Apart from the development of scientific method, inferences depend upon habits that have been built up under the influence of a number of particular experiences not themselves arranged for logical purposes. A says, "It will probably rain to-morrow." B asks, "Why do you think so?" and A replies, "Because the sky was lowering at sunset." When B asks, "What has that to do with it?" A responds, "I do not know, but it generally does rain after such a sunset." He does not perceive any _connection_ between the appearance of the sky and coming rain; he is not aware of any continuity in the facts themselves--any law or principle, as we usually say. He simply, from frequently recurring conjunctions of the events, has associated them so that when he sees one he thinks of the other. One _suggests_ the other, or is _associated_ with it. A man may believe it will rain to-morrow because he has consulted the barometer; but if he has no conception how the height of the mercury column (or the position of an index moved by its rise and fall) is connected with variations of atmospheric pressure, and how these in turn are connected with the amount of moisture in the air, his belief in the likelihood of rain is purely empirical. When men lived in the open and got their living by hunting, fishing, or pasturing flocks, the detection of the signs and indications of weather changes was a matter of great importance. A body of proverbs and maxims, forming an extensive section of traditionary folklore, was developed. But as long as there was no understanding _why_ or _how_ certain events were signs, as long as foresight and weather shrewdness rested simply upon repeated conjunction among facts, beliefs about the weather were thoroughly empirical. [Sidenote: It is fairly adequate in some matters,] In similar fashion learned men in the Orient learned to predict, with considerable accuracy, the recurrent positions of the planets, the sun and the moon, and to foretell the time of eclipses, without understanding in any degree the laws of the movements of heavenly bodies--that is, without having a notion of the continuities existing among the facts themselves. They had learned from repeated observations that things happened in about such and such a fashion. Till a comparatively recent time, the truths of medicine were mainly in the same condition. Experience had shown that "upon the whole," "as a rule," "generally or usually speaking," certain results followed certain remedies, when symptoms were given. Our beliefs about human nature in individuals (psychology) and in masses (sociology) are still very largely of a purely empirical sort. Even the science of geometry, now frequently reckoned a typical rational science, began, among the Egyptians, as an accumulation of recorded observations about methods of approximate mensuration of land surfaces; and only gradually assumed, among the Greeks, scientific form. The _disadvantages_ of purely empirical thinking are obvious.