The Critical Philosophy of Hermann Cohen

The Interpretation of Plato

2 The Interpretation ofPlato 1. THE INITIAL PSYCHOLOGICAL INTERPRETATION Cohen's "further investigation"! of Kant's critical idealism byway of his reading of Plato began with the brief but intense 1878 essay Platons Ideenlehre und die Mathematik. Cohen had been showing interest in Plato since the beginning of his philosophical studies. His first unpublished work, which won him the Breslau University Philosophy Faculty prize in 1863 bears the significant title Uber die Psychologie des Plato und Aristoteles. In the following years, after moving to Berlin, Cohen took part in another competition with an essay on the concepts of' casus' and' contingens ' in Aristotle, which earned the praises of Trendelenburg,2 even though it was not awarded the prize. On the strength of a new version of this essay, however, Cohen was awarded a university lecturing license by the University of Halle in 1865.3 An important part of this dissertation is devoted to the interpretation of Plato (cf. 5 I: 18-21). The last of the theses presented in the final part of the essay, "Platonis Convivium post Phaedonem scriptum esse" (5 I: 29) shows that Cohen's investigation of Plato was already philologically oriented. In 1866 he published his first work dealing exclusively with Plato: Die platonische Ideenlehre psychologisch entwickelt. Nevertheless, even if Cohen's interest in Plato can be documented from his very early philosophical studies, his critical-idealist approach to Plato cannot be traced back to a date earlier than 1878, the year Platons Ideenlehre und die Mathematik was published. With the sole exception of the 1876 article, Friedrich Albert Lange, which already contains an impor21 22 THE CRITIC<\L PHILOSOPHYOF HERMANN COHEN tant theme in Cohen's new interpretation of Plato-the idealist interpretation of Greek philosophy, particularly of the line of thought from Democritus to Plato, which will be dealt with later-Cohen's previous writings show his attachment to the Volkerpsychologie of Lazarus and SteinthaI. These writings, therefore, deal with problems unrelated to critical idealism and posit theses that were either to be abandoned or considerably modified from the 1878 essay onward. Some commentators consider Die platonische Ideenlehre psychologisch entwickell and Platons Ideenlehre und die Mathematik to be components of a coherent interpretation of Plato. This was especially true in the years in which the Marburg school's work on the interpretation of Plato was still in full swing, a fact which made a neutral, comprehensive examination of its development more difficult.4 E. Hohmann,s when criticizing the unilateral nature of Natorp's exclusively logical, nonontological interpretation of Plato's ideas, wrote that, though Cohen was the "father" of this kind of interpretation, he was less radical than Natorp, since in his 1866 essay he recognized that Plato did develop, however negatively, in the direction of a substantialist conception ofideas. Ifwe want to see matters in a correct historical perspective and have a clearer theoretical idea of Cohen's thought, it is better to think of Die platonische Ideenlehre pjychologisch entwickell as a youthful work, which was superseded to all intents and purposes by his later studies, and which is of negligible interest from the standpoint of critical idealism. One striking example of Cohen's early outgrowing his initial interpretation of Plato is the fact that in the 1866 essay he recognized an ontologically oriented development of Plato's theory of ideas, which was stressed by Hohmann, while this recognition is nowhere to be found in his subsequent works. This does not mean, of course, that Cohen had necessarily abandoned this view,6 but that, from 1878 onwards, his interpretation of Plato's ideas changed radically. He was no longer interested in investigating the birth and psychological development of the theory of ideas in Plato's mind, but turned to the analysis of the logical, transcendental, and critical meaning of these ideas, as Plato's truly significant contribution to critical idealism. Die platonische Ideenlehre psychologisch entwickelt was published in the Zeitschrift fur Volkerpsychologie und Sprachwissenschaft co-edited by Lazarus and SteinthaI. It was conceived and written entirely under the influence of the idea of the shared identity of philosophy and psychology that characterizes Cohen's early thought.7 This is true for at least two reasons: principally, because interest in Plato is justified by the recognition of the latter's role as the originator of "psychological philosophy" (S I: 39); second, because the interpretation of the birth of Plato's theory of The Interpretation ofPlato 23 ideas as an original "discovery" (Entdeckung) , for which the way was prepared by earlier and contemporary Greek culture and philosophy, but which really sprang from a psychological process in a genius' mind, is an excellent example of the "psychological" historiographical method theorized by Cohen in the same essay (cf. S I: 30 ff.). There are two themes in this essay that deserve special attention, not so much because they were to be taken up again in later works, but because they were to undergo substantial changes during this development and thus become valuable yardsticks in the evaluation of Cohen's outgrowing his early interpretation of Plato. These two themes, which I shall now briefly discuss, are the distinction between doo; and ioea and the relationship between Plato and Kant. In his 1866 essay Cohen tried to show that it is possible to identify a clear distinction between Plato's use of the terms cloo;and ioea, starting from the dialogues of the second period.8 The first term, often used in the plural, corresponds to the Socratic concept and is synonymous with yevo;, ovvapl;, I/Jvm;, and ovcria. The second term, which is only rarely used in the plural, is the truly original, creative concept discovered by Plato, meaning the act of intuition, the ''vital intellectual activity of contemplation (Schauen) " (S I: 61), which is capable ofgrasping the essence of things. Plato's Idea, which has a close etymological connection with the act of contemplation, is thus the transposition, into the field of general knowledge, of the intuitive act of immediate understanding of essence, which Plato had seen in the work of the artist: So when this question arose: what is the Socratic concept, the essence of a work of art, of a table, of the beautiful? Plato's answer was: what I contemplate as this essence, the image revealed to me in contemplation, that which the artist observes in his creation. and the carpenter or mechanic in his making a chair or shuttle. Did he ask: what is it that attracts us in Phidias' statue of Zeus? Is there a being completely resembling it in the world of bodies? No. Then is it perhaps a purely spiritual essence, a product of Eastern speculation? No, it is the image of the art ofgenuine Greek sculpture. But if it is neither body nor spirit. what is it then? It is the true union of both, which we grasp during contemplation, it is the vision (Gesicht). to use Biblical language, the intuitive sign of a harmonious union between these two extremes, which in discursive thought never again will allow themselves to be united. In the harmonious fusion of the everlasting with the ephemeral. of the necessary with the contingent. of the spiritual with the bodily. of what is spaceless with what is in space, this vision is the true essence of the work ofart we all admire. (S I: 53-4) Ideas as dory are therefore none other than the hypostatized result of the act of spiritual contemplation, from which the object of contempla- 24 THE CRITICAL PHILOSOPHYOF HERMANN COHEN tion has not yet been isolated as a separate substance. This act is the authentic, original meaning of the Platonic ioea. Each substantialization ofan idea, then, is a distancing from and distortion of the original meaning , "which alone can allow access to knowledge of the real sense" (S I: 65), of the Platonic Idea. This distorting process had already begun in Plato's own works, in particular in Parmenides (cf. SI: 65ff.) . Distinguishing eloo~ from ioea is also a permanent feature of Cohen's subsequent interpretation of Plato. However, at least in one fundamental respect, there was a significant change in its meaning. In the 1866 essay the close association of Idea with the psychic act of contemplation is not only a metaphor, but an identification, which led Cohen to think of the Platonic Idea as a forerunner of Fichte's concept ofintellectual intuition: Socrates interpreted essence, the concept as the thing it is (das Seiende) , but left one question open: how do we know this essence, this concept? Plato answered the question, with the conditional originality ofthe discoverer , by pointing to contemplation as the activity peculiar both to thinker and artist, as the phenomenon of all creation, whether it be at a high or low level. Thus he is the original forebear of intellectual intuition, oftranscendental idealism. (S I: 53) It is interesting to note how transcendental idealism is grounded in intellectual intuition, at this stage in Cohen's development. However, in the 1878 essay, he found important new features in Plato's theory of ideas: in the first instance, the function ofmathematics and the meaning of Idea as hypothesis. This was to bring about a radically new interpretation of the difference between eioo~ from ioea. The idea of contemplation is now only thought of as the psychic origin of Idea, or as the occasion of its discovery by Plato, or even as an image inspired by the word's etymology, but in reality misleading, owing to its subjective nature.9 For Cohen, the Platonic iOEa is still a further development of the Socratic eloo~ not as intellectual intuition, but rather as logical foundation : Idea is hypothesis, that is, critical foundation of the concept.1O In a passage in the Asthetik des reinen Gefuhls there is an interesting description of the development ofPlato's reflection on the meaning ofIdea: We have known since the Logik der reinen Erkenntnis that Plato established Idea as hypothesis. As regards the meaning of the word, he saw Idea in contemplation , and thus, in the guise of vision, it earned its plac(~ in the history of cultural problems. However, when his scientific knowledge reached maturity, Idea was no longer limited to vision. The initial psychological origin of Idea, qualified by contemplation, moved on in the direction of an objective, profound, systematic realization, in relation to mathematics. Thus Idea was not limited to vision, but became foundation, (ARC I: 243) The Interpretation ofPlato 25 This passage is more of a reflection of Cohen's progress in interpreting Plato's ideas than ofdevelopment in Plato's own thought. Moving on from the conception of Idea as a psychic-metaphysical act of the knowledge of essence to that of Idea as the logical foundation of concept denies any kind of legitimacy to intellectual intuition. In fact, Cohen now recognized that the latter is a distortion of the Platonic Idea, and was responsible for the metaphysical deviations in the subsequent history of philosophy: Intuition springs from idealism: it is probably the most effective constituent feature of Plato's Idea. Therefore, it may seem that it could have no better references. But it is well known that so far no agreement has been reached on the meaning of Idea. It is thus only too understandable that this ambiguous feature of the concept of Idea has become the fundamental reason behind a kind of idealism, whose falsity is revealed in its hostility to rationalism. Idealism is logically grounded scientific rationalism . If, on the other hand, intuition (Intuition) is not to activate pure contemplation , which is identical to pure thought, if it is to go beyond scientific thought, invalidate it and make it suspect, then the ambiguity of the intuitive moment (Anschauung) manifests itselfin this intuition, which is irreconcilably opposed to the truthfulness ofIdea. (LRE 500) 11 Mter this clarification of the meaning of the Platonic Idea, we can now form a more precise picture of the limits of the relationship between Plato and Kant posited by Cohen in his 1866 essay. Cohen placed Plato on the same line as Kant, inasmuch as the conception of Idea as intuition inevitably leads on to the recognition ofreality as a phenomenon . However, if such a development is coherently followed, it will lead on beyond Kant to Fichte. If things are inasmuch as they communicate ideas, and ideas are nothing other than intuitive acts ofthe subject, then this subject of intuition must be recognized as "that which only exists" (alleinseiend) (cf. S I: 68). As can be seen, Cohen had not only identified a Plato-Kant line, but also a Plato-Kant-Fichte line. Indeed, the interpretation of Plato in this essay is conditioned more by reference to Fichte than to Kant, as the interpretation ofIdea as intellectual intuition and the special meaning attributed to transcendental idealism, to which I have already referred, confirm.12 In subsequent works, especially Platons Ideenlehre und die Mathematik, Cohen developed further the tenuous links between Plato and Kant discovered in 1866. This time the direction was not Fichte oriented but led toward scientific and critical idealism. Particular attention was devoted to Plato's role as a forerunner in the establishment of the transcendental doctrine of knowledge, the discovery of the transcendental a priori, and the evaluation of the function of mathematics. 26 THE CRITICAL PHILOSOPHY OF HERMANN COHEN 2. PlATO IN THE HISTORY OF IDEALISM The new interpretation of Plato in Platons Ideenlehre und die Mathematik consists, in the first place, of a realignment of his position in the history of philosophy. He is now presented as the founding father of critical idealism. This historiographical act, which, as we shall see, has important theoretical implications, was nothing more than an initial response to the need for expanding the historical roots of critical idealism already felt by Cohen when preparing the first edition of Kants Theorie der Elfahrung. There are no traces of this new history of critical idealism in Greek philosophy in the 1866 essay, which followed the completely different historiographical line of the development of "psychological philosophy" (cf. S I: 45-53). The undoubted source and point of reference for Cohen's reconstruction of the history of ancient idealism was Lange's Geschichte des Materialismus. One piece of documentary evidence for this connection is provided by Cohen's obituary of Friederich Albert Lange written for the PreussischeJahrbUcherin 1876. The obituary contains critical discussion of Lange's book and the first references to Cohen's new historiographical approach, before those in Platons Ideenlehre und die Mathematik. Further evidence can be seen in Cohen's clearly antithetical position vis-a-vis Lange in many parts of his discussion.13 Lange considered Democritus to be the real founding father of materialism. It was the latter who, with his atomism, had opened up the road of science, freeing knowledge from the fallacious poetics .of myth and religion. He was the first to provide a rigorous definition of matter (d. ibid., p. 8). He replaced teleology with a mathematical, mechanical explanation of reality (cf. ibid., p. 13). He provided a materialistic interpretation of the soul (d. ibid., pp. 19ff.). He formulated a eudemonist morality which was entirely coherent with his theoretical materialism (d. ibid., pp. 2lf.). The theories of Socrates, Plato and Aristotle, which Lange believed to be part of a coherent whole,14 were a "reaction" (ibid., p. 38) ofreligious and poetic idealism against Democritus's scientific materialism and Protagoras's sensualism, which complements it gnoseologically. The only point in favor of this idealism is its claim that the human spirit has the right to free itself "in a poetic upward thrust above the rough, defective edifice of scientific knowledge" (ibid., p. 60) ,15 but, from the point ofview of the progress of scientific knowledge and of rigorous thought, it was a "great revolution which, for thousands of years, misdirected the world into the maze of Platonic idealism" (ibid., p. 42). In his 1876 obituary Cohen showed that he was already aware of the decisive importance of substantially correcting this interpretation of The Interpretation ofPlato 27 Greek philosophy, not only for historical reasons, but also, from the theoretical point ofview, in the name ofa correct definition ofgenuine idealism as critical idealism: Here there is a historical error, which led Lange to a systematically mistaken view. Did Kant, who put an end to materialism, invent idealism from nothing? Perhaps what Lange considered materialism before Kant was in fact idealism, in the same way as he admitted that there was an element of materialism in Kant himself. In short, how did Lange treat idealism in Plato? This question is of fundamental importance, both for a systematic solution and historical understanding. Though Democritus' work was written before Plato, all subsequent materialism depends on the latter. Was Plato cured of materialism? Did not Democritus depend on the Eleatic philosophers, whose speculations he rendered consequent in his own way? Is not the Eleatic theory one of Plato's main sources? Thus a clear view of the sources of idealism is crucial for a work adopting this approach. Following it, the traditional picture of philosophical viewpoints could be subjected to detailed revision. Judgment on the regulative factor in conflicting world views seems to depend on their delimitation on the basis ofa historical appreciation of Platonism. It must be admitted that the meaning and value of Platonism is unacknowledged in this work. (S II: 378-79) It was the significance of Democritus and his relationship with Plato that was crucial for this reinterpretation of Greek philosophy: "Democritus was the ancient philosopher who lay behind Lange's mistaken view ofPlato. And the understandable prominence he accorded to Democritus had the result of overshadowing Plato, though objectively this need not necessarily have been the case" (S II: 380). In Cohen's view, an opposition between materialism and idealism limited to the dogmatic appropriation ofmatter and spirit, two opposing principles of reality, was too superficial. In this sense one should recognize spiritualism, not genuine idealism, as the opposite of materialism: Materialism is in no way the opposite ofidealism. The two concepts are not mmually exclusive. Lange was well aware of this, and demonstrated the fact with regard to several historical figures. However, he never managed to formulate precisely the inexactitude of the traditional opposition, because he failed to acknowledge either the historical origin of this opposition , or the significance of Platonism. Materialism maintains that matter is the sole concrete reality, the only explanatory principle of every specific real being. In view of this, it is not entirely true to say that idealism posits Idea as corresponding to matter. It is more exact to say that idealism is in search of the meaning of concrete reality, while materialism, starting off from sense perception, places reality in something, for which the senses 28 THE CRITICAL PHILOSOPHY OF HERMANN COHEN cannot satisfactorily vouch. Idealism asks a question about the value of reality (Wirklichkeits-Wert) , of that something, of that matter. On the other hand, there are certain presumed forms of idealism which, to no lesser extent than materialism, dogmatically appropriate an equivalent something , the so-called spirit, as the location and organ of ideas. This deceptive name must be denied those attempts, and, on the basis of their teaching, the term spiritualism should be used to describe them. This is the natural opposite of materialism. (S II: 382-83) However, critical idealism was the genuine idealism, in the light of which the philosophy of Democritus and Plato was to be interpreted. This rational method of foundation provides the basis upon which the concepts of 'matter' and 'spirit' can and must be philosophically justified : On the other hand, idealism, in its classic forms in Plato, Descartes and Kant, is a method. not a doctrine. Its results, both positive and negative, are the consequences of a gnoseological critique. If idealism does acknowledge matter, this is a gnoseologically verified concept, i.e. a concept which has been investigated in its relationship with the conditions on which the certainty of knowledge is founded. If it groups together certain phenomena, considering them to be spiritual, since they would receive a false common name without such delimitation, idealism does not become spiritualism, even if it professes a noumenon: as long as this is verified on the basis ofits cognitive value, or, rather, on the basis of its degree ofvalidity . In this case, however insignificant its invitations to carry out such elaborations , one could ask where true opposition to idealism is located in philosophical practice. I know only one answer: in all thought that does not proceed gnoseologically, whether or not it starts offfrom faith in outer objects or rational things (Vernunftdinge). Immanent idealism is mathematical investigation of nature, if this is not an ambiguous expression. This criterion proves useful indeed for understanding the great historical forms, including materialism. What Lange admired in Democritus is certainly not simply materialism, but a gnoseological endeavor. (S II: 383) In Platons Ideenlehre und die Mathematik, Plato's theory of ideas is presented as a synthesis of multiple trends in previous thought, and, at the same time, as the original reworking of results achieved into a new line of thought (cf. ERW97). The Platonic theory of ideas, for the first time in the history of thought, by positing the problem of knowledge, had established idealism.16 In Cohen's view idealism had existed before Plato, but it had reached maturity, and therefore its true meaning, in the Platonic theory of ideas. Plato was the meeting point of three trends in Greek thought. Two of them had already been developed and summed up in Democritus. Democritus, on the one hand, was the heir of the The Interpretation ofPlato 29 Eleatic philosophers, inasmuch as he took over the principle of the identity between being and thought: "There is no other being than that of thought" (S I 337). However, while Parmenides's 'being which is thought' was still the sensible being of the cosmos, and only its changeable , illusory appearance, its becoming, denied, Democritus put aside the being of things and reinstated the being of pure thought, the void, the nonbeing (j1r, DV) as the true being (trenDv). He was able to achieve something Parmenides had not achieved, because he made use of the mathematical thought of Pythagoreanism and thus filtered off all sensible content from being: "Now there is a being which the Eleatics consider a non-being; this is the true being! This true being consists of mathematical concepts, not only as regards atoms, but also the abstraction ofa separator, in the thought of an interval (Zwischen), or ofa distance (Abstand), in which sensible beings are no longer present" (S I: 340-41). Cohen did not accept Lange's view of Democritus as the founding father of materialism. For him: "Eleatic thought regarding being passedfrom atomism and underwent an idealist transformation" (S I: 340-41). Democritus went beyond the synthesis he had achieved ofEleatic and Pythagorean thought. His achievement was taken over by Plato and grafted onto the Socratic tradition, with important new results. Once having put aside the thesis that the Platonic Idea is to be identified with the Socratic concept, Cohen did not even accept the conception of the Platonic Idea as a substantialization of this concept: "it would only be an Eleatic concept applied to the Socratic being; the result would then be the opinion, recently expressed with fruitful frankness, that the theory of ideas in no way represents Plato's specific philosophy" (S I: 343). Plato's starting point was the fundamental requirement of "skeptical idealism" (S I: 343) (or perhaps it would be better to use the term critical idealism), that is, the problem of error, which any critical theory of true being must face: "How can it come about that the Eleatic being, the validity of Duma should suit this Socratic elOD;, this thought, defined YEVOr,," while sensible phenomena become and change, as Heraclitus showed in his teaching on the cosmos? How can one understand that genuine being consists of something withdrawn, inasmuch as it is withdrawn, from any sensible corroboration?" (S I: 343). In Plato's theory ofideas this "critical constituent feature" (SI: 344) is at the same time a further investigation of the Socratic theory ofconcept and a decisive step forward. The Socratic concept is put to the test: Idea is the established concept, or, to put it better, the very foundation of the concept; it is thought justitying itself. Thus Plato passed from mere reflection on thought to the very formulation of the problem of knowl- 30 THE CRITICAL PHILOSOPHYOF HERMANN COHEN edge (S I: 341) and consequently can be considered the initiator of genuine idealism, that is,. of "gnoseologically founded" idealism (S I: 342). 3. THE THEORY OF IDEAS Cohen's intention, in his interpretation of the Platonic theory of ideas in Platons Ideenlt:hre und die Mathematik, was to deny the validity both of the ontological and of the psychological interpretations. If so many interpreters, over such a long period, had viewed Platonic ideas as separate entities, this was due to the fact that Aristotle's views on the subject had undeservedly carried too much weight. He must take responsibility for the outset of a distorted ontological interpretation of these ideas, which was to dominate the later history of philosophy. It was Aristotle who attributed separation (zmplCiJ1oq) between the world of ideas and the sensible world to Plato (cf. S I: 344), as well as the view ofideas as substances (cf. SI: 347) , and the placing ofmathematical concepts at a lower level than ideas, at a halfway stage between the latter and sensible things (cf. S I: 342) P Cohen also rejected the psychological interpretation of Plato's ideas as mere v017J1ara in the mind of the thinking subject, while he had accepted it in Die platonische Ideenlehre psycholog;isch entwickelt (cf. S I: 66) : If i&a is to be viewed correctly as being conditioned by thought, as in Parmenides, where it is at least mentioned as VOTJtla tv ljIVxai~, this state is not to be understood in the sense that Idea thus becomes the chimera of skepticism. Plato rejected this consequence in no uncertain terms. It is the following meaning of current opinion that is to be accepted without hesitation : that all that is known in pure thought was seen by Plato as genuine being. The opinion that the outcome of this pure thought, that what is revealed during its progress, when released from the illusion of the senses, is the creation of thought itself, instead of being the archetype with which the contemplating, reminiscing spirit complies and which it imitates-this phase of subjective idealism would be impossible to find in Plato. It is not even worth considering whether such a conclusion can be attributed to Phito at all. (S I: 345-46) Cohen's new interpretation thus saw Plato's theory of ideas as a genuine forerunner of Kant's transcendental theory of knowledge. It is true that ideas are the object of thought (vorJJ1ma) , but by going beyond the Eleatic tradition, which identified thought with being, and that initiated by Democritus, which recognized true being in nonbeing, Plato concluded that ideas, precisely because they are thought, are the true being The Interpretation ofPlato 31 (DV'rmq DV), since they are the truth of things: 'The beautiful and the good, the rectilinear and the equal-all these DVTa are Dvrmq DVTa, and are all conditioned by thought, inasmuch as they are v01]ra. Since they are thought, are not grounded in perception and cannot trace their origin in completion or expansion of the latter, they are thus Dvrmq Dvra, having the value ofouma, inasmuch as they are v017para" (SI: 346). The route Plato followed to arrive at this conception of ideas, and at the consequent establishment of knowledge as science (cf. S I: 349), was reflection on mathematics: "The necessary mediation is to be found in Plato's investigation of mathematics, in the essential direction for the modern gnoseological conception, where such a mediation can become a certainty, when it takes mathematical method and objects as its starting point. But this is how Idea itself takes on a new meaning, which reconciles the above mentioned alternative between ouma and v01]pa" (S I: 351). The passage in Plato's Republic (523ff.) where Socrates makes a distinction between sensations which induce intellectual investigation and those which do not were, in Cohen's view, of crucial significance. By pointing out that the Republic singles out mathematics as the type of intellectual investigation stimulated by certain sensations, he was led to the conclusion that Plato had regarded mathematics as the mediator between sensibility and understanding. This was the means by which investigation reached the heights of pure thought from humble empirical beginnings: 'Just as, on the one hand, the object of mathematical thought is linked to the sensible of ordinary perception, on the other hand, it also sets up a relationship with the cognitive value of ideas, and can thus be a mediator between the two extremities ofbeing, between DV and Dvrmq DV" (SI: 353). Thus, by way of mathematics, Plato, like Kant, had reached the fundamental discovery ofKantian critical idealism, that is the a priori ofsensibility itself,18 even if "in Plato's view, the senses in themselves certainly do not contemplate the v01]rov, but contain the impulse for this other kind of knowledge. This logical character distinguishes the types ofperception from one another" (S I: 353). However, in Plato's view, mathematical objects, despite their role as mediator, are in every sense ideas, inasmuch as they are objects of pure thought. In this sense, mathematical investigation already belongs to dialectic: 'Thus Plato is shown to have considered the entire cosmic world, at whose apex is situated the complex labyrinth of patterns in the heavens above, as mere examples for the problems of mathematical investigation. And there can be no doubt that those who devote their attention to the contemplation of mathe- 32 THE CRITICAL PHILOSOPHYOF HERMANN COHEN matical 'ideas,' in this very activity at the same time engage in dialectic" (S I: 357). Cohen did not try to avoid Plato's well known distinction between V01JC1l\; and oUiVOW;19 in fact he made use of it to proceed to hypothesis, a second, fundamental component of the meaning of ideas. Plato was correctly identified as the originator of analysis as a scientific method.20 The essential truth ofhis theory ofideas consists in establishing knowledge as critical method, and in the meaning ofIdea as hypothesis: It is clear that this method of taking what is sought for as found, only to reencounter it by means of deductions and connections between them, is highly productive for philosophical problems. This is why Plato himself advises and has recourse to imitation. However, this analogy is not only effective when single problems are to be dealt with. The fundamental concept of his specific philosophical method - he uses the term p£f:Jooo~ when treating the theory of ideas-springs from this characteristic of geometrical thought, at least for its gnoseological legitimation. Idea itself is thought as hypothesis. (S I: 361 ) Thus the theory of ideas manifests itself as the critical method of foundation: Idea "gives account" of the concept. In this sense the theory of ideas further investigates the Socratic theory of concept and constitutes the ultimate response to the fundamental question which was Cohen's starting point for his investigation: why should the value of ovmabe attributed to the being which is thought, the Socratic ElOO\;, when it is denied to sensible phenomena (cf. S I: 343): Idea, revealed in the guise of hypothesis by documentary evidence, is not only better understandable from the psychological point of view; it has been seen to be the spur to gnoseological criticism. From now on it is possible to understand how Idea, inasmuch as it is V017'l"OV, must, at the same time, be oum.a, and v017rov, inasmuch as it is oum.a: the two constituent features of idealism are intertwined in hypothesis. Only that which is thought as an adequate presupposition ofbeing corresponding to the law is thought as Idea, just as this presupposition can only become productive in the methodical connection ofthoughts, as their root. (S I: 362) Mathematical concepts are ideas, in the sense of hypotheses, in all respects, and the mathematical method is analytical, that is, grounded in hypothesis. However, ilt is not the ultimate degree of pure thought, since it cannot "be more than its hypotheses." Mathematics takes hypotheses as principles ofits reasoning, but indulges in no further discussion of the a priori establishment of these principles. Returning to the foundation of hypotheses to track down "that which is adequate for itself'21 in the The Interpretation ofPlato 33 absolute principle, in civV1roOerov,22 in the idea of good, goes beyond mathematics as a way of thinking; this is the "dialectical" process of the brurnlf.1T/; but both "mathematics and thought of ideas" "remain subsumed under pure rational thought (voT/au;) " (SI: 365).23 4. PLATO'S CRITICAL PHILOSOPHY It has often been said that the Kantian interpretation of Plato by Cohen and the Marburg school in general is unacceptable. As a matter offact, Cohen himself, when defending his interpretation (cf. E 25ff.), does not disavow this opinion at all. Nevertheless, this objection needs re-examing and correcting. Admittedly, Cohen does stress those aspects of Plato's theory of ideas that point forward to Kant's views, and also places Plato at the outset of the history of critical idealism, in which Kant's role is fundamental. Furthermore, I would suggest that Cohen's reading of Plato be seen in the light of his brand of Kantianism. As we have already seen, his interpretation of another philosopher centered on questioning him from the standpoint of the interpreter's present concerns. There is no doubt that in 1878, at the time of writing Platons Ideenlehre und die Mathematik, Cohen saw everything from a Kantian viewpoint . However, the differences Cohen brought out between Plato and Kant are far more significant than the similarities. What is more, these differences led Cohen to investigate critical philosophy further and revise his interpretation of Kant, which implied that a considerable critical distance from Kantian philosophy was set up, even to the extent, perhaps , of authorizing us to speak of Cohen subsequently reading Kant in a "Platonic" light.24 We have already seen that one of the most important "Kantian" motifs that Cohen unearthed in Plato was the identification of the a priori in sensibility, connected with the mediating role of mathematics between thought and sense perception. Admittedly, however, on this point Platonic theory, in Cohen's interpretation of it, contrasts decisively with Kant's, even though this is missed by Cohen himself in his 1878 essay. In Plato's view sense perception is merely a "stimulus" (tyepn"ov, Weckmittel) to "pure mathematical thought" (S I: 353): phenomena are simply examples (7rapaoeiYf.1ara, Beispiele) "for mathematical tasks" (S I: 358), but the latter are an integral part of pure thought (vOT/al~) (S I: 365). Mathematical thought is thus seen as a model and an integral part of knowledge, inasmuch as it is pure, nonrepresentative thought.25 It is to this nonrepresentative character of thought that the interpretation of ideas as hypotheses is linked. Plato's critical theory of knowledge is not only 34 THE CRITICAL PHILOSOPHYOF HERMANN COHEN the grounding of the object in hypothesis, of being in nonbeing, identified with thought, but also the production of content by pure thought. It is only thus that Democritus's 11i! ov, as Cohen interprets it, is fully realized in Plato. This conception of hypothesis as origin (Ursprung) and of the Platonic method of knowledge as method ofpurity (Methode der Reinheit) is clearly still not explicit in the 1878 essay. However it is potentially present , and it can be said that Cohen stops short of drawing all the conclusions from his reading of Plato. But when, in his later writings on Plato, after he had already developed his theory of origin and pure thought, Cohen stresses the meaning of the Platonic hypothesis, and sees it as a forerunner of his theories, this should not be seen as a modification of the views expressed in the 1878 essay, but as a coherent development of themes already to be found there, which have freed themselves from an earlier, cautious approach and been taken to their logical conclusions. The final point to note is the absence of an enquiry into the critical meaning of the idea of good in Platons Ideenlehre und die Mathematik. Cohen only briefly refers, without further development, to the critical function that the idea of good, as avv1Co(Jt:TOV, can exercise, both in the ethical and in the logical fields (cf. S I: 365). Undoubtedly this negligence is partly due to the limitations imposed by the subject of the essay. Nevertheless, when dealing with Plato, Cohen found his greatest difficulties with the interpretation of the idea of good. In 1878 Cohen was still a long way from a clear conception of this idea and from a satisfactory solution to his difficulty. When, several years later, he was to find one, it would bring about a partial break with Plato.26 5. THE INTERPRETATION OF PlATO'S THEORY OF IDEAS IN THE PosT-l878 WORKS The meaning of the Platonic Idea as hypothesis was further investigated by Cohen in the works that followed Platons Ideenlehre und die Mathematik. It became the fundamental historical point of reference for the "method of purity," theorized in Cohen's philosophical system.27 The Platonic Idea is a development of the Socratic concept, which, in its formulation as the question What is it? (ri ean), is a clear assertion of the primacy of thought over being, but not yet the authentic grounding ofbeing in thought. Thought can be the truth of being, that is the being of thought is true being, protected from error, only inasmuch as it is critically grounded in thought itself. Idea as hypothesis is the concept that gives account (AOYOV ()((50vaz, Rechenschaft geben) of itself. The Platonic The Interpretation ofPlato 35 Idea is not merely ground (Grundlage), since thought would then only account for itselfrelatively and the founding hypothesis would in turn be without foundation. Rather, it means the very "method" of "foundation" (Grundlegung). Therefore, the Platonic Idea as hypothesis is identical to critical knowledge, in which science is grounded.2H For this reason, although, on occasion, he identified the Platonic Idea with scientific knowledge (cf. KBE 27f.), Cohen, elsewhere and more important, showed that this identification is unsatisfactory and that Idea has a deeper originative nature.29 What is brzlTrri.u1]?-considering the ambiguity of the Greek word which means both "knowledge" and "science ,"-was for Cohen the crucial Platonic question, which provided its own answer: true knowledge is science, that is, well founded knowledge, for which ideas as hypotheses are the foundation: "We already have our answer here: ideas as foundations make up the content of knowledge, a treasure which can be forever increased by means of new foundations, even if all of them might turn out to be developments of older ones" (E lS).30 On the basis of these developments in the meaning of Idea, Cohen now identified the Platonic conception of mathematics with the model of pure thought, rather than with mediation between thought and perception : This is one of the most profound and daring thoughts in his [Plato's) methodology: that he detects a reason for thought in sensation itself. Thus numbers are posited as "stimulators" and "provokers of thought," which are already at work in sensation itself. Nevertheless. he clearly goes too far in his inclination to play down the conflict between thought and sensation. It was his predecessor Pythagoras who stressed the sense of substance in numbers. Plato also believed that the doctrine of numbers is a science of everlasting entities. Therefore, numbering is nothing other than thinking. However thinking is thinking unity. So producing unity is thinking, not only producing the number two, as in Plato's example. But he certainly believed that it is the moment which starts off this thought that is part of sensation. He did not believe that thinking numbers takes place within sensation. (LRA'472-73) It is not contentless thought that is "pure" (cf. LRE 5), but thought that produces its content by itself: "after Plato's /(a(}ap~, for any entity appearing in itself and for itself, pure signifies rather the means ofscientific production by which the entity as such is realized."31 Thought is therefore the true being inasmuch as it is being of the being. Pure thought is certainly not separate from knowledge, in the sense of being 36 THE CRITICAL PHILOSOPHYOF HERMANN COHEN contentless; on the contrary, it is inseparably tied to knowledge: pure thought is the thought of knowledge. However, the conclusion is to be drawn that real knowledge is pure thought, since the true foundation of the being is grounded in thought, which produces its content in the process of self-foundation: Idea became the justification of concept, in the sense that thought laid its own foundation on it. "Giving account" (;tarov 8z8ovaz) and laying foundation (VltoriBEaBaz) become synonyms. Logos is the same concept. But when it becomes logos by itself, then it becomes self-foundation. And this foundation of Idea means and guarantees the true being. There is no truth, no knowledge, no being beyond this,just as on this side there is no being and no science. (Lm~21 L) It is not difficult to realize how such reflections had already led Cohen far from Kantian transcendental philosophy, as he had interpreted it in the first edition of Kants Theorie der Erfahrung. Plato's pure thought is the opposite of representative thought, and scientific knowledge is the grounding of the object entirely in thought; in fact it is production of the object by thought: "In his critique of knowledge Plato's task was to bring about such a drastic separation as to put the thought of knowledge entirely on one side, while he put representation on the side of sensation. One ought 1:0 believe that representation also partakes of thought, or rather thoUlght of representation. But Plato makes a clean divide and, to all intents and purposes, assigns representation to sensation " (1<-'RW1l3). Cohen now saw the essence of scientific idealism, that is, critical idealism in Platonic pure thought, rather than in the Kantian a priori synthesis . It was Plato who had moved on from Socratic rationalism to idealism, by way of the critical meaning of ideas,32 since Plato's theory that"all laws are onlyfoundations, can only be foundations ... is the principle of every idealist method and thus of all true idealism and rationalism. Foundation is the criterion ofreason" (ARC I: 88). ...