Hume had shown that we naively read causality into this world and think that we grasp necessary succession in intuition. The same is true of everything that makes the body of the everyday surrounding world into an identical thing with identical properties, relations, etc. ( and Hume had in fact worked this out in detail in the Treatise, which was unknown to Kant). Data and complexes of data come and go, but the thing, presumed to be simply experienced sensibly, is not something sensible which persists through this alteration. The sensationalist thus declares it to be a fiction.
He is substituting, we shall say, mere sense-data for perception, which after all places things (everyday things) before our eyes. In other words, he overlooks the fact that mere sensibility, related to mere data of sense, cannot account for objects of experience. Thus he overlooks the fact that these objects of experience point to a hidden mental accomplishment and to the problem of what kind of an accomplishment this can be. From the very start, after all, it must be a kind which enables the objects of prescientific experience, through logic, mathematics, mathematical natural science, to be knowable with objective validity, i.e., with a necessity which can be accepted by and is binding for everyone.
But Kant says to himself: undoubtedly things appear, but only because the sense-data, already brought together in certain ways, in concealment, through a priori forms, are made logical in the course of their alteration – without any appeal to reason as manifested in logic and mathematics, without its being brought into normative function. Now is this quasi-logical function something that is psychologically accidental? If we think of it as absent, can a mathematics, a logic of nature, ever have the possibility of knowing objects through mere sense-data?
Edmund Husserl, The Crisis of European Sciences and Transcendental Phenomenology