Donate books to help fund our work. Learn more→

The Rudolf Steiner Archive

a project of Steiner Online Library, a public charity

Truth and Science
GA 3

II. Kant’s Theory of Knowing’s Basic Questions

Kant is usually cited as the founder of the theory of knowing (Erkenntnistheorie) 24t/n Kant’s Critique of Pure Reason was published in 1781, well before the term epistemology was introduced into the world of philosophy in 1854 by the Scottish philosopher Ferrier. The term epistemology has become more and more popular with English-speaking philosophers, and is usually used by them to translate Erkenntnistheorie, which when properly translated into native English words is Theory of Knowing. in the modern sense of the word. One could rightly object to this view by saying that the history of philosophy before Kant contains numerous investigations that should be viewed as more than just the seeds of such a science. Volkelt also notes in his fundamental work on the theory of knowing that the critical treatment of this science began with Locke. 25Vokelt, Erfahrung und Denken. Kritische Grundlegung der Erkenntnistheorie. Hamburg und Leipzig 1886, S.20. But even in earlier philosophers, even in the philosophy of the Greeks, one finds discussions that are currently brought up again to clarify the theory of knowing. All the problems discussed there were churned and digested in depth by Kant, and following him, numerous thinkers worked through them in such a comprehensive manner, that the earlier attempts at solutions can be found either in Kant himself or in his followers. So, by being purely factual rather than historical, the present study of the theory of knowing will not miss anything of importance, but of course while also including everything of importance since the appearance of Kant's Critique of Pure Reason (Kritik der reinen Vernunft). What was achieved beforehand in this field has been recreated in this epoch starting with Kant.

Kant's basic epistemological question is: How are synthetic judgments possible a priori? 26t/n a priori is knowing something having nothing to do with physical sense-perceptions. a priori knowing includes analytic concepts such as 2+2=4 true by virtue of their meaning alone, and synthetic concepts like “all physical bodies are heavy” (have gravitational attraction to other bodies), true both because of what they mean and because of the way the world is.. Let's look at this question in terms of its lack of presuppositions! Kant raises the issue because he is of the opinion that we can only acquire unconditionally certain knowledge if we are able to prove the justification of synthetic judgments a priori. He says: "Proving this justification must include the possibility of the pure use of reason in the founding and implementation of all sciences, all that that contain a priori theoretical knowledge of objects." 27Kant, Critique of Pure Reason, Kritik der reinen Vernunft, p. 61ff, Kirchmann’s German edition. (All other page numbers in reference to Critique of Pure Reason are from this edition.) “Proving this involves whether philosophy’s first principles (metaphysics) stand or fall, and therefore whether they exist at all.” 28Ibid, Preface Section V

Is this question, as Kant poses it, free of presuppositions? Not at all, because it makes the possibility of an unconditionally certain system of knowledge dependent on the fact that it is built up only from synthetic judgments and from judgments that are gained independently of all experience. Kant calls synthetic judgments those in which the concept of the predicate adds something to the concept of the subject that lies entirely outside of it, “even though it is connected with it” 29Ibid p. 53f. whereas in analytical judgments the predicate only says something that already exists in the subject in a hidden way. This probably is not the place to address Johannes Rehmke's 30Rehmke, Johannes: Die Welt als Wahrnehmung und Begriff, usw., The World as Percept and Concept, etc., Berlin, 1880, p. 161 ff. sharp objections to this structure of the judgments. For our present purpose it is sufficient to see that we can only attain truthful understanding (das Wissen) through judgments which add to a concept a second concept, the content of which, at least for us, was not yet contained in the first. If, with Kant, we want to call this class of judgments synthetic, we can at least admit that knowing, that understanding, can only be gained in the form of judgment if the connection between the predicate and the subject is synthetic. But things are different with the second part of the question, which requires that these judgments be gained a priori, independently of all experience. It is quite possible (by this we mean, of course, the mere possibility of thinking) that such judgments do not exist at all. At the beginning of the theory of knowing it must be considered completely undetermined whether we can come to judgments only through experience or without any prior similar experience. Yes, when viewed without bias, such independence seems impossible from the outset. Whatever becomes known, it must first enter our immediate and individual awareness, it must be a direct experience.

We also acquire mathematical judgments by simply experiencing them individually. Even if you were to believe, as B. Otto Liebmann does,31B. Otto Liebmann, Zur Analysis derWirklichkeit. Gedanken und Tatsachen, On the Analysis of Reality. Thoughts and Facts that mathematical facts are grounded in the specific organization of our consciousness, then the matter would be no different. One can then say that this or that sentence is necessarily valid, because if its truth were to be abolished, consciousness would also be abolished, but we can only know it if it becomes an experience for us, exactly the way a process in external nature is experienced. No matter whether the content of such a sentence contains elements that guarantee its absolute validity, or whether it is secured for other reasons, I cannot get hold of it in any other way than by its confronting me as an experience. This is one thing.

The second concern is that at the beginning of epistemological investigations, one must not claim that knowing something’s absolute validity cannot come from experience. It is quite conceivable that the experience itself could have some characteristic which would guarantee the certainty of the insights gained from it.

There are two presuppositions in Kant's line of questioning. The first is that we must have a way other than experience to know something. The second is that all understanding of experience can only have limited validity. Kant is not at all aware that these propositions need to be examined, that they can be doubted. He simply takes them over as prejudices from dogmatic philosophy and uses them as the basis for his critical investigations. Dogmatic philosophy presupposes them as valid and simply applies them, arriving at the process of knowing corresponding to them. Kant assumes they are valid, and then only asks himself under what conditions can they be valid? But what if they are not valid at all? Then Kant's theory lacks any basis. Everything Kant puts forward in the five paragraphs that precede the formulation of his basic question is an attempt to prove that mathematical judgments are synthetic.32Note added later by Rudolf Steiner: This attempt, incidentally, is one which the objections of Robert Zimmermann (Über Kant's mathematisches Vorurteil und dessen Folgen) show it to be, if not altogether in error, at least highly questionable. 33t/n (Robert Zimmermann, 1824–1898, was Professor of Philosophy in the University of Vienna, 1861–95. His book on Aesthetics was published in 2 volumes, Vienna, 1870. Rudolf Steiner attended lectures by Zimmermann on fundamentals of ethics at the University of Vienna. Steiner's impressions of this great interpreter of Herbart's aesthetics are contained in the 3rd chapter of Steiner’s own autobiography.) 34t/n Analytic truths such as 2 + 2 = 4 are true just due to their meaning, but synthetic truths such as snow is white depends on the meaning in relationship to other perceptions in the world.

But the two assumptions cited remain as scientific prejudices. In Kant’s Critique of Pure Reason, he says “Experience teaches us that something is one way or another, but not that it cannot be otherwise” and “Experience never gives its judgments true or strict ones, only assumed ones and comparative generality (by induction).” 35Kant, Critique of Pure Reason, in Einleitung II, p. 58, Sec. v. In his preface we find, “First, as far as the sources of metaphysical knowledge are concerned, it is already inherent in their concept that they cannot be empirical. Their principles (not only their basic axioms but also their basic concepts) must therefore never be taken from experience, that is, from knowing something from physical sensation, but from metaphysical sources, from knowledge beyond experience.” 36Kant, Critique of Pure Reason, Prolegomena Paragraph 1 Finally, Kant says, “First of all, it must be noted that actual mathematical propositions are always a priori judgments and not empirical, because they entail necessity which cannot be derived from experience. But if you don't want to admit this, then I'll limit my statement to pure mathematics, the very concept of which implies that it does not contain empirical knowing, but only pure a priori knowing.37Kant, Critique of Pure Reason, p. 58, Sec. v. We may open the Critique of Pure Reason wherever we want, and we will find that all investigations within it are conducted under the presupposition of these dogmatic propositions. Cohen 38Hermann Cohen (1842–1918), Kants Theorie der Erfahrung, Kant's Theory of Experience, Berlin, 1871, pp. 90 ff. of the German ed. and Stadler 39August Stadler (1850–1910), Die Grundsätze der reinen Erkenntnistheorie in der Kantschen Philosophie, The Principles of the Pure Theory of Cognition in the Philosophy of Kant, Leipzig, 1876, p. 76 f. of the German ed. try to prove that Kant demonstrated the a priori nature of mathematical and purely scientific propositions. Now everything that is attempted in the criticism can be summarized as follows:

‘Because mathematics and pure natural science are a priori sciences, the form of all experience must be grounded in the subject. So, all that remains is the material of sensations that is empirically given (given through sensory nerves). This is built up into a system of experience through the forms lying in the mind. The formal truths of the a priori theories only have meaning and significance as organizing principles for the material of sensation; they make experience possible, but do not extend beyond it. However, these formal truths are the synthetic judgments a priori, which, as conditions of all possible experience, must therefore reach as far as the latter itself. The critique of pure reason therefore does not prove the apriority of mathematics and pure natural science, but only determines their area of validity. The prerequisite is that its truths should be gained independently of experience.’ Yes, Kant does so little to provide a proof for this a priori, rather he simply excludes it. The part of mathematics, Kant says, in which the same could be doubted, even in his opinion, is limited only to what he says can be deduced from simpler concepts. 40Kant, Critique of Pure Reason, p. 29, line 26f. Johannes Volkelt also finds that “Kant starts from the explicit presupposition that there actually is a general and necessary Wissen (the experience of understanding sense perceptions and non-sensory concepts).” He goes on to say, "This presupposition, which Kant never explicitly examined, is so contradictory to the character of a critical examination of epistemology (kritischen Erkenntnistheorie, critique of theory of knowing) that one must seriously consider the question of whether the Critique of Pure Reason is a valid critique of epistemology." Although Volkelt believes that one can answer this question in the affirmative, for good reasons, "the attitude of critiquing in Kant's epistemology is fundamentally disturbed by this dogmatic presupposition." 41Vokelt, Erfahrung und Denken. Kritische Grundlegung der Erkenntnistheorie. Hamburg und Leipzig 1886, p. 21 But enough, even Volkelt finds that the Critique of Pure Reason is not an epistemology without presuppositions.

The views of O. Liebmann, Hölder, Windelband, Überweg, Eduard von Hartmann 42Otto Liebmann (1840–1912), Analysis, 1880, A. Holder, Kantischen Erkenntnistheorie, Tübingen, 1874, p. 14 ff., Wilhelm Windelband (1848–1915) Phasen der Kantschen Lehre, p. 239, F. Ueberweg, System der Logik, p. 380 f., Eduard von Hartmann (1842–1906), Kritische Grundlegung, Berlin, 1875, p. 142–172 of the 2nd German ed., and Kuno Fischer hold essentially similar views on this point, namely, that Kant bases his whole argument on the assumption that knowledge of pure mathematics and natural science is acquired a priori. and Kuno Fischer 43Kuno Fischer Geschichte der neueren Philosophie VB. S.60 says that Volkelt is wrong when he comments on my own (Kuno Fischer’s) comments on Kant's epistemology (p. 198 f.) that "it is not clear from K. Fischer's presentation whether in his view Kant only supports the psychological actuality of the general and necessary judgments or at the same time presupposes the objective validity and lawfulness of the same. In the passage cited, Fischer says that the main difficulty in the Critique of Pure Reason is due to its “foundations being dependent on certain presuppositions”, “which one must have admitted to accept the following”. For Fischer, these prerequisites of Kant’s argument are that “first the fact of knowing” is established, and then, through analysis, the properties of knowing are found “from which that fact itself is explained”. also essentially agree with my view, that Kant places the a priori validity of pure mathematics and natural theory as a prerequisite at the top of his discussions, that we really know things independently of all experience, and that experience only provide insights of comparative generality, that we could accept only as a corollary of other judgments. These claims must necessarily be preceded by an investigation into the nature of experience and one into the nature of knowing. Only after this could the first and all following sentences follow.

Now one could reply to any objections raised in these reasoned critiques the following: that every theory of knowing must first lead the reader to an unconditional starting point. What we generally know at any point in our lives is far removed from this starting point, so we first must be artificially led back to it. In fact, such a purely didactic instructional intention is necessary for every epistemologist at the start of any consideration. This must be limited to showing to what extent the beginning of knowing in question really is the beginning, for it would have to proceed in purely self-evident analytic logically-reasoned sentences, and unlike Kant’s argument, should not make any supposedly meaningful claims that might influence the content of the following discussions. It is also the responsibility of the epistemologist (Erkenntnistheoretiker) to show that the beginning that he assumes is really without presuppositions. But all of this has nothing to do with the nature of this beginning itself, but stands entirely outside of it, and says nothing about it. Even at the beginning of mathematics lessons, I must try to teach students the axiomatic character of certain truths. But no one will want to claim that the content of axioms is dependent on previous considerations. 44In the chapter titled "The Starting Point of Epistemology," I shall show to what extent my discussion fulfils these conditions. In the same way, the epistemologist should show in his introductory remarks how one can arrive at a beginning without presuppositions of any sort, for the actual content must be free of any prior considerations. The work of Kant, whose initial assertions are specifically dogmatic, is far from a proper introduction to epistemology.