* words preserved in topographical names. It is, indeed, not improbable that the dialect of the Bretons may be, in part, a relic of the idiom of the Armorican Gauls, but there is a degree of uncertainty connected with that supposition which prevents our assuming it as a matter of fact. We have indeed reason to believe, that the native language continued to be spoken in some parts of Gaul nearly to the end of the Roman domination, but we are not sure that this was the case in Britanny. On the other hand, there is historical evidence that Armorica received a colony from Britain about the period of the Saxon invasion of this island. By the older historians of France, the Bretons are described as a particular and distinct people, under the name of Britanni: they claimed a descent from the insular Britons. It cannot be proved that the Celtic language had not entirely ceased to be spoken in the districts which they occupied, and that the dialect which has long prevailed there was not introduced anew, on the arrival of this colony. It is, therefore, in the British isles that we must look for the genuine remains of the Celtic language, preserved by an unbroken succession from early times. In the British isles we have two extant languages, handed down from the earliest ages, each possessing a peculiar literature; these are the respective languages of Britain and of Ireland. Under each of these divisions we may class three cognate dialects—the Welsh, the Cornish, and the Armorican belong to the former; -the Irish, the Manx, and the Scottish Gaelic, which last is supposed to have been spread from Ireland into Scotland some centuries after the Christian era, belong to the second division. The Irish and British languages cannot with propriety be termed dialects of one speech, since each is unintelligible to persons who have learnt only the other. They are sister languages, and perhaps resemble each other as nearly as the English and German: but the difference between them is far too great and too fundamentally interwoven with their grammatical structure, to have arisen since the era of the Roman conquest of Gaul. That either of these languages is a cognate dialect with that of the old Celtic Gauls, is an assertion which requires proof, and perhaps such proof may be furnished, but * Gregory of Tours, Fredegarius Scholasticus and Eginhardt. it is evident that this character can belong to only one of the two languages. Both however claim it. There is a host of writers who take it for certain that the Celtic idiom is preserved in or represented by the British or Welsh language, and there are fully as many who advance the same claim in favour of the Erse or Gaelic. Whatever conclusion may be adopted as to the question agitated between these parties, it will be found to involve consequences which are very important in their bearings on the history of the western nations of Europe. We cannot therefore proceed in the ethnology of the Celtic race without discussing it. But, in the first place, it will be necessary to form a correct idea of some subordinate inquiries which connect themselves with this controversy, and of some results which follow the admission of either of the two suppositions. If we conclude that the Erse or Gaelic is to be considered as the true descendant and representative of the Celtic, or of the idiom of the people of Gallia Celtica, who then were the Britons? They were not originally a colony of Celtic Gauls. Whence came they, and with what people on the continent were they connected? Again, who were the Belgæ, the people of northern Gaul, who, according to Cæsar, had a language of their own, different from the Celtic? These and other inquiries have received different replies from writers, who maintain the same hypothesis as to the Celtic language, and agree in identifying it with the Erse. With respect to the Belgæ, many English writers have adopted the opinion of Mr. Pinkerton, which they maintain on the authority of a well-known passage in Cæsar's Commentaries. They consider the Belge to have been a Teutonic or German race, and altogether separate from the Celtic stock, and suppose that a German dialect was prevalent through the north of Gaul and even on the southern coast of Britain, which the Belgæ from the opposite shore had in some places colonised before the conquest of Gaul by the Romans. The Britons are imagined hy these writers to have emigrated, not from Gaul, with the inhabitants of which they had no immediate connexion, but from Denmark or the north of Germany. The principal argument adduced in proof is the national designation of the Welsh people, who call themselves Cymru or Cymri, a name resembling that of the ancient Cimbri. Other writers, principally in Germany and France, holding the same opinion as to the Celtic language and its relation to the Erse, suppose that the Welsh or Britons were descendants of the Belgæ. The proofs on which they ground this supposition are not so obvious or so much upon the surface, as the arguments adduced by Pinkerton and his followers. I shall take occasion to explain, and shall endeavour to estimate them. The opinion to which I now advert was first advanced, as I believe, by Schloetzer, the learned historian of the north. It was adopted by Gatterer and in part by the authors of the Mithridates, and has been set forth more fully by M. Thierry in his learned work entitled "Histoire des Gaulois." Lastly, it has received the countenance of two writers whose authority cannot be lightly regarded, namely, the modern historians of Rome, Niebuhr and Dr. Arnold. for a On the other hand, we must take up moment the supposition that the Welsh Britons were a colony of the Celtic Gauls, and that the Welsh language represents the idiom of the ancient Celts. We shall find that this, if once admitted, accounts for many facts in ancient history which are otherwise very difficult to explain. It has been a very old and until lately a very general opinion, that the ancient Britons were a Celtic people and a colony from Celtic Gaul. Both nations were governed by the same druidical priesthood. The Belgæ indeed appear likewise to have been subject to the Druids, but the principal domain of that celebrated hierarchy in Gaul was the Celtic part of it. Many other historical facts are easily understood on the supposition of a near relationship between the Britons and the Celts, which can hardly be reconciled with a different opinion. But I am not at present advocating the truth of this supposition. I am only stating it as a subject which must be discussed.* * I fear that some of my readers may think this discussion needless; they may perhaps not be aware that a contrary opinion has been maintained; but those who are acquainted with what has been written by continental writers during the last twenty years, know that the opinion of Schloetzer and Thierry is almost universally received both in Germany and in France. We must further observe that those who look upon the Britons as nearly related to the Gauls, and consider the Welsh language as the modern representative of the Celtic, generally suppose the Belgae to have been another branch of the same stock, and not a people of an entirely different family, namely Germans. Some have thought it probable that the Belgic language was analogous to that of the ancient Irish, and they appeal for proof to an old tradition, which derives the population of Ireland in part from a colony termed by the national Irish poems Fir Bolg or Viri Belgæ. Others, persuaded that the Belgic language was not so remote from the Celtic as is the Irish from the Welsh, conclude that the idioms of both nations were dialects related to each other, as are the Welsh to the Cornish or to the Armorican. It would be a long and fruitless task to discuss all the passages of ancient writers and the other arguments, by an appeal to which the advocates of each of these several opinions have endeavoured to support their favourite notions. The principal hope that I entertain of being enabled to throw a ray of light on this portion of European ethnography is by adopting a different method, and one which has not been pursued to any considerable extent in relation to the history of the Celts. I shall endeavour to proceed on the same path which has been followed with remarkable success by the learned Baron W. von Humboldt, in his inquiries respecting the aboriginal inhabitants of Spain. But in order to render obvious and available to my purpose the evidence that may be collected on this ground, it will be necessary to take a previous survey of the principal divisions of Gaul, and of the mutual relations of the Celtic tribes spread through other parts of Europe. SECTION II.-Earliest Accounts of the Celtic Nations.Original Celtica of the Greeks-of its earlier Inhabitants, the Ligurians. The southern coast of Gaul was long known to the Greeks, before any notice was obtained of its Celtic inhabitants. It was then the country of the Iberians and Ligurians, or of a mixed population descended from both these races. In the Periplus of Scylax we find the observation, "After the Iberians, that is in proceeding eastward, succeed the Ligurians and Iberians, interspersed or mixed as far as the Rhone," that is, from the confines of Spain to the Rhone. "From the river Rhone are Ligurians, who extend as far as the Arno.” This is perhaps the country from which the Ligurians are reported to have expelled the Sicani, who were an Iberian race.* The country about Marseilles, before the Phocæan colony was founded there, is said to have been a part of Liguria. Thus Scymnus Chius : ἐν τῇ Λιγυστικῇ δὲ ταύτην ἔκτισαν πρὸ τῆς μάχης τῆς ἐν Σαλαμῖνι γενομένης. Marcianus Heracleota introduces the same lines, after mentioning the colony of Marseilles, founded by the Phocæans.† Even Herodotus described the country about Marseilles as inhabited by Ligurians. In those fabulous times to which the story of Hercules and his expedition to the Garden of the Hesperides was referred, we find the Ligurians placed by the poets in the same region. Between Marseilles and the Rhone, near Aquæ Sextiæ, now Aix, Hercules is said to have encountered great difficulties in his march. He was opposed by the Ligurians, and was only enabled to overcome them by the aid of Jupiter, who rained down stones from heaven. This story is told variously by Pomponius Mela and Eustathius. It is noticed by Strabo, who describes the spot where the encounter took place, and cites a passage of Eschylus from the lost tragedy of Prometheus Solutus relating to it.‡ * Scylax Caryand. Periplus. Hudson, tom. i. The same meaning is expressed in some verses of Avienus: "Rhodani-alveo, Ibera tellus, atque Ligyes asperi + Marcian. Heracleot. Hudson, tom. i. "Prometheus Herculi exponens iter à Caucaso ad Hesperidas, in hanc loquitur sententiam : ἥξεις δὲ Λιγύων εἰς ἀτάρβητον στρατὸν, |