondern ein inhaltlich zentraler Bestandteil von Kants Morallehre (Cohen 1899, 625). Paulsens Gegenargument gegen Haeckel besteht also in einer eigenen Definition von ‚Neukantianismus‘, den er mit der „Theorie der Erfahrung“ identifiziert, was offensichtlich eine Anspielung ist auf Kants Theorie der Erfahrung, eben das Hauptwerk desjenigen Philosophen, der Paulsen noch vor Vaihinger eine ungebührlich metaphysische Auslegung des höchsten Gutes vorgeworfen hatte: Hermann Cohen. Als Ernst Haeckel im Nachwort der Volksausgabe seines Textes auf diese Kontroverse mit Paulsen zurückkam, identifizierte er gar seine eigene Kantdeutung in den Welträthseln mit dem Neukantianismus, den er nun mit konkreten Namen versah: Max Heinze, Hermann Cohen und Hans Vaihinger (Haeckel 1903, 155 – 157). Die Front gegen metaphysische Kantauslegungen und gegen den Spiritismus, die Haeckel jetzt mit dem Wort ‚Neukantianismus‘ markiert, wird im Gefolge der Kontroverse mit Paulsen nun zu einer positiven Selbstbeschreibung. Und zwar zu einer, die den Neukantianismus als eine konkrete Gegenbewegung zur Mystik du Prels verortet. Die Namen, die Haeckel diesem Neukantianismus nun, abgesehen von sich selber zuordnet, sind auffälligerweise alles Autoren, die die Postulatenlehre radikal ablehnten und umdeuteten. In diesem Kontext wird der Begriff ‚Neukantianismus‘ nun auch mit dem Begriff ‚Kritizismus‘, der, wie gezeigt, bereits vorher eine positive Selbstplatzierung darstellte, identifiziert und mit dem Projekt Hermann Cohens gleichgesetzt, im Fall Vorländers u. a. in Bezug auf Cohens Ethik des reinen Willens, in der dieser eine scharfe Kritik an den Postulaten der praktischen Vernunft äußerte (Vorländer 1908, 424; 427; Heinze 1902, 215; Adickes 1904, 1).

6 Eduard von Hartmann in der Debatte um das höchste Gut Im Kontext dieser Debatte meldete sich auch Eduard von Hartmann mit einem Artikel in den Kant-Studien zu Wort. Kant sei zwar kein absoluter oder metaphy-

72

Hauke Heidenreich

sischer, aber in Bezug auf jede Art von Glückseligkeitslehre zu Recht Pessimist und als solcher eben Vater des Pessimismus gewesen. Die Verschmelzung vom teleologischen und eudämonologischen Pessimismus zum „absoluten metaphysischen Pessimismus“ sei schließlich, so Hartmann in aller Bescheidenheit, sein eigenes Verdienst, das sich nicht bei Kant finde (Hartmann 1901, 22). Das Problem befinde sich dort, wo Kant – entgegen der von Hartmann behaupteten ‚Essenz‘ der kritischen Philosophie, die sich in auffälliger Nähe zu Haeckels und Paulsens Beschreibung „in Bezug auf das Jenseits zu bescheiden und jedes Urteils zu enthalten“ habe – „sittliche Postulate“ und die Lehre vom höchsten Gut dazu benutze, „um eudämonologische Werturteile über Unerkennbares zu erschleichen“ und „den fürs Diesseits verworfenen Eudämonismus fürs Jenseits mittelbar wieder einzuführen“. Die Metaphysik-Vorlesungen würden zudem zeigen, dass Kant „bis in die 90er Jahre“, also bis weit in die kritische Periode hinein, an der „Unräumlichkeit und Unkörperlichkeit der Seelen im Jenseits“ festgehalten habe (Hartmann 1901, 27). Weiterhin bestehe die ethische Inkonsequenz darin, dass Kant die „Schrecklichkeit veränderungslosen Fortbestandes und der Unzufriedenheit über das noch unerreichte Ziel“ im Jenseits wissentlich in Kauf nehme: „Kant wäre also besser bei seiner Ansicht v. J. 1766 stehen geblieben, wonach das Jenseits auch die Entfernung der Zeitalter aufhebt. Kant meint, wenn er später, und sei es nach Jahrtausenden, doch sterben solle, so wolle er lieber bald sterben, als noch länger die Komödie mit ansehen (‚Vorlesungen über die Metaphysik‘, ed. Pölitz, S. 243)“ (Hartmann 1901, 28). Neben diesen von Hartmann behaupteten Irrtümern stünde aber noch Kants ethischer Idealismus, der wiederum von der Postulatenlehre, die ja eigentlich nur eine inkonsequente Position in Kants zweiter Kritik darstelle, völlig unabhängig sei. Kants Sinn sei dann richtig erfasst, „wenn ich den eudämonologischen Pessimismus ein Postulat des sittlichen Bewusstseins nenne, weil er für dieses unentbehrliche Voraussetzung einer eudämonistisch ungetrübten Reinheit des ethischen Idealismus ist“. Diese Positionen, mit denen sich Hartmann in Vaihingers Zeitschrift nun an der Seite von Vaihingers Kritik der Postulate positioniert, brachten ihm denn auch Zustimmung ein durch niemand anderen als Bruno Bauch, der ab 1904 die Herausgebertätigkeit der Kant-Studien übernahm. Nachdem Bauch bereits 1903 die Als-Ob-Deutung der kantischen Postulate Vaihingers anerkannt hatte (Bauch 1903, 20), enthüllt Bauch 1904, warum diese Deutung angemessen sei. Denn mithilfe der Postulate würde Kant widerrechtlich implizite Glückseligkeit als Grundlage der Moral konstruieren, daher müsse mithilfe der Vaihingerschen Deutung des Als Ob die rein „menschlich-psychologische Bedeutung“ des höchsten Gutes klar herausgestellt werden (Bauch 1904a, 110). In einem weiteren Text dieses Jahres bezieht sich Bauch nun explizit auf Hartmann und dessen Text in den Kant-Studien: Kant habe nämlich nicht die Neigungen und das Glückse-

Eduard von Hartmann und die Genese des Neukantianismus

73

ligkeitsstreben als verwerflich verdammt, sondern nur der sittlichen Bestimmung untergeordnet und den geringen sittlichen Wert des Eudämonismus gezeigt, eben darin, dass das eudämonistische Ziel, höchste Glückseligkeit, für menschliches Vermögen unerreichbar sei. Genau deswegen habe Hartmann nun Recht, Kant „irdischer eudämonologischer Pessimist“ zu nennen und wiederum den metaphysischen Pessimismus von Kant fernzuhalten (Bauch 1904b, 208).

7 Fazit: Systematische Einheit als Effekt historischer Differenz Die kontrovers geführten Debatten um das kantische höchste Gut und die Postulate der praktischen Vernunft bildeten genau den Kontext für Bauchs Positionierung im Diskurs. Zur Begründung der Als-Ob-Theorie, welche nun die Postulate ‚richtig‘ deuten soll, bezieht sich aber Bauch nun nicht auf Vaihingers Front gegen Paulsen und du Prel, sondern dezidiert auf Hartmanns Position, die auch Paulsen teilte, nämlich des durch Postulate widerrechtlich in Kants Morallehre eingeschmuggelten Eudämonismus. Offensichtlich erkennt Bauch Hartmann gerade in dieser Gemengelage als zentralen Autor der neukantianischen Debatten. Hartmanns bereits in den 1870er Jahren gegen Vaihinger geäußerte und seinerzeit vielkritisierte Forderung nach Abschaffung des höchsten Gutes und dessen Ersetzung durch ein sittliches Postulat des Pessimismus wird nun, im Gefolge der Spiritismus-Kontroverse Vaihinger / du Prel / Paulsen, offensichtlich zur Signatur dessen, was in dieser Zeit nun positiv gedeutet als ‚Neukantianismus‘ beschrieben wird. Der Kampfbegriff ‚Neukantianismus‘ wird so, vor allem durch Haeckels Vorstoß, zu einer positiven Selbstbeschreibung umgedeutet. Die Einheit des Neukantianismus entsteht also unter Verweis auf die Ablehnung der Postulate und der Tradierung der Vaihingerschen Spiritismus-Front unter Berufung auf Hartmanns Begründung. Es ist hierin auffällig, dass mit Haeckel und Hartmann zwei Autoren zentral in die Debatten um die Existenz eines einheitlichen Neukantianismus involviert sind, die traditionell nicht mit dem Neukantianismus in Verbindung gebracht bzw. dezidiert von diesem abgegrenzt werden. Ihre Polemiken, die einen einheitlichen Neukantianismus im Diskurs beschwören, waren wichtige Referenzpunkte der darauffolgenden Diskussionen, in denen Vaihinger das früher polemisch als neukantianisch beschriebene Konzept Langes von der „Welt der Dichtung“ nun als Fiktionsthese auf die Deutung der Postulatenlehre überträgt und hierbei offensichtlich mit Hartmann auch einen Gegner Langes auf seiner Seite hat. Hartmann hatte Lange und Vaihinger in den 1870er Jahren noch

74

Hauke Heidenreich

verkappten Sozialismus unterstellt, jetzt aber gelten ‚neukantianische‘ Deutungen der Lehre vom höchsten Gut als Spiritismus-Abwehr. Die Wirkmächtigkeit dieser Transformation, um nicht zu sagen: dieser Entstehung eines neukantianischen Diskurses um 1900 sollte nicht unterschätzt werden. Otfried Höffe, 2012 einem durch die Postulatenlehre verursachten „RestEudaimonismus“ in der zweiten Kritik auf der Spur, bezieht sich zur Begründung seiner Verwerfung der Postulate nämlich auf niemand anderen als Bauch (Höffe 2012, 212). Dass Sebastian Gardner die Als-Ob-Deutung der Postulate noch 2015 ausdrücklich akzeptierte, wurde bereits erwähnt. Es zeigt sich also in dem bisher Gesagten, dass die heutige Deutung der Postulatenlehre, durch Nachweis der Rezeptionsvorgänge, auf einer umgedeuteten Spiritismus-Front beruht, in der Eduard von Hartmanns Begründung seiner Kritik an den Postulaten als zentrales Diskurssegment platziert wurde, das heute noch rezipiert wird. In gewisser Weise scheint das Verständnis dessen, was heute als ‚Neukantianismus‘ bezeichnet wird, und der gegenwärtige Kanon zur Deutung der Postulate auf Hartmanns, Vaihingers und Haeckels Kritik am höchsten Gut zu rekurrieren, die die Autoren im Zusammenhang einer polemischen Festschreibung des Begriffes ‚Neukantianismus‘ zuallererst vornahmen und dann in den Diskurs um 1900 implementierten. Offensichtlich setzt sich Hartmanns Sicht auf das höchste Gut um 1900 im durch die Kant-Studien vorgegebenen kanonischen Rahmen der Kantforschung durch und wird nun zur Eigenverortung des Neukantianismus gegen den Spiritismus. Dass der Neukantianismus eine Gegenbewegung zum Atheismus und zur „Mystik“ sei, wurde im Übrigen auch zeitgenössisch erkannt (Troeltsch 1895, 198). Wolfs These der zentralen Verwobenheit Hartmanns in die Entstehung des Neukantianismus konnte so verifiziert werden. Sie ist zu ergänzen dadurch, dass Hartmanns Kritik über die Rezeption Bauchs anscheinend sogar in der gegenwärtigen Kantforschung zur Kritik der Postulate herangezogen wird. Im Rahmen dieses Artikels konnte nur auf wenige Verwobenheiten eingegangen werden. Hartmanns zentrale Verwicklung in spiritistische Debatten, zum Beispiel seine konkreten Vorschläge für den ‚wissenschaftlichen‘ Ablauf einer Geisterbeschwörung und seine diesbezüglich gestellte Forderung nach Geld von der Regierung (Hartmann 1885, 10; 15), hatten ihm bereits zeitgenössisch den Vorwurf eingebracht, eigentlich selber Spiritist zu sein (Wundt 1906, 372). Gerade auch Hartmanns Briefwechsel mit Carl du Prel (Kaiser 2008, 37– 38), auf dessen Modell er sich immer wieder kritisch bezog (Hartmann 1900a, 563 – 570; 1900b, 281), platziert ihn, wie auch Vaihinger, als zentralen Akteur in der Popularisierung von du Prels Spiritismus.

Eduard von Hartmann und die Genese des Neukantianismus

75

Es konnte in dieser kurzen Studie gezeigt werden, dass die Entstehung einheitlicher philosophischer Konzepte nicht zu trennen ist vom historischen Kontext und innerhalb desselben platziert wird. Systeme zu Kants Philosophie werden unter Ausschluss konkurrierender Deutungen generiert, in diesem Falle: des Spiritismus, wobei sowohl das eigene System wie auch der Ausschluss mit Namen versehen werden. Diese Namen sind also von den historischen Praktiken ihrer Konstituierung ebenso wenig zu trennen wie die Ausschließungsprozeduren von der Genese eines Systems. Gerade diese Ausschließungsprozeduren zeigen die Spuren der Historizität eines Systems, die das System selbst zu verschleiern trachtet. Im Falle Vaihingers bedeutet das: es entsteht ein philosophisches System (Philosophie des Als Ob) zur Deutung der Postulate, das in einer dezidierten Front gegen den Spiritismus installiert wurde. Die Platzierung gegen eine konkurrierende Annahme geschieht also zeitlich vor der Konstituierung des Systems, die historische Differenz ist die Bedingung der Behauptung und Konstituierung von Einheit! Gerade die Debatte um Haeckel und Hartmann zeigt zudem, dass ausgeschlossene Deutungen der Postulate nicht in einen diskursfreien Raum abgedrängt werden, sondern subkutan die Entstehung eines Systems, sozusagen als Negativfolie, konstitutiv begleiten. Es trifft also nur dann zu, eine negative Rezeption einer Position als radikalen Bruch mit ebenjener zu deuten, wenn man eine stabile Autoridentität im Diskurs nachweisen könnte. Wie aber gezeigt, erhalten sich Spuren der verdrängten Positionen in den konstruierten Systemen. Die Verdrängung passiert qua Polemik, die als Basis der nun folgenden Systemkonstruktion gesehen werden kann und gerade die Existenz einer starren Autorposition massiv infrage stellt. Autorpositionen und Einheiten entstehen, entsprechend den Beobachtungen in dieser Arbeit, dagegen als Effekte des Umgangs mit historischer Differenz. In Bezug auf Hartmanns Neukantianismus-Definition konnte zudem beschrieben werden, wie Rezeptionen historischer Sachverhalte sich zur Chronologie verhalten. Die Rezeptionsvorgänge sind nicht nur durch eine reine Vorgängigkeit bestimmter Themen strukturiert, sondern die jeweils im Diskurs behauptete Machtposition im akademischen Feld bestimmt über die Rezeption. Inhalte, die als Spuren in einem System vorhanden sind, werden durch die Macht des Sprechenden, wie z. B. durch Vaihinger als Herausgeber der Kant-Studien, beschworen und reifiziert. Dadurch entstehen Begriffe, die dann in der Folge tradiert und unter Bezug auf wissenschaftliche Autorität abgesichert werden.

76

Hauke Heidenreich

Literaturverzeichnis Adickes, Erich (1904): Auf wem ruht Kants Geist? Eine Säkularbetrachtung, in: Archiv für systematische Philosophie 10, pp. 1 – 19. Bauch, Bruno (1903): Vom Prinzip der Moral bei Kant. Einladungsschrift zu der am Donnerstag den 19. November 1903 vormittags 11 Uhr mit Genehmigung der Hohen Philosophischen Fakultät der Vereinigten Friedrichs-Universität Halle-Wittenberg in der Aula zu haltenden Antrittsvorlesung Über Goethes Weltanschauung, Halle. Bauch, Bruno (1904a): Luther und Kant, Berlin. Bauch, Bruno (1904b): Die Persönlichkeit Kants, in: Kant-Studien 9, pp. 196 – 210. Beiser, Frederick C. (2017): The Genesis of Neo-Kantianism 1796 – 1880, New York. Bergunder, Michael (2016): ‚Religion‘ and ‚Science‘ within a Global Religious History, in: Aries. Journal for the Study of Western Esotericism 16, pp. 86 – 141. Bormann, Walter (1899): Kantsche Ethik und Okkultismus, in: Gesellschaft für wissenschaftliche Psychologie (ed.): Beiträge zur Grenzwissenschaft. Ihrem Ehrenpräsidenten Dr. Carl du Prel gewidmet, München Jena, pp. 107 – 139. Bormann, Walter (1900): Zwei Stimmen über Magie vor dem Forum der Wissenschaft, in: Die Übersinnliche Welt. Mitteilungen aus dem Gebiete des Okkultismus. Organ der Wissenschaftlichen Vereinigung Sphinx in Berlin 8, pp. 27 – 37, 55 – 63, 82 – 92. Clark, Christopher (2006): Preußen. Aufstieg und Niedergang 1600 – 1947, Bonn. Cohen, Hermann (1899): Ein Buch über Kant, in: Die Nation. Wochenschrift für Politik, Volkswirtschaft und Literatur 16, pp. 609 – 613, 623 – 626. Cohen, Hermann (1907): Ethik des reinen Willens, 2nd ed., Berlin. Cohen, Hermann (1910): Kants Begründung der Ethik. Nebst ihren Anwendungen auf Recht, Religion und Geschichte, Berlin. Flach, Werner (2012): Zur Neubewertung des Neukantianismus, in: C. Krijnen, A. J. Noras (ed.): Marburg versus Südwestdeutsch. Philosophische Differenzen zwischen den beiden Hauptschulen des Neukantianismus, Würzburg, pp. 9 – 40. Gabriel, Gottfried (2007): Windelband und die Diskussion um die Kantischen Urteilsformen, in: M. Heinz, C. Krijnen (ed.): Kant im Neukantianismus. Fortschritt oder Rückschritt, Würzburg, pp. 91 – 108. Gardner, Sebastian (2015): Postulate der reinen praktischen Vernunft, in: M. Willaschek, J. Stolzenberg, G. Mohr, S. Bacin (ed.): Kant-Lexikon. Band II: Habitus – Rührung, Berlin Boston, pp. 1814 – 1821. Gerhard, Carl (1878): Rezension von Hartmanns Neukantianismus, Schopenhauerianismus und Hegelianismus in ihrer Stellung zu den philosophischen Aufgaben der Gegenwart, in: Philosophische Monatshefte 14, pp. 99 – 105. Graf, Friedrich Wilhelm, Brandt, Dina (ed.) (2007): Ernst Troeltsch. Rezensionen und Kritiken (1894 – 1900), Berlin New York. Graf, Friedrich Wilhelm, Christophersen, Alf (2004): Neukantianismus, Fichte- und Schellingrenaissance. Paul Tillich und sein philosophischer Lehrer Fritz Medicus, in: Zeitschrift für Neuere Theologiegeschichte 11, pp. 52 – 78. Haeckel, Ernst (1899): Die Welträthsel. Gemeinverständliche Studien über Monistische Philosophie, Bonn.

Eduard von Hartmann und die Genese des Neukantianismus

77

Haeckel, Ernst (1903): Die Welträthsel. Gemeinverständliche Studien über monistische Philosophie. Volksausgabe mit einem Nachworte: Das Glaubensbekenntnis der Reinen Vernunft, Bonn. Hartmann, Eduard von (1877): Neukantianismus, Schopenhauerianismus und Hegelianismus in ihrer Stellung zu den philosophischen Aufgaben der Gegenwart, Berlin. Hartmann, Eduard von (1880): Zur Geschichte und Begründung des Pessimismus, Berlin. Hartmann, Eduard von (1885): Der Spiritismus, Leipzig/Berlin Hartmann, Eduard von (1893): Rezension zum zweiten Band von Vaihingers Commentar zu Kants Kritik der reinen Vernunft, in: Preußische Jahrbücher 71, pp. 340 – 346. Hartmann, Eduard von (1900a): Geschichte der Metaphysik. Zweiter Teil: Nach Kant, Leipzig Hartmann, Eduard von (1900b): Zum Begriff des Unbewussten, in: Archiv für systematische Philosophie 6, pp. 273 – 290. Hartmann, Eduard von (1901): Kant und der Pessimismus, in: Kant-Studien 5, pp. 21 – 29. Heidenreich, Hauke (2019): Die Empirie des Übersinnlichen – C. G. Jungs Konzept des kollektiven Unbewussten als Umdeutung Kants zwischen Okkultismus, Religion und Parapsychologie, in: Zeitschrift für Religions- und Geistesgeschichte 71, pp. 41 – 62. Heinze, Max (1894): Vorlesungen Kants über Metaphysik aus drei Semestern, Leipzig, Heinze, Max (ed.) (1901): Ueberweg’s Grundriss der Geschichte der Philosophie. Dritter Theil: Die Neuzeit bis zum Ende des achtzehnten Jahrhunderts, 9th ed., Berlin. Heinze, Max (ed.) (1902): Ueberweg’s Grundriss der Geschichte der Philosophie. Vierter Teil: Das neunzehnte Jahrhundert, 9th ed., Berlin. Höffe, Otfried (2012): Kants Kritik der praktischen Vernunft. Eine Philosophie der Freiheit, München. Kaiser, Tomas H. (2008): Zwischen Philosophie und Spiritismus. Annäherungen an Leben und Werk von Carl du Prel, Saarbrücken. Köhnke, Klaus Christian (1993): Entstehung und Aufstieg des Neukantianismus, 2nd ed., Frankfurt (Main). Kant, Immanuel (1781/1788): Kritik der praktischen Vernunft, in: I. Kant: Gesammelte Werke, ed. königlich preußische (später deutsche) Akademie der Wissenschaften, Berlin 1900 f. Krijnen, Christian (2012): Denken als Ursprung, in: C. Krijnen, A. J. Noras (ed.): Marburg versus Südwestdeutsch. Philosophische Differenzen zwischen den beiden Hauptschulen des Neukantianismus, Würzburg, pp. 63 – 95. Krijnen, Christian & Noras, Andrzej J. (ed.) (2012): Marburg versus Südwestdeutsch. Philosophische Differenzen zwischen den beiden Hauptschulen des Neukantianismus, Würzburg. Lange, Friedrich Albert (1877): Geschichte des Materialismus und Kritik seiner Bedeutung in der Gegenwart. Zweites Buch: Geschichte des Materialismus seit Kant, 3rd ed., Leipzig. Lasson, Adolf (1877): Rezension von Vaihingers Hartmann, Dühring und Lange: Zur Geschichte der Deutschen Philosophie im XIX. Jahrhundert, in: Philosophische Monatshefte 13, pp. 218 – 231. Liebmann, Otto (1865): Kant und die Epigonen. Eine kritische Abhandlung, Stuttgart. Luft, Sebastian (2015): The Neo-Kantian Reader, New York. Medicus, Fritz (1900): Rezension von Walter Bormanns Kantsche Ethik, in: Kant-Studien 4, pp. 333 – 335. Orth, Ernst Wolfgang (1994): Die Einheit des Neukantianismus, in: H. Holzhey, E. W. Orth (ed.): Neukantianismus. Perspektiven und Probleme, Würzburg, pp. 13 – 30.

78

Hauke Heidenreich

Pascher, Manfred (1997): Einführung in den Neukantianismus. Kontext – Grundpositionen – Praktische Philosophie, München. Paulsen, Friedrich (1899): Immanuel Kant. Sein Leben und seine Lehre, 2nd & 3rd ed., Stuttgart. Paulsen, Friedrich (1900): Ernst Haeckel als Philosoph, in: Preußische Jahrbücher 101, pp. 29 – 72. Prel, Carl du (1964): Immanuel Kants Vorlesungen über Psychologie. Mit einer Einleitung: Kants mystische Weltanschauung, Stuttgart. Prunea-Bretonnet, Tinca (2018): From Mysticism to Metaphysics. An ‚Irrational‘ Critic and His Influence on Kant’s Reception at the End of the 19th Century, in: A. Falduto, H. F. Klemme (ed.): Kant und seine Kritiker – Kant and His Critics, Hildesheim, pp. 243 – 255. Renz, Ursula (2002): Die Rationalität der Kultur. Zur Kulturphilosophie und ihrer transzendentalen Begründung bei Cohen, Natorp und Cassirer, Hamburg. Schaarschmidt, Carl (1881): Rezension von Hartmanns Zur Geschichte und Begründung des Pessimismus, in: Philosophische Monatshefte 17, pp. 287 – 292. Sieg, Ulrich (1994): Aufstieg und Niedergang des Marburger Neukantianismus. Die Geschichte einer philosophischen Schulgemeinschaft, Würzburg. Sieg, Ulrich (2016): Gerechtigkeitssinn und Empörung. Die Marburger Schule des Neukantianismus, Marburg. Simon, Gerd (2014): Leben und Wirken Hans Vaihingers, in: M. Neuber (ed.): Fiktion und Fiktionalismus. Beiträge zu Hans Vaihingers ‚Philosophie des Als Ob‘, Würzburg, pp. 21 – 42. Sommer, Hugo (1880): Der Pessimismus und die Bedeutung des höchsten Guts, in: Preußische Jahrbücher 46, pp. 480 – 493. Stengel, Friedemann (2011): Aufklärung bis zum Himmel. Emanuel Swedenborg im Kontext der Theologie und Philosophie des 18. Jahrhunderts, Tübingen. Stengel, Friedemann (2015): Emanuel Swedenborg, in: M. Willaschek, J. Stolzenberg, G. Mohr, S. Bacin (ed.): Kant-Lexikon. Band III: Sache – Zyniker, Zynismus, Berlin Boston, pp. 2220 – 2222. Stolzenberg, Jürgen (1995): Ursprung und System. Probleme der Begründung systematischer Philosophie im Werk Hermann Cohens, Paul Natorps und beim frühen Martin Heidegger, Göttingen. Teichmüller, Gustav (1877): Wahrheitsgetreuer Bericht über meine Reise in den Himmel, verfaßt von Immanuel Kant, Gotha. Troeltsch, Ernst (1895): Atheistische Ethik, in: Preußische Jahrbücher 82, pp. 193 – 217. Troeltsch, Ernst (1912): Die Soziallehren der christlichen Kirchen und Gruppen, Tübingen. Vaihinger, Hans (1876): Hartmann, Dühring und Lange. Zur Geschichte der Philosophie im XIX. Jahrhundert. Ein kritischer Essay, Iserlohn. Vaihinger, Hans (1881): Commentar zu Kants Kritik der reinen Vernunft. Zum hundertjährigen Jubiläum derselben. Band I, Stuttgart. Vaihinger, Hans (1891): Rezension zu du Prels Kants Vorlesungen über Psychologie, in: Archiv für Geschichte der Philosophie 4, pp. 721 – 723. Vaihinger, Hans (1892): Commentar zu Kants Kritik der reinen Vernunft. Zum hundertjährigen Jubiläum derselben. Band II, Stuttgart/Berlin/Leipzig. Vaihinger, Hans (1895): Rezension zu Heinzes Vorlesungen Kants über Metaphysik aus drei Semestern, in: Archiv für Geschichte der Philosophie 8, pp. 420 – 428.

Eduard von Hartmann und die Genese des Neukantianismus

79

Vaihinger, Hans (1899): Rezension von Dippels Der neuere Spiritismus, 2. Auflage, in: Kant-Studien 3, p. 362. Vaihinger, Hans (1900a): Kant – ein Metaphysiker?, in: B. Erdmann et al (ed.): Philosophische Abhandlungen. Christoph Sigwart zu seinem siebzigsten Geburtstage 28. März 1900, Tübingen Freiburg i. Br. Leipzig, pp. 133 – 158. Vaihinger, Hans (1900b): Mitteilungen zu Kant und Swedenborg, in: Kant-Studien 4, p. 134. Vaihinger, Hans (1901a): Bibliografische Notiz zum Erscheinen von du Prels Ausgewählten Schriften, Band I, in: Kant-Studien 5, p. 486. Vaihinger, Hans (1901b): Bibliografische Notiz zu du Prels Der Tod, das Jenseits, das Leben im Jenseits, in: Kant-Studien 6, p. 336. Vaihinger, Hans (1902): Aus zwei Festschriften. Beiträge zum Verständnis der Analytik und der Dialektik in der Krit. d. r. V., in: Kant-Studien 7, pp. 99 – 119. Vaihinger, Hans (1913): Die Philosophie des Als Ob. System der theoretischen, praktischen und religiösen Fiktionen der Menschheit auf Grund eines idealistischen Positivismus. Mit einem Anhang über Kant und Nietzsche, 2nd ed., Leipzig. Volkelt, Johannes (1881): Eduard von Hartmann, in: Nord und Süd 18, pp. 54 – 73. Vorländer, Karl (1908): Geschichte der Philosophie. Zweiter Band: Philosophie der Neuzeit, 2nd ed., Leipzig. Wolf, Jean-Claude (2006): Eduard von Hartmann. Ein Philosoph der Gründerzeit, Würzburg. Wundt, Wilhelm (1906): Essays, 2nd ed., Leipzig. Zeidler, Kurt Walter (2009): Negation, Andersheit und Unendlichkeit im realistischen Kritizismus, in: P. Fiorato (ed.): Verneinung, Andersheit und Unendlichkeit im Neukantianismus, Würzburg, pp. 125 – 136. Zeidler, Kurt Walter (2014): Vorwort, in: K. W. Zeidler, C. Krijnen (ed.): Wissenschaftsphilosophie im Neukantianismus. Ansätze – Kontroversen – Wirkungen, Würzburg, pp. 5 – 7. Ziche, Paul (2000): Die „Scham“ der Philosophen und der „Hochmut der Fachgelehrsamkeit“. Zur philosophischen Diskussion von Haeckels Monismus, in: P. Ziche (ed.): Monismus um 1900. Wissenschaftskultur und Weltanschauung, Berlin, pp. 61 – 79.

Christian Krijnen

The Problem of Schematism in Kant and its Transformation in Southwest Neo-Kantianism Abstract: The meaning and validity of Kant’s Kant’s doctrine of schematism remains contested until today. In neo-Kantianism and post-War transcendental philosophy, Kant’s schematism of the pure concepts of understanding is transformed drastically. Kant’s thesis of heterogeneity is overcome by taking it back into the internal relationships of the structure of cognition. The spontaneity of thought, performing schematizations, is retained, but Kant’s project of conceiving of the foundations of knowledge in the fashion of a theory of apperception of the I as well as the externality of the given and the determination of the given that goes along with it is sublated by an objective order of validity-noematic constitution and regulation. Kant’s doctrine of schematism, then, shows to be methodology.

Introduction In two recent dissertations on Kant,¹ a problem that has always occurred in discussions about Kant plays a major role. Heidegger (1951, 1962) even managed to turn the issue into a vital question: the problem of schematism. The broad spectrum of interpretations of Kant’s doctrine of schematism is rather surprising taking into account that Kant himself labelled the chapter on schematism as one of the most important parts of the Critique of Pure Reason. The interpretations reach from defending that the doctrine of schematism is superfluous because of the transcendental deduction of pure concepts of understanding up to supplying this deduction originally.²

 See Birrer (2017) and Bunte (2016). Both authors also sketch the state of research (Birrer 2017, ch. 1; Bunte 2016, 51 ff.). For the reception of Kant’s doctrine also see: Allison (2004, 202 ff.); Düsing (1995); Höffe (2003, 152 ff).  Against this background, see the interpretation of Caimi (2015). For Birrer (2017, 5 ff.), however, Caimi presents a “revisionist reading” that marginalizes the role of sensibility in Kant’s framework of the heterogenous sources sensibility and understanding. In contrast, Birrer holds that the relationship between sensibility and understanding is continuously redetermined in the course of Kant’s line of argument while at the same time respecting their independence. https://doi.org/10.1515/kantyb-2020-0004

82

Christian Krijnen

It cannot be overlooked that in the discussion about Kant’s schematism the attempts of Southwest Neo-Kantianism and post-war transcendental philosophers influenced by it, like Hans Wagner and Werner Flach, to relativize Kant’s heterogeneity of the two sources of knowledge – sensibility and understanding – by means of a foundational unity, are not dealt with as such.³ Let alone that the approach of Hegel to overcoming Kant’s dualistic conception with a philosophy of self-mediation of the concept is discussed as a truly intellectual challenge. Yet neither in Kantian transcendental philosophy nor in Hegel’s speculative idealism does Kant’s schematism play the important role that it should play according to Kant himself. Rather, they aim to solve Kant’s problem of schematism – by not letting it arise in the first place. In what follows I shall show this with regard to Southwest Neo-Kantianism. In the final section of this essay, post-war transcendental philosophy will enter the stage to press forward the continuity of the results of the Neo-Kantian appropriation of Kant. It is my thesis that Kantian transcendental philosophy offers relief from a weak point of Kant’s transcendental idealism. Although I shall not go into the issue of whether Kantian transcendental philosophy can cope sufficiently with Hegel’s challenge of the concept, it becomes clear that the discussion of Kant’s schematism and its relevance for contemporary philosophy requires perspectives that go beyond the usual lines of interpretation.

(Kant’s Kritik der reinen Vernunft is quoted from the Cambridge Edition Critique of Pure Reason (Kant 1998); all other translations of German sources are mine.)  Bunte (2017) is an exception. Birrer, like Bunte, not interested in the problem of schematism from a merely historical perspective, does not deal with the mentioned tradition but with an early Neo-Kantian like Riehl and in particular Marburgian Neo-Kantians like Cohen and Cassirer (Birrer 2017, 12 ff.). The transcendental philosophy of Wagner and Flach, which is developed in close discussion with Kant, plays no role in his presentation of “contemporary Kant scholarschip” (Birrer 2017, 19 ff.). For him, the point of the Neo-Kantians is to demonstrate that an “intellectual meaning of conceptual necessity” is already part of “sensible intuitive representations”. In contrast to this view, the present article will, among others, show that for the discussed tradition of transcendental philosophy, the dualism of intuition and concept that Birrer presents has already transcended the original togetherness of these moments in the cognitive relationship. In this respect, Bunte (2016) seems more profound. As a consequence of his approach, Kant’s concept of the thing-in-itself turns into a final problem (Bunte 2016, ch. 3.3). After having explained its conceivability (Bunte 2016, ch. 3.3.1), the possibility of its experience turns out to be a “residual problem” (Bunte 2016, 303), or, to be more precise, an “unsolvable” problem (Bunte 2016, 310). This leads to a remarkable tension between the relevance of Kant, asserted continuously by Bunte, and the unsolvable residual problem he diagnosed. Apparently, the presuppositions of Kant’s dualism of knowledge sources are in need of a more in-depth (‘critical’) investigation.

Kant’s Schematism in Southwest Neo-Kantianism

83

1 The Problem for which Kant’s Schematism of Pure Concepts is a Solution There is no disagreement about the fact that Kant’s chapter on schematism deals with a problem of mediation that results from his architectonic of theoretical knowledge. The heterogeneity of intuition (repraesentatio singularis, Einzelvorstellung) and concept (repraesentatio generalis, Allgemeinvorstellung) needs to be brought into a relationship concerning the concrete determination of objects. Schematism generally concerns the relationship between an undetermined content (Inhalt) of cognition and its determining form (hence, not the traditional subsumption of universals and individuals): they have to come together in an adequate way. For this, a schema is needed, be it a schema of pure concepts of sensibility, of empirical concepts, of pure concepts of understanding, or even a schema of the idea.⁴ Schemata function here as rules for the application of rules, or, to be more precise, as conditions of applications of rules. In this respect, for Kant the issue of schematism addresses a necessary principle of mediation regarding form and content. Otherwise, the principle would remain a mere function without being capable of representing an object: There would be no object as a formed content. Without schemata, as Kant says, “the actions of the understanding” as well as the “unity of reason” that results from it, would remain “undetermined” in itself (CPR A 664/B 692 f.). Although the issue of schematism has a general meaning that goes beyond the specific meaning of the mentioned types of schemata, in view of the transformation in subsequent German idealism, I shall focus on the foundational level of the relationship between intuition and category (pure concept of understanding). This focus also concerns the basic structure of the cognitive relationship as conceived of by Kant – a relationship between intuition and concept, form and content. For Kant too, the transcendental schemata are decisive as the conditions of the possibility of knowledge. What problem does the schematism of pure concepts of understanding intend to solve? We could say a problem of concretization. The chapter of schematism is not about justifying the legitimate use of categories as principles of the determinacy of objects, as is the case in the chapter on transcendental deduc-

 Although in the chapter on schematism Kant does not mention the schema of the idea, the first three schemata mentioned do not exhaust Kant’s doctrine of schematism (pace Bunte (2016, 58 note 219)). It is often not taken into account that Kant also integrates a schema of reason (see e. g. Düsing (1995, 50 ff.) or Höffe (2003, 154 f.)). For its execution, the idea too requires a “schema” (CPR B 860). On the relevance of the schema of the idea for the development of a system of philosophy see Krijnen (2008b, ch. 6.1.2).

84

Christian Krijnen

tion; rather, it has to make plausible how categories can be applied to objects (CPR A 138/B 177). The categorical synthesis of understanding does not make an object, but the object is given by sensibility as an object that is in need of determination. The universal form of sensibility is time. Objects appear in time. The category, as Kant holds, is with respect to the object given in time, on which it is applied, totally “heterogenous” (CPR A 138/B 177) or “un-homogenous” (CPR A 137/B 176). Yet “subsumption” of an object under a concept presupposes homogeneity. How can categories, pure logical, timeless concepts, be applied to objects given in time? This requires a mediating “third” factor, i. e. a specific “representation” that should be pure and both “sensible” and “intellectual” (a priori governed): it should be homogeneous with the category and the appearance (CPR A 138/B 177). In the background of the heterogeneity that needs to be overcome by schematism stands the heterogeneity of intuition and concept or sensibility and understanding, continuously emphasized by Kant. As Kant writes in the chapter on schematism, compared to empirical concepts or even sensible intuitions, pure concepts of understanding are “entirely non-homogenous” (CPR A 137/B 176). From this, the heterogeneity of category and appearance results; the latter presupposes the former. The “transcendental determination of time” fulfills, according to Kant, the required condition of homogeneity. Like the category it is “general” and based upon a “rule a priori,” and it is also homogenous with the appearance as “time” is contained in every empirical representation of a manifoldness; the transcendental determination of time is for Kant the schema of the category (CPR A 138/B 177 f.). Herewith concrete knowledge of an object is achieved. Concrete knowledge of an object is both pure synthesis and an immediate relation to an object. Contingency is conceived of as a categorical founded unity, that is to say, concrete determination of an object. In this respect, the application of categories on appearances as objects of cognition is clarified. Kant’s thesis of a complete heterogeneity of category and appearance, involving the need for a mediating schema with regard to any concrete determination of objects, hence knowledge, is sufficient to notice that in subsequent Kantian transcendental philosophy such a schema is missing, at least as a basic element of the definition of knowledge.⁵ Kant’s problem of schematism concerns here, at the most, a subordinate issue, not a fundamental problem of the theory of knowledge. What are the reasons for this significant transformation? To begin with an evergreen of Kant scholarship, it cannot concern an irrelevancy in the sense of Kemp Smith (informed by Curtius (1914)). His argument

 The same applies to Hegel’s speculative idealism.

Kant’s Schematism in Southwest Neo-Kantianism

85

is that intuitions and categories are either not fully heterogenous, otherwise subsumption would be impossible, or the opposite is the case, with the result that schematism is redundant (Kemp Smith 1918, 318). For some interpreters, this criticism is supposed to be not only close to “Neo-Kantian interpretations” but also an expression of an Aristotelian type of ontological hyletic sensualism (Birrer 2017, 2 f. with 17). The latter interpretation touches upon two intrinsically related perspectives that are relevant for an assessment of the way the Southwest Neo-Kantians deal with the problem of schematism. First, for Southwest Neo-Kantians the relationship of form and content qualifies the validity structure of knowledge. In this respect it stands beyond any onticism of a sensible matter that obtains non-sensible forms by thought, resulting in a cosmos, a formed entity. By contrast, the relationship concerns, to speak in Kant’s terms, two types of representation: intuition as a representation of an individual and the concept as a representation of a something general. Hence, it concerns a relationship of capacities of reason, modelled in the fashion of a theory of representations, not an ontology of different types of being. For Kant, the “manifoldness of given representations” (intuitions or concepts) is unified in an “apperception” by means of an “action of understanding” that is the “logical function of judgments” (CPR B 143). Therefore, the manifoldness is “determined” with regard to a function of judgment that brings it to “consciousness” (CPR B 143). Moreover, as the categories are nothing but these functions for judging (in their significance for the content and hence for reality), any manifoldness of a given intuition are subject to the categories (CPR B 143). To formulate it with Hegel, being is a determination of thought, or as the Neo-Kantian Rickert puts it, content is itself a form and objects are always objects of knowledge. Thus, the relationship at issue, the problem of mediation, concerns an intragnoselogical relationship. On top of that, in the first instance, it does not concern the relationship of category and object but of intuition and concept: the object as a constituted entity is always already the result of a categorical synthesis. The issue of schematism concerns the problem of determination of concrete objects or the concrete determinacy of objects (Kant: ‘subsumption of objects under a concept’); categories are attached to concrete objects, and hence pure concepts of understanding applied to sensibility under the conditions of sensibility. In this way it becomes apparent why we are capable of making fundamental statements (synthetical judgements a priori) about nature as an object of knowledge. From the pure concepts of understanding flow synthetic judgments a priori that subsequently function as general principles (Grundsätze) for any knowledge of objects. These principles establish constitutively the concrete determination of objects. Categories, therefore, are not merely related to sensibility in general but

86

Christian Krijnen

related to sensibility in a concretizing way; determinacy is singularized.⁶ Categorical determinacy is transformed into a general determinacy that leads to singularized determinacy. This singularization of determinacy is the subject matter of the chapter of schematism. Something heterogeneous – that is to say the intellectuality of the category and the intuitive manifoldness of the (inner) sense, repraesentatio generalis and repraesentatio singularis – is mediated. The schema is the “third” factor, the a priori-conceptional representation that mediates both. The unity of synthesis, modelled in terms of a theory of apperception, is mediated with a singular intuition. The result is the “unity in the determination of sensibility” (CPR A 140/B 179). This finally leads to Kant’s conception of the empirical world as a world governed by natural laws. The doctrine of schematism develops the laws of the empirical world as laws that are subordinated to the laws of pure understanding. Concrete determination of objects, knowledge of nature, is according to Kant lawful determination. Second, the Southwest Neo-Kantians do not conceive of the cognitive relationship in the fashion of a theory of representation, as Kant does, but in terms of structures of validity: knowledge is addressed regarding the structure of its validity. This structure is a structure not of capacities or powers but of functions of validity. Kant’s idea that knowledge is conducted in judgments and concepts are based upon functions of judging leads in Southwest Neo-Kantianism to a conception that qualifies knowledge as a whole of principles, of functions of objective determinacy in its validity. By contrast, Kant gives this pure determinacy-logical issue of determination of objects, oriented towards judgments, an, as we could say, apperception-theoretical twist: he integrates the competence to determine objects by the ‘I think’ in his conception of the foundations of knowledge.⁷ From a systematic perspective, Kant’s own distinction between a subjective and an objective deduction has led subsequent transcendental philosophy to the ‘primacy’ of determining the foundations of knowledge in a validity-noematic or objective-logical orientation, in an orientation that concerns the content (Gehalt) of knowledge (Krijnen 2008a; 2014a). Consequently, the issue of a cognizing subject that appropriates the object of cognition is at most a subsequent theme. The function of an ‘I think’ as the highest unity of apperception is rendered explicit in a validity-noematic fashion.⁸

 Flach underlined this in his interpretation of Kant’s schematism (Flach 2001; 2002, 160 ff.).  Flach, possibly the most consistent proponent of a pure validity-functional conception of the foundations of knowledge (see Flach 1994), stresses this in his interpretation of Kant (Flach 2002, 112 ff.).  For Hegel, the science of logic is a doctrine of the idea in the abstract element of thought. Constellations of reality, more specifically a knowing subject, also do not play a role here; the func-

Kant’s Schematism in Southwest Neo-Kantianism

87

2 On the Southwest Neo-Kantian Urge for Unifying Kant’s Dualism The Neo-Kantians thematize Kant primarily from a systematic and not from a historical perspective. Notwithstanding the fact that Neo-Kantians like Hermann Cohen, Ernst Cassirer, or Bruno Bauch provide extensive Kant interpretations too (Cohen 1889; 1910; 1918; Cassirer 1994b; 1994a, vol. 2, 585 – 762; Bauch 1923a), the spiritus rector of Southwest Neo-Kantianism, Wilhelm Windelband, phrases it clearly: “Understanding Kant means to surpass him,” (Windelband 1915, IV) that is to develop Kant’s thought further.⁹ Neo-Kantian philosophy basically is about the problem of validity. Kant’s contribution to philosophy is valued regarding his insight into the problem of validity as well as into the method of how to cope with it. Yet Kant’s contribution should not only be reactivated but also re-actualized. For the Neo-Kantians, Kant’s conception of the ‘transcendental’ essentially concerns an entirety of grounds of validity. It cannot be captured by referring to a being beyond the cognitive relationship but only by turning reflexively towards thought as the grounds of all validity. The objective validity of the human accomplishments of meaning, of human objectivations, has its foundations in a set of validity principles, or as Kant would put it: in a set of ‘conditions of the possibility’ of such accomplishments. The objective validity of these validity principles is made plausible by showing that they are validity conditions of such theoretical or non-theoretical (practical, aesthetical, etc.) objectivations, that is productions of phenomena of meaning. Like Kant, both the Marburgian Neo-Kantians and the Southwest Neo-Kantians are eager to determine the determinacy of knowledge by determining the validity determinacy of cognition. By taking Kant’s transcendental philosophy as a theory of validity, possible ‘metaphysical’ residues and ‘psychological’ depravations in Kant’s doctrine are extin-

tion of the ‘I think’ as a transcendental unity of apperception is taken over by the ‘concept’ in its speculative meaning; a real subject, then, is at issue not before Hegel’s philosophy of subjective spirit.  See in this spirit also Cohen (1902, VII), Rickert (1924/25, 163 – 166; 1899, Vorbemerkung), or Natorp (1974, 243; 1912, 194; 196). – Of course, one should not fall victim to the suggestion that the Kant interpretation of both main schools of Neo-Kantianism is homogeneous (Cohen’s Kant interpretation, for example, has been harshly criticized by both Marburgians and Southwest NeoKantians). The relationship between Kant and Neo-Kantianism is addressed in many studies. See, among others, Heinz and Krijnen (2007).

88

Christian Krijnen

guished or at least subordinated to the overarching validity-theoretical content of Kant’s transcendental philosophy.¹⁰ Regarding the first dimension of the schematism problem – the foundational background dualism of intuition and concept or sensibility and understanding –, Kant’s transcendental aesthetic was heavily criticized not only in the early phase of German idealism but in Neo-Kantianism too. For Hegel, Kant does not develop both stems of knowledge “from the concept” but picks them up merely “empirically,” in particular from psychology (Hegel 1971, vol. 20, 339). The Neo-Kantians are basically of the same opinion. Without doubt, the problems Kant’s philosophy aimed to deal with emerged from the context of German metaphysics of his time and British empiricism. In this context, Kant poses the question concerning the “relationship between knowledge and its object” and answers it with his Critique of Pure Reason. The Neo-Kantians, though, operate in a philosophical situation characterized by other problems. In particular it is significant that in their time German idealism had declined and German empiricism arisen. By returning to Kant, the Neo-Kantians intend to sublate both developments. It therefore is not surprising that the Neo-Kantian treatment of Kant leads them to a different systematic conception of philosophy. The first component in Kant’s Transcendental Aesthetic that does not satisfy Southwest Neo-Kantianism exactly concerns the mentioned background dualism, and hence the problem of constitution addressed by Kant’s Transcendental Aesthetic: the problem of the meaning, function, and justification of both “stems” (CPR B 29) of knowledge. For my thesis it suffices to point to the tendency, decisive for Southwest Neo-Kantianism too (Krijnen 2007), to develop Kant’s dualism of stems further towards their unity. This culminates in Heinrich Rickert’s “model of a theoretical object in general” (Rickert 1924, 10 ff.; 1921, 50 ff.). Here, it becomes apparent that form and content are part of the concept of thought as the (theoretical) thought of something. Content itself turns out to be part of the model of an object. On this most fundamental level of self-relation of objective thought to content particular forms like ‘time’ and ‘space’ do not play any role. ‘Content as such’ is, as Rickert articulates it, the ‘logical place for the a-logical’: it guarantees the relatedness of thought to content and with this makes a plurality of specific contents possible. Rickert’s approach results, compared to Kant, in a different, more uniform model of constitution of experience in its original determinacy, its primary con See on the multidimensional character of Kant’s transcendental thought, for instance, the Kant scholar Zocher (1959), who himself stems from Southwest Neo-Kantianism. Recently, Flach (2015, 24) has diagnosed a “double ontological and power-theoretical burden” in Kant’s philosophy that should be avoided by a post-Kantian philosophy of validity.

Kant’s Schematism in Southwest Neo-Kantianism

89

stitution. This epistemological issue at the same time functions as the basis for a subsequent methodological issue, namely the knowledge of the non-philosophical sciences – the second component of Southwest Neo-Kantian criticism of Kant’s transcendental aesthetic. The reproach here is that Kant’s transcendental aesthetic with its forms ‘time’ and ‘space’ restricts what is given immediately to what is given by the senses. For Rickert’s theory of knowledge and science as well as for his ontology, this second component is equally important. Both components are essential parts of the reception and transformation of Kant’s Transcendental Aesthetic of the Critique of Pure Reason by Southwest Neo-Kantianism. This second component concerns not § 1 of the Critique (CPR A 19/B 33 ff.), the function of intuition, but §§ 2 ff. (CPR B 37 ff.), the forms of what has been intuited (Krijnen 2013). Both components are relevant for the problem of schematism. The first concerns Kant’s dualism of stems versus Rickert’s model of an object as such, that is Rickert’s heterology, guiding the Southwest Neo-Kantians. At issue in Kant’s dualism of stems and Rickert’s heterology is the basic structure of knowledge as a model of validity functions of cognition. Both propose a whole that consists of two parts. Kant thinks of this whole as consisting of “two stems of human cognition,” as he continuously emphasizes (even immediately before the Transcendental Aesthetic (CPR B 29)). He develops the determinacy of these stems in § 1 and returns to this at the beginning of the ‘transcendental logic’, talks of “two fundamental sources of the mind” (“receptivity” and “spontaneity”) or “intuition” and “concept” as the “elements of all our cognition” (CPR A 50/B 74). Knowledge is qualified by the cognitive moments of ‘intuition’ and ‘concept’ or rather ‘receptivity’ and ‘spontaneity’. These two sources of knowledge make up its fundamental determinacy; they are shown to be the basic principles of knowledge. They constitute knowledge. Knowledge is an interrelated validity functional whole of the principles of intuition and the concept, of aesthetical and logical conditions. Therefore, knowledge, in its validity structure, oscillates between indeterminacy and determinacy: The aesthetical and logical conditions of knowledge constitute the giveness of an object as well as thinking an object (CPR B 29 f.; A 50/B 74 f.).Via sensibility, the object is given, by understanding, the object is thought. Correspondingly, Kant distinguishes principles or forms that guarantee the knowledge function of intuition, i. e. the forms of time and space; and principles or forms that guarantee the knowledge function of thought, i. e. the pure concepts of understanding or categories. Both elementary functions of knowledge define knowledge in its fundamental structure in two constitutive respects. Both cooperate in the constitution of knowledge: “thoughts without content are empty, intuitions without concepts are blind. […] These two faculties or ca-

90

Christian Krijnen

pacities cannot exchange their functions. […] Only from their unification can cognition arise”. (CPR A 51/B 75 f.) The ‘transcendental aesthetic’ addresses knowledge as a representation of an undetermined object: a singular representation (repraesentatio singularis). Intuition represents the object immediately. The relevance of intuition, sensibility, or receptivity for knowledge is that it supplies material for knowledge; it establishes the undetermined object that needs to be determined, that is a substrate of possible determinations, starting from the “sensation” (sensatio) as the lowest genetic level of “representation with consciousness” (CPR B 376). As an “effect of an object on the capacity to represent, as far as we are affected by it” (CPR B 34), the sensation, taken by itself, concerns a mere “subjective perception,” and is completely validity indifferent, merely material for knowledge (see CPR B 34; A 50/B 74). The intuition Kant deals with, however, is not a merely validity-unrelated sensation of impressions but an already a priori formedness of the supplied material. Empirical intuition is only relevant for knowledge because ‘pure’ intuition guarantees its relation to an object. Regarding this function of representing the object immediately, intuition is both opposed and assigned to the concept as a mediated representation of the object. Together with its validity-functional counterpart ‘thought’, intuition makes up the determinacy of the content of knowledge. Although the forms of intuition ‘space’ and ‘time’ are intuitive and not conceptual in nature, and hence the forms of intuition not identical with the categories, their validity function is to guarantee the validity relevance of the material for knowledge. Within the whole of validity functions of knowledge, they play a constitutive role. The relationship between intuition and concept in Kant’s theory of constitution is that the relatedness to sensibility is a condition of the possibility of knowledge. Pure intuition is for Kant form – thus a principle of relations, an a priori factor of the possible determination of objects (CPR B 34 f.).¹¹ In this respect, intuition and concept do not differ. Only together do pure concept and pure intuition make a priori objective determination possible. Both sources of knowledge are captured in their function for knowledge, in their validity relevance. By implication, they are subordinated to the One principle that is knowledge and interpreted transcendental philosophically. The cognitive relationship is qualified in its meaning not by a hypostasis of two irreducible elements but by the validity functional togetherness of two moments of the One that is knowledge. These two moments are essentially those of indeterminacy

 The form of intuition enables it to ‘order’ the ‘sensations’, ‘matter’, or ‘manifold of appearances’ (i. e. the manifold of intuition in its undeterminacy).

Kant’s Schematism in Southwest Neo-Kantianism

91

(non-understood manifoldness) and determinacy of knowledge. This is the core of the matter Kant is dealing with. The Southwest Neo-Kantians stick to it. With Kant’s implementation of this thought, however, they are deeply dissatisfied.

3 The Southwest Neo-Kantians on Schematism and Original Unity 3.1 Bauch The Southwest Neo-Kantian Bruno Bauch, possibly the best of all Neo-Kantian Kant interpreters, commented both in his comprehensive book on Kant (Bauch 1923a) as well as in his major work on theoretical philosophy (Bauch 1923b) on the problem of schematism.¹² Already from his Kant interpretation decisive aspects concerning the basic structure of knowledge are brought to light, in first instance in the context of his elaborations on Kant’s transcendental aesthetic. On the one hand, Bauch wants to discuss the knowledge functional concern of Kant’s transcendental aesthetic (Bauch 1923a, 152 ff.). The validity function that is at issue is that of establishing a substrate of determinations. On the other hand, Bauch argues against a radical dualism of the two sources or stems of knowledge. For Bauch, both sources are different “validity conditions” or “validity parts”. Kant discusses the principles of sensibility too in the “perspective of objective validity” (Bauch 1923a, 153). Due to its own task, for a “science of all principles of a priori sensibility” (CPR B 35), despite the original duality of thought and intuition, sensibility cannot be fully a-logical. Rather, for Bauch, already here the idea of a “logical in the a-logical” emerges. Intuition only has relevance for knowledge due to the “concept of knowledge”: the concept of knowledge makes possible the task of a transcendental aesthetic as a transcendental aesthetic (Bauch 1923a, 154). Accordingly, Bauch pushes Kant’s conceptual instruments to the sideline and lets the logical or validity functional role of thought in the concept of knowledge take center stage (Bauch 1923a, 155 f.).

 Recently, Pringe (2015) discussed Bauch’s modifications. He underlines the importance of the infinitesimal principle for Bauch’s solution of the problem of schematism. Bauch, however, concludes his argument against Kant’s approach of schematism on p. 270. The infinitesimal principle plays a role not before his interpretation of the ‘Grundsätze’, more precisely of Bauch’s elaboration of the categorial determinacy of the sensation as a quantity (Bauch 1923b, 271 ff.).

92

Christian Krijnen

According to Bauch, although sensation does not determine the object, the material of sensation is relevant for knowledge, and hence the a posteriori part of the logical as the sphere of the form. That the material of a sensation is given, is a condition of the possibility of knowledge and belongs to its formal determinacy – the a posteriori is itself an a priori; form and matter belong together intrinsically (Bauch 1923a, 157 f.). Whereas the matter is “given” to us in its material content, the “form” enables the manifold of the appearance, as Kant says, to be ordered (CPR B 34). For Bauch, this ‘enabling’ also hints towards the relationship of intuition and thought, and hence of transcendental aesthetic and transcendental logic (Bauch 1923a, 159). The a priori means for Bauch the “logical-lawful meaning” qua condition of the possibility of knowledge (Bauch 1923a, 159 f.), thus the relationship between receptivity and spontaneity. This relationship reaches from the matter of sensations to the forms of thought. Only because the form of spontaneity as the sphere of the logical also rules the matter of sensations, does this matter logically belong to the concept of knowledge (Bauch 1923a, 159 f. incl. note 4). Bauch lays down precisely the logical relationship between form and matter in its systematical meaning. This leads him to numerous complaints about Kant, even if Bauch himself at the same time tries to relativize his criticism (at least regarding the fundamental aspects) by a kind of ‘letter – spirit strategy’. From early Kant scholarship on, and for Bauch too, Kant’s doctrine of the two sources of knowledge is reproached for having a certain ontic characteristic, despite the objective-logical meaning of both sources.¹³ In Bauch’s view, the duality remains guiding for the Critique of Pure Reason, notwithstanding the fact that Kant aims to connect both validity functions in his conception of concrete knowledge; the possibility of knowledge itself requires their unity (Bauch 1923a, 148; 153; 156). Although Bauch concedes that Kant’s method to ‘isolate’ form and matter of sensibility as well as sensibility and understanding (CPR B 34– 36) is not meant as an objective separation of different issues that belong together but as a methodological distinction in the sense of a perspective,¹⁴ he holds the reproach of iso-

 Bauch (1923a, 147 f.). Zocher observes, notwithstanding Kant’s transcendental or ‘semantical’ orientation, an “ontic” (an ontical idealism) in no less than three variations: a “metaphysical,” “psychological,” and “vague” idealism (Zocher 1959, 41 ff. with 47; 1954, 180 f.; 190 ff.).  Bauch (1923a, 158 f.). Although Heidemann (2002, 78 ff.) regards Kant’s method of isolation as positive, he must confess that it causes the problem of how to mediate intuition and concept in concrete knowledge, that is to say the problem of schematism, with the consequence that the original heterogeneity and irreducibility of intuition and concept cannot be strict. Yet Heidemann has to face the problem that such a beginning of philosophical concept formation is, in accordance with Hegel’s criticism of Reinhold (Hegel 1971, vol. 9, § 10 N; vol 5, 68 ff.), merely

Kant’s Schematism in Southwest Neo-Kantianism

93

lation in the Critique of Pure Reason to be correct (Bauch 1923b, 2 f.). For Bauch, this isolation leads to an abstraction as in the Critique of Pure Reason. In particular in the Transcendental Aesthetic, Kant sticks to the independence of given empirical material and logical forms.¹⁵ Yet for Bauch there just is no principle, i. e. ‘given material’, that is independent of the logical principle or equally original (Bauch 1923a, 200 f. with 204). As Bauch shows, the respective parts only have their determinacy in relationship to each other. A firm dualism of form and content distorts the issue at stake (Bauch 1923a, 203). As a result, Bauch claims “objective reason” as the missing foundation (Bauch 1923a, 204). That Kant holds on to the thing-in-itself is for Bauch, seen from a logical perspective, nothing but a futile attempt to guarantee the independence of the object from the subject. In this respect too, Bauch wants to guarantee the validity claim of the empirical material by the necessity of the concept. ¹⁶ In short, the initial dualism of sources is the problematic situation. It should be replaced by a more original relationship. Bauch opts, in conformity with Kant’s transcendental revolution in philosophical thought, for reason, or, more precisely, ‘objective reason’, as the ultimate foundation. Objective reason is the foundation. Only in the context of objective reason can sensation obtain its relevance for knowledge, i. e. its logical meaning (Bauch 1923b, 204 ff.; cf. 225 ff., 233 ff.). In line with Bauch’s persistent validity-functional theory of determination, sensations and the like are always already determined categorically in order to be material and sensation at all (Bauch 1923b, 201 ff., 259) – without categorical laws there are no objects to intuit. A constellation like material without categories is for Bauch a totally empty abstraction; such material would not even

hypothetical. Rather, rational foundation of the two stems can only succeed from the concept of knowledge itself in its validity functional meaning. This concept is the ‘higher’ or ‘original’ unity of the sources of knowledge. Interestingly enough, Heidemann finally refers to Brandom’s theory of discursive praxis and the interpretation of concepts as functions and norms that emerge from it, as it seems to offer a possible unity of intuition and concept. This, however, is exactly the project the Southwest Neo-Kantians are heading for.  In contrast to this view, Bauch’s merits consist not least in the development of a strict functional understanding of the concept (Bauch 1914; 1923b; 1926).  See Bauch (1923a, 163 f. note 1 with 164 f.; 1923b, 204). See for Bauch’s criticism of Kant also Bauch (1914). Here too Bauch criticizes Kant’s conception of the relationship between empirical content and categorical form in the Critique of Pure Reason (Bauch 1914, 310) as well as the dualism of intuition and category and a “misplaced dogmatism of the thing-in-itself” (Bauch 1914, 332). According to Bauch, in Kant’s first Critique the “originally ambiguous position of generality and particularity, form and matter of cognition as such as well as within the form of concept and intuition” nolens volens is retained (Bauch 1914, 336).

94

Christian Krijnen

be conceivable as material.¹⁷ Kant’s famous sentence, thoughts without content are empty, intuitions without concepts blind (CPR B 75) for Bauch cannot mean that empty thoughts or blind intuitions do exist, as both are mere abstractions, that is results of a philosophical reflection. As Bauch says, the emptiness of thought and the blindness of intuitions respectively only have meaning in terms of their pureness, and hence, as different objective validity laws (Bauch 1923b, 267; 1982, 266). Regarding the fact that both are objective validity laws, there is no “principal” difference between an intuition and a concept (Bauch 1923b, 268). Intuition, or the laws of intuition, for Bauch have the validity function to include the “content of sensations” (Empfindungsinhaltlichkeit), the “manifold material” in the categorical network of validity in order to determine intuition in conformity with the “law of the concept” (Bauch 1923b, 259 with 267 f.; 1926, 206 f.). Hence, the laws of intuition cannot be fully or “in principle” independent or detached from the laws of the relationship of categories. In any case objective validity is at stake; as law determinacy, the laws of intuition are subjected to categorical laws (Bauch 1923b, 267 ff.). Consequently, Bauch does not model the distinction between types of validity laws as an “isolation” in need of a mediator. Their unity is already apparent in any concrete object of intuition as determined by transcendental laws.¹⁸ The material of knowledge is included in the forms of the validity relations of objective thought. It is in this context that Bauch makes remarks about the problem of schematism in Kant. For Bauch, Kant fails to understand the relationship between intuition and concept adequately, as “most evidently” is the case in schematism (Bauch 1923b, 270). Although time and space are no categories or concepts, in Kant their distinction turns into an isolation in need of a uniting mediator – despite the fact, recognized by Kant, that without such a unity neither an object of intuition nor the intuition of an object is possible. Pure intuition and category or concept are not separated from each other but, for Bauch, belong together in the “determinacy of the concrete object by laws of transcendental logic” (transzendentallogische Bedingtheit des konkreten Gegenstandes). Any “artificial unification” is superfluous here. In a similar fashion Bauch notes in his monograph on Kant that Kant initially isolates intuition and concept fully, although this iso-

 Bauch (1923b, 259). See Bauch (1926, 203 f.), where conceptualizing the content of sensation in an isolating fashion is criticized as an abstract way of handling the issue.  Bauch (1923b, 270). According to Bauch, in the Critique of Pure Reason Kant holds on too much on the “isolation” of form and content (Bauch 1923b, 304, 308). Kant’s talk of a ‘swarm of sensations’ is denounced as an empty abstraction (Bauch 1923b, 259; 1926, 203; 243). (Kant himself speaks of a “swarm of appearances” (CPR A 111).)

Kant’s Schematism in Southwest Neo-Kantianism

95

lation is only possible in abstraction and reflection and not in knowledge (Bauch 1923a, 233). For Bauch, this becomes fatal in Kant’s doctrine of the principles of pure understanding (Grundsätze des reinen Verstandes), as the initial isolation is not taken as a mere product of reflection but as an objective constellation. The problem of “application” is modeled into a problem of uniting intuition and concept by a mediator, a third factor, schematism, although they are already united by the concepts of the transcendental and that of synthesis (Bauch 1923a, 234). In Bauch’s view, the price Kant finally has to pay for his procedure of isolation and abstraction is the introduction of an artificial mediator (Bauch 1923a, 234; 239). Bauch addresses Kant’s chapter on schematism from this perspective (Bauch 1923a, 234 ff.). He rejects the idea of a complete lack of homogeneity of intuition and concept (Bauch 1923a, 235). Yet this this was the constellation that led Kant to the problem of applying categories to appearances. For Bauch, it is a pointless problem; there is no sufficient reason for introducing a mediating third factor, which is the transcendental schema.¹⁹ Although in his interpretation Bauch continuously speaks of a mediation between category and intuition, whereas for Kant the application of categories to appearences is at issue, the direction of impact is clear: Bauch aims to overcome the original heterogeneity by an original unity in the concept of knowledge. Consequently, he renders explicit the internal laws of validity of this unity. By implication, a schema is superfluous. In Bauch’s discussion of Kant’s problem of application, that is the problem of concrete determination of an object, it becomes clear that, formulated with regard to Kant’s schematism, the sensation of quality becomes a “structural element” of the object of intuition by the concept. The concept puts together the object as a categorical unity of a manifoldness of its elements. In order to be a sensation, sensation is always already embedded in a web of categories, of categorical relationships (and not a mere modification of the state of the subject, merely subjective; in order to be subjective in this way, it must already be objective).²⁰ Certainly, Bauch does not want to give up the “positive meaning” of Kant’s schematism, that is its legitimate validity function, but he does so without referring to a third factor: the unity of category and intuition is guaranteed by the category itself (Bauch 1923b, 237). The schema too is categorically conditioned. Its positive meaning boils down to, as Kant says, being a “rule for the determination of our intuition in accordance with a certain general concept” (CPR B 180), a “method” or “procedure” for “providing a con-

 Bauch (1923a, 236). Bauch refers here also to the analysis of Curtius.  Bauch (1923b, 275). See also Bauch’s distinction between the sensation of quality (Qualitätsempfinden) and the quality of sensation (Empfindungsqualität) (Bauch 1923b, 251).

96

Christian Krijnen

cept with its image” (CPR B 179 f.). Intuition is determined by the concept. This determination of intuition by the schema that the category gives to itself in order to give itself an image, that is to determine intuition, for Bauch is the “good sense of schematism” (Bauch 1923a, 239). The schema is the fundamental relationship of unity between the category and the determination of intuition. Without a schema, categories would remain functions of determination without being capable of representing an object. In actual knowledge, category and intuition are what they are only in relation to each other (Bauch 1923a, 240). So far the issue of originality in Bauch! It already points to the logic of determination of a concrete object. I shall come back to this latter issue in section 4, where the Southwest Neo-Kantian alternative for the mediating function of the schema is discussed. For now, a look at the idea of originality as it is conceived by other Southwest Neo-Kantians shows that it concerns a commonly shared idea.

3.2 Other Southwest Neo-Kantians Jonas Cohn’s doctrine of utraquism, ²¹ for instance, concerns the relationship between the ultimate moments of knowledge: form and content of thought, thought form and thought content. Cohn holds that the content is ‘alien to thought’ (denkfremd) insofar as it cannot be deduced from thought but is given to thought. However, in order to be thinkable and theoretically relevant at all it needs to contain ‘form’ (the form of being given). Both ultimate moments of thought correlate reciprocally: neither an unformed content nor a form without content are part of the realm of thought. Therefore, Cohn criticizes any fixed, non-utraquistically conceived opposition of concept and intuition (Cohn 1923b, 255). The dualism of sensibility and understanding for him results from of a “rather primitive psychology,” mixing up concepts of powers and values (Cohn 1908, 99). As in Bauch, the relationship between intuitions and concepts is being developed further towards its unity, which is conceived of as a necessary knowledge-functional relationship (Cohn 1908, 117). Within Southwest Neo-Kantianism, only Emil Lask’s doctrine of judgment is a serious exception from the sketched line of reasoning. Not surprising, then, that Bauch, Cohn, and Rickert all reject Lask’s doctrine as not radical enough (Cohn 1923b, 153 f.; Bauch 1923b, 192 ff., 200 ff.; Rickert 1928, 283 f., 335 f.,

 See for Cohn’s utraquism especially Cohn (1923b). Also compare Cohn (1908, 116 ff.; 1923a, 9 f.; 1932, 36 f.; 1949).

Kant’s Schematism in Southwest Neo-Kantianism

97

cf. 278 – 297, 332 ff.). Explicitly referring to Kant (Lask 1923b, 73 f., 80; 1923a, 328 ff.), Lask introduces a constellation that for the critics mentioned does not exist. Something that is absolutely independent of the logical, something merely “given,” “logical amorphous material” that as “logically naked” and “material substrate” becomes equipped with categorial predicates, a “merely logical addition,” and thus “placed” within categorial determinations (Lask 1923a, 333; 1923b, 73 ff.). By contrast, according to the main doctrine content itself is a logical principle. Material independent of validity, material “not-involved” in validity, “alien to validity” (geltungsfremd) does not exist within the realm of knowledge. It is not surprising that in Lask scholarship, Lask’s theory of validity has been seen as a positive reception of Kant’s schematism (Nachtsheim 1992, 180 – 187). Nevertheless, Lask’s conception of ultimate foundations also contains possibilities for a reconciliatory reading. Such a reading, however, would tie Lask back even more strongly to the tendency of the Southwest Neo-Kantian search for unity.²² The key for this is Lask’s concept of “objective form of structure” as “original structure” consisting of the two parts “form and material” (Lask 1923b, 281). Unsurprisingly, in this respect Lask refers to Rickert. Indeed, Rickert with his ‘model of an object’ has famously and for Southwest Neo-Kantianism decisively tried to conceptualize the relatedness of the parts of the origin. From a systematic point of view, the tendency we saw in Bauch and Cohn to develop Kant’s fundamental dualism of knowledge towards a unity, culminates in Rickert’s model of an object theoretically thought of as such.²³ It addresses the theoretical object as a whole of fundamental logical conditions of all thoughts and thinking related to truth (Rickert 1924, 8 ff.; 1921, 50 ff.). With that, it determines knowledge in its origin, the theoretical realm itself, the ultimate presupposition of any concrete theoretical determination. The origin establishes theoretical objectivity and thus the cognitive relationship itself.²⁴ Kant’s two stems of knowledge are integrated in the model of an object regarding their primaryconstitutive meaning. Accordingly, the model of an object establishes at the

 See Nachtsheim (1992, 231 ff.), who developed this reading further in Nachtsheim (2017). The latter study clearly shows how strongly Lasks holds on to the heterology of Southwest Neo-Kantianism. See Lask’s elaborations of “objective original structure” or “objective form of structure” in his doctrine of judgment (Lask 1923a, 364 ff., 381 ff.). Still, Lask cultivates a hypostatic rhetoric of ‘category – material’, as if both would exist outside the structural unity of knowledge and would be brought together subsequently by cognition.  See on this model extensively Krijnen (2001, ch. 5).  Therefore, I do not think that Rickert’s model of an object offers an “analogon” (Nachtsheim 1992, 186, note 28) for the problem of synthesis of form and content in the empirical object, addressed in Kant’s schematism of pure concepts of understanding.

98

Christian Krijnen

same time a substrate of possible cognitive determinations. Content turns out to be the ‘logical place for the a-logical’ (Rickert 1921, 52; 1924, 12). Obviously, content obtains the status of a principle. Content is a form of logical or theoretical thought. Thought itself posits that to which it relates to and thus involves a selfrelation to content. As the logical place for the a-logical, the form ‘content’ guarantees the relatedness of thought to content, and with this enables a manifoldness of contents. Thought appears to be the encompassing unity within which the parts (Glieder) of the unity are possible themselves. Insofar as the origin constitutes the possibility of any determination of objects, the logical sphere of origin does not yet concern immediately thought in its determination of objects but only in its first beginnings of thinking an object.²⁵ Due to its foundational function, the origin qua beginning of thought is also the beginning of a process of determination; it contains the basic determinacy of the logical sphere of determination too. Within the horizon of knowledge, there is nothing radically alien to thought, no radical heterogeneity of opposites in need of a subsequent mediation. The relationship of form and content is itself a relationship of forms. Only within the theoretical realm does the concept of (theoretical) content make sense. Both are equally originally parts of the cognitive relationship that notwithstanding their unbreakable conceptual relatedness are different from each other. Anything that can enter thought is constituted by thought, and hence given by thought itself: only due to its own principles does thought have content. Only within the cognitive relationship do concepts like form or content have meaning. Instead of referring to some non-validity factor, something not constituted by thought, at most there are levels of relative originality (layers of apriority). This idea of relative originality or layered apriority offers the key for solving exactly the problem Kant was aiming at with his doctrine of schematism of the Critique of Pure Reason. This issue will be discussed in part 4 regarding Rickert and Bauch, before in part 5 I briefly go into contemporary transcendental philosophy. For now, we can see that Southwest Neo-Kantians like Rickert, Bauch, or Cohn do not intend to extract the material of cognition from its form, but they insist that already content as content is characterized by forms. ‘Content’, ‘being’, ‘existing’, ‘material’, and so forth all are determinations of forms without

 In the terminology of his late work on the logic of the predicate, the object as such (Etwas überhaupt) does not yet contain ‘forms of cognition’ but only ‘forms of thought’, that is to say forms of the concept of the subject, not of the concept of the predicate (Rickert 1930, 111 ff.). At most, that which is to be determined is posited; as a subject of predication the object is still undetermined. In the realm of determination, the logical object of the origin takes the place of the concept of the subject in judgments. See on this Krijnen (2008b, ch. 8).

Kant’s Schematism in Southwest Neo-Kantianism

99

which a concrete entity – that in its concreteness always consists of form and content – is formed content by forming forms and would not even be conceivable as ‘material’ and the like. Instead of a harsh dualism of form and content, the intragnoseological turn in conceptualizing the grounds of knowledge requires an understanding of content (being, material, etc.) as a form in its relationship to the ‘content of the content’. The latter we can only, as the Neo-Kantian Rickert puts it, “’undergo’ or ‘intuit’ or experience otherwise a-logically” (Rickert 1921, 53 f.; 1924, 13). As any content is always formed content and any form a form filled with content, from the standpoint of isolation there is no adequate qualification for a content untouched by forms, a pure a-logical content. We cannot think such a concept logically or theoretically and, hence, the intended issue is theoretically irrelevant.²⁶ A complete isolation of form or thought and content is just a residuum of the direct-intentional cognitive attitude. Yet the fabric of form and content for the Southwest Neo-Kantians has, beyond any ontics, a purely semantical (Zocher) or functional (Bauch) meaning.

4 The Southwest Neo-Kantian Alternative for Kant’s Schematism: Methodology 4.1 Rickert (1) One main result so far is that a radical heterogeneity in the origin of knowledge has been parried and a relationship of mutually and necessary related parts established. However, Kant’s problem of schematism does not concern the general relationship between concept and intuition or receptivity and understanding but the more specific problem of applying categories to objects of knowledge, which is the problem of a concrete determination of objects. The problem of concretizing knowledge is not solved by the Southwest Neo-Kantians via a mediating third factor but by making explicit the original unity of knowledge, and hence by the concept, form, or idea of knowledge. This leads to a model of layers or levels of principles of objective determination (layered apriority). These levels reach from the origin of knowledge up to its individualization in concrete objects. Yet this idea of individualizing, concretizing, or singularizing concerns both the subjective-logical dimension and the objective-logical dimension of the object. A closer look at Rickert and Bauch makes clear what this means.

 See Rickert (1921, 53; 1924, 53). Also see Rickert’s extensive criticism of epistemological intuitionism (Krijnen 2001, ch. 5.2.2.5).

100

Christian Krijnen

According to Rickert’s heterology, the theory of knowledge does not start with the separation of form and content, subject and object, and the like. Such a separation appears to result from objective thought itself. All thinkable entities consist of form and content. Ontologically seen, the given sensible and intelligible world consists of sensible objects that contain, seen epistemologically, sensible content and intelligible form, as well as intelligible objects (figurations of meaning) consisting, epistemologically seen, of intelligible content and intelligible form.²⁷ For a transcendental philosophy on the level of Rickert’s reflection, there is no radical externality to thought, no non-validity factor or something not constituted from validity. There are at the most levels of originality within the theoretical realm. Of course, a content must be ‘there’ in order to be recognized, but already ‘content’ is shown to be a principle within the realm of validity. ‘There’ is something only because of thought. The validity structure of thought is logically prior to any knowledge of being that and what something is. Rickert’s levels of relative originality within the theoretical realm reach from the original synthesis of thought in the model of an object as such up to the constitution of concrete objectivity via methodological forms of cognition; they cover the whole spectrum of theoretical determination. By implication, also concepts like reality or that of the given (for cognition) must be constituted. Here, it is not the relatedness of thought to content (Inhalt) that is at issue, but the specific meaning of that which is given as material for, e. g., scientific knowledge (hence, givenness on a rather late level of constitution). In short, the cognitive relationship is being concretized in terms of a theory of principles; it is conceptually compressed to concrete-objective meaning. Rickert’s philosophy addresses the problem of object constitution in two directions: a subjective-logical and an objective-logical direction.²⁸ His epistemological opus magnum, Der Gegenstand der Erkenntnis, is designed subjective-logically. In the course of several revisions, Rickert outlined the objective-logical dimension too. As Rickert propagates a logical primacy of the objective direction, it is not so much about two directions of the theory of knowledge but about the exploration of two dimensions or reflective modes of knowledge. As a doctrine of the validity determinacy of knowledge, the theory of knowledge always addresses the One ‘Gegenstand der Erkenntnis’ in the sense of the standard for knowledge. This standard concerns two dimensions. On the one hand, knowledge is

 Rickert (1939a; 1939b; 1934). This also leads to an important difference with regard to Kant’s Transcendental Aesthetics of the Critique of Pure Reason, §§ 2 ff.: For Rickert (and subsequent transcendental philosophy in general), time and space are not the original forms of what is given immediately. See Krijnen (2013).  Rickert (1909; 1912; 1928). See for the following also Krijnen (2014b).

Kant’s Schematism in Southwest Neo-Kantianism

101

thematical as ‘that which is thought’, and hence as an objective figuration, as an object: the objectivity of knowledge is at issue here. This concerns validity as an entirety of principles that guarantee the objectivity of cognitive performances. Whereas on the objective route the ‘object’ is thematical in itself and hence detached from the cognizing subject, the subjective route deals with knowledge as cognition of an object by a subject: the subjectivity of knowledge is at issue here. This concerns validity in its logical performance. Within the realm of cognition, that ‘what’ is thought as an objective configuration has to be distinguished logically from that ‘through which’ it is thought as a subjective configuration. These two different issues of the theory of knowledge not only lead to different determinations of knowledge, they also have two different points of departure of validity reflection. The theory of knowledge always departs from a fact, a factum, some reality. Rickert’s subjective route, also called the transcendental-psychological route, departs from the fact of cognizing; from the real cognitive act. From this subjective act, the theory of knowledge makes its way gradually to the transcendent object as the ground of all objectivity: from the subject to the object. The objective route, also called the transcendental-logical route, deals as soon as possible and regardless of the psychological act of knowing with the transcendent object in itself: the objective route progresses pure logically. In contrast to the subjective approach, for the objective approach the question about cognizing the object by a subject takes a back seat. Therefore, the starting point of reflection is not the subjective-logical fact of cognizing but the objectivelogical of cognition: the (supposedly) true proposition. In chapter five of Der Gegenstand der Erkenntnis, Rickert offers a discussion of empirical realism, typical of the non-philosophical sciences. He aims to show its compatibility with transcendental idealism. This discussion is of interest insofar as from a systematic point of view the problem of concretizing knowledge is at issue – albeit in a subjective-logical fashion – and hence the issue of Kant’s chapter on schematism. From this subjective-logical perspective, levels of appropriating the object by the subject are elaborated on. The validity ground of knowledge, the ‘object’ qua standard of cognition has already been shown by Rickert to be an “ought,” not a “real existing being” (Rickert 1928, 350). The cognizing subject is regarding its validity subjected to a factor that norms its performances. According to Rickert’s view that the theory of knowledge is a theory of forms, his analysis in chapter five concerns the ‘form’ of knowledge too, and hence the principles of validity. More in particular, it deals with a specific form, which Rickert calls the “category”. Through the category, the “act of cognition becomes objective” (Rickert 1928, 361; 366). Rickert addresses this subjective logically (Rickert 1928, 362 ff.).

102

Christian Krijnen

It turns out to be a complex issue, consisting of many forms that have a function for capturing the object of cognition by the cognitive act (Rickert 1928, 365 ff.). Characteristic of the category from a subjective-logical perspective is that it, so to speak, concerns the “transition from the ought to the realm of real beings,” that is to say that Rickert’s category is the moment that “attaches” or “predicates” the relevant form to the content (Rickert 1928, 366). The category is the form of attaching, “predicating” and thus the form of meaning typical of the “act of recognition”; it concerns the act of capturing the object by the cognizing subject (Rickert 1928, 355 ff.). That the validity of cognition “grounds” (Rickert 1928, 369) in the category as the form of recognition of theoretical normativity is explained by Rickert subjective logically. Knowledge turns out to be a process of providing forms. Rickert’s subjective-logical intragnoseological interpretation leads to a whole of subjective-logical forms of knowledge which makes clear what cognition “presupposes and must presuppose in order to claim objectivity” (Rickert 1928, 369). After it has been clarified subjective logically what it means that cognition becomes objective, Rickert discusses particular problems, showing the compatibility of transcendental idealism and empirical realism. Interesting, from a methodological point of view, for the idea of a subjective-logical treatment of the foundations of knowledge and the promise of reconciling the forms of knowledge and the reality that it contains is the following. The first particular problem concerns the issue of “givenness.” For Rickert, givenness is a category and as such a subjective-logical figuration (Rickert 1928, 371 f.). At stake is the meaning of concrete objective empirical knowledge regarding the problem of givenness. Therefore, Rickert clarifies the form of givenness – Rickert also speaks of facticity (Tatsächlichkeit) – for empirical knowledge. He shows that the validity of judgements of facticity or givenness are based upon the form in as far as the category, as the form of the act of cognition, grants them “objectivity” (Rickert 1928, 372). That what is actually ‘given’ is shown to be determined by forms. The opposition between what is actually given or “experience” and “thought” is replaced by thought as the formal foundation of any cognition of reality (Rickert 1928, 378; 381). This reduction of empirical realism to the issue of givenness (Tatsächlichkeit) is of course insufficient to understand the meaning of empirical knowledge. Its meaning contains further presuppositions, reaching far beyond that which is given. It contains especially the presupposition that facts are always part of a larger context. Indeed, scientific knowledge in particular aims at a coherent whole of cognition (Rickert 1928, 383). This striving for coherent cognition of reality has a subjective-logical foundation too. As every science deals with material at issue, regarding the claim of empirical knowledge Rickert distinguishes two

Kant’s Schematism in Southwest Neo-Kantianism

103

aspects: 1) The material and 2) its cognitive treatment. Rickert shows the compatibility of idealism and realism both for the material and for its treatment. Of course, something like ‘material’ represents a rather advanced level of constitution of the object and thus a concretization of the relationship of form and content. On top of that, what I have just said applies not only to scientific knowledge but to knowledge in general. Within the sketched context, the compatibility of thought and reality boils down to showing that also the material of knowledge, the interrelated real world, presupposed by the cognizing subject, contains, from the perspective of transcendental idealism, “forms of relationships” (Rickert 1928, 389 f.). It always turns out to be the category that forms the transition from the dimension of “the ought to that of reality”; in this case the category of relatedness (Rickert 1928, 393). Cognizing persistently remains recognizing norms that function as the standard also for our judgements of reality (Rickert 1928, 394 f.). More specifically, these concern forms of recognition of the “epistemological subject”. This abstract subject without flesh and blood constitutes the objective reality subjective logically, that is to say the world that from an empirical realist point of view seems to be existing in itself (Rickert 1928, 397 f.). At the same time, however, it is understood how for the real, “cognizing subject” (with flesh and blood) an independent objective world can exist at all, since for the cognizing subject the epistemological subject makes up the “ideal” or norm (Rickert 1928, 397 ff.). Rickert’s idealism even conceives of the objective reality as the “reality untouched by any forms of scientific or pre-scientific concept formation of a real subject” (Rickert 1928, 414). Such a world, however, is from the start determined by forms. After the problem of objective reality has been dealt with, Rickert determines the forms that determine the conceptual determination of objective reality by the cognizing subject (Rickert 1928, ch. 5., § 5). They concern, regarding scientific knowledge, the “concept of the science of reality” (Rickert 1928, 402). These are the so-called methodological forms of knowledge. With these forms, the cognizing subject determines reality conceptually. This conceptual determination of objective reality needs to be understood subjective logically too. Rickert achieves the compatibility on the level of constitution of scientific concept formation about reality.²⁹ Such concept formation again is determined by forms. These concern a process of capturing reality by a transforming performance of the real, cognizing subject (“umbildendes Auffassen”), and with this by “methodological forms” that guide the cognizing subject, which regarding their

 See for pre-scientific knowledge e. g. Rickert (1929, ch. 1.I).

104

Christian Krijnen

validity do not depend on the cognizing subject (Rickert 1928, 403 f.). These methodological forms are also treated by Rickert as forms of recognition. They produce the meaning of statements about reality subjective logically. Consequently, also the “objectivity” of scientific concept formation depends subjective logically on the fact whether their “forms are grounded upon valid norms” (Rickert 1928, 431). A transcendental ought functions as the validity “foundation” of any cognition. So far, the subjective logical reflection. Rickert also reflects objective logically on objectivity (Rickert 1939c; 1921; 1930). In particular his Logik des Prädikats (1930) attempts to begin not with the performance of cognizing but aims to show what forms make up that which is thought (at least in it constitutive dimension as methodological forms are not discussed here)³⁰. It presents a theory of objective-logical object constitution, starting with the “most simple logical meaning and archaic predicates” (Urprädikate) (Rickert 1930, 70 ff.). Interestingly enough, Rickert translates predicate here in a logical sense as κατηγορούμενον, and hence as a form of propositions (Aussagen), propositions not in the subjective logical sense as stating but objective logically as its result, as that what is stated. Of course, here too, it turns out that a ‘logical’ foundation precedes any ‘ontology’: transcendental idealism remains the foundation of any possible realism; logically, that which is real is always that “which is predicated as real” (Rickert 1930, 78 – 80). In his doctrine of forms of thought and forms of knowledge, Rickert (1930, 111 ff.) shows that already the logical subject (the concept of the subject in judgments), and hence that which is thinkable at all and therefore cognizable, is determined by forms. The issue of deictic expressions like ‘this’, dealt with in his book on the Gegenstand der Erkenntnis within the context of constitutive forms of reality, returns in the Logik des Prädikats as the immediate relationship to a sensible intuition, which presupposes “identity” as a form (of thought) (Rickert 1930, 114 ff.; 141). In short, the thesis of a compatibility of, as Kant puts it, category and appearance has proved to be correct – appearances are on any level of their determinacy determined by forms. The logical dependence of everything that in one way or another ‘is’ of the forms of thought and cognition leads Rickert to surpass the subjective-logical determination towards a validity-functional noematic, objective-logical differentiation of principles of theoretical validity. These reach from the level of pure heterogeneity of the origin of cognition via its concretization in judgements up to the constitution of concrete-objective meaning consti-

 Rickert has discussed them extensively in his monographs on methodology (Rickert 1929; 1926).

Kant’s Schematism in Southwest Neo-Kantianism

105

tuted by archaic predicates and subsequent categorical forms of objective reality as well as methodological forms of cognition. Yet it is Bauch who develops a theory of knowledge in a strict objective-logical fashion. Moreover, he does so in discussion both with realism and Kant. Concluding my elaborations on Southwest Neo-Kantianism, I shall show why also for Bauch, Kant’s doctrine of schematism is unnecessary regarding the problem of concretizing cognition.

4.2 Bauch As for Rickert, also for Bauch is reality an important philosophical problem. At stake are questions like what reality is, what it means that reality is given, how reality must be in order to be recognizable, how it relates to the cognizing subject and its thought.³¹ Accordingly, for Bauch too the notion of ‘fact’ becomes an epistemological problem. As in Rickert’s philosophy both ‘that’ (Dasein, existence) and ‘what’ (Sosein, quality) are posited in the realm of knowledge and therefore subject to the validity laws of knowledge, if it should be possible at all that for the subject matter to become an object of cognition, even (logically) to be at all (Bauch 1915; 1923b, 123 ff.). Bauch in particular fulminates against an abstract conception of the concept and conceives of form and content as an intrinsic relationship. Continuously it turns out that being in its possibility is founded in thought. This results, as in Rickert, in a transcendental idealist conception of an object. The object qua standard or measure of cognition is for Bauch (1923b, 91 f.) not an existing reality but a whole of relations of validity (Geltungsbeziehungen). Actual, concrete, real cognition is oriented towards such validity relations and thus obtains its validity. These relations, and hence the concept, is the object qua measure for the cognizing subject (Bauch 1923b, 217 ff.). Both for Rickert and Bauch the “foundation of reality” (Bauch 1923b, 126; 1982, 254) is a non-real realm. Just like Rickert, Bauch too rejects realism as an adequate epistemological position while at the same time accepting “empirical realism”. In his methodology of the empirical sciences, Bauch follows, notwithstanding several modifications in detail and the dominant objective logical perspective, Rickert’s approach. How does Bauch’s theory of knowledge as an objective-logical theory surpass Kant’s schematism as a mediating third factor? To be sure, Bauch also takes the subjective-logical dimension of knowledge into account. He integrates,

 Bauch (1923b, part I, esp. 93 ff.; 1982, 255 ff.). See for Bauch’s theoretical philosophy notably Bauch (1923b). Also see Bauch (1923/24; 1926; 1982).

106

Christian Krijnen

so to speak, both routes of Rickert but in a more uniform way and from the start from an objective-logical perspective. As a consequence, the significance of forms of knowledge as principles of objective determinacy becomes very clear. For Bauch, the problem of knowledge both concerns the dimension of cognizing and cognition. Yet the former has to be developed from the latter, from the theory of knowledge in an objective perspective (Bauch 1923b, 49; 1923/24; 1982). Both the object as well as its cognition underlie mutual conditions: conditions of truth. Whereas cognizing is shown to be a process of relating which, regarding its validity, can only direct itself towards relations, relations of validity, of truth, the object too turns out to be a relation based upon truth relations. Truth relations (and the truth is nothing but the whole of them) therefore are objectivelogical and hence trans-subjective relations. Bauch conceives of them as objective validity functions. To be an object implies to stand in relations and to cognize is to direct oneself towards relations. More precisely, the constitution of the object of knowledge as well as its cognition take place via three types of truth or validity relations: the category, the concept, and the idea. Concerning cognition of real objects, in the first instance the order of the content of sensations by which the object is given is at stake. For being a sensation at all, sensations must be part of a relation (being, identity, difference, contentuality, etc.). This relation is the category. Like Kant, but in contrast to Rickert, Bauch conceives of the category strictly in its objective-logical meaning: categories are principles of objective determinacy, not relations of cognizing but of that which is cognized. Reality presupposes such relationships. The category itself, then, is part of a relationship too, determining the object (thing, property, cause, effect, one, many etc.). The relation between the categories, which determines the object, is the concept. The concept is, as Bauch formulates it, “the objective law of formation of the object as an object of cognition” (“objektives Bildungsgesetz des zu erkennenden Gegenstandes”).³² The concept makes up the relationship between the categories. At the same time, the objects are constituted by concepts and hence categories. The relation between the concepts is the foundation of both our cognition of objects and the objects of our cognition. This philosophy of relationships finally leads to the whole of all relationships, i. e. the idea. The idea is the system of concepts. Hence, it is the objective-validity relationship that, as the whole of conditions of objects, constitutes reality and its cognition.

 Bauch (1982, 265). Bauch also characterizes it as the function of the direction from the general to the particular (Bauch 1923b, 283 ff.; 1926, 101 f.; 131 ff.; 188; 1982, 266).

Kant’s Schematism in Southwest Neo-Kantianism

107

Validity relations are conditions of the objectivity of objects. Consequently, they are themselves not objects. They underlie all being from the start and persistently. Like Rickert, Bauch also elaborates on this with regard to the method of concrete-objective determination qua cognition of reality. According to Bauch’s theory of truth, a part of the fundamental structure of truth is the method. The method is the way to the truth. In short, Bauch holds that knowledge has a validity-noematic structure that covers the whole spectrum of determination: it reaches from the origin of objectivity to the determinacy of concrete objects while at the same time the cognitive relationship remains with itself.

5 Outlook: Post-War Kantian Transcendental Philosophy of Hans Wagner and Werner Flach Essentially, it is the same story in the transcendental philosophy of Hans Wagner (1980) and Werner Flach (1994). To solve the problem of concretization of determinacy, no Kantian schema is necessary. On the basic level of foundations, Kant’s doctrine of schematism does not play a role. As I have discussed this already in detail with regard to Rickert and Bauch, I shall conclude with some general remarks. Both Wagner and Flach develop their philosophy in terms of a validity noematics, i. e. the primacy of Rickert’s objective route. Therefore, systematically, they follow Bauch’s approach. Kant’s dualism of two sources of knowledge is, under influence of Neo-Kantianism and Richard Hönigswald, revised and transformed into a four-pillar cognitive relationship that integrates validity noematic and validity noetic aspects (Wagner 1980, 1 ff.; Flach 1994, 145 ff.). The revision follows Kant’s own idea of the primacy of the validity-noematic perspective. As in Bauch, the analysis of the content (Gehalt) takes center stage. Accordingly, the problem of actualizing, individualizing, or concretizing knowledge is dealt with in a validity-noematic fashion. Wagner solves the problem of concretization by a model of layered apriority. This model provides specific principles for concretizing knowledge, i. e. the so called “regulative” and “systematic” apriority. For the cognizing subject these are normative constraints (Wagner 1980, § 23 f.; 1992, § 9). They emerge in the course of Wagner’s validity-noematic reflection. It is a reflection on the principles of the validity of that which is thought – hence, not a reflection on “that

108

Christian Krijnen

I think” (Wagner 1992, 208 f.; 1980, §§ 4– 7).³³ Thought is conceived of as the principle of objectivity (Wagner 1980, 22 ff.; 1992, 227 ff.). As such a principle, it differentiates itself logically into that which is thought and the object of thought (subject and object, mind and world, etc.). With regard to its validity, thought cannot recur to external instances, but it is on its own and in that sense unconditioned: it contains the validity conditions of its thinking in itself. Like Rickert and Bauch, Wagner holds that any theory of knowledge that takes its point of departure for determining knowledge in the separateness of the two parts subjects and object, mind and world, or the like goes astray (Wagner 1992, 228; 1980, 192). On the level of the origin of cognition, there cannot be any radical dualism of heterogeneous factors. Rather, thought concretizes itself as a principle into thought as concrete instances. The objective validity of thoughts of cognizing subjects is assured by obeying the principles, laws, or norms of knowledge. Here is where the idea of validity-noematic levels of apriority comes into play. The principles of knowledge consist of different types, each with its own grade of fundamentality, reaching from the origin up to concrete objects and their relationships. A radical heterogeneity of parts of this relationship in need of mediation by a third factor in Wagner too does not exist. All moments within a layer of apriority as well as the layers of apriority themselves as layers of a whole cohere in the form of reciprocal implications of the parts (Wagner 1980, 175; see 194 for the regulative apriority). Accordingly, for Wagner, as he formulates it with a view to Kant, sensibility and understanding from a transcendental

 Birrer (2017, 248 f.) is right in emphasizing that schematism is about bringing together intuition (individual case) and concept (rule), but as he stresses the action of “deciding” whether a given object is in accordance with a rule, the decisive logical aspect, namely that the object is determined by the concept, and hence the concretizing function of the schema, disappears. The “argumentative contribution” of the chapter of schematism is, for Birrer, to “examine” the synthesis of understanding with respect to its “justified subsumption” (Birrer 2017, 252). As Birrer (2017, 253) himself quotes Kant’s statement that the schemata of pure understanding enable it “to subject appearances to general rules […] and thereby to make them fit for a thoroughgoing connection in one experience,” (CPR B 185) it is important to point to the difference between examining and subjecting. If my interpretation is correct, Birrer therefore needs to introduce a “double perspective” regarding transcendental schematism: “making subsumption possible” and “constituting experience” (Birrer 2017, 264 ff.). Caimi (2015, 201 f.) simply writes that schematism concerns the representation of the “subsumption of concrete objects […] Hence, categories as concepts of cognition must be applied, they are attached as predicates to concrete, singular appearances.” See for Kant’s parlance of “subsumption” as a relationship of heterogeneity (and not of subsumption of concepts) Flach (2001). Kant’s presentation of schematism in terms of ‘subsumption’ has given rise to numerous critical analyses.

Kant’s Schematism in Southwest Neo-Kantianism

109

perspective are no powers or functions that exist in themselves: they interrelate in the form of reciprocal implication.³⁴ With regard to the main line of thought of a validity-noematic foundation, Flach’s theory of knowledge, the most advanced development of the presented project of foundations of knowledge, does not differ.³⁵

 Wagner (1980, 192). Bunte (2017, 79 ff.) tries to defend Kant against Wagner’s criticism of the origin of the categories. He does so, however, by referring to a Kantian doctrine that Wagner explicitly rejects: the doctrine of “transcendental apperception”. Moreover, Bunte speaks about judging, whereas for Wagner the judgment is at stake (not the noetic dimension). An ‘I think’ and thus Kant’s apperception-theoretical emphasis of the foundations of objectivity is rejected as the highest point of transcendental philosophy. The cognitive relationship itself makes up the foundation. Whereas Wagner relates Kant’s metaphysical deduction and the relationship between judgment and categories addressed here to his transition from primary to secondary apriority, Bunte (2017, 83) holds that in Wagner the “transcendental schema” delivers this transition. This cannot be the case, already because in the course of Kant’s transcendental deduction it is shown that “all sensible intuitions stand under the categories” (CPR B 20). Thus, the transition from primary to secondary apriority has already been conducted. Wagner’s primary apriority is indifferent with respect the type of givenness (sensible, non-sensible): objectivity as such (being of entities, Sein des Seienden) is at stake here (the passage of Wagner Bunte refers to – that the principles of cognition of an object must at the same time be the principles of the object of cognition (Wagner 1980, 169) – is not only about relativizing this convertibility by primary apriority; Wagner also has given up the restriction of knowledge to “experience,” which for Kant is “essential” (Wagner 1980, 168 f.). Hence, Wagner’s secondary apriority cannot immediately concern the dimension of “self-determination of thought in the mode of time” (Bunte 2017, 84). On top of that the transcendental schema is not about the categorical determinacy of objects but about applying categories to objects of knowledge. Wagner consequently does not use Kant’s doctrine of schematism for his determination of secondary apriority, let alone for the transition from the primary to the secondary apriority.  Concerning the details, there are serious differences, linked to Flach’s attempt to advance the conception of the concretization of determinacy. Their systematical significance is high – although, regrettably, within Kant scholarship they are rather unknown. Flach, compared to Wagner, presses forward the validity-functional continuity of thought. A dualism between sensation, giving the object, and category, determining the object, does not exist in his theory. He therefore criticizes Wagner’s distinction between secondary apriority, containing the principles of being, and regulative and systematic apriority determining the process of research, which is heuristic in nature, as the sensible given, then, is conceptualized merely in its function to signify concreteness (heuristic principles, strictly speaking, do not concern the sensuous condition of Kant’s doctrine of schematism), leading to an ontological burden that fails to do justice to the spirit of Kant’s doctrine of schematism. Flach, by contrast, takes Kant’s schematism as a contribution to the method of empirical knowledge: concrete empirical knowledge is knowledge of empirical laws, containing the dimensions of observation, description, and explanation. Wagner’s function of signification and Kant’s function of schematism respectively are transformed by Flach into a function of testing (Erproben) the adequacy of an objective proposition (Sachaussage).

110

Christian Krijnen

To be sure, already from Flach’s early Prinzipienlehre der Anschauung it becomes clear that the problem of schematism belongs to the issue of concretization of knowledge. Flach discusses it in a chapter on intuition and judgment that aims to understand concretization as a structure of judgment (Flach 1963, 146 ff.). In his late Erkenntnislehre, however, schematism is no longer a topic for itself. Apparently, from the perspective of foundations, it is of a subordinated interest. Moreover, as becomes clear from the early work, it belongs to the dimension of subjectivity, that is to say to validity noetics. Concrete noematic determinacy is also for Flach only possible as individualization or concretization of validity. Concretization enriches the structure of the origin towards “meaning that is thematically bounded,” “self-mediation of infinity to finity,” and hence to a “(contingent) judgment filled with content” (Flach 1963, 146 f.). Intuition here functions as the principle of the logical individualization of validity. In this context, sensibility becomes a problem in terms of principles, a problem that Kant tried to solve in his chapter on schematism (Flach 1963, 147). According to Flach, Kant develops the concretizing function of noematic constitution. The schema is shown to be a rule that provides the concept with its image. This implies for Flach that the schema is not so much a mediating third factor between the heterogeneous category and appearances but rather the peculiar deduction structure of validity-logical individualization. Flach argues that in Kant’s doctrine of schematism the idea of a principle transforms that of a mediating third factor. The ‘categories’ turn into ‘principles’ (Grundsätze). This concretizing of knowledge takes place via the “formal-synthetical structure of the judgment,” that is to say it is conceptualized in the fashion of a validity noematics. As in Rickert or Wagner, it leads to clarifying the problem of predication in judgments. The judgment overcomes the heterogeneity of the realm of origin. Logically, a situation of heterogeneity within homogeneity has arisen (Flach 1963, 150 f.). Flach’s Erkenntnislehre discusses this in a much more differentiated way. Kant’s schematism, however, is addressed only marginally. Yet the line exposed until now is confirmed here. Kant’s schematism has its place within Flach’s “doctrine of method” (Methodologie), more in particular, it becomes a theme as soon as the “specific methods” are at issue, that is to say the “regionalization of scientific knowledge” (Flach 1994, ch. 4.4). Whereas Flach’s Logic discusses the “constitutivity of the validity noematic structure” (Flach 1994, 249), the Doctrine of Method deals with what Flach calls the “organization” of knowledge (Flach 1994, ch. 4, cit. 355), complementary to and (validity theoretically) based upon the constitutive structure of knowledge. It deals with the dimension of “regulative apriority” (Flach 1994, ch. 4.1.2). Due to its methodological nature, knowledge becomes concrete determination or determination related to a con-

Kant’s Schematism in Southwest Neo-Kantianism

111

crete issue (sachbezogene und sacherfüllte Bestimmung). It is here, specifically on the level of the regulative function of specific methods,³⁶ which ensure that cognition is determination of a specific issue, that Flach addresses Kant’s doctrine of schematism. It turns out that Kant’s subsumption of appearances under the concepts of understanding, finally under pure concepts of understanding for Flach expresses the methodical determinacy of the description of what is given by observation. On the level of appearances, the determinacy of intuition is conceptually determined (Flach 1994, 626 f.). In sum, in Flach’s theory of knowledge, Kant’s thesis of heterogeneity is completely taken back into the development of the internal relationships of the validity noematic structure of cognition. In Flach’s conception there is only the route leading from the most basic principles of any determinacy to the determined concrete object. Flach holds on to the spontaneity of thought, performing schematizations. Yet he gives up Kant’s project of conceiving of the foundations of knowledge in the fashion of a theory of apperception of the I, and hence a theory of consciousness. He gets rid of the externality (Äußerlichkeit) of the relationship between what is given and its determination. This externality is replaced, we should perhaps say sublated, by an objective order that is an order of validity-noematic constitution and regulation. Taking the sketched historical development from Neo-Kantianism to postwar transcendental philosophy into account, there appears to be agreement about the positive meaning of Kant’s doctrine of schematism. It concerns methodology.

References Allison, Henry E. (2004): Kant’s Transcendental Idealism, rev. and enl. New Haven. Bauch, Bruno (1914): Über den Begriff des Naturgesetzes, in: Kant-Studien 19, pp. 303 – 337. Bauch, Bruno (1915): Idealismus und Realismus in der Sphäre des philosophischen Kritizismus: Ein Verständigungsversuch, in: Kant-Studien 20, pp. 97 – 116. Bauch, Bruno (1923/24): Das transzendentale Subjekt: Eine transzendentalphilosophische Skizze, in: Logos 12, pp. 23 – 49. Bauch, Bruno (1923a): Immanuel Kant, 3rd ed. Berlin, Leipzig. Bauch, Bruno (1923b): Wahrheit, Wert und Wirklichkeit, Leipzig. Bauch, Bruno (1926): Die Idee, Leipzig. Bauch, Bruno (1982): Theoretische Philosophie (1931), in: H.-L. Ollig (ed.): Neukantianismus: Texte der Marburger und der Südwestdeutschen Schule, ihrer Vorläufer und Kritiker, Darmstadt, pp. 243 – 278.

 See on the doctrine of the general and specific methods Flach (1994, ch. 4.3 f.).

112

Christian Krijnen

Birrer, Mathias (2017): Kant und die Heterogenität der Erkenntnisquellen, Berlin/Boston. Bunte, Martin (2016): Erkenntnis und Funktion: Zur Vollständigkeit der Urteilstafel und Einheit des kantischen Systems, Berlin/Boston. Bunte, Martin (2017): Wissen und Bestimmtheit: Primär- und sekundärkonstitutive Apriorität bei Kant und Hans Wagner, in: C. Krijnen, K. Zeidler (ed.): Reflexion und konkrete Subjektivität: Beiträge zum 100. Geburtstag von Hans Wagner (1917 – 2000), Wien, pp. 75 – 89. Caimi, Mario (2015): Der Schematismus der reinen Verstandesbegriffe, in: E. Rainer (ed.): Kants Theorie der Erfahrung, Berlin/München/Boston, pp. 201 – 237. Cassirer, Ernst (1994a): Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit, 4 Bände, Darmstadt. Cassirer, Ernst (1994b): Kants Leben und Lehre, Nachdr. der Ausg. New Haven, Darmstadt. Cohen, Hermann (1889): Kants Begründung der Aesthetik, Berlin. Cohen, Hermann (1902): Logik der Reinen Erkenntnis, Berlin. Cohen, Hermann (1910): Kants Begründung der Ethik: Nebst ihren Anwendungen auf Recht, Religion und Geschichte, 2nd ed. Berlin. Cohen, Hermann (1918): Kants Theorie der Erfahrung, 3rd ed. Berlin. Cohn, Jonas (1908): Voraussetzungen und Ziele des Erkennens: Untersuchungen über die Grundfragen der Logik, Leipzig. Cohn, Jonas (1923a): Selbstdarstellung, in: R. Schmidt (ed.): Die Philosophie der Gegenwart in Selbstdarstellungen, 2. verb. Aufl., Leipzig, pp. 1 – 21. Cohn, Jonas (1923b): Theorie der Dialektik: Formenlehre der Philosophie, Leipzig. Cohn, Jonas (1932): Wertwissenschaft, Stuttgart. Cohn, Jonas (1949) Cohn, Jonas, in: W. Ziegenfuß (ed.): Philosophenlexikon, Bd. 1, Berlin, pp. 193 – 197. Curtius, Ernst R (1914): Das Schematismusproblem in Kants Kritik der reinen Vernunft: Eine philologische Untersuchung, in: Kant-Studien 19, pp. 338 – 366. Düsing, Klaus (1995): Schema und Einbildungskraft in Kants ‘Kritik der reinen Vernunft’, in: L. Kreimendahl (ed.): Aufklärung und Skepsis: Studien zur Philosophie und Geistesgeschichte des 17. und 18. Jahrhunderts, Stuttgart-Bad Cannstatt, pp. 47 – 71. Flach, Werner (1963): Zur Prinzipienlehre der Anschauung. Band 1: Das spekulative Grundproblem der Vereinzelung, Hamburg. Flach, Werner (1994): Grundzüge der Erkenntnislehre: Erkenntniskritik, Logik, Methodologie, Würzburg. Flach, Werner (2001): Kants Lehre von der Gesetzmäßigkeit der Empirie. Zur Argumentation der Kantischen Schematismuslehre, in: Kant-Studien 92, pp. 464 – 473. Flach, Werner (2002): Die Idee der Transzendentalphilosophie: Immanuel Kant, Würzburg. Flach, Werner (2015): Die kantische Hypothek, in: W. Bock (ed): Kant zu Geschichte, Kultur und Recht, Berlin, pp. 15 – 25. Hegel, Georg W. F. (1971): Werke in zwanzig Bänden, edited by E. Moldenhauer and K. M. Michel, Frankfurt am Main. Heidegger, Martin (1951): Kant und das Problem der Metaphysik, 2nd ed., Frankfurt am Main. Heidegger, Martin (1962): Die Frage nach dem Ding. Zu Kants Lehre von den transzendentalen Grundsätzen, Tübingen.

Kant’s Schematism in Southwest Neo-Kantianism

113

Heidemann, Dietmar (2002): Anschauung und Begriff: Ein Begründungsversuch des Stämme-Dualismus in Kants Erkenntnistheorie, in: K. Engelhard (ed.): Aufklärungen (FS K. Düsing), Berlin, pp. 65 – 90. Heinz, Marion and Krijnen, Christian (eds.) (2007): Kant im Neukantianismus: Fortschritt oder Rückschritt?, Würzburg. Höffe, Otfried (2003): Kants Kritik der reinen Vernunft: Die Grundlegung der modernen Philosophie, München. Kant, Immanuel (1998): Critique of Pure Reason, translated and edited by P. Guyer and A. W. Wood, Cambridge. Kemp Smith, Norman (1918): A Commentary on Kant’s ‘Critique of Pure Reason’, London. Krijnen, Christian (2001): Nachmetaphysischer Sinn: Eine problemgeschichtliche und systematische Studie zu den Prinzipien der Wertphilosophie Heinrich Rickerts, Würzburg. Krijnen, Christian (2007): Das konstitutionstheoretische Problem der transzendentalen Ästhetik in Kants ‘Kritik der reinen Vernunft’ und seine Aufnahme im südwestdeutschen Neukantianismus, in: M. Heinz, C. Krijnen (ed.): Kant im Neukantianismus: Fortschritt oder Rückschritt?, Würzburg, pp. 109 – 134. Krijnen, Christian (2008a): Kants Subjektstheorie und die Grundlegung einer philosophischen Anthropologie, in: Zeitschrift für philosophische Forschung 62, pp. 254 – 273. Krijnen, Christian (2008b): Philosophie als System: Prinzipientheoretische Untersuchungen zum Systemgedanken bei Hegel, im Neukantianismus und in der Gegenwartsphilosophie, Würzburg. Krijnen, Christian (2013): Geschichtsphilosophie bei Kant, im Neukantianismus und im gegenwärtigen Kantianismus, in: C. Krijnen, M. de Launay (ed.): Der Begriff der Geschichte im Marburger und südwestdeutschen Neukantianismus, Würzburg, pp. 29 – 57. Krijnen, Christian (2014a): Gegenstandskonstitution bei Husserl und in der klassischen deutschen Philosophie: Eine problemgeschichtliche Deutungslinie, in: F. Fabbianelli, S. Luft (ed.): Husserl und die klassische deutsche Philosophie: Husserl and Classical German Philosophy, Cham/Heidelberg/New York/Dordrecht/London, pp. 115 – 131. Krijnen, Christian (2014b): Transzendentaler Idealismus und empirischer Realismus, in: C. Krijnen, K. W. Zeidler (ed.): Wissenschaftsphilosophie im Neukantianismus: Ansätze – Kontroversen – Wirkungen, Würzburg, pp. 11 – 56. Lask, Emil (1923a): Die Lehre vom Urteil (1912), in E. Lask (1923c), Band II, pp. 283 – 463. Lask, Emil (1923b): Die Logik der Philosophie und die Kategorienlehre (1911), in: E. Lask (1923c), Band II, pp. 1 – 182. Lask, Emil (1923c): Gesammelte Schriften, edited by Eugen Herrigel. 3 Bände, 1 – 182. Tübingen. Nachtsheim, Stephan (2017) : La “struttura formale”: Un’ osservazione sul concetto di fondazione ultima, in: E. Lask: Un secolo dopo, edited by S. Besoli and R. Redaelli, Macerata, pp. 11 – 22. Natorp, Paul (1912): Kant und die Marburger Schule, in: Kant-Studien 17, pp. 193 – 221. Natorp, Paul (1974): Zum Gedächtnis Kants (1904), in: J. Kopper, R. Malter (ed.): Immanuel Kant zu Ehren, Frankfurt am Main, pp. 236 – 260. Pringe, Hernán (2015): Bruno Bauch and the Comprehensibility of Nature, in: Kriterion 56, pp. 355 – 370.

114

Christian Krijnen

Rickert, Heinrich (1899): Fichtes Atheismusstreit und die Kantische Philosophie, Berlin. Rickert, Heinrich (1909): Zwei Wege der Erkenntnistheorie: Transcendentalpsychologie und Transcendentallogik, in: Kant-Studien 14, pp. 169 – 228. Rickert, Heinrich (1912): Urteil und Urteilen, in: Logos 3, pp. 230 – 245. Rickert, Heinrich (1921): System der Philosophie: Erster Teil: Allgemeine Grundlegung der Philosophie, Tübingen. Rickert, Heinrich (1924/25): Alois Riehl, in: Logos 13, pp. 162 – 185. Rickert, Heinrich (1924): Das Eine, die Einheit und die Eins: Bemerkungen zur Logik des Zahlbegriffs, 2nd. umg. Aufl., Tübingen. Rickert, Heinrich (1926): Kulturwissenschaft und Naturwissenschaft, 6. u. 7. durchg. u. erg. Aufl., Tübingen. Rickert, Heinrich (1928): Der Gegenstand der Erkenntnis: Einführung in die Transzendentalphilosophie, 6. verb. Aufl., Tübingen. Rickert, Heinrich (1929): Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische Einleitung in die historischen Wissenschaften, 5. verb. u. erw. Aufl, Tübingen. Rickert, Heinrich (1930): Die Logik des Prädikats und das Problem der Ontologie, Heidelberg. Rickert, Heinrich (1934): Grundprobleme der Philosophie: Methodologie, Ontologie, Anthropologie, Tübingen. Rickert, Heinrich (1939a): Die Erkenntnis der intelligiblen Welt und das Problem der Metaphysik. Erster Teil (1927), in: H. Rickert (1939c), pp. 97 – 138. Rickert, Heinrich (1939b): Die Erkenntnis der intelligiblen Welt und das Problem der Metaphysik. Zweiter Teil (1929), in: H. Rickert (1939c), pp. 139 – 185. Rickert, Heinrich (1939c): Unmittelbarkeit und Sinndeutung: Aufsätze zur Ausgestaltung des Systems der Philosophie, edited by August Faust, Tübingen. Wagner, Hans (1980): Philosophie und Reflexion, 3rd ed., München/Basel. Wagner, Hans (1992): Die Würde des Menschen: Wesen und Normfunktion, Würzburg. Windelband, Wilhelm (1915): Vorwort (1883), in: Präludien: Aufsätze und Reden zur Philosophie und Ihrer Geschichte. 5th ed., 2 Bände, III–IV, Tübingen. Zocher, Rudolf (1954): Kants transzendentale Deduktion der Kategorien, in: Zeitschrift für philosophische Forschung 8, pp. 161 – 194. Zocher, Rudolf (1959): Kants Grundlehre, Erlangen.

Roberto Horácio de Sá Pereira

Cassirer and Kant on the Unity of Space and the Role of Imagination Abstract: The focus of this paper is Cassirer’s Neo-Kantian reading of Kant’s conception of unity of space. Cassirer’s neo-Kantian reading is largely in conformity with the mainstream of intellectualist Kant-scholars, which is unsurprising, given his own intellectualist view of space and perception and his rejection of the existence of a ‘merely sensory consciousness’ as a ‘formless mass of impression’. I argue against Cassirer’s reading by relying on a Kantian distinction first recognized by Heinrich Rickert, a neo-Kantian from the Southwest school, between Kenntnis (roughly knowledge by acquaintance) and Erkenntnis (roughly propositional knowledge). Correspondingly, I claim that concepts and categories are conditions for Erkenntnis of objects as such, namely for thinking of and apprehending the pre-existing unity as an object, rather than for the ‘constitution’ of this very unity.

1 Cassirer’s Place in Neo-Kantianism Classical neo-Kantianism officially begins with Hermann Cohen’s 1885 book, Kants Theorie der Erfahrung. Historians generally distinguish between two main schools of Neo-Kantianism. The first is the so-called Marburg School, whose great exponents are Cohen and Paul Natorp (first generation) and later Ernst Cassirer (second and last generation)¹. The second is the so-called Southwest School (also known as the Baden or Heidelberg School), whose great exponents are Wilhelm Windelband and Heinrich Rickert (first generation) and then Emil Lask (second generation)².

 It is noteworthy that although Cassirer clearly belongs to the Marburg School, nevertheless he did not limit himself to investigating the foundations of the natural sciences, but also developed the transcendental analysis of Cohen’s logic of science in the direction of a theory of culture in which science is regarded as just one among other forms of symbolization. Cassirer’s intellectual development thus shows that the transcendental analysis of the natural sciences can be naturally extended and generalized to a transcendental analysis of culture.  The sociologist Max Weber also belongs to the school. But as Beiser (2015) reminds us, Berlin also became a center of neo-Kantianism later in the 19th and early 20th centuries with Friedrich Paulsen, Alois Riehl, and Benno Erdmann. https://doi.org/10.1515/kantyb-2020-0005

116

Roberto Horácio de Sá Pereira

Otto Liebmann (1912) remains the starting point of any description of neoKantianism. He attacked Fichte, Schelling, Hegel, and Schopenhauer and their successors as speculative dogmatists and is credited with proposing the famous mantra: “zurück zu Kant!” He believed that the fundamental ideas of Kant’s philosophy had been eclipsed by Schopenhauer and the speculative metaphysics of the German absolute idealists. However, Frederick Beiser claims that neo-Kantianism traces back to the 1790s, even before Kant’s death (1804)³. Even at the end of his life, Kant’s ideas had already been distorted. Jakob Fries, Johann Herbart, and Friedrich Beneke were proto-neo-Kantian pioneers by means their repudiation of the excesses of speculative idealism. All defined themselves as Kantians, and all called for a return to the spirit of Kant’s teachings. They anticipated, and laid down the foundations for, defining the distinguishing features of classical neo-Kantianism. The first unavoidable question is, in which sense we must understand the famous mantra “zurück zu Kant!”? In other words, in which sense are neo-Kantians ‘Kantians’ or in which sense do they belong to the Kantian tradition? Windelband, for example, claims that Kant’s “immortal achievement” is the idea that the world is a product of “synthetic consciousness” (Windelband 2015). In a similar vein, Cohen claims that the “transcendental method” is at the core of Kant’s philosophy, which consists in the discovery of “the formal conditions for the possibility of experience.” The key idea can be traced back to Kant’s “Copernican Revolution” whereby he claims, “that we know a priori of objects only what we put into them” (CPR B XVIII). Instead of regarding a priori knowledge as dependent on the object, neo-Kantians propose that we think of the object as dependent on conditions under which such a priori knowledge is possible, namely as a logical construction. And their reason for proposing this is that the object is nothing more than a logical construct of their “transcendental method.”⁴ Given this, the two main schools of classical neo-Kantianism are divided by two distinct concerns rather than by two different methods or views about the nature of philosophy. The Marburg School is concerned with the logic and foundations of the natural sciences. In contrast, the main concern of the Southwest School is

 See Beiser (2015, 3).  Two examples (chosen randomly) illustrate the point. In his main work of 1910, Cassirer states: “Physical concepts are valid not insofar as they reproduce a given rigid being but insofar as they comprise a project for possible postulations of unity” (1910, 427). In a similar vein, Windelband adds that the idea is of rejecting the commonsense view that the world is something external to us that we can at best hope to represent or “mirror” in consciousness. Instead, “the world that we experience is our deed” (2015, 318).

Cassirer and Kant on the Unity of Space and the Role of Imagination

117

with the investigation of the theory of value and the foundations of the cultural sciences. The second pressing question is in which sense they are ‘neo-Kantians’ rather than Kantians simpliciter. In this regard, there is a relative consensus in the secondary literature. To start with, every scholar agrees that the neo-Kantians emphatically reject the assumption that “synthetic consciousness” is to be understood in psychological terms. Neo-Kantians reject all forms of psychologism. However oddly it may sound today, the idea is that “synthetic consciousness” is not a mental achievement of the conceptual operations of the human cognitive apparatus. A much more important distancing from Kant concerns the rejection and reinterpretation of Kantian notions of the ‘thing in itself’. For Kant, the thing in itself, or noumenon in the negative sense, is essentially the unknown object as it exists independently of the mind, affects our sensibility, and hence manifests to us humans in space and time as appearances or mere representations. Neo-Kantians have two distinct but not necessarily exclusive reactions. Tacitly assuming a two-world view (or two-objects view) reading of Kantian transcendental idealism, several neo-Kantians simply reject the existence of noumena insofar as they reject the existence of “another world” as something useless.⁵ Windelband, for example, says that “the thing-in-itself [is] incoherent and unnecessary” (Windelband 1911, 323). Yet, even rejecting noumena as a separate reality, several other neo-Kantians understand noumena (in the negative sense) charitably as the attempt to trace the bounds of possible knowledge. We need a negative notion of noumena to remind us that scientific research never ends. Yet, the third and more important distancing from Kant concerns the topic of this paper, namely the Kantian dualism between sensible intuitions and discursive concepts, or between sensibility and understanding. As every Kantian philosopher knows, this duality lies at the very heart of the Critique. ⁶ The majority of Marburg neo-Kantians reject Kantian dualism simpliciter. ⁷ More precisely, they reject the Kantian distinction between understanding and what they call “pure intuition”, that is, a seeing that is not cognitively penetrated by intellectual pow-

 See Windelband (2015, 323). See also Windelband (1892; 1894; 1900; 1911; 1915; 1956; 1958a; 1958b; 1980).  Strawson put it in this way: “the duality of intuitions and concepts is in fact but one form or aspect of the duality which must be recognized in any philosophy which is concerned with human knowledge, its object, its expression or communication. These are three directions of the same philosophical concern rather than three different concerns” (1966, 47).  In this regard, Sellars (1968) and McDowell (1994) consciously or unconsciously follow the path traced by the Marburg school.

118

Roberto Horácio de Sá Pereira

ers. Again, they conceive objects as merely logical constructions by means of the transcendental method (the discovery of the formal conditions of experience). In contrast, the Southwest Neo-Kantians accept the duality between intuitions and concepts, but also reject the idea that intuitions are the result of things-in-themselves affecting sensibility. Moreover, they also hold that there is no pure seeing that is not cognitively penetrated by intellectual faculties.⁸ The fourth and equally important distancing from Kant is a direct consequence of the third, namely the rejection of the Kantian notion of ‘the given’.⁹ According to the members of the Marburg School, what is given to thought is already synthesized and as such already takes a propositional form.¹⁰ Therefore, neo-Kantians must embrace some form of epistemic holism or coherentism in regard of perception.¹¹ To use Rickert’s distinction (Rickert 2015), what they reject is Kenntnis (acquaintance). The ‘given’ always takes the form of Erkenntnis, that is, a propositional knowledge of truths. The fifth distancing from Kant is also noteworthy. From the neo-Kantian perspective, Kant makes a basic mistake in conceiving of knowledge in terms of the ‘representation’ of reality. Against this, members of both schools of Neo-Kantianism insist that knowledge is formative or creative, rather than representative in character, meaning that knowledge does not simply ‘copy’ an externally subsisting reality, but instead actively shapes the objects of the reality with which it is concerned. In the light of all of these distancings, it becomes quite clear that one of the distinguishing marks of Neo-Kantianism is what we today call “conceptualism”

 In this regard, Rickert has an insightful view. He recognizes the key Kantian opposition between Kenntnis (knowledge by acquaintance) and Erkenntnis (propositional knowledge), assuming that Kenntnis results from the intuition of particulars: “the task of Kenntnis is evidently to be viewed as that which has reflected as faithfully as possible what is given to us in intuition” (2015, 389). However, he accuses Martin Heidegger, his own disciple, of attempting to reduce Erkenntnis to mere Kenntnis, as if only Kenntnis could capture the essence of things. That is what he calls “theoretical intuitionism” (2015).  In this sense, the neo-Kantians anticipate Sellars (1968) and McDowell’s (1994) criticism of the so-called “myth of the given”.  Cohen, for example, claimed that thinking accepts nothing as given, but rather discovers in each generation. Taking these cues from the study of the history of differential calculus, Cohen ends up arguing that each fact is generated by thought and determined by its position in a logically necessary system. Regarding this, see Natorp (1882a; 1882b; 1887; 1888; 1894; 1901; 1904a; 1904b; 1910; 1912a; 1912b).  See Cassirer (1957, 11).

Cassirer and Kant on the Unity of Space and the Role of Imagination

119

or “intellectualism”.¹² In this regard, it is noteworthy, however, that neo-Kantians never charged Kant for not being an intellectualist. Instead, their implicit charge is that Kant is not intellectualist enough. We can easily locate Cassirer’s great contributions within this broad framework. As a member of the second generation of the Marburg School, Cassirer distances himself from Kant in all five of the ways I just described. He rejects Kant’s mentalism, Kant’s notion of noumena, Kantian dualism between intuitions and concepts,¹³ Kant’s notion of ‘the given’, and the Kantian conception of knowledge as representation rather than formation and creation. In a nutshell, Cassirer’s neo-Kantianism is by far the best illustration of conceptualism or intellectualism in philosophy of mind and epistemology, namely, as a top-down account of knowledge according to which objects are ‘constituted’ by the ‘transcendental method’. From a systematic viewpoint, the evidence favoring non-intellectualism over intellectualism is overwhelming.¹⁴ But that is not my concern here. I have two basic reasons for this. First, that would require another paper, leading me far afield.¹⁵ Second, it makes little sense (at least to me) to criticize neo-Kantianism

 McLear suggests the word “sensiblism” in opposition to intellectualism. “Sensiblism” is the claim that sensible intuition places us (or represents or presents) object independently of the intervention of concepts. Intellectualism, in contrast, is the claim that our intentional relation with objects crucially depends on conceptualization of the given. See McLear (2011; 2016). I use both labels ‘nonconceptualism’ and ‘non – intellectualism’ interchangeably but meaning what McLear has in mind.  Cassirer’s view is that there is no “merely sensory consciousness, that is, a consciousness remaining outside of any determination by the theoretical functions of signification and preceding them as an independent datum” (1957, 8). He also goes further and insists that even space and time are on the same footing as the synthetic functions generally attributed to thought, or the understanding. Rather than passively given “intuitions”, they are parts of the active, cognitive process of determination from which the object of scientific experience is born. See Cassirer (1957, 182; 412; 433; 439).  On the phylogenetic scale, I assume as an empirical fact that genuinely perceptual systems appear in animal species well before belief and propositional attitudes. Although bees, frogs, pigeons, goldfish, and octopi, for example, lack beliefs and, hence, lack demonstrative concepts and singular thoughts, they have low-level visual representational systems. In fact, the best explanations of some of these low-level representational systems attribute perceptions of physical objects in space and rudimentary properties.  In regard to Kantian nonconceptualism, see my papers (2001; 2004; 2010; 2012; 2013; 2014; 2015; 2016a; 2016b; 2017a; 2017b; 2018; 2019). Hanna has listed at least seven compelling pieces of evidence against intellectualism. Here they are: (1) From infant and non-human animal cognition: Normal infants and some non-human animals are capable of perceptual cognition but lack possession of concepts. Therefore, normal infants and some non-humans are capable of non-conceptual cognition with non-conceptual con-

120

Roberto Horácio de Sá Pereira

in general and Cassirer in particular from a systematic viewpoint since neo-Kantianism is one of the great milestones in the history of contemporary continental philosophy and Cassirer is certainly its most influential and important exponent. So, what is the aim of this essay? Even when distancing himself from Kant on many different points, Cassirer never ceased to be a reader of Kant. Given this, I’ve selected as the focus of this paper Cassirer’s reading of Kant’s conception of unity of space and the role of imagination. Cassirer’s neo-Kantian reading is largely in conformity with the mainstream of intellectualist Kant-scholars, which unsurprising, given his own intellectualist view of space and perception and his rejection of the existence of a ‘merely sensory consciousness’ as a ‘formless mass of impression’. I argue against Cassirer’s reading by relying on a Kantian distinction first recognized by Rickert, between Kenntnis (roughly knowledge by acquaintance) and Erkenntnis (roughly propositional knowledge). Corre-

tent. (2) From phenomenological fineness of grain: Our normal human perceptual experience is so replete with phenomenal characters and qualities that we could not possibly possess a conceptual repertoire extensive enough to capture them. Therefore, normal human perceptual experience is always to some extent non-conceptual and has non-conceptual content (See Sect. III). (3) From perceptual discrimination: It is possible for normal human cognizers to be capable of perceptual discriminations without also being capable of re-identifying the objects discriminated. But re-identification is a necessary condition of concept-possession. Therefore, normal human cognizers are capable of non-conceptual cognitions with non-conceptual content (See Sect. III). (4) From the distinction between perception (or experience) and judgment (or thought): It is possible for normal human cognizers to perceive something without also making a judgment about it. But non-judgmental cognition is non-conceptual. Therefore, normal human cognizers are capable of non-conceptual perceptions with non-conceptual content. (5) From the knowing-how versus knowing-that (or knowing-what) distinction: It is possible for normal human subjects to know how to do something without being able to know that one is doing it and without knowing precisely what it is one is doing. But cognition that lacks knowing-that and knowing-what is non-conceptual. Therefore, normal human subjects are capable of non-conceptual knowledge-how with non-conceptual content. (6) From the theory of concept-acquisition: The best overall theory of concept-acquisition includes the thesis that simple concepts are acquired by normal human cognizers on the basis of non-conceptual perception of the objects falling under these concepts. Therefore, normal human cognizers are capable of non-conceptual perception with non-conceptual content. (7) From the theory of demonstratives: The best overall theory of the demonstratives ‘this’ and ‘that’ includes the thesis that demonstrative reference is fixed perceptually, essentially indexically, and therefore non-descriptively by normal human speakers. But essentially indexical, nondescriptive perception is non-conceptual. Therefore, normal human speakers are capable of non-conceptual perception with non-conceptual content (Hanna 2008, 44). See also Hanna (2006; 2008; 2011; 2015).

Cassirer and Kant on the Unity of Space and the Role of Imagination

121

spondingly, I claim that concepts and categories are conditions for Erkenntnis of objects as such, namely for thinking of and apprehending the pre-existing unity as an object, rather than for the ‘constitution’ of this very unity. In the first step of the B-Deduction, Kant proves that we cannot think of something given to a sensible intuition in general as an object without categories, while in the second step he proves that without categories we cannot apprehend the pre-existent unity of space as an object. Even though nonconceptualism or anti-intellectualism is a controversial claim as a reading of Kant’s philosophy, I am not going to defend it directly: the textual evidence favoring the nonconceptual reading of Kant is overwhelming.¹⁶ My support of the nonconceptualist reading here is a case of the inference-to-the-bestexplanation. The novelty of this paper is that the nonconceptualist reading provides the solution of the second step of the B-Deduction (and what makes the ADeduction clearer). How will I proceed? After this historical digression, where I have tried to scrutinize the relations between Kantians and Neo-Kantians and locate Cassirer’s contribution, the remainder of this paper is divided into three major sections. The first one is devoted to an exposition of Cassirer’s conceptualist reading of Kant’s theory of intuition. From the viewpoint of a nonconceptualist reader, every statement by Cassirer about the relation between intuition and concepts is unacceptable, in particular his reading of metaphysical deduction. Here I take the side of the nonconceptualist reader.

 Hanna (2011) enlisted the following key passages: “Objects can indeed appear to us without necessarily having to be related to functions of the understanding. (CPR A 89/B 122; emphasis added) Appearances can certainly be given in intuition without functions of the understanding. (CPR A 90/B 122; emphasis added) Appearances might very well be so constituted that the understanding would not find them in accordance with the conditions of its unity. […] [and] in the series of appearances nothing would present itself that would yield a rule of synthesis and so correspond to the concept of cause and effect, so that this concept would be entirely empty, null, and meaningless. Appearances would none the less present objects to our intuition, since intuition by no means requires the functions of thought. (CPR A 90 – 1/B 122 – 23; emphasis added) That representation which can be given prior to all thinking is called intuition. (CPR B 132) The manifold for intuition must already be given prior to the synthesis of the understanding and independently from it. (CPR B 145; emphasis added) Concept differs from intuition by virtue of the fact that all intuition is singular. He who sees his first tree does not know what it is that he sees.” (V-Lo/Wiener AA 24: 905). See also Hanna (2011, 404– 403).

122

Roberto Horácio de Sá Pereira

However, the real bone of contention is Kant’s concept of unity of representations.¹⁷ The second section is devoted to refuting Cassirer’s reading.

2 Cassirer’s Intellectualist Reading of Kant Cassirer’s account of experience can be divided into two complementary phases. In his influential book (1910), Substance and Function, he understands the Kantian notion of experience in the sense of scientific experience. Here, he clearly follows Cohen’s interpretation of the first Critique, Theorie der Erfahrung. Accordingly, perception is seen as nothing more than observation in scientific experiments. In this regard, in Substance and Function, Cassirer assumes a hyper-intellectualism. First, observations are theory laden. An observation in scientific experiments can only have a determinate content if the relevant concepts of a theory (and the inferential relations between them) are represented in the particular observational content in question. But the key claim is the following. There is no duality between concepts and observations. Instead, concepts must be seen as constitutive elements of what is itself given – of what the very observation in question consists in. In the second phase of Cassirer’s argument, that is, in the third volume of his Philosophy of Symbolic Forms, Cassirer extended this hyper-intellectualist view of sense experience to his account of perception outside science. Consider this: For there is no seeing and nothing visible which does not stand in some mode of spiritual vision, of ideation. A seeing and a thing-seen outside of this “sight,” a “bare” sensation preceding all formation, is an empty abstraction. The “given” must always be taken in a definite aspect and so apprehended, for it is this aspect that first lends it meaning. This meaning is to be understood neither as secondary and conceptual nor as an associative addition: rather, it is the simple meaning of the original intuition itself. (Cassirer 1957, 134)¹⁸

 In view of space limitations, I cannot consider Kant’s conception of the ‘given’ and Cassirer’s reading of it.  Cassirer’s claim gains some plausibility when we consider what is today called the hypothesis of cognitive penetration. The hypothesis that perception is cognitively penetrable holds that cognitive states such as beliefs, desires, and possibly other states can causally influence perceptual processing in such a way that they end up determining subjects’ perceptual contents or experiences. The philosophical significance of penetrability is easy to grasp: if perception is cognitively penetrable, then what we think literally affects how we see the world. But as I said, my concern in this paper is not primarily Cassirer’s own philosophy, but rather Cassirer’s intellectualist reading of Kant, in particular his reading of the key Kantian concept of a unity of representations in B-Deductions.

Cassirer and Kant on the Unity of Space and the Role of Imagination

123

The fundamental idea is again quite clear. First, there is no seeing without concepts. We can see something only with ‘the eyes of the mind’. A bare sensation is nothing but an empty abstraction. Therefore, there is no given simpliciter. What we can be ‘given’ is already something sensible but conceptually determined. There is no intuition without concepts. The assumption that we could intuit without concepts is an “empty abstraction” or, to put it again in Sellars’s words, a myth, the so-called “myth of the given” (Sellars 1968). Thus, the extension of Cassirer’s hyper-intellectualism to the non-scientific field of perception in the third volume of his Philosophy of Symbolic Forms is this: Applied to the problem of perception, this means that where we are concerned with distinguishing the world of prescientific consciousness from the constructive determinations of scientific cognition, we may look on perception itself as something relatively simple and immediate. In relation to these constructive determinations, it may appear as a simple datum, as something “given in advance.” But this by no means deprives us of the possibility, or relieves us of the obligation, of recognizing it in another context as something thoroughly mediated and conditioned. (Cassirer 1957, 10)

His idea that perception is deprived of conceptual determination is illusory. Even in the non-scientific field of perception there is always conceptual determination. Now let us turn to Cassirer as a reader of Kant. First, he rightly claims that the objective Deduction is not only concerned with the foundation of empirical ordinary knowledge, but it “is also directed essentially toward the form of objective knowledge as we find it in the exact sciences, aiming at those principles through which the mere rhapsody of perceptions becomes a tightly enclosed unity, a system of empirical knowledge” (Cassirer 1957, 7). Yet, quoting the key passage of the metaphysical and transcendental B-Deduction (CPR B 131– 32)¹⁹, he claims that: The ‘I think’, the expression of pure apperception, must be able to accompany all my representations: “for otherwise something would be represented in me which could not be thought; in other words, the representation would either be impossible, or at least be, in relation to me, nothing.” (Cassirer 1957, 8)²⁰

 B 131– 32 echoes the famous Kantian slogan: without thoughts or concepts, intuitions are blind (A 51/B 75).  Kant’s original texts are a little different: “The I think must be able to accompany all my representations; for otherwise something would be represented in me that could not be thought at all, which is as much as to say that the representation would either be impossible or else at least would be nothing for me.” (B 131– 32).

124

Roberto Horácio de Sá Pereira

It seems obvious to me that on B 131– 32, Kant is not claiming that without the ‘I think’ I would not represent any object, but rather that this representation would be nothing for me, that is, a blind intuition that would not amount to cognition (Erkenntnis). Quite a strong piece of evidence for this is the sentence that follows Cassirer’s quote: “that representation that can be given prior to all thinking is called intuition” (CPR B 132). Likewise, by far the most natural and reasonable reading of the metaphysical deduction is the idea that we derive the categories of understanding when we think of the unity of representations in judgment as the reflection of the pre-existent unity of the manifold of sensible representations. For example, when “I carry a body, I feel the pressure of its weight”. Now when I think of this unity between the particular body and the property “weight,” I think “it, the body, is heavy” (CPR B 142); that is, I think of it on the basis of the categorical judgment S is P. What emerges is the category of substance: an object with one or several properties. However, in support of his intellectualist reading of the metaphysical deduction Cassirer quotes the troublesome text that constitutes the second step of the B-Deduction: When, then, for example, I make the empirical intuition of a house by apprehension of the manifold contained therein into a perception, the necessary unity of space and of my external sensuous intuition lies at the foundation of this act, and I, as it were, draw the form of the house conformably to this synthetical unity of the manifold in space. But this very synthetical unity remains, even when I abstract the form of space, and has its seat in the understanding, and is in fact the category of the synthesis of the homogeneous in an intuition; that is to say, the category of quantity, to which the aforesaid synthesis of apprehension, that is, the perception, must be completely conformable. (Cassirer 1957, 11– 12)²¹

Yet, this passage echoes back to the even more troublesome footnote at B 160: Space, represented as object (as is really required in geometry), contains more than the mere form of intuition, namely the comprehension of the manifold given in accordance with the form of sensibility in an intuitive representation, so that the form of intuition merely gives the manifold, but the formal intuition gives unity of the representation. In the Aes-

 Kant’s original text: “Thus if, e. g., I make the empirical intuition of a house into perception through apprehension of its manifold, my ground is the necessary unity of space and of outer sensible intuition in general, and I as it were draw its shape in agreement with this synthetic unity of the manifold in space. This very same synthetic unity, however, if I abstract from the form of space, has its seat in the understanding, and is the category of the synthesis of the homogeneous in an intuition in general, i. e., the category of quantity with which that synthesis of apprehension, i. e., the perception, must therefore be in thoroughgoing agreement” (CPR B 162).

Cassirer and Kant on the Unity of Space and the Role of Imagination

125

thetic I ascribed this unity merely to sensibility, only in order to note that it precedes all concepts, though to be sure it presupposes a synthesis, which does not belong to the senses but through which all concepts of space and time first become possible. For since through it (as the understanding determines the sensibility) space or time are first given as intuitions, the unity of this a priori intuition belongs to space and time, and not to the concept of the understanding (Sect. 24) (CPR B 160n; original emphasis in bold)

In this troublesome footnote, Kant reminds us that in his Aesthetic he claimed that unity of space and time precedes all discursive concepts, including the discursive concepts SPACE and TIME, as the form of the sensible intuition. Yet, he now adds that the unity of space and time presupposes a synthesis that cannot be given by the senses. The product of this synthesis is what he calls a ‘formal intuition’, that is, the result of the determination of the sensibility by the understanding. The question is: does this ‘formal intuition’ mean the same as the previous ‘pure intuition’? Has Kant changed his mind in-between the two editions of the Critique? Cassirer’s view follows the mainstream of intellectualist readers of Kant in this regard: For it is now the function of knowledge to build up and constitute the object, not as an absolute object but as a phenomenal object, conditioned by this very function. What we call objective being, what we call the object of experience, is itself only possible if we presuppose the understanding and its a priori functions of unity. We say then that we know the object when we have achieved synthetic unity in the manifold of intuition. (Cassirer 1957, 5; emphasis added) […] without the synthesis of apprehension, reproduction, and recognition, we should have neither a perceiving nor a thinking ego – there would be an object neither of pure thought nor of empirical perception. (Cassirer 1957, 8; emphasis added)

Now, by quoting this passage from B 162 of the B-Deduction, Cassirer is claiming that the unity of space underlies the synthesis of the homogeneous according to the category of quantity. According to this intellectualist reading, there is no difference between ‘formal intuition’ and ‘pure intuition’. In both cases, without the unity performed by the apperception according to concepts, the world of perception would be a “mere formless mass of impression” (Cassirer 1957, 8). The careful reader must remember that in the Transcendental Aesthetic, Kant not only claims that space and time are the forms of sensible intuition. He also claims to have proven that space and time are pure intuitions; that is, they are not only the form of what appears to our outer and inner sense, but also immediate and singular representations of space (CPR A 25/B 39) and of time (CPR A 32/B 47), that is, immediate and singular representations of the spatiotemporal forms. In the particular case of space, Kant quite clearly claims that without

126

Roberto Horácio de Sá Pereira

any concepts whatsoever, including the concept of space, we are already able to represent an ‘‘infinite magnitude’’ (CPR B 40), the object of our outer sense. The pure intuition of space is a paradigmatic case of nonconceptual content: without the category of quantity or any other spatial concept whatsoever, the subject is able to represent an infinite magnitude, but unable to represent it as an infinite magnitude, that is, without recognizing what ‘‘an infinite magnitude’’ means in discursive terms. Kant goes beyond this and wonders how such pure intuitions are possible. It is at this moment that he introduces a further crucial concept: forms of human sensibility. We can only immediately represent a priori the forms of what appears to our outer sense and inner sense because those forms of appearances lie a priori in us as formal constitutions of our human sensibility (CPR B 41). Be that as it may, the intellectualist/conceptualist reader has the onus of explaining why Kant characterizes the Deduction as unavoidable as follows: Objects can indeed appear to us without necessarily having to be related to functions of the understanding. (CPR A 89/B 122; emphasis added) Appearances could after all be so constituted that the understanding would not find them in accord with the conditions of its unity. […] [and] in the succession of appearances nothing would offer itself that would furnish a rule of synthesis and thus correspond to the concept of cause and effect, so that this concept would be entirely empty, nugatory, and without significance. Appearances would nonetheless offer objects to our intuition, for intuition by no means requires the function of thinking. (A 90 – 1/B 122 – 23; emphasis added)

To begin with, a distinction must be made here between strong and weak forms of conceptualism. According to the first, sensible intuitions already involve concepts, that is, there is nothing sensible that is not conceptually determined. In contrast, according to the second, there are sensible intuitions without concepts. However, without concepts, sensible intuitions are nothing more than a manifold of sensations without reference and devoid of representational content. While the Southwest School seems to embody a weak form of intellectualism, the Marburg School clearly endorses a strong form of intellectualism. Yet, in this regard, Cassirer seems to hesitate. On one hand, he claims that there is no seeing without the eyes of the intellect, clearly assuming a strong intellectualist view and reading of Kant’s philosophy. On the other hand, he seems to admit that the existence of a “mere formless mass of impressions” (Cassirer 1957, 8) is at least conceivable. The question now is: what does Cassirer have in mind with a “mere formless mass of impressions” (Cassirer 1957, 8)? In his own words:

Cassirer and Kant on the Unity of Space and the Role of Imagination

127

For the reality of the phenomenon cannot be separated from its representative function; it ceases to be the same as soon as it signifies something different, as soon as it points to another total complex as its background. It is mere abstraction to attempt to detach the phenomenon from this involvement, to apprehend it as an independent something outside of and preceding any function of indication. For the naked core of mere sensation, which merely is (without representing anything), never exists in the actual consciousness; if it exists at all, it is the prime example of that illusion which William James called ‘the psychologist’s fallacy’. (Cassirer 1957, 141)

The ‘mere formless mass of impressions’ is a case of what James called the psychological illusion, namely the fallacy, of reading into the mind he is examining what is true of his own; especially of reading into lower minds what is true of higher ones. That entitles us to assume that for Cassirer, Kant’s statements on A 90 – 1/B 122– 23 show that he’s not contemplating a real metaphysical possibility (nonconceptualism), but only an epistemic possibility (namely that the world could appear to us as a “mere formless mass of impression” without categories) that must be excluded at the end of the B-Deduction. Without this assumption the intellectualist cannot make sense of what Kant states on A90 – 1/B122– 23, let alone of the second step of the B-Deduction.²² Given this, the main role of the Deduction is to show that even if it is conceivable that objects can appear to us without necessarily having to be related to functions of the understanding (the existence of a “mere formless mass of impression” (Cassirer 1957, 8), this epistemic possibility is only a conjecture to be ruled out in the Deduction. That echoes Allison’s reading of the Deduction (Allison 2015). Allison’s assumption here is that without categories, our experience would undergo a radical phenomenological change. It would be reduced to a “mere formless mass of impressions” or, to use the famous words of William James, to a great blooming, buzzing confusion.²³ Why does Allison think so? Because as a conceptualist, he truly believes that the understanding is not only the power that makes us understand what is given to our senses but also the power that makes us understand that what we intuit and perceive exists mind-independently as an object. As the rule-giver for a synthesis of imagination, the understanding is also the power of

 According to Gomes, Kant is contemplating “a mere epistemic possibility to be eliminated later (in the Deduction) as an unreal metaphysical possibility” (Gomes 2014, 6). Gomes reminds us (2014, 6) that Kant uses the indicative “can” (können) in the formulation on A 89/B 122, as opposed to the subjunctive “could” (könnten) on A 90 – 1/B 122– 23. The first is a stark hint signaling that he takes the possibility of objects appearing without categories as real, while the second is a mere epistemic possibility to be eliminated later.  That is exactly what Strawson (1966) called the “sense-datum theory” or “sense-datum hypothesis”.

128

Roberto Horácio de Sá Pereira

creating intentional objects out of the chaotic sensory manifold given to our senses. It is as if the unification of the manifold of sensory states in accordance with rules were a real mental act that assembles the pieces of a puzzle to form a picture of reality. However, the intellectualist hypothesis that without concepts our cognitive life would be reduced to a “mere formless mass of impression” lacks textual support. There are only a few passages in the A-Deduction that could, when misread, suggest Allison’s skeptical scenario. One of them is Kant’s statement on A 107 that “inner perception is empirical and forever variable”. But this certainly does not mean that without apperception and categories our introspective selfknowledge would be a chaotic manifold of sensory states. Nevertheless, the most important passage is this one: Unity of synthesis in accordance with empirical concepts would be entirely contingent, and, were it not grounded on a transcendental ground of unity, it would be possible for a swarm of appearances (ein Gewühle von Erscheinungen) to fill up our soul without experience (Erfahrung) ever being able to arise from it. But in that case all relation of cognition (Erkenntnis) to objects also disappear, since the appearances would lack connection in accordance with universal and necessary laws, and would thus be intuition without thought, but never cognition (Erkenntnis), and would therefore be as good as nothing for us. (CPR A 111; emphasis added)

On a closer look, however, Kant’s swarm of appearances is not Cassirer’s “mere formless mass of impressions” devoid of representational content. Kant is clearly assuming that a swarm of appearances can fill up our souls, that is, that objects can appear to our senses without Erfahrung and Erkenntnis. Cassirer’s mistake is to take Erkenntnis to mean mere representations of objects. Instead, ‘Erkenntnis’ is a technical term.²⁴ Erkenntnis is neither the representation of objects nor the representation of mind-independent particulars. Instead, it is the realization that what we represent nonconceptually and mind-independently by the senses in fact exists mind-independently.

 See Rickert (1888; 1896; 1899a; 1899b; 1900; 1902; 1907; 1909; 1910; 1911; 1913; 1914; 1921; 1924a; 1924b; 1926; 1928; 2015).

Cassirer and Kant on the Unity of Space and the Role of Imagination

129

3 The Nonconceptualist reading of the B-Deduction Let me now provide you with a sketch of the B-Deduction in light of my nonconceptualist reading of Kant. The starting point must be the exact statements on A 89/B 122 and A 90 – 1/B 122– 23 taken as a real metaphysical possibility. However, independently of any concepts, we do not experience a “mere formless mass of impressions” devoid of reference. Instead, we mind-dependently directly represent (kennen) mind-independent particulars, albeit without knowing that they exist mind-independently. The first thing to note is that if this really is Kant’s starting point, Cassirer’s assumption that on A 89/B 122 and A 90 – 1/B 122– 23 Kant is not contemplating an epistemic possibility to be ruled out later makes littles sense. Moreover, since according to nonconceptualism we do possess direct access to objects, it makes little sense to assume that for Kant the relation to object is always mediated by a synthesis according to concepts. Let us remember why the Deduction seemed necessary to Kant. Since we do not possess an intellectus archetypus, there is no direct link between the categories of the understanding and our sensible intuition: the understanding cannot create an object, which means that its concepts can be empty. Likewise, sensibility cannot make sense of what it represents, which means that it can represent blindly. That is the problem with the Deduction: since categories are not conditions of the nonconceptual representation of objects by sensible intuition and vice-versa, how can we prove that categories apply to the object nonconceptually represented by the senses? How can we prove that objects nonconceptually represented by the senses fall under categories? In the face of the heterogeneity of intuitions and concepts, the transcendental Deduction requires a tertium quid (third or intermediate term) that links categories to the appearances of the sensibility: cognition (Erkenntnis) or the possible experience of objects. In the first step of the B-Deduction, this tertium quid first assumes the form of the transcendental apperception. If mind-independent objects could indeed be represented nonconceptually by sensible intuitions without necessarily having to be related to functions of the understanding, then “something would be represented in me that could not be thought at all, which is as much as to say that the representation would either be impossible or else at least would be nothing for me”(CPR B 132). But the pressing question is: why does the B-Deduction need a second step? Let us take a further look at the passage of §24:

130

Roberto Horácio de Sá Pereira

The pure concepts of the understanding are related through the mere understanding to objects of intuition in general, without it being determined whether this intuition is our own or some other but still sensible one, but they are on this account mere forms of thought, through which no determinate object is yet cognized. (CPR B 150; emphasis added)

In the first step of the B-Deduction, the tertium quid that links the categories of the understanding to objects nonconceptually represented by the senses is the propositional thought that those objects represented by the senses exist objectively or mind-independently. Thus, concluding the first step of the B-Deduction, I could think of those objects represented by the senses only as existing objectively or mind-independently, e. g., by judging that bodies are heavy if I think of them according to categories. For, as Kant put it in his Prolegomena, categories are just “the condition for determining judgments as objectively valid” (Prol, §39, AA 04: 324). According to the example provided by Kant, my categorical judgment that bodies are heavy can only be objectively true or false if I think of bodies as material substances in space and heaviness as one of their properties (CPR B 142). Until now, Kant has proven (if anything) that the nonconceptually directly represented objects of a sensible intuition in general must fall under categories whenever I think about them and make judgments about them. Now, according to Kant, the new tertium quid is the so-called figurative synthesis or synthesis speciosa “as an effect of the understanding on the sensibility” (CPR B 154). Now, if intellectualists have the problem of making sense of Kant’s statement on A 90 – 1/B 122 – 23, the non-intellectualist finds an insuperable obstacle in the role of the imagination in the B-Deduction. For one thing, for nonintellectualist readers, every synthesis of Kant’s must be nonconceptual and as such independent from categories. Indeed, in the transcendental B-Deduction, Kant places imagination under the control of the understanding. In § 24 Kant refers to a pure and a priori synthesis of sense representations, which he names figurative synthesis, or alternatively, transcendental synthesis of the imagination, and states that this synthesis of the imagination is an “effect” (Wirkung) resulting from the action of the understanding on sensibility, meaning with this that the transcendental synthesis necessarily conforms to concepts of the understanding (CPR B 152). Commenting on the parallel passage of the A-Deduction, Cassirer is clear in this regard: Yet this transcendental function of the imagination is not grasped in its core even where an attempt is made to reduce it to apperceptive rather than mere reproductive processes. True, this seems to constitute a decisive step beyond any sensationalist foundations, for apperception not only signifies the apprehension and subsequent synthesis of given impressions but also represents a pure spontaneity, a creative act of the spirit. (Cassirer 1957, 159)

Cassirer and Kant on the Unity of Space and the Role of Imagination

131

Moreover, even in the metaphysical deduction, Kant seems already to subordinate the syntheses of imagination to the understanding. In this regard, Cassirer states: The same action which imparts unity to the different representations in a logical judgment also gives to the mere synthesis of different representations in an intuition a unity which, expressed in general terms, is the pure concept of the understanding. The categories on which the system of mathematical and physical cognition is founded are accordingly the same as those on which our concept of the natural world rests […] This right is grounded in the supposition that every synthesis – even the synthesis which makes objective perception, the perception of ‘something’ possible – is subordinated to the pure concepts of the understanding. (Cassirer 1957, 11; emphasis added)²⁵

If Cassirer is right, then (i) Kant’s doctrine on imagination in the first edition does not seem to be coherent with his doctrine in the second edition, and (ii) even in the first edition, there seems to be a contradiction between the metaphysical and the transcendental Deductions: while in first Kant claims, according to Cassirer, that every synthesis is subordinate to categories, in the A-Deduction he seems to admit the possibility of apprehension and reproduction without recognition through concepts.²⁶ The easiest way to deal with these incongruent scenarios consists in claiming that Kant simply realized that his A-Deduction was a mistake: there is no apprehension without concepts. Yet, for some reason he missed the opportunity to revise the metaphysical deduction. However, the revision had to go deeper. If the claim that objects can appear to us in space and time without necessarily having to be related to functions of the understanding is metaphysically impossible, why did he write his Transcendental Aesthetic? Longuenesse (1998) is the only conceptualist reader who is coherent in this respect. She clearly sees that if Kant’s statements on A 89/B 122 and A 90 – 1/B 122– 23 are metaphysically impossible, we face the challenge of rereading the Transcendental Aesthetic. Now, considering that Kant rewrote his Deduction several times and his Refutation dozens of times, why did he never rewrite his Transcendental Aesthetic? Pace Longuenesse (1998), any reading of the second step of the B-Deduction that entails a rewriting of the Transcendental Aesthetic is self-refuting.

 Kant’s original texts are a little different: “The same function that gives unity to the different representations in a judgment also gives unity to the mere synthesis of different representations in an intuition, which expressed generally, is called the pure concepts of understanding” (CPR B 104– 5).  In this scenario, the synthesis would be possible, albeit “in vain” (CPR A 103 – 4).

132

Roberto Horácio de Sá Pereira

However, the obstacle is removed when we bear in mind that what is in question in the Deduction is not the possibility of representing objects or the possibility of representing what is subjectively given to the senses as existing mind-independently. Instead, what is in question is the possibility of Erfahrung and of Erkenntnis. In the first step of the B-Deduction, Kant states that categories are necessary for thinking that something given to an intuition in general is an object. Yet, Erkenntnis requires more than thinking. It requires intuition. Thus, in the second step, this Erkenntnis requires that the sensible form of the apprehension of something given to our senses as something that exists objectively or mind-independently is subordinate to categories. This is Kant’s figurative synthesis, defined metaphorically as “an effect of the understanding on the sensibility” (CPR B 154). The key phrase for making sense of the footnote is what Kant emphasizes in boldface: “as an object”. Kant must provide the grounds for natural science and geometry. Without showing that categories are conditions for the sensible apprehension of objects as objects, natural science and geometry would be groundless. That is what in the second step he calls formal intuition (in opposition to pure intuition), that is, the sensible apprehension of a pre-existing unity as a unity (as an object). From this, it follows that: (i) formal intuition and pure intuition are different concepts, (ii) Kant never changed his mind about the role of the imagination, and (iii) even though the sensible apprehension of space as an object requires that the synthesis of the homogeneous is determined by the category of quantity, we can apprehend unities without categories. Kant’s main argument of the second step of the B-Deduction can be formulated in a very simple and persuasive way. The first premise is the factual one according to which we do in fact apprehend space as existing mind-independently (figurative synthesis). The second is conditional: we do apprehend space as a mind-independent particular if we represent it as a homogeneous magnitude according to the category of quantity. Now, by applying modus ponens to both premises, we are entitled to conclude that the category of quantity applies to space and a fortiori to everything in it. This insight also provides an easy reading of the troublesome footnote. What Kant had in mind with “space, represented as an object as is really required in geometry” (CPR B 160n, Kant’s own emphasis in bold) is not space as the intentional object of our outer sense, neither is it space as a particular existing mindindependently. Instead, what he meant is the apprehension of space as something existing mind-independently. Likewise, “the formal intuition that gives unity of the representation” (CPR B 160n) is not a replacement for pure intuition, the representation of the form of intuition, but rather the apprehension that the representation of space is as a mind-independent object.

Cassirer and Kant on the Unity of Space and the Role of Imagination

133

Bibliography References to Kant’s works are given in the German Academy edition: Gesammelte Schriften, herausgegeben von der Königlich Preussischen Akademie der Wissenschaften, 29 vols. (Berlin: 1902 – 1983; 2d ed., Berlin: De Gruyter, 1968, for vols. I–IX). They are indicated as follows: abbreviation of the title of the work, followed by AA., volume, and page. For the Critique of Pure Reason, the references are shortened, in keeping with current practice, to the pagination of the original edition indicated by A for the 1781 edition and B for the 1787 edition. All translations in this essay are taken from P. Guyer and A. Wood’s translation of Kant’s Critique of Pure Reason (Cambridge: Cambridge University Press, 1998) Allison, Henry E. (2015): Kant’s Transcendental Deduction. An Analytic-Historical Commentary, Oxford. Beiser, Frederick C. (2015): The Genesis of Neo-Kantianism, New York. Cassirer, Ernst (1923): Substance and Function, Chicago/London: Open Court. Cassirer, Ernst (1957): The Philosophy of Symbolic Forms. Vol. 3: The Phenomenology of Knowledge. Translated by Ralph Manheim. New Haven/London: Yale University Press. Gomes, Anil (2014): Kant on Perception: Naive Realism, Non-Conceptualism, and the B-Deduction, in: Philosophical Quarterly, 64 (254), pp. 1 – 19. Hanna, Robert (2006): Kant, science, and human nature, Oxford. Hanna, Robert (2008): Kantian Nonconceptualism, in: Philosophical Studies, 137 (1), pp. 41 – 64. Hanna, Robert (2011): Kant’s Non-Conceptualism, Rogue Objects, and The Gap in the B Deduction, in: International Journal of Philosophical Studies 19 (3), pp. 399 – 415. Hanna, Robert (2013): Review: Förster, Eckart, The twenty-five years of philosophy, in: Kantian Review 18 (2), pp. 301 – 315. Hanna, Robert (2015): Cognition, content, and the a priori: A study in the philosophy of mind and knowledge, Oxford. Liebmann, Otto (1912): Kant und die Epigonen: Eine kritische Abhandlung, Berlin. McDowell, John (1994): Mind and World, Cambridge. Natorp, Paul (1882a): Descartes’ Erkenntnistheorie. Eine Studie zur Vorgeschichte des Kritizismus, Marburg. Natorp, Paul (1882b): Galillei als Philosoph, in: Philosphische Monatshefte 18, pp. 193 – 229. Natorp, Paul (1887): Über objektive und subjektive Begründung der Erkenntnis, in: Philosophische Monatshefte 23, pp. 257 – 286. Natorp, Paul (1888): Einleitung in die Psychologie nach kritischer Methode, Freiburg/Leipzig. Natorp, Paul (1894): Religion innerhalb der Grenzen der Humanität, Freiburg/Leipzig. Natorp, Paul (1901): Zur Frage der logischen Methode. Mit Beziehung auf Edmund Husserl’s Prolegomena’ zur reinen Logik, in 6 Bänden, pp. 270 – 283. Natorp, Paul (1904a): Allgemeine Psychologie, in Leitsätzen zu akademischen Vorlesungen, Marburg. Natorp, Paul (1904b): Sozialpädagogik: Theorie der Willenserziehung auf der Grundlage der Gemeinschaft, Stuttgart. Natorp, Paul (1910): Die logischen Grundlagen der exakten Wissenschaften, Leipzig/Berlin. Natorp, Paul (1912a): Allgemeine Psychologie nach kritischer Methode, Freiburg/Leipzig. Natorp, Paul (1912b): Kant und die Marburger Schule, in: Kant-Studien 17, pp. 192 – 221.

134

Roberto Horácio de Sá Pereira

Pereira, Roberto (2001): Referência e Juízo in Kant, in: Analytica, 6, pp.79 – 117. Pereira, Roberto (2004): Conteúdos não-conceituais Kantianos, in: O que nos faz Pensar 28, pp. 25 – 50. Pereira, Roberto (2010): Nem conceitualismo nem não-conceitualismo, in: Manuscrito 33 (2), pp. 467 – 495. Pereira, Roberto (2012): Disjuntivismo em Kant, in: Aurora, 24, pp. 129 – 155. Pereira, Roberto (2013): What is Nonconceptualism in Kant’s philosophy, in: Philosophical Studies 164, pp. 223 – 254. Pereira, Roberto (2014): Nonconceptualism, in: Compêndio em Linha de Filosofia Analítica, pp. 1 – 32. Pereira, Roberto (2015): Relativizing the Opposition between Content and State Nonconceptualism, in: Abstracta 8 (2), pp. 17 – 30. Pereira, Roberto (2016a): The Constitutional View, in: Principia 20 (2), pp 165 – 177. Pereira, Roberto (2016b): Combining the representational and the relational view, in: Philosophical Studies 137, pp. 255 – 269. Pereira, Roberto (2017a): Nonconceptualism or de re Sense: a new Reading of Kantian Intuition, in: Abstracta 10, pp. 45 – 64. Pereira, Roberto (2017b): A nonconceptualist reading of the B-Deduction, in: Philosophical Studies 174, pp. 425 – 442. Pereira, Roberto (2018): Misunderstanding the role of Concepts in Kant, in: Kant-e-Prints, pp. 6 – 25. Pereira, Roberto (2019): What Kant Has to Teach Us About What Mary Learns, in: Borderless Philosophy 2, pp. 257 – 276. Rickert, Heinrich (1888): Zur Lehre von der Definition, Tübingen. Rickert, Heinrich (1896): Die Grenzen der naturwissenschaftlichen Begriffsbildung: Eine logische Einleitung in die historischen Wissenschaften, Tübingen/Leipzig. Rickert, Heinrich (1899a): Fichtes Atheismusstreit und die Kantische Philosophie, Berlin. Rickert, Heinrich (1899b): Kulturwissenschaft und Naturwissenschaft, Tübingen/Leipzig. Rickert, Heinrich (1900): Psychophysische Causalität und psychophysischer Parallelismus, in: B. Erdmann et al. (eds.): Philosophische Abhandlungen. Christoph Sigwart zu seinem siebzigsten Geburtstage, Tübingen/Leipzig, pp. 59 – 87. Rickert, Heinrich (1902): Die Grenzen der naturwissenschaftlichen Begriffsbildung. Eine Logische Einleitung in die historischen Wissenschaften, Tübingen/Leipzig. Rickert, Heinrich (1907): Geschichtsphilosophie, in: W. Windelband (ed.): Die Philosophie im Beginn des 20. Jahrhunderts. Festschrift für Kuno Fischer, Heidelberg, pp. 321 – 342. Rickert, Heinrich (1909): Zwei Wege der Erkenntnistheorie. Transcendentalpsychologie und Transcendentallogik, in: Kant-Studien 14, pp. 169 – 228. Rickert, Heinrich (1910): Vom Begriff der Philosophie, in: Philosophische Aufsätze, Tübingen/Leipzig, pp. 3 – 36. Rickert, Heinrich (1911): Lebenswerte und Kulturwerte, in: Logos 2, pp. 131 – 166. Rickert, Heinrich (1913): Vom System der Werte, Tübingen/Leipzig. Rickert, Heinrich (1914): Über logische und ethische Geltung, in: Logos 19, pp. 182 – 221. Rickert, Heinrich (1921): System der Philosophie. Erster Teil: Allgemeine Grundlegung der Philosophie, Tübingen/Leipzig. Rickert, Heinrich (1924a): Das Eine, die Einheit und die Eins. Bemerkungen zur Logik des Zahlbegriffs, Tübingen/Leipzig.

Cassirer and Kant on the Unity of Space and the Role of Imagination

135

Rickert, Heinrich (1924b): Die Probleme der Geschichtsphilosophie. Eine Einführung, 2. Aufl., Heidelberg. Rickert, Heinrich (1926): Kulturwissenschaft und Naturwissenschaft, Tübingen/Leipzig. Rickert, Heinrich (1928): Der Gegenstand der Erkenntnis. Einführung in die Transzendentalphilosophie, 6th ed., Tübingen/Leipzig. Rickert, Heinrich (2015): Knowing and Cognizing. Critical Remarks on Theoretical Intuitionism, in: S. Luft (ed.): The Neo-Kantian Reader, New York, pp. 384 – 395. Sellars, Wilfrid (1968): Science and metaphysics. Variations on Kantian Themes, New York. Strawson, Peter F. (1966): The Bounds of Sense, London. Windelband, Wilhelm (1892): Geschichte der Philosophie, Freiburg. Windelband, Wilhelm (1894): Geschichte und Naturwissenschaft. Rede zum Antritt des Rectorats der Kaiser- Wilhelms, Strasbourg. Windelband, Wilhelm (1900): Platon, Stuttgart. Windelband, Wilhelm (1911): Die Geschichte der neueren Philosophie in ihrem Zusammenhange mit der Allgemeinenkultur und den besonderen Wissenschaften, Leipzig. Windelband, Wilhelm (1915): Präludien. Aufsätze und Reden zur Philosophie und ihrer Geschichte, Band I, Tübingen. Windelband, Wilhelm (1956): History of Ancient Philosophy, New York. Windelband, Wilhelm (1958a): A History of Philosophy, Vol. I, Greek, Roman, Medieval, New York. Windelband, Wilhelm (1958b): A History of Philosophy, Vol. II, Renaissance, Enlightenment, and Modern, New York. Windelband, Wilhelm (1980): History and Natural Science, in: History and Theory 19, pp. 165 – 185. Windelband, Wilhelm (2015): Introduction to A History of Philosophy with Special Reference to the Formation and Development of its Problems and Conceptions, in: S. Luft (ed.): The Neo-Kantian Reader, New York, pp. 299 – 313.

Hernán Pringe

Cohen’s Logik der reinen Erkenntnis and Cassirer’s Substanzbegriff und Funktionsbegriff Abstract: This paper compares Cohen’s Logic of Pure Knowledge and Cassirer’s Substance and Function in order to evaluate how in these works Cohen and Cassirer go beyond the limits established by Kantian philosophy. In his Logic, Cohen seeks to ground in pure thought all the elements which Kant distinguishes in empirical intuition: its matter (sensation) as well as its form (time and space). In this way, Cohen tries to provide an account of knowledge without appealing to any receptivity. In accordance with Cohen’s project of reformulating the Kantian theory of sensibility, Cassirer undertakes in Substance and Function the task of developing an alternative doctrine of pure and empirical manifolds. But whereas Cohen analyzes the laws of pure thought, Cassirer aims to highlight the functional character of concepts in the development of modern mathematics and physics. I will discuss these two different approaches to the problems raised by Kantian philosophy and I will argue that Cassirer went further than Cohen in the project of critical idealism.

Introduction In the preface to the first edition of his Logic of Pure Knowledge, published in 1902, Hermann Cohen announces that a second part to this investigation is already planned. The second volume of the book would complement the first one by taking into account the contemporary debate on the philosophical and scientific issues raised by Cohen’s Logic (Cohen 1922, XI). Ultimately, this second part never appeared. But in 1910, after reading Substance and Function, Cohen writes a letter to Cassirer, where he declares:

The project leading to this paper has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 777786. The investigation is also part of the project CONICYT/FONDECYT Regular Nº 1190965. The author would like to thank the anonymous referees for helpful comments on a previous version of this paper. https://doi.org/10.1515/kantyb-2020-0006

138

Hernán Pringe

I congratulate you and all the members of our philosophical community on your new and great achievement. If I shall not be able to write the second part of my Logic, no harm will be done to our common cause, since my project is to a large degree fulfilled in your book (Gawronsky 1949, 21).

The aim of this paper is to analyze the relationship between Cohen’s and Cassirer’s investigations on the critique of knowledge and to evaluate the way in which each of them goes beyond the limits established by Kantian philosophy. In particular, I will try to show that in Substance and Function Cassirer went further than Cohen in the project of critical idealism.

1 Cohen In his Logic of Pure Knowledge, Hermann Cohen aims to carry out the Copernican turn which, according to him, Kant fails to achieve. On Cohen’s reading, if objects must conform to our cognition because knowledge produces the object (Cohen 1907, 4), then this cannot just amount to the determination of the mere form of objectivity in general. On the contrary, the spontaneity of thought must also generate the matter of cognition. For this reason and in opposition to Kant, Cohen claims that in cognition thought does not face any given matter, not even a pure one (Cohen 1922, 26 – 27). What Kant calls given is nothing but a product: a product of thought. The Kantian distinction between thinking and cognizing an object (CPR B 146), which relies precisely on the consideration of intuition as a non-conceptual representation, is thus abandoned in favor of a doctrine of thought that is at the same time a doctrine of knowledge. Though, in Cohen’s sense, thought does not depend on any receptivity that would provide it with a sensible content. According to Cohen, only in this way can the Copernican turn that Kant prescribed to metaphysics be finally executed successfully.¹ One of the key aspects of Cohen’s doctrine is the relationship that his Logic establishes between pure knowledge and infinitesimal calculus. The Cohenian interpretation of calculus seeks to ground in pure thought all the elements

 In a recent book, Beiser claims: “The first question to ask about this doctrine is why Cohen attributes ‘autonomy’ or ‘independence’ to pure thinking. Why does he have confidence in the power of pure thinking to generate its own content? Why must it be independent of all content outside itself? The answers to these questions cannot be readily found in Cohen’s book, where he simply states his results without explaining how he arrived at them.” Beiser (2018, 196). On the contrary, in this section we will attempt to present the arguments that ground Cohen’s theses in the Logic.

Cohen and Cassirer

139

which Kant distinguishes in empirical intuition: its matter (sensation) as well as its form (time and space). In this way, by means of infinitesimal calculus, Cohen tries to provide an account of knowledge without appealing to any receptivity.

1.1 Pure thought and infinitesimal calculus For Cohen, thought is pure if nothing is given to it, while knowledge is pure if its origin is pure thought, thus understood. From this viewpoint, there is no room for a pure element of knowledge that does not have its origin in thought. Accordingly, against the Kantian notion of pure intuition, Cohen maintains that “purity can only be of a single kind.” (Cohen 1922, 150)². Intuition cannot be pure precisely because it is irreducible to thought. But the purity of thought is only negatively expressed by the affirmation that nothing is given to it. The positive claim is that pure thought generates its own content. Thus, in opposition to Kant’s doctrine, thought is not to be applied to any object heterogeneous in comparison to it. Consequently, the question concerning the conditions of the application of the pure concepts of the understanding to sensible appearances – the motivation behind the Kantian schematism (CPR A 137/B 176 f.) – no longer arises on Cohen’s account. Against Kant, Cohen rejects any alleged heterogeneity between thought and being (Cohen 1922, 151). Whereas Kant puts all his effort into proving that thought applies necessarily to appearances in spite of the fact that thought is pure, Cohen claims that thought applies necessarily to the object of knowledge precisely because thought is pure. The purity of thought amounts to its applicability (Cohen 1883, 131). Yet the peculiarity of Cohen’s doctrine is the role it ascribes to infinitesimal calculus in the generation of being by thought.³ The Cohenian interpretation of the infinitesimal is the key to his logic of pure knowledge. Cohen claims that Kant did not properly understand the meaning of calculus for transcendental philosophy and therefore he was unable to recognize the productive capacity

 All translations of Cohen’s work are my own. Kant′s translations are taken from Kant (1998), Kant (2000) and Kant (2002).  Cohen’s interpretation of infinitesimal calculus undergoes a substantial development starting from Cohen (1883), through Cohen (1885) and Cohen (1902), up until Cohen (1914). Whereas in his 1883 work, intuition still plays a role in the foundation of infinitesimals, in 1902 Cohen claims that the infinitesimal is a product of mere thought. The discussion of this development goes beyond the scope of this paper, which is focused on the mature doctrine of the Logic. For a comprehensive historical reconstruction of Cohen’s infinitesimal method, see Giovanelli (2016). See also Gordin (1929, 33 – 34).

140

Hernán Pringe

of thought. For Cohen, “if the infinitesimal principle had found in the Critique the place it deserved, then the sensibility would not have been put before thought [and] pure thought would not have been undermined in its autonomy” (Cohen 1922, 35). Kant did not see the importance of the infinitesimal principle. In contrast, the logic of pure knowledge underlines “the crucial logical meaning of the infinitesimal principle” and it may even be characterized as “the logic of the principle of infinitesimal calculus” (Cohen 1922, 35). Thought generates by itself its own content according to this principle, i. e. by means of infinitesimal calculus thought produces the matter which for Kant could only be given by sensibility. In cognition, Kant maintains, pure concepts of the understanding are applied to sensible appearances. The system of those synthetic judgments that flow a priori from pure concepts of understanding under the application conditions contained in the schemata is the so-called system of all principles of pure understanding (CPR A 136/B 175). “Kant”, Cohen claims, “distinguished pure thought from pure intuition. Thus, he needed the combination of both heterogeneous conditions.” (Cohen 1922, 477). Such a combination is to be found in those principles. Contrary to this, Cohen’s logic contains no element not originated in thought. For this reason, the transcendental foundations of knowledge will not be principles [Grundsätze], but plain and simply judgments [Urteile].⁴ The logic of pure knowledge should show that these judgments of pure thought do have objective reference, although they do not contain any alogical element. There are four kinds of judgments of pure thought (Cohen 1922, 77– 78): the judgments of the laws of thought, the judgments of mathematics, the judgments of mathematical natural science and the judgments of methodology. The first class corresponds to the judgments of quality, the second to those of quantity, the third to those of relation and the fourth to those of modality. The Cohenian interpretation of infinitesimal calculus appears in the discussion of the judgments of mathematics, i. e. in the judgments of quantity.⁵ These judgments are those expressing the transcendental foundations of mathematical cognition and include the judgments of reality, multiplicity and totality. In what follows, I will analyze the role of each of these in the generation of pure knowledge.

 On the relationship between Kant’s system of principles and Cohen’s system of judgments, see Edel (1988, 516 – 523).  Regarding Cohen’s table of judgments, Natorp asks: “Ist es fortan noch zulässig, die Qualität der ‘logischen‘ als einer eignen, vormathematischen Erkenntnisart zuzuweisen?” Natorp (1986a, 15 – 16).

Cohen and Cassirer

141

1.2 The judgment of reality and sensation Mathematics is the best example of pure thought. However, this is so not because mathematical knowledge does not depend on experience. It is true that mathematics is an a priori science, but this is not the reason why it is pure. The purity of mathematics does not consist in its independence from experience, but rather in its applicability to empirical objects. Mathematics, as Cohen discusses in his Logic, is the “mathematics of the mathematical science of nature.” (Cohen 1922, 121).⁶ Cohen agrees with Kant on the claim that mathematical cognition satisfies the demands of the Copernican turn, but he does not share the reasons that Kant puts forward in favor of it. According to Kant, mathematical objects are constructed in pure intuition by a synthesis governed by the concepts of the understanding. Mathematical objects would therefore conform to our cognition and we could cognize a priori in such objects what reason has put into them. In this way, the Copernican turn would be accomplished. However, this mathematical construction is executed on a manifold, which, as far as it is sensible, is merely given, and thus imposes a limit on the spontaneity of thought. For this reason, from Cohen’s viewpoint, the Kantian doctrine of construction in intuition does not give a proper account of the Copernican turn in mathematics. In particular, infinitesimal calculus shows that mathematical cognition is not only independent from sensation, but moreover is independent from intuition, no matter how pure this intuition might be for Kant (Cohen 1922, 124). It is important to stress that this does not imply that mathematical cognition is formal or empty. Rather, as already said, mathematics is pure knowledge in Cohen’s sense, i. e. mathematics generates its own object and this generation guarantees the possibility of a mathematical science of nature. Cohen will try to show that the first stage in the generation of the object of natural science is the production of the content of natural science by means of mathematics and, in particular, by means of infinitesimal calculus. The generation of being by thought will be a generation by means of infinitesimal calculus. In this way, the physical application of calculus and, in general, of number, will be well-founded. The generative power of calculus reveals itself in the notion of “infinitesimal number”, both in the Newtonian fluxion and in the Leibnizian infinitesimal (Cohen 1922, 124). Newton and Leibniz show how thought generates the finite

 Mormann underlines this point: “Jede Philosophie der Mathematik, die zum Anwendungsproblem der Mathematik nichts zu sagen hatte, war fü r Cohen, und fü r die Marburger Schule insgesamt, grundsätzlich unbefriedigend.” Mormann (2018, 115).

142

Hernán Pringe

from the infinitesimal, the extended from the non-extended (Cohen 1922, 126). Infinitesimal calculus is based on the idea that “the finite must have its origin in the non-sensible.” (Cohen 1922, 135). Thus, “infinitesimal calculus just aims at being a means, the adequate means, for the determination of the finite.” (Cohen 1922, 179). The transition from quality to quantity takes place through the infinitesimal. The infinitesimal enables the generation of a quantitative something from a quantitative nothing, which is nevertheless qualitatively determinate. The structure of this movement from nothing to something is that of the judgment of origin. But the infinitesimal has moreover a specific meaning, according to which it generates a qualitatively and quantitatively determinate content that is the condition of the possibility of a mathematical science of nature. For Cohen, the infinitesimal corresponds to a particular kind of judgment that Cohen calls the judgment of reality [Urteil der Realität] (Cohen 1922, 128). The logical foundation of infinitesimal calculus is thus a kind of judgment that will guarantee its applicability to nature, since reality [Realität], which is the basic presupposition of physics, will be at the same time the basic presupposition of calculus. According to Cohen, the history of infinitesimal calculus shows that there are three fundamental problems that the notion of infinitesimal enables us to solve. Firstly, the geometrical problem of tangents; secondly, the algebraic problem of series and, finally, the dynamical problem of velocity and acceleration. In the solution of these problems one can see how “the motif of origin” gets specified in “the motif of reality”, i.e. how the peculiar character of the logical foundation of calculus reveals itself progressively. In the problem of tangents, the “the motif of origin” is present because the point where the tangent touches the curve is the generating point of the curve itself. This meaning of the point is incompatible with the old definition of point as the limit of a line. “Now,” Cohen argues, “the point means something different, something positive. The point is not just the end of the line, but rather the beginning of it.” (Cohen 1922, 129). This beginning is not an arbitrary one. More precisely, it is rather an origin than a beginning, since the point contains the generation law of the curve. As Natorp states, “the law can be conceived, intensively, to be concentrated in a point or, extensively, to be extended in the trajectory.” (Natorp 1910, 220). The curve is generated from its tangent. This generation does not just take place at the beginning of the curve, but all along it. The curve is generated from its tangent at every point. “This absolute character of the point,” Cohen states, “in so far as the curve is generated from it uninterruptedly” is called reality (Cohen 1922, 130).⁷ In the same sense, in an infinite

 For Nelson, Cohen understands the infinitesimal in a mystical sense by affirming the exis-

Cohen and Cassirer

143

series, the general term expresses the law of the series and contains the rule of the generation of the series. Finally, in the dynamic problem, the presence of a generating element is even more notable. In this case, Cohen claims, “the demand of an absolute point cannot be avoided, since only thereby can a real [reale] movement be definitely distinguished from a subjective succession of representations.” (Cohen 1922, 130). Such a point, which corresponds to acceleration, is the ultimate ground of movement. Regarding the form of movement, the point is “valid as absolute.” The point represents a “being in becoming” or, better, a “being for becoming” (Cohen 1922, 130). The unity of the accelerated point is defined, Cohen claims, “as reality.” (Cohen 1922, 131). As is well known, Kant sharply distinguishes reality [Realität] from actuality [Wirklichkeit] (CPR A 219/B 266). Reality is a category of quality, whereas actuality is a category of modality. If we judge that a certain object is actual, we are not thereby adding any new content to our representation. Rather, we just indicate that there is something that corresponds to our concept. In fact, even “if the concept of a thing is already entirely complete, I can still ask about this object whether it is merely possible, or also actual, or if it is the latter, whether it is also necessary?” (CPR A 219/B 266). Modal categories “express only the relation to the faculty of cognition.” (CPR A 219/B 266). In particular, the category of actuality does not express any content-related determination of the object, but just the positing of the object (i. e. its existence) with all its formal and material determinations. While actuality expresses that the object exists, reality expresses what the object is, no matter whether it exists or not. According to Kant, “reality (thinghood)” is something, “the concept of which in itself already expresses a being […] because through it alone, and only so far as it reaches, are objects Something (things).” (CPR A 574/B 603). Reality is a positive determination. Reality is “that which can be thought only through an affirmative judgment.” (CPR A 246/B 302). The category of reality corresponds to the act of understanding in such a judgment. The same function carried out by the category of reality in the connection of representations in an affirmative judgment is at play in the synthesis of the empirical manifold of an intuition, which provides the latter with a determinate content. When the category of reality is applied to the matter of intuition, this

tence of infinite small things. Nelson (1905). To this criticism, Cassirer responds: “Cohen […] betont, daß das Infinitesimale nicht als Ding, sondern als Bedingung, nicht als eine irgendwie vorhandene Wirklichkeit, sondern als ein gedankliches Instrument zur Entdeckung und zum Aufbau des wahrhaften Seins zu gelten habe.” Cassirer (1906, 32 footnote). On Nelson’s criticism see Peckhaus (2014).

144

Hernán Pringe

matter turns into a determinate content, i. e. a certain quality, which may be the predicate of an affirmative judgment. Cohen follows Kant in emphasizing the distinction between reality [Realität] and actuality [Wirklichkeit] (Cohen 1922, 128). On Cohen’s view, while the judgment of reality belongs to the judgments of mathematics,⁸ the judgment of actuality belongs to the judgments of methodology, together with the judgments of possibility and necessity. Cohen indicates that for Kant the reality and the actuality of an object depend on a different relation to sensation, considered in the corresponding principles of pure understanding. In the anticipations of perception, the real [das Reale] is the object of sensation, while the second postulate of empirical thinking establishes that the actual is that which is connected with sensation (Cohen 1922, 462). In the case of actuality, Cohen acknowledges that there is a justified “claim of sensation [Anspruch der Empfindung]” (Cohen 1922, 436, 450, 463), which nevertheless cannot be satisfied by sensation itself: “sensation itself cannot achieve what it demands.” (Cohen 1922, 436). The claim may only be satisfied by the connection with the pure conditions of knowledge (Cohen 1922 463). This means that all categories have “a tendency to sensation” (Cohen 1922, 437), i. e., they take part in the process of satisfying its demand, by incorporating sensation in time and space and legitimating it as infinitesimal reality (Cohen 1922, 437).⁹ In order to answer the justified demand of sensation involved in the determination of actual objects, Cohen argues that the role Kant assigns to sensation has to be carried out by pure thought. Pure thought accomplishes this task by means of the judgment of reality and the positing of the infinitesimal. For Cohen, the differential (dx) represents what it is [das Seiende], and only through it may that something (x) of which the differential is differential, in general, be (Cohen 1922, 136).¹⁰ But whereas for Kant reality is “that in empirical intuition which corresponds to the sensation” (CPR A 168/B 209)¹¹–and elsewhere even goes as far as to claim that “Sensatio [est] realitas phaenomenon” (CPR A 146/B 186)–for Cohen reality does not depend on sensation. Rather, it is the infinitesimal reality and not sensation that is the ultimate condition of the content

 As already indicated, the judgment of reality belongs to the judgments of quantity, while for Kant reality is a category of quality.  For a discussion of the relation between sensation and actuality see Holzhey (1986, 159 – 160), Dufour (2002).  Natorp points out: “Aber es [das x] muss allerdings, und wäre es nur als Fragezeichnen, im Denken augestellt sein, ehe von dx überhaupt mit sicherem Sinn die Rede sein kann; und die Beziehung auf das x als das Zuerzeugende ist von dessen Begriff allerdings unabtrennbar.” Natorp (1986b, 53).  See also CPR A 143/B 182.

Cohen and Cassirer

145

of thought and, in this way, of knowledge. Only through the infinitesimal may thought be the thought of being. But such infinitesimal has its origin in thought itself and not in any alleged receptivity. Thus, the reality of the infinitesimal presupposes that we get free of the cognitive task of sensation (Cohen 1922, 128).¹² The Kantian category of reality determines a being independently given by sensibility. The category of reality is applied to the received matter that “fills time.” (CPR A 143/B 183).¹³ Contrary to this, through Cohen’s judgment of reality, thought produces being as an infinitesimal reality.¹⁴ While the judgment of origin claims that “thought must begin with the origin, as far as thought is thought of knowledge and therefore of being” (Cohen 1922, 134), the judgment of reality establishes that such being is an infinitesimal reality. The infinitesimal is the non-sensible ground of the finite. Consequently, mathematics and natural science become intimately related. By means of the infinitesimal, mathematics provides natural science with the real and thus the applicability of mathematics, i. e. its purity, is finally achieved (Cohen 1922, 135). According to Kant, the spontaneity of thought unifies the manifold given by sensibility. In this act of synthesis, the understanding imposes its laws on appearances, since such synthesis takes place in accordance with the categories. In contradistinction to this, for Cohen the object of knowledge conforms to pure thought because pure thought produces it. In the judgments of mathematics the multiplicity [Mehrheit] that Kant called given manifold is rather produced by thought and acquires a unity that makes it fit to receive the determinations of the mathematical science of nature. The multiplicity and the unity of this multiplicity (its totality) will be produced by thought. Multiplicity and totality will be grounded in the infinitesimal reality. In this regard, Cohen claims that “the infinitesimal is the true unity.” (Cohen 1922, 137). Infinitesimal reality will be the foundation of number.¹⁵ But not only will the judgment of reality be involved

 On this issue, Flach claims: “Das Infinitesimale und nicht die Sinnlichkeit steht für die Mannigfaltigkeit.” Flach (2009, 40).  Due to this application of the category, the real, which is an object of sensation, acquires intensive magnitude, i. e., a degree. (CPR A 165/B 206). Even though in other works Cohen claims that “the identity of intensive and infinitesimal magnitude was a general assumption in Kant’s times” (Cohen (1883, 14)) and that “intensive magnitude means nothing but differential magnitude,” (Cohen (1885, 427)) in his Logic he rather maintains: “There is no intensive magnitude.” Cohen (1922, 492). See also Holzhey (1986, 251).  Cohen claims that while in the mere differential dx the origin of the finite is just defined, only by means of the differential quotient (as relation between two differentials) may this origin be determined. Cohen (1922, 182).  Cohen’s philosophy of mathematics (and, in general, of science), in which the concept of the infinitesimal plays a central role, may be characterized as a “nichtarchimedischer Ansatz” Mor-

146

Hernán Pringe

in the generation of number. Other judgments will be necessary as well. Thought will generate number by means of the production of a multiplicity. The judgment of multiplicity will carry out this task. According to Cohen, in this production time will be generated by thought. The multiplicity so produced will then be unified by the judgment of totality. This totality corresponds to the integral. Such integration will be the origin of space. Differential and integral calculus will therefore express the way in which pure thought generates the content of the mathematical science of nature, i. e., for Cohen, the content of experience. We will now see how this generation progresses from the infinitesimal reality to the spatio-temporal content.

1.3 The judgment of multiplicity and the category of time The pure origin of number is the primordial unity of reality. Pure thought will then generate a multiplicity according to the corresponding judgment. Cohen distinguishes this multiplicity [Mehrheit] from what he calls a mere plurality [Vielheit]. A plurality is a set of elements that precede any generation by thought. While the elements of a plurality are merely given, the elements of a multiplicity are produced by pure thought. More precisely, the multiplicity is the outcome of an adding in which the addends do not logically precede the addition: The very adding in which the addends in the first place arise is the task and its possibility is the problem (Cohen 1922, 149).

The judgment of reality generates the absolute unity that Cohen represents by the number 1. This unity is pure being, without any other determination. It is ab-

mann (2018, 108). From this perspective, new light may be shed on Russell’s criticism of Cohen (1883) in Russell (1903). Russell’s criticism, Mormann claims, is false in a twofold way: a) infinitesimals are not contradictory pseudo-concepts and b) infinitesimals were not erradicated from mathematical discourse because of the works of Cantor, Dedekind and Weierstrass (Mormann 2018, 107). In this regard, Mormann argues: “Aus der Tatsache, dass solche Größen [infinitesimals, HP] mathematisch “vernü nftige” Entitäten sind, lässt sich natü rlich keineswegs ableiten, dass auch Cohens “nichtarchimedische” Wissenschaftsphilosophie dieses Prädikat verdient. Immerhin aber lässt sich behaupten, dass sein Ansatz nicht von vornherein eine Totgeburt war, wie dies die starken Thesen Russells, Carnaps, und Quines glauben machen wollten. Ich glaube sogar, man kann zeigen, dass eine mathematische Wissenschaftsphilosophie, die in gewisser Weise an den Marburger Ansatz anschließt, einen Beitrag auch zur aktuellen wissenschaftsphilosophischen Diskussion leisten kann” (Mormann 2018, 108). Natorp regrets that Russell just considers the doctrine contained in Cohen (1883) and not the mature version of Cohen (1902) (Natorp 1910, 221– 222, footnote 1).

Cohen and Cassirer

147

solute being, presupposed by any determinate being. Cohen calls this unity the “number of reality.” To this, Cohen opposes the “number of multiplicity”, generated by the judgment of multiplicity. This number is the unity understood as a term of a series. The unity is now represented by a variable x or by a determinate A, whose only determination is being such a term. The judgment of multiplicity expresses the serialization by means of which pure thought generates the determinate terms of a series from the general one. These terms are thus not given but produced by pure thought. Time finds its foundations in this serialization. For Cohen, time is not a pure intuition, but a category. According to Cohen, the fundamental mode of time is not the sequence [Folge]. In a sequence, we arrange representations that are independent from such order. Temporality is in this case merely their form. Regarding sequence, Cohen affirms: But, more deeply, the error relies on the fact that in this mode of time determination representations themselves are already there and it merely remains to determine their form, their temporal order (Cohen 1922, 153).

According to Cohen, the temporal sequence cannot satisfy the demands of pure thought, because the elements of a sequence are merely given. And this, Cohen claims, “puts purity in danger.” (Cohen 1922, 151). Conversely, thought generates time by means of anticipation. This is the pure foundation of time. Cohen maintains that time is “the category of anticipation.” (Cohen 1922, 155). While sequence “looks backwards” by ordering what already is, anticipation “looks forward” by generating what follows as something that must follow.¹⁶ “In this way the first form of multiplicity arises: by the differentiation of the past from the originary act of the future.” (Cohen 1922, 155). This anticipation does not act on what is anticipated as a form imposed on a given matter. Rather, anticipation as a form generates the matter.¹⁷ Serialization generates a determinate content that is nothing but the point from which anticipation is made. This is its only determination. In this way, the future generates the past. Time as a category, Cohen maintains, “does not have the task or competence of generating any other kind of content than this correlation between the future and the past.” (Cohen 1922, 156). Thus, Cohen claims, “the category of time has generated the unities of multiplicity and thereby has generated in pure thought the content that otherwise would count as given.” (Cohen 1922, 193). In other words, for Cohen, pure

 “In view of the not yet, the not any longer arises,” Cohen says. Cohen (1922, 154– 155).  In Cohenian terms: generation itself is what is generated.

148

Hernán Pringe

thought generates by the act of anticipation the temporal multiplicity that Kant considered to be passively received by the subject.

1.4 The judgment of totality and the category of space While the judgment of reality generates the absolute unity of the infinitesimal, the judgment of multiplicity determines this infinitesimal as a term of a temporal series. Pure being, firstly determined as infinitesimal reality, is secondly deployed in a temporal multiplicity, produced through anticipation by pure thought. This multiplicity, however, needs to get unified. The culmination of the series is the task of a new act of pure thought, corresponding to a third kind of judgment: the judgment of totality. This kind of judgment relates the infinitude of the series to the infinitesimal character of its terms. The totality of the series will be the sum total of infinite infinitesimal terms. This sum corresponds to the mathematical operation of integration. “The integral”, Cohen maintains, “is nothing but the totality in which the infinite series combines with the infinitesimal.” (Cohen 1922, 182– 183). For this reason, according to Cohen, only from a formal viewpoint may integration be considered as the inverse operation of differentiation. Rather, from the perspective of pure thought, Cohen argues, integral calculus provides differential calculus with its true significance, since the invention of the infinitesimal had the ultimate aim of determining the finite being as an infinite sum of infinitesimals. “Therefore”, Cohen stresses, “totality is the true end and the true business of the infinitesimal.” (Cohen 1922, 183). Integration will carry out the unification of multiplicity, necessary for the determination of the content of natural science. For this purpose, a new category, additional to number and time, is required: namely, the category of space. In this connection, Cohen claims: “Being must turn into something external to thought. This does not injure the identity [of thought], because thought itself generates this external [something] and thereby thought turns into the thought of nature and therefore being.” (Cohen 1922, 188). Just as in the case of time, Cohen rejects in this way the Kantian characterization of space as intuition, i. e. as a given representation. Space is rather a category, grounded in the judgment of totality. The temporal multiplicity is the multiplicity of a series that cannot be conceived of as a line. In this multiplicity there are only anticipations, it is “a tireless coming and going, a tending forwards and looking backwards.” (Cohen 1922, 193). But nothing persists beyond this monotonous mode of generation. The unceasing temporal flux prevents the constitution of a permanent content. In order for such a result to be reached, coexistence [Beisammen] is required. And this requirement will be fulfilled by space (Cohen 1922, 194). The infinite elements of the temporal

Cohen and Cassirer

149

series are unified in the totality of space. Spatiality is not given, like Kantian pure intuition, but it is rather produced by pure thought. Coexistence, generated by pure thought through the judgment of totality, is simply exteriority. The conservation of coexistent things is their exteriority (Cohen 1922, 196). According to Cohen, space is therefore not a pure intuition, a sensible condition for appearances to be given, as it was for Kant. Rather, space is a category through which nature is generated by pure thought (Cohen 1922, 197). The generation of spatio-temporal content by the judgments of mathematics may be illustrated if we consider the simple dynamical problem of a moving point. The infinitesimal reality is that of the instantaneous velocity of the point, which thought generates through the judgment of reality. The judgment of multiplicity generates the temporal series of instantaneous velocities. The judgment of totality combines these anticipations in a spatial unity. More precisely, through the judgment of totality thought integrates the instantaneous velocities in the temporal interval generated by the judgment of multiplicity. The result of such integration will be the trajectory of the moving point, i. e. the covered space. In this connection and regarding planetary orbits as an example, Cohen claims: The most general contents, the planets themselves, may be determined through conic sections. The presupposition of totality is to be found in them as well. And infinitesimal calculus has arisen in their computation. The integration of the equation is the integration of the curve. Therefore, the determination of space is the purpose of integral calculus. The diversity of curves is the diversity of the content that can be defined for science (Cohen 1922, 198).

The judgments of mathematics generate the content of nature and thereby provide the conditions of mathematical natural science. The judgment of reality generates infinitesimal reality. The judgment of multiplicity generates the temporal series of these realities. The judgment of totality composes this multiplicity in space.¹⁸ This generation of content takes place by means of differential and integral calculus.

 In CPR B 154, Kant claims that “we cannot represent the three dimensions of space at all without placing three lines perpendicular to each other at the same point. […] Motion, as action of the subject (not as determination of an object), consequently the synthesis of the manifold in space, if we abstract from this manifold in space and attend solely to the action in accordance with which we determine the form of inner sense, first produces the concept of succession at all. The understanding therefore does not find some sort of combination of the manifold already in inner sense, but produces it, by affecting inner sense.” The discussion of space in the chapter on the judgments of mathematics does not deal with the problem of its dimensions. Cohen never-

150

Hernán Pringe

Paraphrasing Galileo, Cohen claims that the judgments of quantity provide philosophy with the necessary characters to write the book of nature (Cohen 1922, 207). The logic of pure knowledge determines nature as an object of applied mathematics in this way. However, for the book of nature to be written, not only do we need an alphabet, but also a proper syntax. This syntax will finally make the formation of sentences possible: Once the judgments of mathematics have created the alphabet on the base of the judgments of the laws of thought, the judgments of the mathematical science of nature enter into scene in order to write the philosophy of nature with those letters (Cohen 1922, 588).

The syntactic rules according to which the sentences of natural science are formulated will be provided by the judgments of relation.

1.5 Empirical intuition and pure thought In sum, according to Cohen all the elements of Kant’s empirical intuition find their foundations in pure thought. More precisely, both the matter and form of empirical intuition have their origin in the judgments of mathematics. On the one hand, Cohen does not conceive the content that Kant calls matter of intuition to be sensation. Such content is rather the infinitesimal reality, which is not passively received by an alleged sensibility as the effect of an affection, but spontaneously produced by pure thought through the judgment of reality. On the other hand, time and space, the Kantian forms of intuition, are not given either. Time and space are generated by pure thought through the judgments of multiplicity and totality respectively. In this generation, moreover, the category of number arises. The judgments of reality, multiplicity and totality are the foundations of infinitesimal calculus as the method by which pure thought generates the content of experience. Cognition not only rules the form of the object of experience, but its matter as well. “Thought cannot have any origin outside itself, if its purity is not to be restricted or tarnished. Pure thought in itself, excluding any other

theless assumes that space has three dimensions, by claiming that time is to be included as a fourth dimension in a physical coordinate system (Cohen 1922, 199). The introduction of coordinate axes is a result of the operation of the judgments of mathematical natural science, in particular of the judgment of substance, as logical condition of motion (Cohen 1922, 232– 233). While for Kant the representation of the three dimensions of space depends on the action of understanding and not solely on sensibility, for Cohen this problem belongs to the logical foundations of mathematical physics and not merely of mathematics.

Cohen and Cassirer

151

thing, must exclusively generate pure cognitions.” (Cohen 1922, 13). According to Cohen, infinitesimal calculus makes such a task first possible.

2 Cassirer In accordance with Cohen’s project of reformulating the Kantian doctrine of sensibility,¹⁹ Cassirer undertakes in Substance and Function the task of showing how those elements of knowledge that for Kant are merely given by sensibility are rather constructions of thought [Denksetzungen]. As we have just seen, in his Logic of Pure Knowledge, Cohen first deals with the problem of the matter of intuition and later with the problem of its form (time and space). In Substance and Function, Cassirer discusses the problem of time and space first and then he turns to the problem of the empirical content of knowledge. The philosophical investigation on the foundations of arithmetic (2.1) and geometry (2.2) will reveal the true transcendental status of the Kantian forms of sensibility. Then, Cassirer will analyze the developments of modern physics in order to establish the constructive character of the empirical manifolds of natural science (2.3).

2.1 Number and time Cassirer puts forward a theory of number which differs from the Kantian doctrine. Kant states that number is the pure schema of magnitude (quantitatis) […], which is a representation that summarizes the successive addition of one (homogeneous) unit to another. Thus, number is nothing other than the unity of the synthesis of the manifold of a homogeneous intuition in general, because I generate time itself in the apprehension of the intuition (CPR A 142 /B 182 f.).

For Kant, number is the concept of an addition executed upon the given manifold of pure time. By means of this synthesis, such a manifold is apprehended as a unity. In contrast, for Cassirer “the intuition of pure time, upon which Kant based the concept of number, is indeed unnecessary here.” (Cassirer 1923, 40). Following Dedekind, Cassirer states that the number is rather a position in a pro-

 In relation to Cohen’s Logic, Cassirer maintains: “Wie die “Logistik”, so ist auch die moderne kritische Logik ü ber Kants Lehre von der “reinen Sinnlichkeit” hinweggeschritten. Auch ihr bedeutet die Sinnlichkeit zwar ein erkenntniskritisches Problem, nicht aber einen selbständigen und eigentü mlichen Quelle der Gewißheit mehr.” Cassirer (1907, 31).

152

Hernán Pringe

gression.²⁰ A certain position is determined only by the relations it bears to any other position. Dedekind states: If in the consideration of a simply infinite system N set in order by a transformation [Abbildung] Φ we entirely neglect the special character of the elements; simply retaining their distinguishability and taking into account only the relations to one another in which they are placed by the order-setting transformation Φ, then are these elements called natural numbers or ordinal numbers or simply numbers, and the base-element 1 is called the base-number of the number series N (Dedekind 1963, 68).

The crucial point is for Cassirer that numbers “gain their whole being […] first in and with the relations which are predicated of them.” (Cassirer 1923, 36). The relata connected by the transformation do not preexist their relation. Rather the relation is logically prior to them. Numbers are terms of a relation whose continuous application generates them. Therefore, the concept of number does not presuppose any given manifold, as the Kantian doctrine maintains.²¹ Number is the product of mere thought.²² Dedekind claims: With reference to this freeing the elements from every other content (abstraction) we are justified in calling numbers a free creation of the human mind. The relations or laws which […] are always the same in all ordered simply infinite systems, whatever names may happen to be given to the individual elements […], form the first object of the science of numbers or arithmetic (Dedekind 1963, 68).

By means of a generating relation, thought produces an ordered manifold not only regarding their form, but also regarding their matter. While for Kant, matter ordered under the representation of number (the schema of magnitude) was the pure temporal multiplicity given by sensibility, for Cassirer it is pure thought which generates its own “matter”, by producing the manifold of positions in a progression. In numeration, “the “material,” which is presupposed, is not to be thought of as outwardly given, but as arising through free construction.” (Cassirer 1923, 65). But this pure progression, or better, this “order in progression” is

 For a recent interpretation of Dedekind’s theory from the viewpoint of Cassirer’s philosophy, see Yap (2017).  Heiss nevertheless points out that “Cassirer uses unmistakable Kantian language to describe Dedekind’s approach.” Heiss (2011, 772 footnote 44).  More precisely, Cassirer argues that the notion of ordinal number is the fundamental one, and that the notion of cardinal number always presupposes it. “The theory of ordinal number thus,” Cassirer claims, “represents the essential minimum, which no logical deduction can avoid.” Cassirer (1923, 53). For a discussion on the logical priority of ordinal numbers in the context of the Marburg school see González Porta (2011, 103 – 144).

Cohen and Cassirer

153

pure time. “Thus”, Cassirer argues, “time (if we understand by it the ‘concrete form’ of the ‘inner sense’) presupposes number, but number does not, conversely, presuppose time.” (Cassirer 1923, 40). Time is now understood as a product of thought, not because its unity is the result of a synthesis of understanding (as Kant would maintain in the case of a formal intuition (CPR B 160 f., footnote)), but because its manifold is generated according to a serial intellectual principle. In his analysis of the concept of number, Cassirer puts forward a notion of mathematical cognition which differs from the Kantian one.²³ According to Kant, mathematical cognition is rational cognition from the construction [Construction] of concepts. To construct a concept means to exhibit a priori the intuition corresponding to the concept. Therefore, a non-empirical intuition is required for the possibility of mathematics (CPR A 713/B 741). In contradistinction, for Cassirer in mathematics “the “construct” [das Gebilde] is to gain its total constitution from the relations in which it stands.” (Cassirer 1923, 46). No intuition is thereby involved. In contradistinction to the Kantian doctrine, time finds its foundations in arithmetic, now understood as purely intellectual cognition.²⁴

2.2 Geometry and space In his discussion of geometry, Cassirer aims to show how the spatial concepts are ultimately included in the schema of pure serial concepts. As has been argued in the case of time, space, far from being given to thought, is a product of it. In the case of analytic geometry, the individual point of the plane is determined by the values of its coordinates (x, y). By means of a differential equation – f(x, y, y’) = 0 – a definite direction of movement is assigned to every point, so that the problem consists in reconstructing from the system of these directions the whole of a certain curve. The geometrical concepts of direction and curvature “are in their

 For a comparison between Cassirer’s and Rickert’s theory of numbers, see Oliva (2015). Oliva claims that Rickert’s position is closer to Kant’s than “the relationism of the Marburg Neo-Kantians.” Oliva (2015, 306).  Even though Cassirer criticizes the Kantian theory of intuition, he nevertheless underlines that “the formation of the objects of mathematics is constructive, and hence synthetic, because it is not concerned simply with analyzing a given concept into its characteristics, but because, starting from certain basic relations, we advance and ascend to ever more complex ones, where we let each new totality of relations correspond to a new realm of objects. The development of the realm of number fully confirm this meaning, which Kant associated with the idea of mathematical synthesis.” Cassirer (1950, 75). Thus, if we leave aside the Kantian theory of intuition, Hamilton’s definition of algebra as the “science of pure time or order in progression” may be understood “in conformity with a fundamental idea of Kant.” Cassirer (1950, 77).

154

Hernán Pringe

more general expression nothing else than simple serial principles, which we comprehend in their totality and in their transformation according to law.” (Cassirer 1923, 74). As we have seen, Cohen based his account of time and space precisely on the interpretation of differential and integral calculus as expressions of the way in which pure thought generates the Kantian forms of sensibility. Cassirer indicates that this generating character of thought may also be found in any other field of mathematical cognition: The construction of curves out of the totality of their tangents, as shown in infinitesimal geometry, is only an example of a procedure of more general applicability (Cassirer 1923, 75).

The development of mathematics shows that the task of this science “does not consist in comparing, dividing or compounding given magnitudes, but rather in isolating the generating relations themselves, upon which all possible determination of magnitude rests, and in determining the mutual connection of these relations.” (Cassirer 1923, 95). This supreme and universal principle is finally reached in the theory of groups, which provides a unitary perspective from which to overview the whole field of mathematics. Various examples may be indicated of the principle according to which the connected elements are not given prior to their connection, but rather result from the original rules of connection. For our discussion of the relation between Cohen and Cassirer, Grassmann’s extension theory [Ausdehnungslehre] is nevertheless of particular importance. In the Ausdehnungslehre, Grassmann claims that pure mathematics is “the science of the particular existent that has come to be by thought.” (Grassmann 1995, 24). Grassmann calls this particular existent produced by thought form. Therefore, pure mathematics is the theory of forms. Each particular existent can be produced either through a simple act of generation or through a twofold act of placement and conjunction. The continuous form or magnitude emerges in the first way, the discrete form in the second (Grassmann 1995, 25). Moreover, the particular existent may be coordinated with other particulars by the concept of the different or by the concept of the equal. In the first case, the algebraic form is what emerges, and in the second case, it is the combinatorial form. By combining these two oppositions, four types of forms emerge, together with the corresponding branches of the theory. The discrete particular existent unified with others as equal, i. e. the algebraic discrete form, is number. The discrete particular existent unified with others as different, i. e. the combinatorial discrete form, is combination. The sciences of the discrete are therefore number theory and combination theory (Grassmann 1995, 26). Analogously, the continuous form arising through generation of equals is the algebraic continuous form or in-

Cohen and Cassirer

155

tensive magnitude. The continuous form arising through the generation of the different is the combinatorial continuous form or extensive magnitude. The science of intensive magnitudes is function theory, i. e. differential and integral calculus, and that of extensive magnitudes is extension theory. Extension theory [Ausdehnungslehre] raises the science of space to a pure science of form, since geometry, according to Grassmann, does not possess this character. Grassmann maintains that in geometry space is taken as given and not as a product of thought (Grassmann 1995, 24). Rather, by means of extension theory, “all principles expressing spatial intuitions are omitted and consequently the origins of this science are as immediate as those of arithmetic.” (Grassmann 1995, 11).²⁵ This elimination of intuition from the foundations of mathematics could have been a major reason for the poor reception of Grassmann’s work, since the Kantian views on the necessary role of intuition in mathematics were still dominant. For instance, in a letter to Möbius, Ernst Friedrich Apelt, professor of philosophy at Jena, asks him: Have you read Grassmann’s strange Ausdehnungslehre? I know it only from Grunert’s Archiv; it seems to me that a false philosophy of mathematics lies at its foundation. The essential character of mathematical knowledge, its intuitiveness [Anschaulichkeit], seems to have been expelled from the work. Such an abstract theory of extension as he seeks could only be developed from concepts. But the source of mathematical knowledge lies not in concepts but in intuition (Crowe 1994, 79).²⁶

To this letter, Möbius responded: You ask me whether I have read Grassmann’s Ausdehnungslehre. To this I answered that Grassmann himself presented me with a copy and that I have on numerous occasions attempted to study it but have never gotten beyond the first pages, since, as you mentioned, intuitiveness [Anschaulichkeit], the essential character of mathematical thought, is not to be found in the work (Crowe 1994, 79).²⁷

In contrast to such negative opinions on Grassmann’s theory, Cassirer praises the Ausdehnungslehre precisely because Grassmann goes back from the consideration of an extensive manifold as merely given to its rule of generation, thereby confirming that the elements of mathematical calculus are not magnitudes but relations. Mathematics does not presuppose any given manifold, as Kant claims, but rather any measurable and divisible magnitude presupposes different types

 Translation slightly modified.  Quoted by Crowe. Translation slightly modified.  Quoted by Crowe. Translation slightly modified.

156

Hernán Pringe

of connection and arrangement, as Leibniz maintains. From this viewpoint, Cassirer claims that “the logic of mathematics, as Grassmann understands it, is, in fact, in a strict sense a ‘logic of origin’ [Logik des Ursprungs].” (Cassirer 1923, 99). The development of modern mathematics, which Cohen intended to treat in a never written second volume of his Logic, was finally dealt with by Cassirer in Substance and Function. Cassirer indicates that this development confirms the core of Cohen’s doctrine, namely the claim of the generative power of pure thought, expressed by the concept of origin. Modern mathematics enables us to broaden the scope of Cohen’s investigation, focused on infinitesimal calculus, to other fields of mathematics, so that the generation by pure thought may now be understood in terms of a serial principle: Cohen’s Logik der reinen Erkenntnis developed its fundamental thought of the origin in connection with the principles of infinitesimal calculus. Here, in fact, is the first and most striking example of the general point of view, which leads from the concept of magnitude to the concept of function, from “quantity” to “quality” as the real foundation. In advancing to the other fields of modern mathematics, the logical principle here established gains new confirmation. However different these fields may be in content, in structure they all point back to the fundamental concept of origin. The postulate of this concept is fulfilled wherever the members of a manifold are deduced from a definite serial principle and exhaustively represented by it (Cassirer 1923, 99; my emphasis).

2.3 The matter of intuition: mathematical and physical manifolds By considering the development of modern mathematics, Cassirer shows how time and space may finally be understood as products of pure thought and that therefore those manifolds that Kant identified as forms of intuition have their origin rather in spontaneity. But what about the matter of intuition? Just as Cohen, Cassirer argues that the fundamental issue of the critique of knowledge is the problem of the relation between logical and mathematical concepts and the objects of experience. ²⁸ As we have seen, Cohen finds the key to solving this problem in the judgment of reality. Reality [Realität] is the presupposition of both empirical natural science and pure mathematics. This reality, understood as infinitesimal, grounds the possibility of the empirical application of mathemati-

 “Die logischen und mathematischen Sätze mögen von rein hypothetischer Geltung sein: aber ist es lediglich ein “glü cklicher Zufall”, daß diese Hypothesen sich zureichend erweisen, die empirischen “Tatsachen” zu meistern und ihren Verlauf im voraus zu bestimmen?” Cassirer (1907, 44).

Cohen and Cassirer

157

cal cognition. In contrast, Cassirer argues that the principle of the infinitesimal method is just an example of a serial principle.²⁹ The discussion of the developments of modern physics will show that not only pure mathematical manifolds but also the empirical manifolds of natural science evolve entirely from serial principles. This will enable the application of the former to the latter. What is the epistemological status of physical concepts? Cassirer maintains that the theory of abstraction does not provide a proper answer to this question. The theory of abstraction describes the concept as a generic representation, i. e. as a representation of those features that are common to several individuals. This theory presupposes the existence of those individuals in their multiplicity and the power of the mind to identify their shared properties (Cassirer 1923, 4). Three functions of thought are thereby involved. First, comparison of the given manifold of individuals. Second, reflection on the features in which they agree. Third, abstraction of those dissimilar elements, so that clear consciousness of the common features is finally achieved (Cassirer 1923, 5). Against the abstraction theory, Cassirer maintains that a concept is not a group of properties shared by the individuals falling under it, but a universal rule that enables us to survey a total series of possible determinations at a single glance: We do not proceed from a series aα1β1 aα2β2 aα3β3 […] directly to their common constitutive a, but replace the totality of individual members α by a variable expression x, the totality of individual members β by a variable expression y. In this way we unify the whole system in the expression a x y […] which can be changed into the concrete totality of the members of the series by a continuous transformation, and which therefore perfectly represents the structure and logical divisions of the concept (Cassirer 1923, 23).

The unity of the concept does not correspond to a fixed set of features, but to a rule which represents the manifold of elements in accordance with a law. The introduction of the manifold of sensation into such a universal serial connection is the first step in the construction of a mathematical theory of nature.³⁰ Once empirical appearances are brought under concepts as serial principles the physical theory may be finally established. In the terms of contemporary philosophy of science, Cassirer’s claim is that physical theories are not directly related to em-

 For a detailed discussion of the Leibnizian roots of this argumentative move, see Ferrari (1988, 181– 252).  The sensible manifold of sensation is substituted by the intellectual manifold of measure and number. The latter is the conceptual or ideal limit of the former. For a discussion of Cassirer’s methodology of idealization in mathematics see Mormann (2008).

158

Hernán Pringe

pirical appearances, but rather to models, which we put in place of the “facts of perception”: We investigate the impact of bodies by regarding the masses, which affect each other, as perfectly elastic or inelastic; we establish the law of the propagation of pressure in fluids by grasping the concept of a condition of perfect fluidity, we investigate the relations between the pressure, temperature and volume of gas by proceeding from an “ideal” gas and comparing a hypothetically evolved model to the direct data of sensation (Cassirer 1923, 130).

According to Kant, “the logical actus of the understanding, through which concepts are generated as to their form, are: 1. comparison […], 2. reflection, and 3. abstraction.” (Log 09: 94). As to their form, empirical concepts are so generated. By contrast, as to their matter, empirical concepts are given a posteriori.³¹ For Kant, the formation of empirical concepts by means of comparison, reflection and abstraction depends on the condition that the matter of appearances be homogenous enough for the understanding to discover certain common features and to form with them generic representations.³² Without the homogeneity of appearances in regard to their matter “no empirical concepts and hence no experience would be possible.” (CPR A 654/B 682). While the homogeneity of appearances is necessary for the formation of empirical concepts, the fact that this sameness of the kind of the appearances actually obtains is contingent. The homogeneity of the matter of appearances cannot be established a priori, because it concerns that aspect of the appearance which can only be determined a posteriori. In that situation, Kant claims that reason “prepares the field for the understanding” (CPR A 657/B 685) by presupposing this homogeneity as a subjective condition for the use of the understanding. Such a presupposition consists in the representation of nature as if appearances (regarding their matter) were homogeneous enough to make the formation

 Regarding the Kantian doctrine of concepts, Cassirer claims that the analytical universal is not the universal of generic concepts but of functional ones. See Cassirer (1920, 373 n1). But, Ihmig points out, although pure concepts may be understood in that way, Cassirer′s interpretation is not valid in the case of empirical concepts. See (Ihmig 1997, 224).  “If among the appearances offering themselves to us there were such a great variety – I will not say of form (for they might be similar to one another in that) but of content, i. e., regarding the manifoldness of existing beings – that even the most acute human understanding, through comparison of one with another, could not detect the least similarity (a case which can at least be thought), then the logical law of genera would not obtain at all, no concept of a genus, nor any other universal concept, indeed no understanding at all would obtain, since it is the understanding that has to do with such concepts.” (CPR A 654/B 681 f.).

Cohen and Cassirer

159

of empirical concepts possible.³³ This transcendental presupposition is necessary for the understanding to search for the conceptual unity under which a manifold of appearances is brought, for without such a presupposition reason “would set as its goal an idea that entirely contradicts the arrangement of nature.” (CPR A 651/B 679). Kant also approaches the problem of the possibility of empirical concepts from the perspective of the third critique.³⁴ The task of finding empirical concepts, under which given particulars are to be subsumed, is that of the reflecting power of judgment (CJ, AA 05: 179). This task depends on transcendental conditions that go beyond those of pure understanding. For, in order to build empirical concepts not only are the categories to be applied to appearances, but one must moreover presuppose that, given a certain appearance, it has some empirical features common to other appearances as well.³⁵ Without the presupposition that such common features can be established, the reflecting power of judgment would not even commence its activity, since it would not attempt a task which is considered to be impossible. Thus, “the principle of reflection on given objects of nature is that for all things in nature empirically determinate concepts can be found.” (EEKU, AA 20: 211).³⁶  Along with the principle of homogeneity, the principles of specification and continuity of forms are presupposed as well in CPR A 657/B 685.  On the relationship between the appendix to the transcendental dialectic and the Critique of the Power of Judgment see: Thöle (2000).  “Now of course pure understanding already teaches (but also through synthetic principles) how to think of all things in nature as contained in a transcendental system in accordance with a priori concepts (the categories); only the (reflecting) power of judgment, which also seeks concepts for empirical representations, as such, must further assume for this purpose that nature in its boundless multiplicity has hit upon a division of itself into genera and species that makes it possible for our power of judgment to find consensus in the comparison of natural forms and to arrive at empirical concepts, and their interconnection with each other, through ascent to more general but still empirical concepts.” (EEKU, AA 20: 213).  Kant explains: “On first glance, this principle does not look at all like a synthetic and transcendental proposition, but seems rather to be tautological and to belong to mere logic. For the latter teaches how one can compare a given representation with others, and, by extracting what it has in common with others, as a characteristic for general use, form a concept. But about whether for each object nature has many others to put forth as objects of comparison, which have much in common with the first in their form, it teaches us nothing; rather, this condition of the possibility of the application of logic to nature is a principle of the representation of nature as a system for our power of judgment, in which the manifold, divided into genera and species, makes it possible to bring all the natural forms that are forthcoming to concepts (of greater or lesser generality) through comparison.” (EEKU, AA 20: 211– 212). Cassirer claims that “in establishing the principle of formal purposiveness” Kant “spoke as the logician of Linnnaeus’ descriptive science.” Cassirer (1950, 127).

160

Hernán Pringe

However, there is no guarantee of the success of that search for empirical concepts. Whether appearances have in fact these common empirical features or not cannot be a priori determined, because, as empirical, these features must be given a posteriori. By means of the presupposition of the homogeneity of appearances no determination of appearances as homogeneous is made. Rather, we just assume a maxim for our reflection upon them. The sameness of kind is not necessary for the object but just for the subject, in so far as the latter searches for empirical concepts. That we are indeed able to form such empirical concepts and that we can subsume a manifold of appearances under them is a mere coincidence between a subjective requirement and an ultimate fact in regard to which the subject is completely passive: the sensible matter as a homogeneous empirical content. Thus, according to Kant, the possibility of empirical concepts cannot be established a priori and we are rejoiced when we can form them, “as if it were a happy accident which happened to favor our aim.” (CJ, AA 05: 184). While this reference to a lucky chance may be considered a valid way to formulate the problem of the coordination of empirical concepts and spatiotemporal objects, by no means does it express a satisfactory solution to it.³⁷ On the contrary, if we realize that physical concepts are serial principles, the coordination problem does not arise: Thus no content of experience can ever appear as something absolutely strange; for even in making it a concept of our thought, in setting it in spatial and temporal relations with other contents, we have thereby impressed it with the seal of our universal concepts of connection, in particular those of mathematical relations. The material of perception is not merely subsequently molded into some conceptual form; but the thought of this form constitutes the necessary presupposition of being able to predicate any character of the matter itself, indeed, of being able to assert any concrete determinations and predicates of it (Cassirer 1923, 150).

According to Cassirer, the function of numbering and measuring is required in order first to produce the empirical facts upon which any physical theory is to be built. While the theory of abstraction presupposes facts that subsist for themselves, the functional theory of concepts denies that there are physical concepts and physical facts in separation: “we possess “the facts” only by virtue of the totality of concepts, just as, on the other hand, we conceive the concepts only with reference to the totality of possible experience.”(Cassirer 1923, 147). Physical concepts of measurements are “the real and necessary apperceptive concepts for all empirical knowledge in general.” (Cassirer 1923, 149 – 150). Without this refer See Cassirer (1920, 13). See also Bauch (1917, 413 ff.).

Cohen and Cassirer

161

ence of the empirical fact to the totality of experience, the “unity of consciousness” itself would be destroyed, i. e. the “fact” would no longer belong to “my” reality. Any empirical fact must necessarily belong to some series, for it would otherwise not count in general as a fact. Kant claims that reason, when approaching nature, “with its principles in one hand, according to which alone the agreement among appearances can count as laws, and, on the other hand, the experiments thought out in accordance with these principles” is not instructed by nature like a pupil, who listens to all that his master chooses to tell him, “but like an appointed judge who compels witnesses to answer the questions he puts to them.” (CPR B XIII). The functional theory of the concept explains this crucial feature of scientific knowledge. When performing an experiment, the physicist not only receives the matter of intuition in space and time or subsumes this sensible manifold under pure concepts of understanding. In an experiment, an anticipatory schema of empirical concepts as serial principles is already put forward, so that the experiment just determines which of the possible series is applicable to the case at hand.³⁸ Precisely for this reason, the possibility of a total heterogeneity of nature is no longer open. While, for Kant, the sensible manifold might be so heterogeneous that it could turn out not to be fit for subsumption under any empirical concept,³⁹ for Cassirer the formation of empirical concepts is not a “happy accident” anymore. Kant maintains that the distinction between empirical and pure is a metaphysical one which rests on the different origin of the representations.⁴⁰ The matter of knowledge, sensation, is empirical, because it has its origin in experience, as the “effect of an object on the capacity for representation, insofar as we are affected by it.”(CPR A 19 – 20/B34). The form of knowledge is pure, because it has its origin in our cognitive capacity, either in our sensibility or our understanding. In contradistinction, for Cassirer, “the possibility of separating the ‘matter’ of knowledge from its ‘form’ by referring them each to a different origin

 “The system of ‘possible’ relational syntheses already developed in mathematics affords the fundamental schema for the connections, which thought seeks in the material of the real. As to which of the possible relational connections are actually realized in experience, experiment, in its result, gives its answer.” (Cassirer 1923, 257). In particular, experience determines which of the possible hypothetical-deductive systems of geometry corresponds to “actual space.” See Cassirer (1957, 418).  In that situation, “every particular could here stand alone, in a sense; and to be known at all it would have to be apprehended and described simply by itself.” Cassirer (1950, 125). On this issue, see Jáuregui (2013).  “It is easy to see that in the distinction between empiricus and purus what matters is the origin of the concept, and this is already a metaphysical investigation, then.” (V-Lo/Wiener, AA 24: 905).

162

Hernán Pringe

in absolute being disappears; as when, for example, we seek the origin of one factor in ‘things,’ and the other in the unity of consciousness.” (Cassirer 1923, 310). The difference between matter and form is rather to be understood as a difference regarding the role that each of these moments plays in the problem of knowledge. The empirical in our scientific knowledge “means ultimately nothing but the proposition that universal rules of connection and universal equations of natural phenomena do not suffice for the construction of its object; but the knowledge of particular constants is also needed, such as can only be gained by experimental observation.” (Cassirer 1923, 311). No difference in origin is here assumed. Neither can the peculiar qualitative character of a certain sensation count as a proof of the dependence of sensation on an absolute transcendent ground, because this determinateness is only achieved by reference to some possible order and thus to a serial principle. If we insisted on abstracting even from this condition, we “would absolutely annihilate” the matter of knowledge. In such a case there would be neither a matter nor a form into whose ground and origin we could enquire (Cassirer 1923, 311). The type of correlation between the matter and form of knowledge which Cassirer puts forward also modifies another central distinction in Kant’s doctrine. For Kant, sensibility provides us with a manifold to be unified by the synthesis of understanding: the intuitive manifold must be brought under the unity of the concept. But, from the viewpoint of the functional theory of the concept, the manifold is not given by any receptivity but produced by the spontaneity of thought: The ‘real’, as it is grasped in the sensuous impression, is not in and for itself a ‘sum’ of various elements, but first appears to us as an absolutely simple and unanalyzed whole. This original ‘simplicity’ of intuition is only transformed into a manifold by the logically analytic work of the concept. The concept is here just as much a source of plurality as it elsewhere appears to be a source of unity (Cassirer 1923, 258).

Both mathematical and physical manifolds, i. e. pure and empirical manifolds, are constructed from serial principles.⁴¹ But, “while in mathematics the construction reaches a fixed end, in experience it is in principle incapable of completion.” (Cassirer 1923, 254). The empirical character of physical manifolds does not consist in its being passively received but in its always being under construc-

 In this connection, Biagioli claims that intuition “retains a role in determining the task of showing that mathematical and natural concepts are of the same kind.” Biagioli (2020, 37).

Cohen and Cassirer

163

tion.⁴² Pure mathematics may be applied to nature because this application is not the imposition of a structure of thought on a being independent of it. Rather, the mathematization of nature is the construction of her being according to a rational order, in a never-ending process.⁴³ This process is what we call experience.⁴⁴

Conclusions Cohen and Cassirer share the project of developing the critique of knowledge beyond the stage reached by Kant. For this purpose, both agree that the Kantian distinction between sensibility and understanding must be reformulated. Against the Kantian doctrine, they argue that we can no longer maintain, as Kant does, that “appearances can certainly be given in intuition without functions of the understanding” (CPR A 90/B 122). As is well known, in the transcendental deduction Kant shows that although “intuition by no means requires the functions of thinking,” (CPR A 91/B 123) the sensible manifold must nevertheless be brought under the transcendental unity of apperception in order for this manifold to be included into my experience. In this sense, the understanding establishes a condition that sensibility must satisfy if intuitions are to be something for me (CPR B 132). But, as we have seen, with the satisfaction of this condition, which guarantees the applicability of the pure concepts of the understanding to sensibility, the problem of the formation of empirical concepts arises, for the subsumption of appearances under categories is not sufficient to

 “Accordingly, the concept formation of physics and chemistry proves to be just as genuinely genetic as that of pure mathematics. But this genesis may be said to be of hypothetical rather than categorical character.” Cassirer (1957, 443 – 444). As Mormann claims: “In other words mathematical concepts are closed while empirical ones are open: the implicit definition of a point in Euclidean geometry fixes the meaning of this concept once and for all while the key concepts of empirical science have a ‘serial form’ in that their meaning is not fixed by a single theoretical framework. Rather, it emerges in a series of theoretical stages in the ongoing evolution of scientific knowledge.” Mormann (2008, 163).  “Number stands at the threshold of a series of relationships which, further pursued and conceived more concretely, should lead finally to the determination of ’reality’ and be included in it.” Cassirer (1950, 66).  “Thus in its relation to the empirical datum the thread of thought never breaks off; but neither can it be spun to its end: such a conclusion would not mean the completion of the fabric, but its veritable destruction, because it would run counter the meaning of experience as a progressive process of determination.” Cassirer (1957, 421– 422). See Moynahan (2003, 47).

164

Hernán Pringe

make such empirical concepts possible.⁴⁵ The doctrine of the regulative use of reason (or reflective judgment) attempts to address this issue, but the possibility of empirical concepts still remains an open problem. Cohen and Cassirer put forward different proposals in order to explain the relation between thought and being, which move away from the Kantian doctrine. This vaguely formulated relation acquires a more determinate expression in the question concerning the possibility of natural science, and even more precisely, in the problem of mathematical physics or applied mathematics. Cohen finds in the infinitesimal reality the intellectual element that may carry out the transcendental role that Kant ascribes to sensation. As we have seen, since this element is not only to be found in natural science, but already in pure mathematics, it makes the applicability of mathematics to experience first possible. Cassirer does not replace the Kantian sensation for an intellectual element, but for a member of an intellectual series. ⁴⁶ Mathematical and physical manifolds are of the same kind, but they differ in the fact that the construction of the empirical manifold is an infinite task. Cohen criticizes Cassirer’s position for not acknowledging that the series presupposes elements to be connected: Yet, I admittedly confess, that after my first reading of your book I still cannot discard as wrong what I told you in Marburg: you put the center of gravity upon the concept of relation and you believe that you have accomplished with the help of this concept the idealization of all materiality. The expression even escaped you that the concept of relation is a category; yet it is a category only insofar as it is function, and function unavoidably demands the infinitesimal element in which alone the root of the ideal reality can be found (Gawronsky 1949, 21).

Cohen subordinates the concept of relation to his logic of origin and emphasizes that the category of function is based on the judgments of origin and reality. In

 See Cassirer (1920, 11– 12). Heis overlooks this problem in his reconstruction of Cassirer’s argument in Heis (2014, 251– 252). The subsumtion of appearances under categories by means of an a priori synthesis is necessary for the possibility of the process of “abstraction”, but it is not a sufficient condition.  On this issue, Giovanelli claims: “Cohen’s idea that the ‘differential’ intensively contains in itself the capacity to generate magnitudes is replaced with the idea of the prevalence of the relation over the terms that it connects. […] The passing of a magnitude into another (e. g. from extensive to intensive magnitude, from finite to infinitely small magnitudes) is not at stake here [i.e for Cassirer], but rather the passing from magnitude as such to the functional relation between magnitudes.” (Giovanelli 2011, 210). In this respect it should be underlined that in his Logic, as already mentioned, Cohen rather maintains: “There is no intensive magnitude.” (Cohen 1922, 492).

Cohen and Cassirer

165

particular, he claims that the equation y = f(x) expresses a dependence of y on x which presupposes that the x variations correspond to the principle of infinitesimal continuity (Cohen 1922, 280).⁴⁷ Contrary to this, Cassirer assumes the relational nature of thought and from this viewpoint tries to overthrow the abstraction theory of the concept. Whereas Cohen analyzes the laws of pure thought, Cassirer aims to highlight the functional character of concepts in the development of modern mathematics and physics.⁴⁸ Cohen claims to have founded reality on pure thought, without reference to any given element. The role played by Kantian sensation is now assumed by the infinitesimal, which in turn grounds the logical foundation of time, in the first place, and space, in the second place. For Kant, the content of knowledge was given by sensation, whereas for Cohen it is a product of mere thought. Infinitesimal reality, which provides said content, is the most basic position of pure thought. On the contrary, for Cassirer the role played by Kantian sensibility is not a point of departure, but the ideal goal of the logical determination. The relational nature of thought is shown first in the concepts of number and time and then in the concept of space and the science of geometry. Empirical manifolds are the expressions of the infinite task of logical determination by means of relational concepts.⁴⁹ Whereas for Cohen the infinitesimal is the logical presupposition of the whole edifice of mathematics and mathematical physics, Cassirer not only idealizes all content of thought, but he also highlights the relational character of the infinitesimal reality itself. Since infinitesimals also depend on a serial principle, for Cassirer the infinitesimal reality cannot be considered an absolute element,

 In the same sense, Gawronsky indicates that the concept of function presupposes the magnitudes that it connects without being able to legitimize their generation (Gawronsky 1910, 82).  In 1914 Cohen discussed Leibniz’s infinitesimal analysis and cited Cassirer’s “Leibniz’ System in seinem wissenschaflichen Grundlagen” and the two volumes of “Geschichte des Erkenntnisproblems in der Philosophie und Wissenschaft der neueren Zeit,” but he omitted any reference to Substanzbegriff und Funktionsbegriff (Cohen 1914, 24). According to Mormann and Katz, this omission suggests that the differences of 1910 between the two philosophers had not been resolved in the meantime (Mormann and Katz 2013).  In contradistinction to Kantian schematism, Cassirer argues that the coordination of concepts and empirical objects is not achieved by means of a third element (the schema), which albeit intellectual is nevertheless also sensible. Rather, in Cassirer’s view, this coordination takes place through a specification of concepts that should be sought “within the domain of concepts itself.” (Cassirer 1956, 166). Such domain of physical knowledge makes up a system of invariants of experience, in which the causality principle constitutes the ultimate condition upon which the possibility of the coordination of concepts and spatio-temporal objects depends. For a discussion of this issue, see Pringe (2014).

166

Hernán Pringe

as Cohen claims.⁵⁰ Rather, even the infinitesimal reality gets dissolved into a system of relations. Thus, if “the mystery of idealism is to dissolve the variety of things into differences of ideas,” (Cohen 1871, 270) as Cohen declares, then Cassirer seems to be closer than Cohen to its disclosure.⁵¹

Bibliography Bauch, Bruno (1917): Immanuel Kant, Berlin. Beiser, Frederick (2018): Hermann Cohen. An Intellectual Biography, Oxford. Biagioli, Francesca (2020): Ernst Cassirer’s transcendental account of mathematical reasoning, in: Studies in History and Philosophy of Science 79. pp. 30 – 40. Cassirer, Ernst (1906): Der kritische Idealismus und die Philosophie des ‘gesunden Menschenverstandes’, edited by H. Cohen and P. Natorp, Giessen. Cassirer, Ernst (1907). Kant und die moderne Mathematik. (Mit Bezug auf Bertrand Russells und Louis Couturats Werke über die Prinzipien der Mathematik.). Kant-Studien, 12(1 – 3), pp. 1 – 49. Cassirer, Ernst (1920): Das Erkenntnisproblem in der Philosophie und Wissenschaft der neueren Zeit, 3. Band, Berlin. Cassirer, Ernst (1923): Substance and Function, Chicago. Cassirer, Ernst (1950): The Problem of Knowledge. Philosophy, Science and History since Hegel, New Haven. Cassirer, Ernst (1956): Determinism and Indeterminism in Modern Physics, New Haven. Cassirer, Ernst (1957): The Philosophy of Symbolic Forms, vol. 3, New Haven. Cohen, Hermann (1871): Kants Theorie der Erfahrung, 1. Aufl., Berlin Cohen, Hermann (1883): Das Princip der Infinitesmal-Methode und seine Geschichte. Ein Kapitel zur Grundlegung der Erkenntnisskritik, Berlin.

 Gawronsky claims that Cohen’s attempt to put infinitesimal numbers “as an absolute element before the whole number and to derive the latter from the former” is both logically and mathematically “an impossible undertaking.” He explains that “Cassirer’s ‘function’, as contrasted with ‘substance,’ meant just that: it is impossible to adscribe an absolute value to a mathematical element, since this value is determined by different relations to which it may belong.” (Gawronsky 1949, 21). It should be noted that Gawronsky argued for Cohen’s doctrine of the differential as an absolute unity in 1910. See Gawronsky (1910, 99). On the relationship between Gawronsky and Cassirer, see Ferrari (2011).  Contrary to this, Rudolph claims: “Cassirer hat, so lässt sich das Fazit der Cohenschen Kritik zuspitzen, den Boden einer Philosophie der Letzbegründung verlassen und bewegt sich nicht mehr im Rahmen einer Position, die in welchem Sinne auch immer, als ‘Idealismus’ bezeichnet werden kann.” (Rudolph 2009, 232). Luft has recently put forward an interpretation of Cassirer’s philosophy of symbolic forms as an absolute idealism. According to Luft, Cassirer’s difference from Hegel is “that the philosopher’s standpoint is no standpoint of its own but parasitic upon the individual standpoints of the symbolic forms (Luft 2015, Chapter 4, 25). For Cassirer′ s discussion of the differences between critical and absolute idealism, see Cassirer (1920, 362 ff).

Cohen and Cassirer

167

Cohen, Hermann (1885): Kants Theorie der Erfahrung, 2. Aufl., Berlin. Cohen, Hermann (1902): Logik der reinen Erkenntnis. System Der Philosophie, 1, 1. Aufl., Berlin. Cohen, Hermann (1914): Einleitung mit kritischem Nachtrag zur “Geschichte des Materialismus” von F. A. Lange. 3. Aufl, reedited by H. Holzhey, Hildesheim, 1984. Cohen, Hermann (1922): Logik der reinen Erkenntnis. System der Philosophie, 1. 3. Aufl., Berlin. Crowe, Michael (1994): History of Vector Analysis, New York. Dedekind, Richard (1963): Essays on the Theory of Numbers, New York. Dufour, Eric (2002): Le statut du singulier. Kant et le néokantisme de l’École de Marbourg, in : Kant-Studien 93 (3), pp. 324 – 350. Edel, Geert (1988): Von der Vernunftkritik zur Erkenntnislogik, Freiburg/München. Ferrari, Massimo (1988): Il giovane Cassirer e la scuola di Marburgo, Milan. Ferrari, Massimo (2011): Dimitrij Gawronsky und Ernst Cassirer: Zur Geschichte der Marburger Schule zwischen Deutschland und Russland, in: C. Krijnen und K. Zeidler (Hrsg.): Gegenstandsbestimmung und Selbstgestaltung. Transzendentalphilosophie im Anschluss an Werner Flach, Würzburg, Königshausen & Neumann, 2011, pp. 89 – 106. Flach, Werner (2009): Hermann Cohens systemtheoretische Subjektslehre, in: C. G. Paolo (ed.): Unità della ragione e modi dell’esperienza. Hermann Cohen e il neokantismo, Soveria Mannelli, pp. 35 – 45. Giovanelli, Marco (2011): Reality and Negation. Kant’s Principle of Anticipations of Perception, Dordrecht. Giovanelli, Marco (2016): Hermann Cohen’s Das Princip der Infinitesimal-Methode. The history of an unsuccessful book, Studies in History and Philosophy of Science 58, pp. 9 – 23. Gawronsky, Dimitry (1910): Das Urteil der Realität und seine mathematischen Voraussetzungen, Marburg. Gawronsky, Dimitry (1949): Ernst Cassirer: His Life and Work. A Biography, in: P. Schilpp (ed.): The Philosophy of Ernst Cassirer, pp. 3 – 37. González Porta, Mario (2011): Estudos neokantianos, São Paulo. Gordin, Jakob (1929): Untersuchungen zur Theorie des unendlichen Urteils, Berlin. Grassmann, Hermann (1995): A New Branch of Mathematics, Chicago/La Salle. Heis, Jeremy (2011): Ernst Cassirer’s Neo-Kantian philosophy of geometry, in: British Journal for the History of Philosophy, 19 (4), pp. 759 – 794. Heis, Jeremy (2014): Ernst Cassirer’s Substanzbegriff und Funktionsbegriff, in: HOPOS: The Journal of the International Society for the History of Philosophy of Science 4, pp. 241 – 270. Holzhey, Helmut (1986): Cohen und Natorp, 2 Bände, Basel/Stuttgart. Ihmig, K.-N. (1997): Cassirers Invariantentheorie und seine Rezeption des ’Erlanger Programms’, Hamburg. Jáuregui, Claudia (2013): Finalidad y uniformidad: el problema de las regularidades empíricas en el contexto del idealismo trascendental kantiano, Estudios de Filosofía, 48, pp. 99 – 108. Kant, Immanuel: Gesammelte Schriften. Hrsg.: Bd. 1 – 22 Preussische Akademie der Wissenschaften, Bd. 23 Deutsche Akademie der Wissenschaften zu Berlin, ab Bd. 24 Akademie der Wissenschaften zu Gö ttingen. Berlin 1900 ff.

168

Hernán Pringe

Kant, Immanuel (1998): Critique of Pure Reason [Kritik der reinen Vernunft], Cambridge: Cambridge University Press. Kant, Immanuel (2000): Critique of the Power of Judgment [Kritik der Urteilskraft] (1790), Cambridge: Cambridge University Press. Kant, Immanuel (2002): Theoretical Philosophy after 1781, Cambridge: Cambridge University Press. Luft, Sebastian (2015): The Space of Culture, Oxford. Moynahan, Gregory B. (2003): Hermann Cohen’s Das Prinzip der Infinitesimalmethode. Ernst Cassirer, and the Politics of Science in Wilhelmine Germany, in: Perspectives on Science, 11 (1), pp. 35 – 75. Mormann, Thomas (2008): Idealization in Cassirer’s Philosophy of Mathematics, in: Philosophia Mathematica (III) 16, pp. 151 – 181. Mormann, Thomas and Katz, Mikhail (2013): Infinitesimals as an issue of neo-Kantian philosophy of science, in: Hopos: The Journal of the International Society for the History of Philosophy of Science 3 (2) (2013), pp. 236 – 280. Mormann, Thomas (2018): Zur Mathematischen Wissenschaftsphilosophie des Marburger Neukantianismus, in: D. Christian: Philosophie und Wissenschaft bei Hermann Cohen, Cham, pp. 101 – 134. Natorp, Paul (1910): Die logischen Grundlagen der exakten Wissenschaften, Leipzig/Berlin. Natorp, Paul (1986a): Zu Cohens Logik, in: Holzhey, H. Cohen und Natorp, Band 2, pp. 6 – 40. Natorp, Paul (1986b): Zu Cohens Logik, in: Holzhey, H. Cohen und Natorp, Band 2, pp. 43 – 78. Nelson, Leonard (1905): Review of Cohen, Logik der reinen Erkenntnis, in: Göttingische gelehrte Anzeigen, 8, pp. 610 – 630. Oliva, Luca (2015): Kant and the Neo-Kantians on mathematics, in: N. De Warren and W. Staiti, (eds.): New Appoaches to Neo-Kantianism, Cambridge, pp. 285 – 306. Peckhaus, Volker (2014): Das Erkenntnisproblem und die Mathematik. Zum Streit zwischen den Marburger Neukantianismus und dem Neofriesianismus, in: C. Krijnen and K. Zeidler: Wissenschaftsphilosophie im Neukantianismus, Würzburg, pp. 233 – 257. Pringe, Hernán (2014): The Principle of Causality and the Coordination of Concepts and Spatio-Temporal Objects in Cassirer’s Philosophy, in: Idealistic Studies 44 (1), pp. 51 – 66. Rudolph, Enno (2009): Hermann Cohen – Ernst Cassirer: Lehrer und Schüler im Widerstreit. Zu Cassirers Cohenkritik, in: G. P. Cammarota (ed.): Unità della ragione e modi dell’esperienza. Hermann Cohen e il neokantismo, Soveria Mannelli, pp. 231 – 241. Russell, Bertrand (1903) The Principles of Mathematics, London: Routledge & Kegan Paul. Thöle, Bernhard (2000): Die Einheit der Erfahrung. Die regulativen Prinzipien bei Kant, in: R. Enskat (ed.): Erfahrung und Urteilskraft, Würzburg, pp. 113 – 134. Yap, Audrey (2017): Dedekind and Cassirer on mathematical concept formation, in: Philosophia Mathematica, pp. 369 – 389.

List of Contributors Brigitte Falkenburg, Prof. Dr. Dr., Institute of Philosophy and Political Sciences, Technische Universität Dortmund Dustin Garlitz, PhD Candidate, University of South Florida Hauke Heidenreich, Dr. des., Martin-Luther University, Halle-Wittenberg Christian Krijnen, Dr., Associate Professor at the Faculty of Humanities, Ancient, Patristic and Medieval Philosophy, Vrije Universiteit Amsterdam Roberto Horácio de Sá Pereira, Dr., Distinguished Professor at the Department of Philosophy, Universidade Federal do Rio de Janeiro (UFRJ) Hernán Pringe, Dr., Researcher at CONICET, Institute of Philosophy, University of Buenos Aires (Argentina) and Full Professor at the Institute of Philosophy, Universidad Diego Portales (Santiago, Chile)

Topics of the Kant Yearbook 2021, 2022 and 2023 The Kant Yearbook dedicates its next three issues to the following topics:

2021: Political Philosophy 2022: Kantian Constructivism 2023: Philosophy of Language

https://doi.org/10.1515/kantyb-2020-0008