THE PRIMITIVE FAMILY.

INTRODUCTION.

The comparative method-The materials for inquiry-The hypotheses of evolution and degradation-Question of uniformity of primitive state -Anthropology and philology-Plan of inquiry-Border-line between institutions originally fluctuating-Groups of kinship-The primitive tie of blood.

THE researches which we propose to make in the present work lead us into paths which have of late years been trodden by many men of more or less importance. In earlier times an attempt was often made to construct the process of man's physical development in accordance with the conceptions of abstract speculation. The worthlessness of this process is now established, and it is only through the analytical method of comparative history that we can hope to understand the historical facts of this development. The importance of the comparative method is well known to philologists. In his valuable work on "Comparative Politics," Freeman asserts that the discovery of this method is an event of sufficient importance to distinguish our century as one of the great eras in the history of mankind. Even if this estimate is exaggerated, it must be admitted that the comparative method has diffused light over many things which previously lay in the deepest darkness. The relationship between different races has been ascertained by the

B

method in question, and it has taught us how to form a conception of the stage of culture attained by the primitive race.

The fitness of this method is explained by the special character of the subject. The sound-symbols which designate given objects were for the most part quite arbitrarily chosen, and it is therefore improbable that two distinct peoples should ever have chosen the same symbol for the same object. Whenever, therefore, we find such agreement, we may almost certainly infer, either that one people has borrowed the word from the other, or that both peoples are derived from one and the same primitive stock. A careful analysis will establish the store of words which were used by this primitive stock. By means of these words we learn what ideas were common to the race, or, in other words, what was their state of civilization.

When we have to do with matters which do not involve such an arbitrary choice, the use of this method cannot afford equally good results. Similar legends and myths, similar social institutions, usages, and customs cannot be taken as a proof of common origin, since these resemblances may result from causes which occur quite independently in different places and at different periods. The comparative method here serves another purpose, by making it possible to discover the definite causes from which given myths and institutions necessarily result. In this case the comparative method only serves for the application of the ordinary methods of experimental inquiry to a field which is withdrawn from direct experimental control. When, for example, races which we have reason to believe are uninfluenced by each other, possess any institution in common, it may be surmised that the cause must be sought in the relative conditions which are common to all races. As soon as this preliminary assumption has been made, we must inquire whether there is any race in which the institu

tion is found without these relative conditions, or if the conditions occur without the institution. And, finally, we must inquire whether, during the development of a race, the institution and the relative conditions can vary, independently of each other. The results which may be obtained in this way will be almost of equal value with those of the exact sciences. The conditions of social development are generally so complex that the possibility of error must be admitted, and it is well to remember this. Yet I believe that too great allowance has been made for such error, since, although we have to do with forces which cannot be included within the range of experience already mentioned, yet it is most probable that these forces are so insignificant that they need not be taken into account.

We have already said that many scientific men have made use of the comparative method in the study of the primitive forms of the social life of man, and of its modes of development, as well as of its limits and stages. Yet we find a wide difference of opinion among them, both with respect to the more important points of view, and to smaller questions of detail. In such cases it may be supposed that facts which are interpreted by competent inquirers in quite opposite ways are absolutely useless for the purposes of science. This would, however, be an error, since it is the common fate of every dawning science to advance gradually from daring hypothesis to truth. I propose in this work to attempt to make a slight step in advance, and if errors should occur, the fault is not to be found in the material, but in the use I make of it.

The material may be divided into several sections: (1) We have direct historical accounts of the development of a single community during a long period; (2) the narratives of travellers, who have described an isolated phase of the life, as it existed at the time, of a race either now extinct or still living; (3) ancient laws

and customs, recorded in writing; (4) ancient myths and legends, and archæological remains, by means of which we may reconstruct periods of which we have no direct observation. It is also absolutely necessary to be acquainted with psychology, in order not to be misled by deceptive appearance. It must be taken for granted that we are not, in this case, bound to prove the trustworthiness of our material. Such proof, which is very difficult, must be assumed to have been given elsewhere. We must now subjoin the following fundamental considerations.

In order to become acquainted with the development of a given community, we can desire no better material than the reports of its condition at different times, as they have come down to us. In fact, however, this mode of establishing laws, or general rules of development, is only possible to a limited extent. For the most part we are without the means of constructing such abstractions as are necessary to establish laws, since the material is far from being sufficient to entitle us to draw conclusions with respect to the non-existence of negative instances. There are comparatively few communities of which we possess sufficient records, and these moreover only include a limited portion of the life of the community in question; we must remain absolutely ignorant of the long pre-historic period. We may indeed admit the possibility of discovering an apparent solidarity in the social development. For we may become acquainted with the primitive condition and earliest development of a community by means of the report of another community of higher civilization, which goes back to the period in which the former had its origin. Yet this solidarity is only apparent. No primitive community can be carefully observed for any length of time by a civilized community without being placed in such relations to the latter as to modify it in many ways, so as to affect its actual development in a way which will not