Referring to the centripetal and centrifugal movements in ethnoses discussed in Chapter I, we may now state that these movements may have a direct influence on the language. Yet, since the language itself is one of varying elements, it may also have its bearing upon the intensity of movements. O. Jespersen, as linguist, has pointed out certain conditions which are essential to the change of languages. Some of his observations I shall now quote.
He points out that the splitting of languages amongst «primitive tribes» is greater than it is in «civilized countries,» which is quite true in some cases, but the difference is not due to the «civilization» or the «primitiveness.» As shown, the size of the units, whence that of the linguistical units, is a function of the cohesion between the populations occupying a certain territory and the methods of adaptation. The instances of very «primitive» groups whose language is not «split» do exist; e.g., the Polynesian dialects are distributed over an enormous territory. Then there are, besides, a great diversity of languages; for example, as reported by travellers in reference to New Guinea, where nearly every few miles one finds a new language. For such diverse conditions various reasons may exist. One of them is the geographical position; another one may be the interrelations between the ethnoses, etc., without taking into account the historic formation of the ethnoses and other conditions which have nothing to do with the cultural state «civilized» and «savage,» of common classification. A great «splitting» of languages is found in the most «civilized» western Europe, where ethnoses, although numerous, occupy very small territories.
With good reason and a near approach to the conception of ethnos, O. Jespersen points to «human geography, which is a decisive factor in the formation of dialects» («Mankind,» op. cit., p. 42). In «human geography» he includes, for example, the differentiation of local groups due to the former administrative church divisions (in France) [39]. In the same group of phenomena, one must include all distinct elements resulting from the centrifugal movement, a study of special importance which, being underestimated, is left without further discussion. He points out (id., p. 43) the existence of «two opposing tendencies, the one in the direction of splitting, the other in the direction of larger and larger units.» The discussion regarding the question of which tendency is stronger and which is weaker is rather interesting from the ethnographical point of view, for it reflects the behaviour of the authors, some of whom do see the centripetal movement and some of whom do not see it, for some of them approve of it, while some of them disapprove of it. O. Jespersen finds that what I call here the centripetal movement as to the language is gaining a larger population (the formation of larger units) («Mankind,» op. cit., p. 45) [40]. This is, of course, a great overestimation of the geographical factor where O. Jespersen uses such an expression as «the local dialects purely conditioned by geographical factors» (id., pp. 45,46). As shown and pointed out by O. Jespersen himself, it is also characteristic of groups in which the geographical condition cannot be taken to be responsible [41]. As to the factors of the centripetal movement, to which O. Jespersen pays special attention on expense of factors of the centrifugal movement, he points out (1) the war, which produces a mingling of population; (2) the annual market; (3) intermarriage [42]; (4) religion; (5) literature and the theatre, (6) political divisions; and (7) the formation of great towns. It is difficult to say why these factors have been selected from hundreds of other manifestations of the centripetal movement. Since there are mixed-up factors such as «war» (mobilization), which has but very small importance in particular cases of not yet completely formed ethnos, and «political division,» which alone may be responsible for both «splitting» and the formation of a «larger unit,» one may see that the importance of these two movements escapes his attention. However, under O. Jespersen's penetrating analysis, the importance of the centripetal movement comes out quite clearly. Indeed, this has been done from the linguistical point of view. If one turns the problem around, i.e., what influence has language on the above-quoted seven factors, one may also see that without a previous ethnical total or partial fusion, they may have no place. In other words, language is an important factor in the process of ethnical differentiation (the formation of smaller ethnoses) and integration (the formation of larger ethnoses), but it cannot be taken alone to be responsible, and it cannot be isolated from the cultural complex and even supplied with magic power over all other manifestations of eihnical adaptation. Thousands of investigations may be written with the consideration of various influences over certain isolated phenomenon, the variations of which actually are conditioned by that of the nature of the ethnoses and particularly by the equilibrium of the centripetal and centrifugal movements. When the mechanism of these changes is not clear, the explanation of variations of certain isolated phenomena cannot be successful. Hence we have psychological, sociological, geographical, economical, mathematical, historical, and other approaches [43] to the minor problems, while the attention must be directed to the mechanism of changes. The description of all cases of gravitation is naturally impossible, just as the description of all cases of cultural variations is also impossible, for we do not know what existed before, except a very short period of reliable records, and what new cases of variations will exist later.
Referring once more to the centrifugal movement, it may be pointed out that the language in this respect is an extremely sensitive phenomenon. In fact, the formation of new sounds is an individual invention, the formation of new complexes oi starters is a very common phenomenon in particularized groups, as, for instance, specialized social groups. When the centripetal movement is not strong enough to oppose particularization, then the new dialect comes into existence. Since a new dialect is in the process of formation, it accumulates round itself new elements of the psycho-mental complex and the whole group of individuals may form a particularized group which limits its communications with other groups. In this way dialects and, further, other new ethnoses may be formed. Since this process of specialization of smaller units is one of vital importance for the existence of larger units, and since this process sometimes varies fast, the classification of dialects and ethnoses presents difficulties which cannot be overcome. As a matter of fact, the limits between the existing dialects and ethnoses sometimes cannot be detected at all — the old limits disappear, the new limits appear. This is a continuous process and its representation in a static form may bring the investigator to commit further mistakes. For instance, the limits between the dialects may happen to be in the process of disappearance, while other new limits are in the process of appearance. In the static treatment both will appear of equal value, while they are not so actually. It is naturally true, not only of dialects, but of all existing phenomena in ethnoses, cultural and biological in a narrow sense of the word.
39. A. Meillet («La Methode,» op. cit., pp. 55, 56), in reference to the ecclesiastic administrative divisions, points out that they were a continuation of the Roman administrative units, which in their turn had not been incidental to, but based upon, the consideration of existing relations amongst the people. Yet, in general, he does not give great importance to the political divisions as a factor influencing languages, but he sees deeper reasons for the existence of political divisions. No doubt, he is absolutely right in principle; however, there are cases when absolutely arbitrary divisions (like that between some states in North America) at last may produce their political effects upon the population in the sense of directing and enforcing centrifugal movement. Indeed, a political division as such is not an important factor, but still it is one of the elements having some weight in the system of centrifugal movement. Cf., also, J. Vendryes, op. cit., p. 307, on influence of political division.
40. This conclusion is one of those conclusions which are laid down as the theory of a future international and unique language for all mankind — an old and well-known theory, inspired by the desire of growing ethnoses who want to assimilate all «mankind,» at least «civilized.» But it is remarkable that this behaviour is also characteristic of ethnical units which are in the process of disintegration and whose leaders realize the impossibility of further keeping the old modus vivendi. Yet the same idea originates amongst the ethnical units which by this means may defend themselves against the assimilative movement. In fact, the adoption of an alien language does not yet mean a complete loss of ethnical independence, especially in the case where the differentiation of units operates in non-linguistical elements, e.g., «religious,» «physical,» «social,» etc. So the same idea may be conditioned by various causes. A. Meillet, in reference to the European countries (pays), says: «II y a la un etat de choses nouveau et qui n'est pas susceptible de durer a la longue: la multiplication des 'langues communes' dans l'Europe d'aujourd'hui, en un temps ou il y a an fond unite de civilisation materielle et intellectuelle, est une anomalie» («La Methode,» p. 20). Here we may note (1) the idea of unity of civilization, and (2) the conception of anomaly. The idea of unity is an organical conception, for we are allowed to speak only of the «seeming similarity.» The idea of anomaly is rather interesting, for it reveals a new complex. In fact, either the state of things is in conflict with the observation of a great number of similar facts (e.g., the anomaly of embryological development), which is not the case, for the present situation in Europe is unique, having never occurred before, or it is in conflict with certain theoretical presumptions, which is just the case. A third supposition may be made; namely, that of the desire of a certain definite achievement, but it must not be, I believe, discussed in the case of A. Meillet. There is a simple rule that, if there are anomalies of such a type, the attention must be drawn to the revision of premises — in this case, the establishment of «norms,» which ought to be scrutinized. As a matter of fact, what is now seen in Europe is well understood as absolutely «normal» effects of strong centripetal and centrifugal movements under intensively varying interethnical pressure. The case of O. Jespersen is different, for he wants to introduce an international language, which operation naturally must be «rationalized.» Cf. also N. Marr's theory of pyramidal unification of language which is also a rationalization of a credo, but in this case, one may guess, professed ex officio.
41. Reference to the geographical conditions is very often made when the other reasons of existence of phenomena cannot be easily understood. Indeed, it is a very soothing hypothesis: since geography is a responsible factor, the analyses and investigations may at least be postponed.
42. It is interesting that this particular case has attracted the attention of many authors.
43. I omit all cases of «socialistic,» «communistic,» «religious,» «moral,» and other ways of attacking the problems, which as such have naturally nothing to do with the science.