Einstein on Kant

John D. Norton
Department of History and Philosophy of Science
University of Pittsburgh

The main text already lays out how Einstein's discovery of the general theory of relativity proved a devastating obstacle to the Kantian notion of the synthetic a priori. This was Einstein's view of the matter.

For completeness, I repeat here Einstein's famous pronouncement in his 1921 essay "Geometry and Experience":

"... an enigma presents itself which in all ages has agitated inquiring minds. How can it be that mathematics, being after all a product of human thought which is independent of experience, is so admirably appropriate to the objects of reality? Is human reason, then, without experience, merely by taking thought, able to fathom the properties of real things?

In my opinion the answer to this question is, briefly, this: as far as the propositions of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality..."

It is a thinly veiled repudiation of Kant. The veil would have been all but completely transparent to Einstein's German language readers of the early 1920s. They were steeped in Kant almost from birth. Einstein is insisting that mathematical proposition, which include those of geometry, can be synthetic ("refer to reality") or a prior ("certain"), but not both.

Shortly afterwards, Einstein gave a fuller accounting of his negative view of the Kantian synthetic a priori. The account appeared in a 1924 review by Einstein of a text, Alfred C. Elsbach, Kant und Einstein. Einstein wrote:

"Until some time ago, it could be regarded as possible that Kant’s system of a priori concepts and norms really could withstand the test of time. This was defensible as long as the content of later science held to be confirmed*) did not violate those norms. This case occurred indisputably only with the theory of relativity. However, if one does not want to assert that relativity theory goes against reason, one cannot retain the a priori concepts and norms of Kant’s system.

[Footnote] *) To refute Kant’s system it actually suffices to indicate a logically conceivable theory (corresponding to conceivable observational material) that conflicts with Kantian norms. Whether non-Euclidean geometries accomplished this remained controversial.

For starters, this does not exclude, at least, the retention of Kant’s way of posing the problem, as Cassirer, for instance, does. I am even of the opinion that this standpoint cannot be strictly refuted by any scientific development. For, one will always be able to say that critical philosophers had hitherto erred in setting up the a priori elements and one will always be able to set up a system of a priori elements that does not conflict with a given physical system. I surely may briefly indicate why I do not find this standpoint natural. Let a physical theory consist of the parts (elements) A, B, C, D, which together form a logical whole that correctly connects the pertinent experiments (sensory experiences). Then the tendency is that less than all four elements, e.g., A, B, D, still say nothing about the experiences, without C; no more so A, B, C, without D. One is then free to regard three of these elements, e.g., A, B, C, as a priori and only D as empirically determined. What always remains unsatisfactory in this is the arbitrariness of the choice of elements to be designated as a priori, even disregarding that the theory could be replaced at some point by another theory that substitutes some of these elements (or all four of them) with others. One could be of the view, though, that through direct analysis of human reason, or thought, we would be in a position to recognize elements that would have to be present in any theory. But most researchers would probably agree that we lack a method for recognizing such elements, even if one were inclined to believe in their existence. Or should one imagine that the search for a priori elements was a kind of asymptotic process that advances along with the development of science?"

Albert Einstein, "Elsbach's Buch: Kant und Einstein," Deutsche Literaturzeitung, 1 (1924): cols. 1685–169. Doc. 321 in Collected Papers of Albert Einstein: Volume 14: The Berlin Years: Writings & Correspondence, April 1923-May 1925. Ed. Diana Kormos Buchwald et al., Princeton: Princeton University Press, 2015. Translation from English Translation Supplement. pp. 324-25.

For further discussion of Einstein's views on Kant (and more) see:

Don Howard, "'Let me briefly indicate why I do not find this standpoint natural.' Einstein, General Relativity, and the Contingent A Priori," Ch. 15 in Mary Domski and Michael Dickson, eds., Discourse on a New Method. Chicago and La Salle, IL: Open Court, 2010.