Schopenhauer on Perception and Reason

In World as Will and Representation, #12:

Thus, for example, an experienced billiard-player can have perfect knowledge of the laws of impact of elastic bodies on one another, merely in the understanding, merely for immediate perception, and with this he manages perfectly. Only the man who is versed in the science of mechanics, on the other hand, has a real rational knowledge of those laws, that is to say, a knowledge of them in the abstract. Even for the construction of machines such a merely intuitive knowledge of the understanding is sufficient, when the inventor of the machine himself executes the work, as is often seen in the case of talented workmen without any scientific knowledge. On the other hand, as soon as several men and their coordinated activity occurring at different times are necessary for carrying out a mechanical operation, for completing a machine or a building, then the man controlling it must have drafted the plain in the abstract, and such a cooperative activity is possible only through the assistance of the faculty of reason.

But it is remarkable that, in the first kind of activity, where one man alone is supposed to execute something in an uninterrupted course of action, rational knowledge, the application of reason, reflection, may often be even a hindrance to him. For example, in the case of billiards-playing, fencing, tuning an instrument, or singing, knowledge of perception must directly guide activity; passage through reflection makes it uncertain, since it divides the attention, and confuses the executant. Therefore, savages and uneducated persons, not very accustomed to thinking, perform many bodily exercises, fight with animals, shoot with bows and arrows and the like, with a certainty and rapidity never reached by the reflecting European, just because his deliberation makes him hesitate and hang back. For instance, he tries to find the right spot or the right point of time from the mean between two false extremes, while the natural man hits it directly without reflecting on the wrong courses open to him. Likewise, it is of no use for me to be able to state in the abstract in degrees and minutes the angle at which I have to apply my razor, if I do not know it intuitively, in other words, if I do not know how to hold the razor. In like manner, the application of reason is also disturbing to the person who tries to understand physiognomy; this too must occur directly through the understanding. We say that the expression, the meaning of the features, can only be felt, that is to say, it cannot enter into abstract concepts. Every person has his own immediate intuitive method of physiognomy and pathogony, yet one recognises that signatura rerum more clearly than does another. But a science of physiognomy in the abstract cannot be brought into existence to be taught and learned, because in this field the shades of difference are so fine that the concept cannot reach them. Hence abstract rational knowledge is related to them as a mosaic is to a picture by a van der Werft or a Denner. However fine the mosaic may be, the edges of the stones always remain, so that no continuous transition from one tint to another is possible. In the same way, concepts, with their rigidity and sharp delineation, however finely they may be split by closer definition, are always incapable of reaching the fine modifications of perception, and this is the very point of the example I have taken here from physiognomy.

This same property in concepts which makes them similar to the stones of a mosaic, and by virtue of which perception always remains their asymptote, is also the reason why nothing good is achieved through them in art. If the singer or virtuoso wishes to guide his recital by reflection, he remains lifeless. The same is true of the composer, the painter and the poet. For art the concept always remains unproductive; in art it can guide only technique; its province is science.