ABSTRACT
Law and Generalization. That the field of politicai history and observation is capable of being gathered up into a science is a bold claim which is too often advanced frivolously, and as frequently dismissed for superficial reasons. Such phrases as a "political law" or "a fundamental political principle" are common enough, but it is usual to explain that these high-sounding words imply no more than that their user believes that he has been able to make certain empirical generalizations, of which the spectacular form often compensates for the modesty of their attainments in precision. They are, on the whole, true of the past; it is hinted prophetically that they may be true of the future. But they have scarcely been taken au pied de la lettre, or, indeed, very seriously. A law, however, is more than a generalization. It is a generalization so well according with the rest of our grounded and systematic knowledge, e.g. the deductions which may be made from mathematical axioms or from the law of gravitation, that we prefer to seek for a secondary law, when confronted with recalcitrant phenomena, rather than to abandon our original formula. Every law is a generalization, but a generalization is other than a law in this respect that the generalization is merely a statement of an observed uniformity, without any necessary conviction of its future validity. That the sun moves from the east to the west is a generalization which does not preclude the sun from standing still upon Gibeon. It happens to be a highly misleading generalization. But the principle "that the force of attraction of a mass varies inversely with the square of its distance," the formula of the rate of acceleration of falling bodies in units of time, and the principle of gravitation are laws, being founded upon observation confirmed by deductions from well-established facts, and systematically consonant with the rest of our experience. The law has, thanks to its coherence with the rest of our knowledge, a quality of certainty which makes it also govern the future, because it implies not only a generalization in terms of a specific group of facts observed in the past, but also an explanation in terms of other certainties. On the other hand, the 5I : 49 ratio of male to female births is only a generalization founded on well-known facts, at least of modern European civilization—a generalization for which we have .at present no adequate biological explanation.
