Page:Folk-lore - A Quarterly Review. Volume 29, 1918.djvu/110

 lOO Magic and Religion

the necessity of enquiring whether the magical is always the illicit or the illicit always the magical.

Dr. Jevons lays down that we should " reserve the term ' magic ' exclusively for the proceedings which excite the disapproval of the community." Taken literally, this definition includes under the head of magic not only all crime, but also offences against etiquette ! Even if we limit ourselves to the " magico-religious " sphere despised and rejected by Dr. Jevons, the formula gives surprising results. The breach of a ritual prohibition is a proceeding that excites disapproval ; it may set in motion the machinery of religion ; for its expiation a sacrifice may be needed. There is therefore no doubt as to the nature of the act ; it is not an ordinary crime but a violation of the moral code which entails non-natural sanctions. But it can hardly be termed magical without using the term in an altogether unusual sense. Let us look at some of the ritual prohibitions of the Ibo ! A child may not crawl out of the house by the gutter-hole ; is this to be classed as magic ? True, there is no intention in the mind of the child ; but a child may equally carry out a rain-making ritual without ulterior motive.

The proof, however, that a definition requires further limitation is evidence, at most, of a lack of experience or of clarity of vision on the part of the definer ; it does not mean that the lines on which the definition is drawn are totally wrong. But before we can give an adequate definition of a class of facts, there must be an adequate classification ; until a zoologist has within his purview all the genera, he cannot state the differentia of a family ; it is by a study of the species and genera that the wider concept is reached, not by an a priori definition, as though the idea existed, pre-formed, and the facts had to be cut to fit a Procrustean bed.

A scientific classification is not necessarily identical with the arrangement of the facts that suggests itself to the