It would be difficult, if not impossible, to assign to the various subdivisions of the Indo - European stock their respective areas, or, where admixture has taken place, their relative proportions. Language and race are, as already observed, by no means coextensive. The Celtic race, for instance, which occupied Gaul, Northern Italy, large portions of Spain and Germany, and even penetrated across the Hellespont into Asia Minor, where it gave name to the province of Galatia, is now repre- sented linguistically by the insignificant populations among whom the Welsh and the Gaelic or Erse lan- guages retain a lingering existence. The Italian race, on the other hand, which must have been well-nigh an- nihilated by, or absorbed in, the overwhelming masses of the Northern hordes, has imposed its language out- side the bounds of Italy over the peninsula of Spain, France, and Wallachia. But. while the races have so intermingled as in many instances to lose all trace of their original individuality, the broad fact of their de- scent from one or other of the branches of the Indo- European famih' remains unaffected. It is, indeed, im- possible to affiliate all the nations whose names appear on the roll of history to the existing divisions of that family, in consequence of the absence or the obscurit}' of ethnological criteria. Where, for instance, shall we place the languages of Asia Minor and the adjacent dis- tricts? The Phrygian approximates perhaps to the Greek, and yet it differs from it materially both in form and vocabulary (Rawlinson, Ilerod. i, 666) ; still more is this the case with the Lycian, which appears to pos- sess a vocabulary wholly distinct from its kindred lan- guages (ibid, i, 669, 677-679). The Armenian is ranged under the Iranian division ; yet this, as well as the Ian-
TONGUES, CONFUSION OF 478 TONGUES, CONFUSION OF
guage of the Caucasian Ossctes, whose iiiiiigeiious name of /r or Iron seems to vindicate for them the same rcla- tionsliip, is so distinctive in its features as to render the connection duhious. The languages prevalent in the mountainous district answering to the ancient I'ontus are equally peculiar (l)iefenbacli. Oriii. Kurop. p. 51). Passing to the westward, we encounter the Thracians, reputed by Herodotus (v, 3) the most powerful nation in the world, the Indians excepted ; yet but one word of their language {bria =" town") has survived, and all historical traces of the people have been obliterated. It is true that they are represented in later times by the (leta?. and these in turn by the Daci ; but neither of these can be tracked either by history or language, unless we accept Grimm's more than doubtful identi- lication whicli would connect them with the Teutonic branch. The remains of the Scythian language are sutiicicnt to establish the Indo-European affinities of tliat nation (Rawlinson, llerod. iii, 196-2()3), but insuf- licient to assign to it a detinite place in the family. The Scythians, as well as most of the nomad tribes as- sociated with them, are lost to the eye of the ethnolo- gist, having been either absorbed into other nationali- ties or swept away bj' the ravages of war. The Sar- niatiB can be traced down to the lazyges of Hungary and Podlachia, in which latter district they survived until the 10th century of our aira (Smith, Diet, of Geog. ii, S), and then they also vanish. The Albanian lan- guage presents a problem of a different kind : materials for research are not wanting in this case, but no detinite conclusions have as yet been drawn from them. The people who use this tongue (the Skipetai-eg, as they call themselves) are generally regarded as the representa- tives of the old Illyrians, wlio in turn appear to have been closely connected with the Thracians (Strabo, vii, 315; .Justin, xi, 1), the name Dardani being found both in lUyria and on the shores of the Hellespont; it is not, therefore, improbable that the Albanian may contain whatever vestiges of the old Thracian tongue still sur- vive (Uiefenbach, Ori(j. Europ. p. C8). In the Italic pe- ninsula the Etruscan tongue remains as great an enigma as ever: its Indo-European character is supposed to be established, together with the probability of its being a mixed language (Bunsen, PAiV. o/" ///*■/. i, 85-88). The result of researches into the Umbrian language, as repre- sented in the Eugubine tablets, the earliest of which (late from about B.C. 400; into the Sabellian, as rejire- sented in the tablets of Velletri and Antino; and into the Oscan, of whicli the remains are numerous, have de- cide.! their position as members of the Italic class {ibid. i, 90-94). The same caimot be asserted of the Mesajn- an or lapygian language, which stands apart from all neighboring dialects. Its Indo-European character is affirmed, but no ethnological conclusion can as yet be drawn from the scanty information afforded us {ibid, i, 94). Lastly, within the Celtic area there are ethnolog- ical problems w-hich we cannot pretend to solve. Tlie Ligurians, for instance, present one of these problems: were they Celts, but belonging to an earlier migration than the Celts of history ? Their name has been refer- red to a Welsh original, but on this no great reliance can be placed, as it would be in this case a local ( =co(ist- tiien) and not an ethnical title, and might have been im- posed on them by the Celts. They evidently hold a posterior place to the Iberians, inasmuch as they are said to have driven a section of this people across the Alps into Italy. Thai they were distinct from the Celts is asserted by Strabo (ii, 128), but the distinc- tion may have been no greater than exists between the British and the (Jaelic branches of that race. The ad- mixture of the Celts and Iberians in the Spanish penin- sula is again a somewhat intricate cpiestion, which Dr. Latham attempts to explain on the groimd that tlio term Celt {KtXrai) really meant Iberian {Edin.nf Eur. p. 35). That such questions as these should arise on a subject which carries us back to times of hoar antiqui- ty forms no ground for tloubting the general conclusion