KELIGIONS 359 to what must be the final result of this historical as well as comparative study, a morphological classification of religions. Here the study of religions reaches its goal, and the task of the philosophy of religion, the other main branch of the so-called science of religion or general theology, begins. It need scarcely be said that the basis of the compara- tive historical study of religions must be a patient and critical examination of the sources from which the know- ledge of the various religions of the world is to be drawn, viz., written documents and traditions, monuments and works of art, sacred writings and heretical books, and, when we wish to inquire into the religions of the uncivi- lized tribes that have no history at all, an impartial weighing of the evidence brought by travellers and settlers from different parts of the globe, in short, an unbiassed ascertaining of facts. Genealogical Classification. There is no difficulty in determining the descent and relationship of religions which have taken rise in historical times, such as Con- fucianism, Buddhism, Judaism, Christianity, Mohamme- danism, and some others of minor importance. But the great majority of ancient religions had their origin in pre- historic times, of which neither documents nor trustworthy traditions are extant. In that case their mutual relation has to be established by reasoning from the myths, ideas, rites, and characteristics common to them. Professor Max M tiller (Lectures on the Science of Religion, pp. 154 sq.) suggests that, whatever classification has been found most useful in the science of language ought to prove equally useful in the science of religion. Now it may be true in general, at least for the most ancient times, that where the languages of a group of nations are proved to belong to one family their religions too most probably "hold together by the same relationship." But this hypothesis requires proof, and that proof is not to be obtained otherwise than by the comparative study of the religions themselves. Only when the religions of two independent nations agree in doctrine and mode of wor- ship, above all in the notion of the relation between God and man, between the divine and the human, to such a degree and in such a manner that this agreement cannot be accounted for by the universal aspirations and wants of human nature, then only may we feel sure that the one of these religions is the parent of the other, or that both have come from a common stock. If not only two but several religions agree in like manner, or nearly so, we get a family of religions. At present we can go no farther. The mutual relations of the different families cannot be determined yet ; the problem is too difficult and too com- plicated to be solved in the present state of science. That religions belonging to different families have borrowed myths and customs from one another and have been sub- jected to one another's influence may easily be proved. But whether the families themselves are branches of one and the same old tree is an open question to which a satisfactory answer cannot be given now. It would be equally premature to venture on drawing up a complete genealogical table of religions. For some families of religions such a classification may be sketched with tolerable certainty; the genealogy of by far the greater number of them can be given in mere outlines only, leaving the fixing of details for further inquiry. We start from what may be held the most certain. Aryan or Indo- Germanic Family. Comparative mytho- logy and the history of religion leave no doubt that all the religions of the Aryan or Indo-Germanic nations, viz., Eastern Aryans (or Indians, Persians, and Phrygians) and Western Aryans (or Greeks, Romans, Germans, Norsemen, Letto-Slavs, and Celts), are the common offspring of one primitive OLD-ARYAN^- religion. That the same name of the highest heaven-god, Dyaus, Zeus, Ju(piter), Zio (Ty), is met with among Indians, Greeks, Italiotes, Germans, and Norsemen, however great the difference of the attri- butes and dignity ascribed by each of them to the god thus named may be, is a fact now generally known. Where this name has been lost, as is the case with the Persians, the Slavs, and the Celts, there are other divine names which they have in common with their kindred nations. Still more important is the fact that most Aryans show a tendency to call their supreme god " father," as is proved by the very common forms Dyaus pitar, Zeus -n-ar^p, Jupiter, Diespiter, Marspiter, Alfocfr. The supreme god in the Avesta, Ahuramazda, is often called father. Moreover many divine names used by different Aryan nations, though varying in form, are derived from the same root, which proves the original unity of their conception. Takp as examples the root di (div), "to shine," and its derivatives Dyaus, Deva, and their family, Diti, Aditi, Dione, Pandion, Dionysos, Diovis, Dianus (Janus), Diana, Juno ; or the root man, " to think " (perhaps equally signifying originally " to shine "), and its derivatives Manu, Minos, Minerva, (Juno) Moneta ; or the roots sur (svar), sar, mar, vas. Especially startling is the use of the same general word for " god " among several Aryan nations, viz., Skr. deva, Iran, daeva, Lat. deus, Litth. dewas (deiwys), Old Norse tivar (plur.), to which belong perhaps also Greek 0eos, Irish dia, Cymr. dew. Daeva and deiwys are used in a bad sense, but this cannot be original. So too the word asura (ahura), which, though it too was used by the Indians in relatively modern times in a bad sense, was the name which the East-Aryans gave to their highest gods, and the Norse asa, pi. sesir (orig. ans), are both to be derived from the root as, anh. If we add to this the remarkable conformity of the myths and customs in all Aryan religions, if, above all, by com- paring them with those of other races, especially of the Semites, we find that the leading idea embodied in these Aryan myths and rites is everywhere the same, however different the peculiar character of each religion may be, namely, the close relation between God and man, the real unity between the divine and the human economy, 2 so that we may call them the " theanthropic " religions. if we remember this, there can be no doubt that all of them have sprung from one primitive OLD-ARYAN religion. However, the degree in which the Aryan religions are mutually related is not always the same. None of them came directly from the OLD-ARYAN religion. They consist of five pairs, each of which must have been first a unity: the Indo-Persian, the Graeco-Roman, the Letto-Slavic, the Norse-Teutonic, and the Gaelo-Cymric. The fact that the members of those pairs are more closely allied with one another than with the other members of the family obliges us to assume five prehistoric Aryan religions : the OLD EAST-ARYAN, the OLD PELASGIC? the OLD WINDIC, the OLD GERMAN, and the OLD CELTIC religions, forming so many links between those historical religions and the common parent of all, the primeval ARYAN worship. Space forbids us to give the complete proof of this con- clusion. We only mention that the Indian and Iranian religions havemanygods incommon, unknown to the Western Aryans, and therefore probably such as arose after the eastern and western branch of the family had separated, e.g., Mitra Mithra, Aryaman Airyaman, Bhaga Bagha Baga (comp. also Aramati Armaiti, Sarasvati Hara- 1 This special type indicates prehistoric religions. 3 This is why they call the Godhead "father," or even "brother, friend, companion." Compare the names Mitra Mithra, "friend," Aryaman Airyaman, "companion," &c. 3 The name is not exact. It is only chosen as the most convenient.
