Magic is a way of controlling things. Imitate the act desired, in a certain way, and it will come to pass. Talismans and amulets, again, possess a secret power for good and evil. Ancient societies built up a lore of this kind, adding to material objects the agency of demons under the control of magicians. This lore is practically a feature of the past. Even white magic is no longer good form, no longer accredited by the dominant social mind. It slinks into out-of-the-way places beyond the public eye. Yet research is showing that these seemingly discredited beliefs and points of view seldom completely disappear. They smolder beneath the surface and flame up now and then in a startling way to remind us that society in its evolution does not carry all its members along at the same rate. The historian is surprised to find that rites which are given an exalted place in various religions are magical at heart, and go back to beliefs which have long been discredited in other settings.
Some who have specialized in folklore and anthropology are very pessimistic as to the degree in which the scientific outlook upon nature is replacing the more primitive attitude associated with magic. One of the greatest authorities upon primitive beliefs and customs writes as follows: "We seem to move on a thin crust which may at any moment be rent by the subterranean forces slumbering below.... Now and then the polite world is startled by a paragraph in a newspaper which tells how in Scotland an image has been found stuck full of pins for the purpose of killing an obnoxious laird or minister, how a woman has been slowly roasted { 47} to death as a witch in Ireland, or how a girl has been murdered and chopped up in Russia to make those candles of human tallow by whose light thieves hope to pursue their midnight trade unseen." The danger to civilization foreseen by the specialist in uncouth customs is undoubtedly exaggerated, but his warning should remind us that education has a very valuable function to perform in training an ever increasing number in scientific habits of thought.
One of the assumptions which underlie magic is the idea that two things are connected in nature because they are like one another. Space is not looked upon as a barrier to this connection. So far as can be seen, anything can affect anything else; and the slightest suggestion of such a relation leads to the belief in its reality. There is almost entire absence of any conception of systematic testing: any accidental association may lead the savage to be assured of an important sign. Thus, if a man went out hunting and saw a rabbit cross his path, and then had bad luck, he would be sure that a rabbit is a sign of bad luck. Moreover, since individuals were on the lookout for hoodoos, they would not tempt providence a second time. This example illustrates the psychology rather than the sociology of the process. It must be remembered that social groups developed what we call superstitions by way of social contagion and suggestion. The laughing acquiescence of the present in hoodoos, mascots and lucky objects cannot be traced back to the credulity of any one individual. Such things come to pass by a process of accretion just as does the belief that a particular house is haunted.