The evidence furnished by their languages is not unfavorable to the supposition that the Basques and Celts may have been of African origin.

Though by Humboldt, and some other eminent writers, the Basque has been regarded as distinct from other languages, the examples which occur at the close of this Introduction must, I conceive, serve to remove all doubt as to the identity of the Basques or Iberians with the other branches of the Human Race. Of these examples grammatical differences cannot serve to diminish the force. (See p. [89] and the chapter on the Chinese Language.) The Basque also shows some [pg xxxvi] traces of a peculiar connexion with the African tongues. Thus its numerals are nearly identical with those of the North African nations, and the formative particle Er is used for similar purposes in the Basque and Egyptian, and in both is placed before the word, a characteristic which distinguishes the African from the European languages. (See p. [142].) Thus we have Juan, “To go,” Er-uan, “To cause to go,” (Basque.) Ouini, “Light,” Er-ouini, “To cause Light,” or “To enlighten,” (Egyptian.) Instances of words formed in the same manner, which are common to the Egyptian and the Celtic, will be found at p. [38], Appendix A.

A striking example of the connexion of the Celtic languages with those of Africa occurs in the region where the respective Physiological peculiarities of North Africa and Negro-land meet. In the vicinity of the river Senegal the line of separation may be said to divide the Iolofs, a Negro nation, from the Fulahs and Phellatahs, whose physical characteristics are of an intermediate nature. Now it is remarkable, that by comparing and as it were uniting the dialects of the Iolofs, the Fulahs, and the Phellatahs, some of the most common Welsh words are obtained essentially unchanged, as in Le oure, “The Moon,” (Fulahs,) Gour, and Gourgne, “A Man,” (Iolofs,) Gourko, “A Man,” (Phellatahs,) Loho, “The Hand,” (Iolofs,) Bourou, “Bread,” (Iolofs,) Bouron, “Bread,” (Fulahs.)

Consistently with the principles on which the origin of languages is hereafter explained in this work, I cannot suggest that these coincidences, striking as they are, afford any proof of a specific connexion between the Celtic and African races. But they tend to prove, nevertheless, that language furnishes no positive ground for inferring that the Celts are more nearly allied to the Asiatic than they are to the African races. Hence, since the evidence of Physiology on this subject is also of a negative character, it may be [pg xxxvii] affirmed, with regard both to this race and the Basques, that the opinion that they are of Asiatic descent—opposed as it is by the evidence of history in one, if not in both cases—and by the inferences which Geographical considerations, in both instances, appear to suggest—requires reconsideration.

In this place I may observe, that in the course of the following inquiries it will be found true as a general principle, that in direct proportion as the proofs of the General Unity of the different races of the Globe are observed to become more distinct, the evidence which has frequently been relied upon as demonstrative of a specific connexion between particular races will also be observed to become more doubtful, for both the affinities and differences which exist between the languages of contiguous—and those of the most distant—nations, are for the most part so nearly alike in character, and so nearly equal in degree, as to favour the inference that the dispersion of the Human Race must have been exceedingly rapid, and that many ancient nations, such as the Basques and Celts, who in subsequent times were found closely contiguous, must, in the first eras of the world, have been isolated from each other by incessant war and nomade habits, almost as early as the most distant nations were! It is certain that the language of the Welsh does not present either to the Basque or to the Teutonic—dialects of nations located contiguously to their Celtic forefathers—examples of affinity more striking than those just adverted to. Nor are the examples above noticed of the connexion between the Welsh and the African dialects by any means more remarkable than the instances of resemblance between the former tongue and the dialect of the Mandans, a North American Indian Tribe, which have been pointed out by Mr. Catlin! In both cases the same observation applies—an observation based on a principle that will be more fully understood hereafter—viz., that [pg xxxviii] these coincidences are unequivocal proofs of a generic, but not of that kind of specific relation, which implies that these nations were at one time united more intimately than the other families of mankind.

Various miscellaneous considerations connected with the primitive migrations of mankind may now he adverted to.

Neither the extent nor the physical features of our Globe are such as imply that the spread of population over its surface must necessarily have been the work of many ages. To traverse the habitable earth from the Southern extremity of Africa to the North of Asia, and thence to the extreme Southern point of the American continent, is a task which would require only a small fraction of one man's life! And in the first ages of the Race, Man was probably a Nomade, a Wanderer! It may be inferred, therefore, that in the early ages of the world the diffusion of population was very rapid in the warmer latitudes, while towards the North it was obstructed rather by climate than by any other cause. As population became more dense in the more favoured regions, weaker tribes, it may be surmised, were gradually driven into the steppes of Asia and the wilds of Siberia, whence they may be supposed to have penetrated into Europe on the one hand, and across Behring's Straits into America on the other. With the exception of America, all the great Continents are connected together by districts easily traversed by Man; and Behring's Strait, which is interposed between America and the North-east of Asia, might be passed in the canoes of some of the most barbarous tribes with which we are acquainted.

The peopling of Islands is a subject that has been discussed very satisfactorily by Dr. Prichard, and after him by Mr. Lyell. Their conclusion is, that the occasional drifting of canoes by storms and currents, is sufficient to account for the existence of Human population in the most remote [pg xxxix] islands, as is proved by facts related by Kotzebue and others. Several reasons have however been suggested in the following pages, for the conclusion that Australia is a recently peopled country.

The geographical distribution of the various languages of the globe seems to render Adelung's arguments for regarding Central Asia as the birth-place of our species eminently convincing. The languages of China and the South-east of Asia are either Monosyllabic, or Tongues that partake of that character; Languages having the same features are spoken through the long chain of islands in the Pacific as far as New Zealand. All the other Tongues of the Globe are Polysyllabic. Now if the birth-place of Man and the focus of migration was in Central Asia, on the borders of Cashmire and Tibet, this division of Languages would necessarily have followed, for it will be observed that Tibet, which is the source of the rivers of the regions to the South-east, would in that case have given inhabitants to the countries of South-eastern Asia, countries which are isolated from all others, for not only are they cut off from Europe, Africa, and Western Asia, by the system of Table-lands and its Mountains, they are also separated from Northern Asia and therefore from America by the Great Desert of Gobi or Shamo. To the Steppes of Northern Asia, and consequently to America as well as to Europe and Africa, the territory of Persia or Iran, which, as has been seen, forms the opposite slope of the system of Table-lands, is the natural route.

The relations which the Parsian, the Pehlwi, and the Zend, the ancient dialects of Persia, bear to those of the surrounding countries, seem to be in a highly interesting manner confirmatory of Adelung's views. The Parsian, which was spoken in the South of Persia in the provinces near to India, approaches so closely to the Sanscrit, the ancient language of that country, that Sir William Jones considered the Parsian [pg xl] to have been the parent of the Sanscrit. The Pehlwi, the language of the Parthians who occupied the centre of Persia, a territory that adjoins the Semetic countries, appears very decidedly to be a connecting link between the Semetic languages on the one hand and the Parsian and Zend and the Indo-European tongues, viewed as a class, on the other. The Zend, the dialect of ancient Media, or North Persia, is supposed to be closely allied to the Armenian. The Parsian, Pehlwi, and Zend, respectively bearing these relations to the languages of the neighbouring countries, are closely connected as sister dialects among themselves. These facts tend to show—from the summit of the Western Table-land viewed as a centre, through Persia viewed as a medium—a radiation of language from which a radiation of population may reasonably be presumed.