VARIETIES OF LANGUAGE Dialects.—Whenever a language is spoken by a great number of people, it is inevitable that there should be within its boundaries greater or lesser differences, partly of a local and partly of a non local character. The former constitute what is generally under stood by the word dialects. The existence of sharp dialect bound aries has been denied by some philologists (Schuchardt, Gaston, Paris and others), who maintain that there are boundaries for each separate linguistic phenomenon (each sound law or morpho logical peculiarity, etc.), but that each of these boundaries ("isoglosses," as they are called) is independent of the boun daries for other phenomena ; a village will thus always in some respects go with its neighbour to the North, in others with that to the East, etc. There is some truth in this, but only where one and the same population has been living for a long time in the same district in a flat country, where there have been no natural hindrances to continued intercourse, and where no great migra tions have ever taken place. But as a matter of fact conditions have very seldom been so simple, and consequently we do find sharp dialect boundaries here and there. This is one of the in disputable facts that have been brought to light especially by so-called linguistic geography, which has of late years flourished more particularly in France, where a great Atlas linguistique de la France consisting of 2,000 maps has provided scholars with incomparable material for these studies.
Dialect boundaries are very often due to natural obstacles to free intercourse; not infrequently these consist in great forests which in olden times were impenetrable, though they may have disappeared later; or again marshy districts, etc., whereas rivers form boundaries only where the stream is so rapid that it is not navigable; in mountainous countries the boundaries do not always follow the highest ranges of mountains, as there is often much traffic through defiles which connect places on either side of the watershed. On the whole we see that it is not physical, but
human geography that is decisive. An instructive case has been studied by Gauchat in Switzerland. The two villages La Ferriere and Les Bois are situated on about the same level and only an hour's walk distant ; yet their dialects are mutually unintelligible, one agreeing with Franco-Provençal, the other with North French. The inhabitants of one village are Protestants and live by indus try; those of the other are Catholics and chiefly pastoral. They therefore look askance at one another, and no marriages take place between them. Now the difference of dialect dates further back than the religious difference, and it is possible to show that Les Bois was founded in the 14th century through immigration from Franche Comte, while La Ferriere was founded a couple of centuries later, when some villagers from Chaux-de-Fonds cleared the forest here to obtain pasturage for their cattle. Only recently the two neighbouring villages have been joined administratively, and they are now beginning to have a little intercourse.
The chief law of linguistic biology is this, that intercourse breeds similarity, and want of intercourse dissimilarity. In former times a great many splittings took place, because means of communica tion were inferior to those of our own day; it was not long of ter the colonization of Iceland, that Icelandic began to show traces of differentiation from Norwegian. Nowadays linguistic agreement can be much better preserved, accordingly the English spoken in Australia and New Zealand is not widely different from that of the mother country, and the difference in speech between Bos ton and San Francisco is much less than what may be observed between two villages in Great Britain that are only a few miles apart. There is no definite point at which we may say that "a dialect" has developed such great differences as to have become a "language"; some will use the former, others the latter term of "Afrikaans," the Dutch "Taal" of South Africa.