CONCEPTS AND IDEAS, therefore, comprise part of the given and at the same time lead beyond it. This makes it possible to define what other activity is concerned in attaining knowledge.
Through a postulate we have separated from the rest of the given world-picture a particular part of it; this was done because it lies in the nature of cognition to start from just this particular part. Thus we separated it out only to enable us to understand the act of cognition. In so doing, it must be clear that we have artificially torn apart the unity of the world-picture. We must realize that what we have separated out from the given has an essential connection with the world content, irrespective of our postulate. This provides the next step in the theory of knowledge: it must consist in restoring that unity which we tore apart in order to make knowledge possible. The act of restoration consists in thinking about the world as given. Our thinking consideration of the world brings about the actual union of the two parts of the world content: the part we survey as given on the horizon of our experience, and the part which has to be produced in the act of cognition before that can be given also. The act of cognition is the synthesis of these two elements. Indeed, in every single act of cognition, one part appears as something produced within that act itself, and, through the act, as added to the merely given. This part, in actual fact, is always so produced, and only appears as something given at the beginning of epistemological theory.
To permeate the world, as given, with concepts and ideas, is a thinking consideration of things. Therefore, thinking is the act which mediates knowledge. It is only when thinking arranges the world-picture by means of its own activity that knowledge can come about. Thinking itself is an activity which, in the moment of cognition, produces a content of its own. Therefore, insofar as the content that is cognized issues from thinking, it contains no problem for cognition. We have only to observe it; the very nature of what we observe is given us directly. A description of thinking is also at the same time the science of thinking. Logic, too, has always been a description of thought-forms, never a science that proves anything. Proof is only called for when the content of thought is synthesized with some other content of the world. Gideon Spicker is therefore quite right when he says in his book, Lessings Weltanschauung, (Lessing's World-View), page 5, “We can never experience, either empirically or logically, whether thinking in itself is correct.” One could add to this that with thinking, all proof ceases. For proof presupposes thinking. One may be able to prove a particular fact, but one can never prove proof as such. We can only describe what a proof is. In logic, all theory is pure empiricism; in the science of logic there is only observation. But when we want to know something other than thinking, we can do so only with the help of thinking; this means that thinking has to approach something given and transform its chaotic relationship with the world-picture into a systematic one. This means that thinking approaches the given world-content as an organizing principle. The process takes place as follows: Thinking first lifts out certain entities from the totality of the world-whole. In the given nothing is really separate; everything is a connected continuum. Then thinking relates these separate entities to each other in accordance with the thought-forms it produces, and also determines the outcome of this relationship. When thinking restores a relationship between two separate sections of the world-content, it does not do so arbitrarily. Thinking waits for what comes to light of its own accord as the result of restoring the relationship. And it is this result alone which is knowledge of that particular section of the world content. If the latter were unable to express anything about itself through that particular relationship established by thinking, then this attempt made by thinking would fail, and one would have to try again. All knowledge depends on man's establishing a correct relationship between two or more elements of reality, and comprehending the result of this.
There is no doubt that many of our attempts to grasp things by means of thinking, fail; this is apparent not only in the history of science, but also in ordinary life; it is just that in the simple cases we usually encounter, the right concept replaces the wrong one so quickly that we seldom or never become aware of the latter.
When Kant speaks of “the synthetic unity of apperception” it is evident that he had some inkling of what we have shown here to be an activity of thinking, the purpose of which is to organize the world-content systematically. But the fact that he believed that the a priori laws of pure science could be derived from the rules according to which this synthesis takes place, shows how little this inkling brought to his consciousness the essential task of thinking. He did not realize that this synthetic activity of thinking is only a preparation for discovering natural laws as such. Suppose, for example, that we detach one content, a, from the world-picture, and likewise another, b. If we are to gain knowledge of the law connecting a and b, then thinking must first relate a to b so that through this relationship the connection between them presents itself as given. Therefore, the actual content of a law of nature is derived from the given, and the task of thinking is merely to provide the opportunity for relating the elements of the world-picture so that the laws connecting them come to light. Thus there is no question of objective laws resulting from the synthetic activity of thinking alone.
We must now ask what part thinking plays in building up our scientific world-picture, in contrast to the merely given world-picture. Our discussion shows that thinking provides the thought-forms to which the laws that govern the world correspond. In the example given above, let us assume a to be the cause and b the effect. The fact that a and b are causally connected could never become knowledge if thinking were not able to form the concept of causality. Yet in order to recognize, in a given case, that a is the cause and b the effect, it is necessary for a and b to correspond to what we understand by cause and effect. And this is true of all other categories of thinking as well.
At this point it will be useful to refer briefly to Hume's description of the concept of causality. Hume said that our concepts of cause and effect are due solely to habit. We so often notice that a particular event is followed by another that accordingly we form the habit of thinking of them as causally connected, i.e. we expect the second event to occur whenever we observe the first. But this viewpoint stems from a mistaken representation of the relationship concerned in causality. Suppose that I always meet the same people every day for a number of days when I leave my house; it is true that I shall then gradually come to expect the two events to follow one another, but in this case it would never occur to me to look for a causal connection between the other persons and my own appearance at the same spot. I would look to quite different elements of the world-content in order to explain the facts involved. In fact, we never do determine a causal connection to be such from its sequence in time, but from its own content as part of the world-content which is that of cause and effect.
The activity of thinking is only a formal one in the upbuilding of our scientific world-picture, and from this it follows that no cognition can have a content which is a priori, in that it is established prior to observation (thinking divorced from the given); rather must the content be acquired wholly through observation. In this sense all our knowledge is empirical. Nor is it possible to see how this could be otherwise. Kant's judgments a priori fundamentally are not cognition, but are only postulates. In the Kantian sense, one can always only say: If a thing is to be the object of any kind of experience, then it must conform to certain laws. Laws in this sense are regulations which the subject prescribes for the objects. Yet one would expect that if we are to attain knowledge of the given then it must be derived, not from the subject, but from the object.
Thinking says nothing a priori about the given; it produces a posteriori, i.e. the thought-form, on the basis of which the conformity to law of the phenomena becomes apparent.
Seen in this light, it is obvious that one can say nothing a priori about the degree of certainty of a judgment attained through cognition. For certainty, too, can be derived only from the given. To this it could be objected that observation only shows that some connection between phenomena once occurred, but not that such a connection must occur, and in similar cases always will occur. This assumption is also wrong. When I recognize some particular connection between elements of the world-picture, this connection is provided by these elements themselves; it is not something I think into them, but is an essential part of them, and must necessarily be present whenever the elements themselves are present.
Only if it is considered that scientific effort is merely a matter of combining facts of experience according to subjective principles which are quite external to the facts themselves, — only such an outlook could believe that a and b may be connected by one law to-day and by another to-morrow (John Stuart Mill).  Someone who recognizes that the laws of nature originate in the given and therefore themselves constitute the connection between the phenomena and determine them, will not describe laws discovered by observation as merely of comparative universality. This is not to assert that a natural law which at one stage we assume to be correct must therefore be universally valid as well. When a later event disproves a law, this does not imply that the law had only a limited validity when first discovered, but rather that we failed to ascertain it with complete accuracy. A true law of nature is simply the expression of a connection within the given world-picture, and it exists as little without the facts it governs as the facts exist without the law.
We have established that the nature of the activity of cognition is to permeate the given world-picture with concepts and ideas by means of thinking. What follows from this fact? If the directly-given were a totality, complete in itself, then such an elaboration of it by means of cognition would be both impossible and unnecessary. We should then simply accept the given as it is, and would be satisfied with it in that form. The act of cognition is possible only because the given contains something hidden; this hidden does not appear as long as we consider only its immediate aspect; the hidden aspect only reveals itself through the order that thinking brings into the given. In other words, what the given appears to be before it has been elaborated by thinking, is not its full totality.
This becomes clearer when we consider more closely the factors concerned in the act of cognition. The first of these is the given. That it is given is not a feature of the given, but is only an expression for its relation to the second factor in the act of cognition. Thus what the given is as such remains quite undecided by this definition. The second factor is the conceptual content of the given; it is found by thinking, in the act of cognition, to be necessarily connected with the given. Let us now ask: 1) Where is the division between given and concept? 2) And where are they united? The answers to both of these questions are undoubtedly to be found in the preceding discussion. The division occurs solely in the act of cognition. In the given they are united. This shows that the conceptual content must necessarily be a part of the given, and also that the act of cognition consists in re-uniting the two parts of the world-picture, which to begin with are given to cognition separated from each other. Therefore, the given world-picture becomes complete only through that other, indirect kind of given which is brought to it by thinking. The immediate aspect of the world-picture reveals itself as quite incomplete to begin with.
If, in the world-content, the thought-content were united with the given from the first, no knowledge would exist, and the need to go beyond the given would never arise. If, on the other hand, we were to produce the whole content of the world in and by means of thinking alone, no knowledge would exist either. What we ourselves produce we have no need to know. Knowledge therefore rests upon the fact that the world-content is originally given to us in incomplete form; it possesses another essential aspect, apart from what is directly present. This second aspect of the world-content, which is not originally given, is revealed through thinking. Therefore the content of thinking, which appears to us to be something separate, is not a sum of empty thought-forms, but comprises determinations (categories); however, in relation to the rest of the world-content, these determinations represent the organizing principle. The world-content can be called reality only in the form it attains when the two aspects of it described above have been united through knowledge.