But this disciplinary training, efficacious as it is within certain limits, does not carry us beyond a restricted boundary. Logical attainment in one direction is no bar to extravagant conclusions in another. A savage expert in judging signs of the movements and location of animals that he hunts, will accept and gravely narrate the most preposterous yarns concerning the origin of their habits and structures. When there is no directly appreciable reaction of the inference upon the security and prosperity of life, there are no natural checks to the acceptance of wrong beliefs. Conclusions may be generated by a modicum of fact merely because the suggestions are vivid and interesting; a large accumulation of data may fail to suggest a proper conclusion because existing customs are averse to entertaining it. Independent of training, there is a "primitive credulity" which tends to make no distinction between what a trained mind calls fancy and that which it calls a reasonable conclusion. The face in the clouds is believed in as some sort of fact, merely because it is forcibly suggested. Natural intelligence is no barrier to the propagation of error, nor large but untrained experience to the accumulation of fixed false beliefs. Errors may support one another mutually and weave an ever larger and firmer fabric of misconception. Dreams, the positions of stars, the lines of the hand, may be regarded as valuable signs, and the fall of cards as an inevitable omen, while natural events of the most crucial significance go disregarded. Beliefs in portents of various kinds, now mere nook and cranny superstitions, were once universal. A long discipline in exact science was required for their conquest. [Sidenote: Superstition as natural a result as science] In the mere function of suggestion, there is no difference between the power of a column of mercury to portend rain, and that of the entrails of an animal or the flight of birds to foretell the fortunes of war. For all anybody can tell in advance, the spilling of salt is as likely to import bad luck as the bite of a mosquito to import malaria. Only systematic regulation of the conditions under which observations are made and severe discipline of the habits of entertaining suggestions can secure a decision that one type of belief is vicious and the other sound. The substitution of scientific for superstitious habits of inference has not been brought about by any improvement in the acuteness of the senses or in the natural workings of the function of suggestion. It is the result of regulation _of the conditions_ under which observation and inference take place. [Sidenote: General causes of bad thinking: Bacon's "idols"]