Truth and Knowledge
GA 3
iv. The Starting Point of Epistemology
As we have seen in the preceding chapters, an epistemological investigation must begin by rejecting existing knowledge. Knowledge is something brought into existence by man, something that has arisen through his activity. If a theory of knowledge is really to explain the whole sphere of knowledge, then it must start from something still quite untouched by the activity of thinking, and what is more, from something which lends to this activity its first impulse. This starting point must lie outside the act of cognition, it must not itself be knowledge. But it must be sought immediately prior to cognition, so that the very next step man takes beyond it is the activity of cognition. This absolute starting point must be determined in such a way that it admits nothing already derived from cognition.
Only our directly given world-picture can offer such a starting point, i.e. that picture of the world which presents itself to man before he has subjected it to the processes of knowledge in any way, before he has asserted or decided anything at all about it by means of thinking. This “directly given” picture is what flits past us, disconnected, but still undifferentiated.*Differentiation of the given, indistinct, would picture into distinct entities is already an act of thought activity. In it, nothing appears distinguished from, related to, or determined by, anything else. At this stage, so to speak, no object or event is yet more important or significant than any other. The most rudimentary organ of an animal, which, in the light of further knowledge may turn out to be quite unimportant for its development and life, appears before us with the same claims for our attention as the noblest and most essential part of the organism. Before our conceptual activity begins, the world-picture contains neither substance, quality nor cause and effect; distinctions between matter and spirit, body and soul, do not yet exist. Furthermore, any other predicate must also be excluded from the world-picture at this stage. The picture can be considered neither as reality nor as appearance, neither subjective nor objective, neither as chance nor as necessity; whether it is “thing-in-itself,” or mere representation, cannot be decided at this stage. For, as we have seen, knowledge of physics and physiology which leads to a classification of the “given” under one or the other of the above headings, cannot be a basis for a theory of knowledge.
If a being with a fully developed human intelligence were suddenly created out of nothing and then confronted the world, the first impression made on his senses and his thinking would be something like what 1 have just characterized as the directly given world-picture. In practice, man never encounters this world-picture in this form at any time in his life; he never experiences a division between a purely passive awareness of the “directly-given” and a thinking recognition of it. This fact could lead to doubt about my description of the starting point for a theory of knowledge. Hartmann says for example:
“We are not concerned with the hypothetical content of consciousness in a child which is just becoming conscious or in an animal at the lowest level of life, since the philosophizing human being has no experience of this; if he tries to reconstruct the content of consciousness of beings on primitive biogenetic or ontogenetic levels, he must base his conclusions on the way he experiences his own consciousness. Our first task, therefore, is to establish the content of man's consciousness when he begins philosophical reflection.”108Hartmann, Grundproblem, p. 1. See note 4, above.
The objection to this, however, is that the world-picture with which we begin philosophical reflection already contains predicates mediated through cognition. These cannot be accepted uncritically, but must be carefully removed from the world-picture so that it can be considered free of anything introduced through the process of knowledge. This division between the “given” and the “known” will not in fact, coincide with any stage of human development; the boundary must be drawn artificially. But this can be done at every level of development so long as we draw the dividing line correctly between what confronts us free of all conceptual definitions, and what cognition subsequently makes of it.
It might be objected here that I have already made use of a number of conceptual definitions in order to extract from the world-picture as it appears when completed by man, that other world-picture which I described as the directly given. However, what we have extracted by means of thought does not characterize the directly given world-picture, nor define nor express anything about it; what it does is to guide our attention to the dividing line where the starting point for cognition is to be found. The question of truth or error, correctness or incorrectness, does not enter into this statement, which is concerned with the moment preceding the point where a theory of knowledge begins. It serves merely to guide us deliberately to this starting point. No one proceeding to consider epistemological questions could possibly be said to be standing at the starting point of cognition, for he already possesses a certain amount of knowledge. To remove from this all that has been contributed by cognition, and to establish a pre-cognitive starting point, can only be done conceptually. But such concepts are not of value as knowledge; they have the purely negative function of removing from sight all that belongs to knowledge and of leading us to the point where knowledge begins. These considerations act as signposts pointing to where the act of cognition first appears, but at this stage, do not themselves form part of the act of cognition. Whatever the epistemologist proposes in order to establish his starting point raises, to begin with, no question of truth or error, but only of its suitability for this task. From the starting point, too, all error is excluded, for error can only begin with cognition, and therefore cannot arise before cognition sets in.
Only a theory of knowledge that starts from considerations of this kind can claim to observe this last principle. For if the starting point is some object (or subject) to which is attached any conceptual definition, then the possibility of error is already present in the starting point, namely in the definition itself. Justification of the definition will then depend upon the laws inherent in the act of cognition. But these laws can be discovered only in the course of the epistemological investigation itself. Error is wholly excluded only by saying: I eliminate from my world-picture all conceptual definitions arrived at through cognition and retain only what enters my field of observation without any activity on my part. When on principle I refrain from making any statement, I cannot make a mistake.
Error, in relation to knowledge, i.e. epistemologically, can occur only within the act of cognition. Sense deceptions are not errors. That the moon upon rising appears larger than it does at its zenith is not an error but a fact governed by the laws of nature. A mistake in knowledge would occur only if, in using thinking to combine the given perceptions, we misinterpreted “larger” and “smaller.” But this interpretation is part of the act of cognition.
To understand cognition exactly in all its details, its origin and starting point must first be grasped. It is clear, furthermore, that what precedes this primary starting point must not be included in an explanation of cognition, but must be presupposed. Investigation of the essence of what is here presupposed, is the task of the various branches of scientific knowledge. The present aim, however, is not to acquire specific knowledge of this or that element, but to investigate cognition itself. Until we have understood the act of knowledge, we cannot judge the significance of statements about the content of the world arrived at through the act of cognition.
This is why the directly given is not defined as long as the relation of such a definition to what is defined is not known. Even the concept: “directly given” includes no statement about what precedes cognition. Its only purpose is to point to this given, to turn our attention to it. At the starting point of a theory of knowledge, the concept is only the first initial relation between cognition and world-content. This description even allows for the possibility that the total world-content would turn out to be only a figment of our own “I,” which would mean that extreme subjectivism would be true; subjectivism is not something that exists as given. It can only be a conclusion drawn from considerations based on cognition, i.e. it would have to be confirmed by the theory of knowledge; it could not be assumed as its basis.
This directly given world-content includes everything that enters our experience in the widest sense: sensations. perceptions, opinions, feelings, deeds, pictures of dreams and imaginations, representations, concepts and ideas.
Illusions and hallucinations too, at this stage are equal to the rest of the world-content. For their relation to other perceptions can be revealed only through observation based on cognition.
When epistemology starts from the assumption that all the elements just mentioned constitute the content of our consciousness, the following question immediately arises: How is it possible for us to go beyond our consciousness and recognize actual existence; where can the leap be made from our subjective experiences to what lies beyond them? When such an assumption is not made, the situation is different. Both consciousness and the representation of the “I” are, to begin with, only parts of the directly given and the relationship of the latter to the two former must be discovered by means of cognition. Cognition is not to be defined in terms of consciousness, but vice versa: both consciousness and the relation between subject and object in terms of cognition. Since the “given” is left without predicate, to begin with, the question arises as to how it is defined at all; how can any start be made with cognition? How does one part of the world-picture come to be designated as perception and the other as concept, one thing as existence, another as appearance, this as cause and that as effect; how is it that we can separate ourselves from what is objective and regard ourselves as “I” in contrast to the “not-I?”
We must find the bridge from the world-picture as given, to that other world-picture which we build up by means of cognition. Here, however, we meet with the following difficulty: As long as we merely stare passively at the given we shall never find a point of attack where we can gain a foothold, and from where we can then proceed with cognition. Somewhere in the given we must find a place where we can set to work, where something exists which is akin to cognition. If everything were really only given, we could do no more than merely stare into the external world and stare indifferently into the inner world of our individuality. We would at most be able to describe things as something external to us; we should never be able to understand them. Our concepts would have a purely external relation to that to which they referred; they would not be inwardly related to it. For real cognition depends on finding a sphere somewhere in the given where our cognizing activity does not merely presuppose something given, but finds itself active in the very essence of the given. In other words: precisely through strict adherence to the given as merely given, it must become apparent that not everything is given. Insistence on the given alone must lead to the discovery of something which goes beyond the given. The reason for so insisting is not to establish some arbitrary starting point for a theory of knowledge, but to discover the true one. In this sense, the given also includes what according to its very nature is not-given. The latter would appear, to begin with, as formally a part of the given, but on closer scrutiny, would reveal its true nature of its own accord.
The whole difficulty in understanding cognition comes from the fact that we ourselves do not create the content of the world. If we did this, cognition would not exist at all. I can only ask questions about something which is given to me. Something which I create myself, I also determine myself, so that I do not need to ask for an explanation for it.
This is the second step in our theory of knowledge. It consists in the postulate: In the sphere of the given there must be something in relation to which our activity does not hover in emptiness, but where the content of the world itself enters this activity.
The starting point for our theory of knowledge was placed so that it completely precedes the cognizing activity, and thus cannot prejudice cognition and obscure it; in the same way, the next step has been defined so that there can be no question of either error or incorrectness. For this step does not prejudge any issue, but merely shows what conditions are necessary if knowledge is to arise at all. It is essential to remember that it is we ourselves who postulate what characteristic feature that part of the world-content must possess with which our activity of cognition can make a start.
This, in fact, is the only thing we can do. For the world-content as given is completely undefined. No part of it of its own accord can provide the occasion for setting it up as the starting point for bringing order into chaos. The activity of cognition must therefore issue a decree and declare what characteristics this starting point must manifest. Such a decree in no way infringes on the quality of the given. It does not introduce any arbitrary assertion into the science of epistemology. In fact, it asserts nothing, but claims only that if knowledge is to be made explainable, then we must look for some part of the given which can provide a starting point for cognition, as described above. If this exists, cognition can be explained, but not otherwise. Thus, while the given provides the general starting point for our theory of knowledge, it must now be narrowed down to some particular point of the given.
Let us now take a closer look at this demand. Where, within the world-picture, do we find something that is not merely given, but only given insofar as it is being produced in the actual act of cognition?
It is essential to realize that the activity of producing something in the act of cognition must present itself to us as something also directly given. It must not be necessary to draw conclusions before recognizing it. This at once indicates that sense impressions do not meet our requirements. For we cannot know directly but only indirectly that sense impressions do not occur without activity on our part; this we discover only by considering physical and physiological factors. But we do know absolutely directly that concepts and ideas appear only in the act of cognition and through this enter the sphere of the directly given. In this respect concepts and ideas do not deceive anyone. A hallucination may appear as something externally given, but one would never take one's own concepts to be something given without one's own thinking activity. A lunatic regards things and relations as real to which are applied the predicate “reality,” although in fact they are not real; but he would never say that his concepts and ideas entered the sphere of the given without his own activity. It is a characteristic feature of all the rest of our world-picture that it must be given if we are to experience it; the only case in which the opposite occurs is that of concepts and ideas: these we must produce if we are to experience them. Concepts and ideas alone are given us in a form that could be called intellectual seeing. Kant and the later philosophers who follow in his steps, completely deny this ability to man, because it is said that all thinking refers only to objects and does not itself produce anything. In intellectual seeing the content must be contained within the thought-form itself. But is this not precisely the case with pure concepts and ideas? (By concept, I mean a principle according to which the disconnected elements of perception become joined into a unity. Causality, for example, is a concept. An idea is a concept with a greater content. Organism, considered quite abstractly, is an idea.) However, they must be considered in the form which they possess while still quite free of any empirical content. If, for example, the pure idea of causality is to be grasped, then one must not choose a particular instance of causality or the sum total of all causality; it is essential to take hold of the pure concept, Causality. Cause and effect must be sought in the world, but before we can discover it in the world we ourselves must first produce causality as a thought-form. If one clings to the Kantian assertion that of themselves concepts are empty, it would be impossible to use concepts to determine anything about the given world. Suppose two elements of the world-content were given: a and b. If I am to find a relation between them, I must do so with the help of a principle which has a definite content; I can only produce this principle myself in the act of cognition; I cannot derive it from the objects, for the definition of the objects is only to be obtained by means of the principle. Thus a principle by means of which we define objects belongs entirely to the conceptual sphere alone.
Before proceeding further, a possible objection must be considered. It might appear that this discussion is unconsciously introducing the representation of the “I,” of the “personal subject,” and using it without first justifying it. For example, in statements like “we produce concepts” or “we insist on this or that.” But, in fact, my explanation contains nothing which implies that such statements are more than turns of phrase. As shown earlier, the fact that the act of cognition depends upon and proceeds from an “I,” can be established only through considerations which themselves make use of cognition. Thus, to begin with, the discussion must be limited to the act of cognition alone, without considering the cognizing subject. All that has been established thus far is the fact that something “given” exists; and that somewhere in this “given” the above described postulate arises; and lastly, that this postulate corresponds to the sphere of concepts and ideas. This is not to deny that its source is the “I.” But these two initial steps in the theory of knowledge must first be defined in their pure form.