Of the two branches of the Keltic stock the British will be considered first, and that in respect to its origin.

It is rare that the population of an island is without clear, definite, and not very distant affinities with that of the nearest part of the nearest continent. The Cingalese of Ceylon can be traced to India; the Sumatrans to the Malayan Peninsula; the Kurile Islanders to the Peninsula of SagalÍn; the Guanches of Teneriffe to the coast of Barbary. The nearest approach to isolation is in the island of Madagascar, where the affinities are with Sumatra, the Moluccas and the Malay stock rather than with the opposite parts of Africa, the coasts of Mozambique and Zanguibar. But Madagascar has long been the great ethnological mystery. Iceland, too, was peopled from Scandinavia and not from Greenland.

It is in Gaul, then, that we must look for the mother-country of Kelts; at least in the first[59] instance, for Gaul is the nearest point—the white cliffs of Folkstone being within sight of the opposite shore. Yet (as an example of the extent to which one ethnological question depends upon another) the Gallic origin of the earliest Britons has been objected to. For a Keltic population, indeed, it has been admitted to be the natural area; but we have seen that a population other and earlier than the Keltic has been inferred from the shape of the skulls, and other phenomena of the Stone period. Now for such a population as this, Jutland or Sleswick has been considered the more likely locality, since the skulls in question have been compared to those of the Laplanders and Finns; and, if this be true, the further north we carry the home of the British aborigines, the less we find it necessary to bring the Finn or Lap families southward. This reasoning is valid if the original fact of any pre-Keltic population be true. Those, however, who doubt the premises, have no need to refine upon the current notion of Gaul being the original home of the Britons. Gaul, then, is the ground from which we take our view of the great Keltic division of the human species in its integrity; for, hitherto, we have seen but the western offsets of it.

That the country between the Seine and Garonne, corresponding with the provinces of[60] Normandy, Brittany, Maine, Anjou, Poitiers, the Isle of France, and the Orleannois, was Keltic, has never been doubted. The evidence of CÆsar is express; and there is neither objection nor cavil to set against it. There it is, where, at the present moment, the Keltic Breton of Brittany continues to be the language of the common people.

The central and south-eastern parts of France—the Nivernois, Burgundy, the Bourbonnois, the Lionnois, Auvergne, Dauphiny, Languedoc, Savoy, and Provence—were chiefly Keltic. Perhaps they were wholly so; but as the Ligurians of Italy, and Iberians of Spain are expressly stated to have met on the lower Rhone, it is best to qualify this assertion. At the same time, good reasons can be given for considering that the Ligurians were but little different from the other Gauls.

South of the Garonne the ancient population was Iberic.

Switzerland, or the ancient Helvetia, was Keltic, and beyond Switzerland, along the banks of the Danube, and in the fertile plains of Northern Italy, intrusive and conquering Kelts were extended as far east as Styria, and as far south as Etruria; but these were offsets from the main body of the stock, whose true area was Gaul and the British isles.[61]

The parts between the Seine and Rhine, the valleys of the Marne, the Oise, the Somme, the Sambre, the Meuse, and the Moselle were Belgic. Treves was Belgian; Luxembourg, Belgian; the Netherlands, Belgian. Above all, French Flanders, Artois, and Picardy—the parts nearest Britain—the parts within sight of Kent—the parts from whence Britain was most likely to be peopled—were Belgian.

Now, as Britain was originally Keltic, unless Belgium be Keltic also, we shall meet with a difficulty.

In my own mind Belgium was originally Keltic; and, perhaps, nine ethnologists out of ten hold the same opinion. At the same time, fair reasons can be given for an opposite doctrine, fair reasons for believing the BelgÆ to have been German—as German as the Angles of old, as German as the present Germans of Germany, as German as the Dutch of Holland, and, what is more to the purpose, as German as the present Flemings of Flanders, possibly occupants of the ancient, and certainly occupants of the modern, Belgium.

Upon the latter fact we must lay considerable weight. Modern Belgium is as truly the country of two languages and of a double population as Wales, Ireland, or Scotland. There is the French, which has extended itself from the[62] south, and the Flemish, which belongs to Holland and the parts northwards; a form of speech which differs from the true Dutch less than the Lowland Scotch does from the English, and far less than the Dutch itself does from the German. More than this. South of the line which separates the French and Flemish, traces of the previous use of the latter language are both definite and numerous, occurring chiefly in the names of places such as Dunkirk, Wissant, &c.

Now, as the French language has encroached upon the Flemish, and the Flemish has receded before the French, nothing is more legitimate than the conclusion than that, at some earlier period, the dialects of the great Germanic stock extended as far south as the Straits of Dover; and, if so, Germans might have found their way into the south-eastern counties of England 2000 years ago, or even sooner. Hence, instead of the Angles and Saxons having been the first conquerors of the Britons, and the earlier introducers of the English tongue, BelgÆ of Kent, BelgÆ of Surrey, BelgÆ of Sussex, and BelgÆ of Hampshire, may have played an important, though unrecorded, part in that long and obscure process which converted Keltic Britain into German England, the land of the Welsh and Gaels into the land of the Angles and Danes, the clansmen of Cassibelaunus, Boadicea, Caractacus and Galgacus[63] into the subjects of Egbert, Athelstan, and Alfred.

Such views have not only been maintained, but they have been supported by important testimonies and legitimate arguments. Foremost amongst the former come two texts of CÆsar, one applying to the well-known BelgÆ of the continent, the others to certain obscurer BelgÆ of Great Britain. When CÆsar inquired of the legates of Remi, the ancient occupants, under their ancient name, of the parts about Rheims, what States constituted the power of the BelgÆ, and what was their military power, he found things to be as follows—"The majority of the BelgÆ were derived from the Germans (plerosque Belgas ortos esse ab Germanus). Having in the olden time crossed the Rhine, they settled in their present countries, on account of the fruitfulness of the soil, and expelled the Gauls, who inhabited the parts before them. They alone, with the memory of our fathers, when all Gaul was harassed by the Teutones and Cimbri, forbid those enemies to pass their frontier. On the strength of this they assumed a vast authority in the affairs of war, and manifested a high spirit. Their numbers were known, because, united by relationships and affinities (propinquitatibus ad finitatibusque conjuncti), it could be ascertained what numbers each chief could bring[64] with him to the common gathering for the war. The first in numbers, valour, and influence were the Bellovaci. These could make up as many as 100,000 fighting men. Of these they promised 40,000; for which they were to have the whole management of the war. Their neighbours were the Suessiones, the owners of a vast and fertile territory. Their king Divitiacus was yet remembered as the greatest potentate of all Gaul, whose rule embraced a part of Britain as well. Their present king was Gallus. Such was his justice and prudence, that the whole conduct of the war was voluntarily made over to him. Their cities were twelve in number; their contingent 50,000 soldiers. The Nervii, the fiercest and most distant of the confederacy, would send as many; the Attrebates 15,000, the Ambiani 10,000, the Morini 22,000, the Menapii 9,000, the Caleti 10,000, the Velocasses and Veromandui 10,000, the Aduatici 29,000; the Condrusi, Eburones, CÆrasi, and PÆmani, who were collectively called Germans (qui uno nomine Germani appellantur) might be laid at 40,000."—Bell. Gall., ii. 4.

Let us consider this as evidence (to a certain extent) of the north of Gaul having been German, without, at present, asking how far it is conclusive. If we look to CÆsar's description of Britain we shall find the elements of a second proposition, viz., that "what is true of the northern coast[65] of Gaul, is true of the southern coast of Britain."[5] So that if the BelgÆ were Germans in the time of CÆsar, the populations of Kent, Surrey, and Sussex were German also.

CÆsar's statement is, "that the interior of Britain is inhabited by those who are recorded to have been born in the island itself; whereas the sea-coast is the occupancy of immigrants from the country of the BelgÆ, brought over for the sake of either war or plunder. All these are called by names nearly the same as those of the States they came from, names which they have retained in the country upon which they made war, and in the land whereon they settled."—B. G., v. 12.

I submit that these two statements would give us unexceptionable evidence in favour of the BelgÆ being Germans, and the south-eastern Britons being BelgÆ, in case they stood with no conflicting assertions to set against them, and no presumptions in favour of an opposite doctrine; in which case the inference that Kent was German would be irrefragable, and would stand thus—

The BelgÆ were Germans[66]—

The south-eastern Britons were the same class with the BelgÆ—

Therefore they were Germans.

Such a syllogism, I repeat, would be in proper form, and the inference satisfactory.

But there is a great deal to set against both: so much as to make it extremely probable that the utmost that can be got from the first statement is, that a part of the BelgÆ, and more especially the Condrusi, Eburones, CÆrasi, and PÆmani were Germans only in the way that the people of Guernsey and Jersey are English, i.e., politically but not ethnologically; and that the second only proves that certain national names occurred on both sides of the channel.

If we look at the numerous local, national, and individual names of the BelgÆ, we find that they agree so closely in form with those of the undoubted Gauls, as to be wholly undistinguishable. The towns end in -acum, -briva, -magus, -dunum, and -durum, and begin with Ver-, CÆr-, Con-, and Tre-, just like those of Central Gallia; so that we have—to go no farther than the common maps—Viriovi-acum, Minori-acum, Origi-acum, Turn-acum, Bag-acum, Camar-acum, Nemet-acum, Catusi-acum, Gemini-acum, Blari-acum, Mederi-acum, Tolbi-acum; Samaro-briva; Novio-magus, Moso-magus; Vero-dunum; Marco-durum, Theo-durum; Ver-omandui; CÆr-asi; Con-drusi; Tre-viri—all[67] Gallic compounds on Belgian ground, and all forms either wholly foreign to any German area, or else exceedingly rare. Now it is no objection to this remarkable and exclusive preponderance of Gallic names in Belgian geography, to say that there is no proof of the designations in question being native; and that, although they existed in the language of CÆsar's informants, who were Gauls, they were strange to the BelgÆ, even as the word Welsh is strange to a Cambro-Briton—being the name by which he is known to an Englishman, but not the true and native denomination. I say that all argument of this kind, valid as it is in so many other cases where it is never applied, has no place here; since CÆsar's informants about the Belgic populations were the BelgÆ themselves, and it is inconceivable that they should have used nothing but Gallic terms when they spoke of themselves, if they had not been Gauls.

The names of the individual Belgic chiefs are as Gallic as those of the towns and nations, e.g., Commius and Divitiacus, and so are those of such Britons as Cassibelaunus.

I submit that this is, as far as it goes, a reason for limiting rather than extending all such statements as the ones in question. And it is by no means a solitary one. A statement of Strabo confirms it:—"The Aquitanians are wholly different"[68] (i.e., from the other Gauls) "not only in language, but in their bodies,—wherein they are more like the Iberians than the Gauls. The rest are Gallic in look; but not all alike in language. Some differ a little. Their politics, too, and manners of life differ a little."—Lib. iv. c. 1.

With the external evidence, then, of Strabo, coinciding with the internal evidence derived from the geographical, national, and individual names, it seems illegitimate to infer from the text of CÆsar more than has been suggested.

Unless we believe the BelgÆ of Picardy to have been Germans, the second fact stated by CÆsar, viz., the Belgic origin of the south-eastern Britons is comparatively unimportant, since it merely shews that between the Britons of the south-eastern coast, and those of the interior, there were certain points of difference, the former being recent immigrants, and Belgium being the country from which they migrated. Nevertheless, this introduces a difficulty; since, by drawing a distinction between the men of Kent, and the men of the Midland Counties, we are precluded from arguing that the Britons in general belonged to the same class as the Gauls; inasmuch as CÆsar's description may fairly be said to apply to the Belgic Britons only.

I think, myself, that CÆsar's statement must[69] be taken as an inference rather than as evidence; in other words, he must not be considered to say that certain Attrebates and BelgÆ crossed the Straits of Dover and settled in Britain, but that, as certain portions both of Belgium and Britain bore the same names, a migration had taken place; such being the explanation of the coincidence. Or, if we suppose CÆsar himself to have been too acute a reasoner to confound a conclusion with a fact (as, perhaps, he was), we may attribute the inference to his informants. Whoever is in the habit of sifting ethnological evidence, is well aware that a confusion of kind in question is one of the commonest of the difficulties he must deal with.

At the same time, that there were some actual BelgÆ in Britain is likely enough; but that they were a separate substantive population, of sufficient magnitude to be found in all the parts of Britain where Belgic names occurred, and still more that they were Germans, is an unsafe inference; safe, perhaps, if the two texts of CÆsar stood alone, but unsafe, if we take into consideration the numerous facts, statements, and presumptions which complicate and oppose them.

The Belgic names themselves, which occurred in Britain, were as follows:—

a. Attrebates.—There were Attrebates both in Belgium and Britain; the Gaelic ones in Artois,[70] which is only Attrebates in a modern form. Considerable importance attaches to the fact, that before CÆsar visited Britain in person, he sent Commius, the Attrebatian, before him. Now, this Commius was first conquered by CÆsar, and afterwards set up as a king over the Morini. That Commius gave much of his information about Britain to CÆsar is likely; perhaps he was his chief informant. He, too, it was who, knowing the existence of Attrebates in Britain, probably drew the inference which has been so lately suggested, viz., that of a BelgÆ migration, or a series of them. Yet the Attrebates of Britain were so far from being on the coast, that they must have lain west of London, in Berkshire and Wilts; since CÆsar, who advanced, at least, as far as Chertsey, where he crossed the Thames, meets nothing but Cantii, Trinobantes, Cenimagni, Segontiaci, Ancalites, Bibroci and Cassi. It is Ptolemy who first mentions the British Attrebatii; and he places them between the Dobuni and the Cantii. Now, as the Dobuni lay due west of the Silures of South Wales, we cannot bring the Attrebatii nearer the coast than Windsor.

b. The BelgÆ.—These—like the Attrebatii, first mentioned by Ptolemy—are placed south of the Dobuni, and on the sea-coast between the Cantii and Damnonii of Devonshire; so that Sussex,[71] Hants, and Dorset, may be given them as their area.

c. The Remi are mentioned by no better an authority than Richard of Cirencester, as Bibroci under another name.

d. The Durotriges, too, or people of Dor-set, are stated by the same authority to have been called Morini.

e. f. In Ireland we have two populations with German names; the Menapii and the Chauci, both in the parts about Dublin, and in the neighbourhood of one another. And these are mentioned by Ptolemy.

Now, whatever these Belgic names prove, they do not prove CÆsar's statement that it was the maritime parts of Britain which were Belgic; since the Menapii and Chauci must have been wholly unknown to him, and the Attrebatii lay inland.

At the same time, they prove something. They also introduce difficulties in the very simple view that Britain was solely and exclusively British. This leads to a further consideration of the details. The Remi may be disposed of first. They stand on bad authority, viz., that of a monk of the twelfth century.

So may the Morini. Though I admit the ingenuity and soundness of the doctrine that the existence of a double nomenclature such as that[72] by which the Durotriges are called Morini, and the Morini, Durotriges, is well explained by the assumption of a second language, and the notion that the inhabitants of certain districts were sometimes called by a British, sometimes by a German, name, the hypothesis is not valid where the facts can be more easily explained otherwise. No one would thus explain such words as Lowlander and Borderer applied to the people of the Cheviot Hills. Yet both are current; one being given when their relation to England, the other when their difference from the Highland Gaels, is expressed.

Now, it so happens that Morini and Durotriges are words that can as little be considered as synonymous terms belonging to different languages as Lowlander and Borderer; since good reasons can be given for referring them both to the Keltic. Their exact import is difficult to ascertain; but if we suppose them to mean coasters and watersidemen, respectively, we get a clear view of the unlikelihood of one being German and the other Keltic. Thus—

Duro-triges coincides with the Latin compound ponticolÆ, since dwr in Welsh, Cornish, and Armorican means water, and trigau means to remain or to inhabit; trig-adiad denoting dwellers, or inhabitants, as is well remarked by Prichard, v. iii. 128.[73]

MÔr, in Morini, is neither more nor less than the Latin word mare.[6] Surely this sets aside all arguments drawn from the supposed bilingual character of the words Morini and Durotriges.

The Cauci and Menapii of Ireland tell a different[74] tale. One name without the other would prove but little; but when we find Cauci in Germany not far from Menapii, and Menapii in Ireland not far from Chauci, the case becomes strengthened. Yet the likelihood of Menap, being the same word as the Menai of the Menai Straits in Wales, suggests the probability of that word being a geographical term. Nevertheless, the contiguity of the two nations is an argument as far as it goes.

And here I must remark, that the process by which words originally very different may become identified when they pass into a fresh language is not sufficiently attended to. Cauci is the form which an Irish, Chauci that which a German, word takes in Latin. And the two words are alike. Yet it is far from certain that they would be thus similar if we knew either the Gaelic original of one, or the German of the other. A dozen forms exceedingly different might be excogitated, which, provided that they all agreed in being strange to a Roman, would, when moulded into a Latin form, become alike. Still the argument, as far as it goes, is valid.

Such are the reasons for believing, at one and the same time, that the Britons came from Belgic Gaul, and that the BelgÆ from whence they came were Kelts.

We cannot, however, so far consider the origin[75] of the British branch of the Keltic stock to be disposed of, as to proceed forthwith to the Gaelic; another population requires a previous notice. This is the Pict.

[5] These are the exact words of one of the ablest supporters of the Germanic origin of the south-eastern Britons, Mr. E. Adams, in a paper entitled, "Remarks on the probability of Gothic Settlements in Britain Previously to A.D. 450."—Philological Transactions, No. ciii.