which it operates: on the contrary, it is only their applicability to the concrete facts originally abstracted from that is held to justify their use and to establish their ‘truth.’ The mathematical abstractions which enable astronomers to calculate the path of a star are justified by their approximate correspondence with its observed position, and if there were any extensive or persistent divergence between the calculation and experience, astronomers would be quite ready to revise their assumptions to the extent even of changing their fundamental notions concerning the nature of space. But in the case of metaphysics the same principle is not, apparently, to apply. If the Dialectic of the Absolute Idea does not accord in its results with the facts of life, we are not to suspect the Dialectic. It possesses an intrinsic certainty by right divine which no failure can be admitted to impair. If the logical (or rather psychological) development of the Idea fails to account for the development in time, we may at the utmost postulate an “unknown synthesis.” This may be philosophy, but it does not look like science.
In the second place, let us ask why science abstracts from the particularity of reality. Not, certainly, because it does not ob.serve it. Nor yet because it ascribes to the deductions from its universal laws a precision which they do not possess. On the contrary, it cheerfully admits that all the laws of nature are hypotheses, represent not the facts but tendencies, are to be used merely as formulas for calculating the facts. But why should we want to calculate the facts by such universal formulas? The answer to this question brings us to the roots of the matter. We make the fundamental assumption of science, that there are universal and eternal laws, i.e. that the individuality of things together with their spatial and temporal context may be neglected, not because we are convinced of its theoretic validity, but because we are constrained by its practical convenience. We want to be able to make predictions about the future behaviour of things for the purpose of shaping our own conduct accordingly. Hence attempts to forecast the future have been the source of half the superstitions of mankind. But no method of divination ever invented could compete in ingenuity and gorgeous simplicity with the assumption of universal laws which hold good without reference to time; and so in the long run it alone could meet the want or practical necessity in question.
In other words that assumption is a methodological device, and ultimately reposes on the practical necessity of discovering formulas for calculating events in the rough, without awaiting or observing their occurrence. To assert this methodological character of eternal truths is not, of course, to deny their validity — for it is evident that unless the nature of the world had lent
