"We have to rest satisfied with anthropomorphism once and for all," answered Einstein, "and there is no sense in wishing to escape from it, for the arguments about anthropomorphism are necessarily also diffused with it, itself. We are thus moving in a circle if we imagine we can deduce something outside of human knowledge. As soon as we have argued around the circle, we find ourselves again at the starting-point, and so we are compelled to mark clear lines of division between instinctive knowledge, derived directly by perception, from conceptual knowledge, derived by processes of abstraction and reflection; in this way we award the palm of supremacy to the human mind."

"But what if the following contradiction were to assert itself? Suppose that the logical 'circle' is not a circle at all, but a spiral, so that the final point of the argument lies just a trifle above the initial point. I feel instinctively that such apparently fruitless circuitous arguments might finally lead to a definite piece of knowledge. For example, a certain insect, the ichneumon-fly, although devoid of a knowledge of science in our sense, infallibly plants its sting in a definite point in the rings of a caterpillar, at just the point that serves its purpose of paralysing the caterpillar without killing it. It acts instinctively, and it is open to me to interpret this occurrence in other words. The fly discloses that it 'knows' the anatomy of the foreign creature, although it has no conceptual knowledge of it in our sense. But it immediately follows from this analogy that, from the point of view of the fly, its perceptual intelligence stands higher than our conceptual intelligence—that is, by changing the perspective, I am led to declare the anatomical knowledge of the fly to be of higher rank than the analogous knowledge of the most learned anatomist. In the same way I might persuade myself that the mathematics of a bird of passage stands above the cartographic knowledge of any human explorer. The migratory bird that flies from the interior of Africa in a straight fine to its nest in Mecklenburg must have something in the nature of a co-ordinate system in its organism. The real reason that we assign a higher position to our conceptual knowledge is that we are equally proud of our intelligence as of our science; this is perhaps a deception depending on some compromise, a sort of illicit deal in which the mind draws bills of exchange on science, and, as a return, science meets its obligations by paying in cheques drawn on the mind!"

I must confess that these hazardous suggestions received no welcome from Einstein, and were not even met with the friendly smile with which he usually accompanies his refutations. Nor do I disguise from myself that the question of conceptual or perceptual knowledge can in no way serve as a basis of proof; we may at most base certain conjectures on the difference of these types of knowledge, conjectures that suggest in words what eludes our clear comprehension. Einstein's refusal to allow this possibility certainly rests on much firmer ground than the somewhat Bergsonian views that I tried to present. Perhaps they are of a hair-splitting nature, and deal with things lying on different planes; and are deduced by unjustifiably altering the perspective with a sort of sophistic somersault; perhaps I may be reproached with seeking, like Münchhausen, to reach a higher standpoint without having a support from which to start. Yet how is it that I find it impossible to free myself from this chain of thought? No reason is forthcoming, for it is a purely metaphysical question, and there has never yet been a clear system of metaphysics free from ambiguities and sophism.

Let us rather confine ourselves to the conceptual intelligence characteristic of human beings, with which, according to Einstein, so many pleasures of theory are available. I asked him whether he would recognize differences of degree in these pleasures, dependent on their intensities. Although I rightly felt that he would answer in the affirmative, his answer took a totally different turn from what I had expected. It was, indeed, a great surprise, for in the matter of happiness of spirit he expressed a view, according to which he—a great discoverer!—does not regard Science as the deepest source of happiness!

"Personally," said Einstein, "I experience the greatest degree of pleasure in getting contact with works of Art. They furnish me with happy feelings of an intensity such as I cannot derive from other realms."

"This is indeed a remarkable revelation, Professor!" I exclaimed. "Not that I have ever doubted your receptivity for products of art, for I have often enough observed how you are affected by good music, and with what interest you yourself practise music. But even at such moments when you gave yourself up to the pleasures of the Muses, and were soaring in regions far removed from the earth, I used to say to myself: This is a delightful arabesque in Einstein's existence; but I should never have surmised that you regard this decorative side-issue as the greatest source of happiness. But your confession seems to go further, perhaps even beyond music?"

"At the moment I was thinking particularly of literature."

"Do you mean literature in general? Or had you a definite writer in mind, when you were speaking of the felicitous effect of works of art?"

"I meant it generally, but if you ask in whom I am most interested at present, I must answer: Dostojewski!" He repeated the name several times with increasing emphasis. And, as if to deal a mortal blow at every conceivable objection, he added: "Dostojewski gives me more than any scientist, more than Gauss!"

"If, Professor," said I, after a pause that may easily be accounted for—"if you mention in the same breath the names of two such powerful but essentially different intellects, you open the way to a discussion that cannot be settled by a mere positive assertion. It is possible to admire intensely Dostojewski as one who moulds personalities and who analyses the inner struggles of the soul, and yet to deny him perpetual fame. This depends on individual judgment, and, as for my own, I believe that Dostojewski, in spite of his direct artistic appeal, will not have his name perpetuated through the centuries like that of many another member of Parnassus. It seems to me to be a more important matter whether a common measure can be found for Art and Discovery at all. Perhaps the test of how far a work can be replaced may be regarded as valid for each. When you say that Dostojewski gives you more than Gauss, this probably corresponds with the feeling that without Dostojewski you would have no 'Karamasoffs' and hence would lack a certain life-value that cannot be replaced. But if Gauss had failed to produce one of his fundamental theorems of Algebra, probably some other Gauss would have appeared, who would have achieved this result. According to this, then, our instinct increases the value of a work of art, as we feel that we are dependent on one being alone for its creation."