The fictitious or conventional character of mathematics is still more apparent than that of the natural sciences; and we shall not add anything to what we have said in the preceding chapter about the abstract concept, the non-concrete universal, which is the distinctive process of mathematical thought. The application of the mathematical processes, through the empirical concepts, to the historical datum, gives origin to what we have called the judgment of numeration (and mensuration), and to the mathematical sciences of nature. All that has been observed of the natural sciences in general is valid for these also. Their truth is still only the truth of the intuitive, historical datum of which the empirical concepts are practical elaborations; the addition of a further practical elaboration, the abstract concept, can add to their mnemonic or, as it is more often called, technical efficiency, but not to the value of their original content. This process, as the purely naturalistic one, can be applied to the human as well as to the natural reality, but it is evident that its usefulness decreases in the passage from the one to the other, following the same standards that apply to the natural sciences in general, those of the relative perceptibility and importance of the individual happening. It is at its highest in physics or astronomy, less notable in biology or economics; practically inexistent in psychology or sociology, the two sciences that suffer not less from the delusions of misapplied statistics than from the invasions of cheap philosophy. Croce's theory of science, as we have already remarked, differs from the generally accepted methodology of modern science only in its context, which is usually agnostic in, the pure scientist, while, in Croce, it consists in the affirmation of the pure concept, or of the autonomy of philosophy: a proposition with which the scientist qua scientist has no reason to quarrel. In both cases, the autonomy of scientific thought is only relative, and the difference of context is a difference in the determination of its limits. In both cases, scientific thought is recognised as thoroughly legitimate only within limits. The cry of the bankruptcy of science, of which we heard so much a few years ago, is as meaningless for Croce as for the pure scientist; science cannot become bankrupt except by over-stepping its logical limits, that is, by first ceasing to be science and becoming the ape of philosophy. [Footnote 1: See _Logica_, part ii, La filosofia, la storia, etc., pp. 171-269.] VII. THE THEORY OF ERROR[1] The practical origin of theoretical error--Confirmations of this doctrine--The forms of error--Æstheticism and empiricism; mathematicism--Philosophism: the philosophy of history and the philosophy of nature--Mythologism: philosophy and religion--Dualism, scepticism, mysticism--The conversion to truth--The function of error.