parative study in the strict sense of the term may. twhnic-aily, be static. The value of such an inv"c.sli;;ati(m is. however, so much inferior scientilically Id cvohitiuiiury comparison that it may practically be if;nored. By far the prcater nuuiher of investigations make their studies of comparative religion historical or evolutionary. Such discussions alone lead to the final ohjcct of the science. This object is to invcstisate the nature and development of religious beliefs and to disiovcr if (lossible the origin of religion itself. Comparative religion therefore ranks as one of the historical sciences.
Dki'I.nitio.n of Religion. Nowhere is definition more dilHcult than in the sphere of religion. The manifestations are so varied, and the causes, where they can be traced, arc so complex, that almost no definition of religion can be given which is altogether free from objection. The definition which seems on the whole least open to adverse criticism is as follows: Religion is the view held by man of an intelligent being or beings which is or are. or which he conceives to be, superhuman, and of the relation, modifiable by his own agency in certain respects and by certain means, which he sustains toward the being or beings in question. This definition recognizes the two-fold aspect of religion, which accoimts for part of the complexity of religious phenomena, the theoretical and the practical. The theoretical side of religion is the view held by man regarding the nature and character of the superhuman being or beings in which he believes. The ])ractieal side is the power which such being or beings exert over him and the power which he has or may acquire over them.
Univers.vi.ity of Religion. Tlie statement which is still made, though with decreasing frequency, that there are races or tribes without religion is almost certainly false. It may be true that many peoples seem to possess no religion, either because these religions are so obscure as to j)ass unnoticed, or because they are contemptuously rejected as false. Yet even the scientific investigator may be misled in this matter. In the first place, the person questioned as to his religion may not. and in the case of savages often does not. understand the questions which are asked him. His answers, therefore. are misleading. It must also be remembered that one questioned concerning his religion will sometimes willfully give a false answer, either because he regards the inquiry as foolish, impertinent, or tiresome, or because he does not dare to reveal his religion. In giving a knowledge of his religion, one gives a part of himself, and the power thus gained by another may be used to his own detriment. If these ol)staeles to an accurate knowledge of a religion exist in the case of living faiths, they hold to a far greater degree in the study of extinct religions or extinct phases of them. The ancient writers who describe foreign religions were either little acquainted with them and unscientifically trained, as were even Herodotus and Plutarch, or were contemptuous and unsympathetic, as was Tacitus. Moreover, the sacred books of religions give tis only a partial view of their own faiths. They represent only the officially sanctioned religion, while the popular deviations from these hook-religions are either ignored altogether, or mentioned with disparagement and hostility, or must be partially reconstructed from chance allusions scattered through the canonical works. There is also a marked tendency both in ancient and in modern investigators "of religion to interpret foreign religions in terms of their own, thus leading to false identifications and to attributions of concepts to religions whieh they may not contain. The stiulent of religion must also guard against mistaking an idea previously imported into a religion from a foreign source for an integral ])art of the faith which he studies. C'L..ssiric.vnox of Religions. Still more difficult than definition of religions is classification. While one may make rough distinctions between various religions, hard and fast lines of demarcation can seldom be drawn, for religions often overlap one another in attributes common to both. Nor is it easy to classify a religion according to its salient feature. To say that Greek religion is the cult of the beautiful, or that Roman religion is legalistic, or that American religions are animistic, does not adequately describe them, for they represent the culminations of long evolutions, in which many factors from many sources have cooperated. Animism, totemism. ghost-worship, nature-worship, and other factors, all combine in difTerent proportions. Several classifications, however, have been proposed. Of them all. the most simjjle and the most worthless divides religions into true and false, the first class holding the single religion adopted by him who makes the classification, the latter class containing all the rest. There are a number of other classifications which are equally unsatisfactory. Among these may be enumerated the division of religions on a linguistic basis into Indo-Germanic. Semitic, and Turanian. This classification, proposed by Jlax Mtiller. proceeds on the false premise that religion and linguistic affinity are one. A religion may indeed originate in a certain race, but it may spread with equal facility to other races, and even to other climates and entirely different systems of civilization. The division into national and universal or international religions, advocated by Kuenen. is also objectionable, since there is no real line of demarcation between the two. Comparatively ]irimitive religions may be international, on account both of their catholicity and of the absence of any national idea, while very advanced religions, as Zoroastrianism, may be strictly national, although apparently intended to be international in scope. Hegel attempted a classification, in four divisions. The most primitive faiths, or nature religions, he regarded as spontaneous. To them he opposed the religions of spiritual individuality, which was differentiated from nature religions by the presence of reason and meditation. Within the religions of spiritual individualities there are three divisions: First, the religion of majesty, where the divine overwhelms the human; second, the religion of beauty, where the divine blends with the human: and third, the religion of design, where a divine ptupose in the universe is recognized. The increased knowledge of the evolution of religion since Hegel's time has rendered this view of his useless. A marked advance was the classification of Hartniann. who made a broad division into naturalism and supranaturalism. Naturalism is characterized by a belief that deities rule in the world and require material representations, while in supra-
