Chapter 10

“But the oldest remains of the Danish language are to be found on our Runic stone monuments, and here at length it perfectly coincides with the earliest Swedish, Norwegian, and Icelandic.“The Danish is closely allied to the Swedish, and both, in the earliest times, lapse into the Icelandic, which, according to all ancient records, was formerly universal over all the North, and must therefore be considered as the parent of both the modern Scandinavian dialects.”45On the subject of the differences of dialect in the differentprovincesof the Northern Kingdoms he says that,“In Norway as well as in Denmark one province terminates its verbs[pg 045]ina, another distinguishes allthe three genders, while a third has preserved a vast number ofoldwordsand inflectionswhich to the othersare unintelligible.”We have thus a proof that even in the provinces of the same kingdom there are differences of“words, grammar,and inflections.”The differencein the numberofgendersis a very remarkable one.The researches of Professor Rask will be found distinctly to warrant the following conclusions. These conclusions are in the nature of results that legitimately flow from his researches; they do not represent the inferences which he himself has thence deduced. With regard both to the languages of England and of his native Scandinavia, this learned writer seems evidently to have been perplexed by the extent and variety of the changes he has described. Hence, in both instances, he has shown an inclination to ascribe to the influence of War and Social disturbance changes which his own researches clearly prove to have been the effects neither of transient nor of local influences, but of causes progressively at work through a series of ages, and embracing large groups of nations and languages in their action.1. The differences which now exist between the various Scandinavian Languages extend to all those features in which it is possible that one Language, or one Class of Languages, can differ from another; viz. to Words, Grammar,Inflections,46and to thearrangementof Words in sentences,47orIdioms.2. Not only do differences of this nature present themselves in the various ScandinavianKingdoms—but also in the various[pg 046]Provinces of the sameKingdom, which in many instances are distinguished by the most marked differences in Words, Grammar, &c. Thus the Dialect of Dalecarlia in Sweden is very ancient and distinct, and approaches to the Gothic.483. These characteristic features of the various languages and dialects of Scandinavia have arisen progressively during the course of ages.4. These differences principally consist in the abandonment in one Kingdom or Province of a portion of the Words, Idioms, Grammar, &c. of the Parent Speech—that part of the elements of the Original Tongue which have become obsolete in one dialect having generally been preserved in the dialects of other kingdoms and provinces—which have at the same time generally lost other distinct portions of the Vocabulary, Grammar, &c. of their common Original. In other words, the“Disjecta Membra”of the old Scandinavian, or“Danska Tunge,”when not preserved in the Danish, have been retained for the most part in the Swedish, Icelandic, and Norwegian, or in some of the Provincial dialects of Scandinavia, and vice versâ.In the various provinces in which it was once spoken different portions of the Parent speech have been abandoned or preserved.5. Hence it follows that the Primitive Language of Scandinavia, or“Danska Tunge,”does not exist in any one—but is dispersed inallits derivative dialects. (Compare the motto from Grotius on the title-page.)6. It is a necessary consequence of the third and fourth propositions that the more ancient remains of the derivative dialects approach more nearly to the Parent Speech, and—in the ratio of their superior antiquity—unite a greater proportion of the distinctive peculiarities of all the sister-dialects,[pg 047]which, as previously stated, have arisen in consequence of certain portions of the Parent speech having been abandoned in some provinces and retained in others, and vice versâ.An interesting illustration of this maxim occurs in a passage from Professor Rask's preface already quoted, in which, after giving a specimen of old Danish, which approaches closely to the Icelandic, he adds,“The few deviations from the Icelandic bear for the most parta strong resemblanceto theSwedish.”In other words, the older specimens of the Danishunitethose peculiarities by which the modern collateral Tongues of Iceland, Denmark, and Sweden are distinguished from each other.Let it be borne in mind, that the lapse of one thousand years has produced these changes, and the instructive nature of this example will be fully apparent. Of the accuracy of the data on which the previous deductions rest, all doubt must be removed by reference to one remarkable event. It is historically certain that the Island of Iceland is inhabited by a nation descended from emigrants from the opposite Norwegian coast. It is historically certain, also, that previously to the Ninth Century these warlike adventurers had not established themselves on the Icelandic soil. Anterior to that period, therefore, it is self-evident that, inasmuch as the Icelanders had no existence as a nation, the Icelandic Tongue could not have had a separate existence as a language. Yet it is certain that in the present day the Icelandic deviates at least as widely from the language of the adjoining Norwegian Coasts as that language deviates from the other Scandinavian Tongues.The evidence furnished by Professor Rask and the writers whose views he has combated, will be found, when fairly balanced, distinctly to support a very important Conclusion, contemplated by neither. The facts adduced on both sides conspire to show a rapid approximation of the Teutonic and[pg 048]Scandinavian branches of the Gothic as we ascend into remote ages.Of this approximation, the features of identity between the Anglo-Saxon and the Icelandic, pointed out by the writers whose views Professor Rask combats, furnish a reasonable presumption, which is converted into positive proof by the evidence collected by Professor Rask himself, that the same features occur in all the ancient, though they do not in the modern, specimens of the Languages of the Scandinavian Peninsulas. It is true, this learned writer, of whose researches I have chiefly availed myself in this Section, maintains that there are some features in which all the Scandinavian differ from the Anglo-Saxon and other Teutonic Dialects, a conclusion, however, but feebly supported by the examples he has adduced, and scarcely reconcilable in any way with the resemblance which the primitive Swedish dialect of Dalecarlia is said to bear to the Gothic. But, assuming the occurrence of some features of difference, even in the earliest specimens we possess, this assumption leaves untouched the proposition that these specimens show a rapid rate of approximation, which, if equally rapid prior to their date, implies that at an era not many ages anterior the identity of the languages of Germany and Scandinavia must have been complete.[pg 049]Section V.The Origin of the Irish Nation. The original Language of the British Isles was a Union of Welsh and Irish. Union of the Irish, Welsh, &c. in the ancient Local Names in the Celtic Countries of Gaul, &c. These Names a connecting Link between the existing Celtic Dialects and the Oriental, Greek, and other Languages, &c.The origin of the Irish nation, or Gael, forms—for numerous reasons—a highly interesting and important subject of inquiry. Of this Nation the very same theories have been maintained as those which have been adopted in some quarters with respect to the North American Indians, the Negroes, and other branches of the Human Family; viz., that they are of a stock aboriginally inferior and distinct, by nature incapable of the virtues of civilization. Let the views advocated by Pinkerton with respect to the Gaelic race—views received with no slight degree of favour in his time—be compared with the doctrines of many modern writers on the subject of the native African and American Races, and an instructive lesson will be learnt on the force of prejudice and the uniformity of error!On the other hand, it must be allowed that the opinions which have been generally espoused on the subject of the origin of the Gael by many of the Historians and Scholars of Ireland and of the Highlands of Scotland, can scarcely be said to possess a better claim to the approbation of a calm and dispassionate judgment. Eminently distinguished as the Irish are by Literary genius, there is probably no subject on which their native talent has appeared to less advantage than in the investigation of the early History of their own[pg 050]Country. Fictions the most extravagant, borrowed from the Chronicles of the dark ages, have been credulously adopted by their first Scholars in lieu of those solid truths to which a calm and sober inquiry alone can lead. Thus we find Mr. Moore, at once the Poet and the Historian of Ireland, lending the sanction of his name to the Fable that the Irish are of Spanish origin; and citing, in answer to the more reasonable hypothesis of a British origin, a variety of Irish writers of no mean note, and some Welsh writers also, in favour of the assertions: 1, that the Irish Language is almost totally unlike the Welsh or Ancient British; and 2, that the Welsh is not a Celtic but a Gothic Tongue! There is every reason to conclude that Mr. Moore—unacquainted, probably, with any of the Celtic dialects himself—resorted to those authorities which he might naturally have deemed most deserving of confidence. But this only renders more lamentably conspicuous the credulity, carelessness, and ignorance of those to whose labours he has appealed. The assertions, 1, that the Welsh and Irish are unlike; and 2, that the Welsh is a Gothic dialect, are contradictions of the plainest facts.Influenced by national feelings Gaelic Scholars have also advanced various other theories, calculated to exhibit the antiquity of their language and race in a favorable point of view. The Gaelic has been maintained to be the Parent, at least in part, of the Latin, the Welsh, &c.; while to the first Colonists of Ireland a Carthaginian or Phœnician origin has been assigned.These conclusions cannot be sustained. But it is highly probable, notwithstanding, that the proofs on which they have been based will be found, in many instances, to contain the germs of important truths, though blended with an admixture of error. The traces of affinity between the Irish[pg 051]and other ancient languages which have been collected by Gaelic Scholars, may be open in many cases to the same remark, which is clearly applicable to the examples of affinity pointed out by Mr. Catlin between the dialect of the North American Indian tribe the Mandans and the Welsh; viz., these features may consist of clear and genuine traces of a generic, though they may afford no proofs of a specific, affinity of race. There can be no doubt that the Irish preserves many primitive forms which the kindred Celtic of Wales has lost; there can be no doubt also that the Irish approximates to the Latin, to the Greek, and to the Egyptian,49&c. in many features which the Welsh no longer exhibits. The examples adduced in Appendix A of the connexion of the Irish language with the Hebrew, Egyptian, &c. are sufficient to show that the Irish are a nation of Oriental origin. But on the other hand it must be borne in mind, that inasmuch as the Welsh, Latin, &c., have also preserved primitive forms which the Irish has lost, there is no ground for concluding that the Gaelic is a Parent rather than a Sister of these venerable Tongues; and inasmuch as the evidence of the Eastern origin of the Gael, however unequivocal, is not clearer or closer than the accompanying50evidence with respect to the Welsh, English, and other European nations, there are no peculiar grounds for referring the first colonization of Ireland to a direct migration from the shores of Palestine or Africa, rather than to the gradual diffusion of population from a central point.The following comparison presents examples of features in which the Irish approximates to the Gothic and other Languages, at the same time that it differs more or less from the Welsh.[pg 052]Words in which the Gaelic resembles the Gothic, and other European Languages, more closely than it resembles the Cymraeg or Welsh.English.Gaelic.Illustrations.Cymraeg.1. Father.Ath-air, (Ir.)Atta, (Gothic.), Ayta, Aydia, (Basque.), Attia, (Hung.), Otek, (Russ.), Fader, slightly varied in all the Gothic dialects, except the Gothic properly so called, Pater, (Greek & Latin.)Tad, (W.)2. Mother.Math-air, (Ir.)Mater or Mutter (with some trifling variations) in Latin, Greek, and all the Teuto-Scandinavian dialects except the Gothic—also in the Sclavonic and Bohemian. Ath-ei, (Gothic.)Mymmog, (Manx dialect.Mam, (W.)3. Brother.Brathair, (Ir.)The Irish form,Brathair,occurs in the Latin and Teuto-Scandinav. tongues; the Welsh form,Brawd,in the Sclavonian tongues.Brawd, (W.), Bredar, (Cornish.)Breur, (Manx dialectBreur, (Arm.)4. Sister.|Siur, (Ir.)The Irish form prevails in the Latin, Teuto-Scand. and Sclavonic.Chwaer, (W.)Piur, (Scotch.)Hor, Huyr, (Cornish.)5. A Company.Drong, (Ir.)Drang, a Throng, a Crowd, (German.)Torv.6. Mock.Magom, (Ir.)Mock, (English.)Gwatwor, (W.)7. Evil.Neoid, (Ir.)Naughty, (Eng.)Droug, (W.)Olk, (Ir.)Ill, (Eng.)8. The Bank of a stream.Rang, (Ir.)Rand,51(Germ.)Glan, (W.)9. A Step.Beim, (Ir.)Bēm-a, a Step, (Greek.), Bain-o, to go, Bahn, a Path, (Germ.)Cam.10. To bear.Beir-im, (Ir.)Fero, (Latin.) Ge-Bähr-en, (Germ.)Dwyn.11. Jeering, Delight, A Desire.Fon-amhad (Ir.), Foun, (Ir.)Fun, (Eng.), Vonne, Delight, (Germ.), Vunsch, a Wish, (Germ.)Vynn, or Mynn, a Wish, (W.)12. A Woman.Geon, (Ir.)Cwen, (Ang.-Sax. & Icel.)Gen-eth, a Girl, (W.)13. To know.Fis-ay-im, Fod-am, (Ir.)Viss-en, (Germ.), Vit-an, (Ang.-Sax.),“I wot,”(Eng.)Wys, or Gwys, Wyth, or Gwyth, Knowledge (W.)14. To heat, or warm.Gorm, (Ir.)Warm, (Eng.)Gwresogi, (W.)15. A Shadow.Sgath, (Ir.)Skia, Skiad-on, (Greek.), Schatten, (Germ.)Cysgod, (W.)16. To speak.|Raid-him, (Ir.)Read-en, (Germ.)Siarad, (W.)Some of these examples would furnish a more plausible argument to show that the Irish are a Gothic race than any which have been advanced to prove that the Welsh are of Gothic origin! It is singular, for instance, that the Irish terms expressive of the Domestic relations are so near the English as to excite in the first instance a suspicion that they must have been borrowed from the followers of Strongbow! But this impression must be dispelled by the reflection that terms of this class are never borrowed from its conquerors by a nation that continues to retain its primitive language. Moreover, it will be observed, that the Irish, in the instance of these words, approaches much more nearly to the Gothic,[pg 054]Hungarian, and Russian, &c. than it does to the English. Again, the Irish word“Gorm,”To heat or warm, islikethe English“Warm.”But, on the other hand, its genuineness is rendered indisputable by itsabsolute identitywith the word 'Gorm' in Persian and Egyptian, (See Appendix A, p.21.) Finally, the resemblances manifested above by the Irish to the Greek are quite as close as those which the former language displays to the English and other Gothic Tongues. In these examples, therefore, we may recognize proofs not of any partial results or specific connexions, but of the more complete approximation of the European languages as we enlarge our range of inquiry, and obtain more ample specimens of each Class.But, notwithstanding the occurrence of some features of difference, it is indisputable that there exists a close specific affinity between the Irish and Welsh Languages, which renders the common origin of the nations who speak them evident. The original identity of the Irish and Welsh Languages was established as far back as the commencement of the eighteenth century, by the investigations of the excellent Archæologist, Edward Lhuyd, who spent five years in travelling through the various Celtic regions, and whose comparison of the dialects of Wales, Cornwall, Armorica, the Highlands of Scotland, and the Isle of Man, is not inferior either in soundness of reasoning, or in patient, extensive, and honest research, to the best German works of the present day. But although the writings of Lhuyd may be said to have established the original unity of the Welsh and Irish races, since the publication of his work, a peculiar opinion has been adopted by some learned men with regard to thetimeof their original separation. Of this opinion, Edward Lhuyd was himself the first advocate; his conclusion was that though the Irish and British Celts were both descendants from one stock, they must have been separated into two[pg 055]distinct Tribes before their arrival in the British Islands. The Gaelic or Irish Tribe he supposes to have preceded the Welsh or British Tribe, by whom he conceives them to have been gradually driven to the West, as the Britons were by the Saxons in subsequent ages. Lhuyd's grounds are as follows:The most ancient names of Rivers and Mountains in the Island of Britain are very generally composed of terms still preserved in the Welsh or Ancient British Tongue. But there are some remarkable exceptions, and in these instances it frequently happens that the Names may be clearly identified with Words still preserved in the Irish or Gaelic branch of the Celtic. For example, the names of the British rivers, the Usk and the Esk, are particularly noticed by Lhuyd; these names are identical with“Uisge, Eask,”the Irish term for“Water.”This word, he observes, does not exist in the Welsh, and he had looked for it in vain in the sister dialect of Armorica; but, he adds, it is still retained by the Irish or Gaelic. Hence, he suggests that the Irish or Gaelic branch of the Celts must have colonized the Island of Britain before the arrival of the Cymry or Welsh branch, by whom, as he conceives, they were expelled, after having conferred names on the principal localities.The evidence of language will be found sufficient to show not merely the common origin of the Welsh and Irish, but also to fix a much more recent date for their separation than that which has been assigned by Lhuyd. It will thence appear that the Irish are descendants of Colonists of the Welsh or British race, not of a distinct Celtic sept, and that the commencement of the separate existence of the Irish nation must be referred to a comparatively recent date, propositions of much interest, of which the proofs about to be advanced will probably be deemed to be at once clear and simple.Lhuyd's reasoning in favour of his theory, that the Irish or Gael existed in Britain as a separate Tribe, prior to the arrival[pg 056]of the Britons who fought against Cæsar, the ancestors of the modern Welsh, is founded on a false analogy not unnatural to a first inquirer.The proposition that the most important local names in every country for the most part consist of terms belonging to the language of the very first inhabitants, is one of which I conceive the truth will be evident. For a proof of this principle, I may refer to Chalmers'52admirable analysis of local names in the Lowlands of Scotland, where, in spite of a succession of Conquests, and the utter extinction in that part of Britain of the language of the original inhabitants, viewed as a vernacular dialect, Welsh and other Celtic names are still preserved, after the lapse of ages, for the most prominent features of the country. This result, it may be observed, is one that flows from the very nature of things. Even the most fierce and ruthless invaders are compelled to hold sufficient intercourse with the first population to enable them to learn the proper names of their localities, and these names, from obvious motives of convenience, they almost universally adopt.Now, had Lhuyd shown that the most ancient Local names in Britain areexclusivelyIrish, there can be no doubt that, consistently with the principle just noticed, his theory would have been supported by the facts to which he adverts. But the most ancient local names in Britain are notexclusivelyorprincipallyIrish; in an equal number, perhaps in a majority, of cases they are Welsh.Moreover, it may be observed that the names of localities in this Island furnish highly instructive evidence, not merely with respect to the different races by whom it has been successively peopled, but also of the order in which they arrived. Thus the names of Rivers and Mountains, and other natural[pg 057]objects, at least of the most conspicuous, are Celtic; the names of the most ancient Towns are Latin, or Latin grafted on British words; more modern Towns and Villages have Saxon appellations; those of more recent origin have frequently Norman designations; and last of all come those places which have names derived from our present English. These various classes of names cannot be nicely distinguished in each particular instance. Of the correctness of the general principle, however, there is no doubt.But the terms noticed by Lhuyd as significant in the Irish language do not belong to a different class of appellations from those which are obviously of British or Cymraeg origin. The Irish and Cymraeg terms are both found to predominate most in the names of the most ancient Class, viz. in those of Rivers, Mountains, &c., and to be thus applied in conjunction. Hence the natural inference that flows from his facts is not that these names were conferred by two distinct and successive races, but that they were imposedcontemporaneouslyand by the same People!Further it may be noticed, that if British Topography presents words extant only in the Irish Tongue, Irish Topography also presents names which cannot be explained by means of the Irish, though their meaning is preserved in Welsh; for example: There is a place near the head of a Stream in Roscommon, called“Glan a Modda,”(from Glan,“The bank of a Stream,”Welsh.) There is a place in Wales, called“Glan a Mowdduy.”There is a place called“Glan-gora,”in a Creek at the head of Bantry Bay; and another place in Ireland called“Glan-gort.”“Ben-heder,”the ancient Irish name for“The Hill of Howth,”interpreted by Mr. Moore“The Hill of Birds.”(Adar,“Birds,”Welsh. The word does not exist in Irish.)Arran, A mountainous Island. (Arran, a Mountain,Welsh. This word does not exist in Irish,) &c. &c.[pg 058]Mr. Chalmers in his Caledonia states that the prevalent ancient names of localities in Britain and Ireland are essentially the same.The conclusions to which these facts legitimately and necessarily lead are, that the British Islands were originally colonized by Settlers, who, at the time of the first occupation of Great Britain and Ireland, spoke one uniform language, in which the Welsh, Irish, and other living Celtic Dialects were combined. We may infer, and I conceive most clearly, that these dialects must be viewed in the light of“Disjecta Membra”of the speech of the old British and Irish Celts, just as the Icelandic, Norwegian, &c. are fragments of the ancient“Danska Tunge,”as noticed in the previous section.It has been shown by Dr. Prichard that the population of Islands has been derived from the neighbouring Continents, and that the population of the more distant Islands has been derived in like manner from those which are nearer to the common source of migration. It is highly unreasonable to assume that Ireland has formed an exception to this general rule, considering that the common basis of the Irish and ancient British or Welsh languages are confessedly the same, unless it can be proved that the accompanying differences are such as to require the solution Lhuyd has suggested. Here, then, the question arises, are the features of difference between the Welsh and Irish languages more numerous or more fundamental, in relation to the interval of time that has elapsed since the Roman Invasion of Britain, than the varieties of dialect among the Scandinavian nations are in relation to the period that has elapsed since the colonization of Iceland? They are not! It will thence be seen that Lhuyd's theory, as to the remote date of the separation of the Gaelic or Irish from the British or Cymraeg branch of the Celts, is founded on an exaggerated conception[pg 059]of the stability of Human Tongues; and that the abandonment by various septs of different synonymes used conjointly by their common forefathers will satisfactorily account for the differences between the Welsh and Irish, to which he attaches so much weight. It will be perceived, for example, that in the Icelandic, of which the existence commenced in the ninth century, and the Continental Scandinavian from which it then sprang, totally different terms are used for“Water,”the very instance to which Lhuyd especially adverts, as regards the languages of the Welsh and Irish, whom we know to have existed as separate nations in the time of Cæsar eighteen centuries ago!Another highly instructive test of the correctness of his theory may be derived from the investigations of Lhuyd himself, who, in his comparison of the Welsh and Irish languages, uniformly distinguished the current terms from the obsolete synonymous words that occur only in ancient MSS. This comparison proves distinctly that the Irish and Welsh languages approximate, as we ascend, at a rate which, if as rapid previously as we know it to have been up to the date of the earliest MSS., would imply that these languages must have been identical about the era of the Roman invasion. As the changes which languages undergo in their infancy are more rapid than those which occur at later stages of their growth, it is possible that the unity of these Tongues may be ascribed even to a much later period, an opinion which has been maintained by a very judicious and excellent writer, Mr. Edward Davies, who in his“Claims of Ossian”has published an early specimen of Irish Poetry, which in Language and Style he regards as identical with the most ancient productions of the Welsh Bards. Making every allowance for the irregularity of the changes which occur in Languages, I do not conceive it possible that the Welsh and Irish could have differed very essentially[pg 060]in the time of Cæsar. This leads directly to another conclusion, viz. that the first colonization of Ireland could not have taken place a great many centuries before the Roman invasion. Had such been the case, the differences between the Welsh and Irish Languages must have been proportionately more extensive. In the time of the Romans we learn that an Irish traitor arrived in Britain, who stated that Ireland might be kept in subjection by a single legion, an incident which tends, however slightly, to favour the opinion that the sister Island was at that period but thinly, perhaps because but recently, peopled.Of the extent of the changes which the Celtic languages have undergone since the first arrival of the Celts in Europe, we possess proofs of far more ancient date than the earliest literary specimens of the living dialects of the Celtic in the Local names of Celtic regions, as preserved in Roman Maps, and in the existing languages of the French, English, and other nations, who occupy countries of which the Celts were the first inhabitants. These names I shall show to consist of three elements: A union of 1, Welsh, Cornish, &c.; 2, Irish, Highland Scotch, &c.; and 3, Terms not extant in any Celtic Tongue, but preserved in the Oriental, Greek, and other languages.As regards the Names of the 1st and 2d Classes, it will abundantly appear from the ensuing examples that, in the Topographical Nomenclature of Gaul, Britain, and other Celtic regions of Europe,53words derived from all the various Celtic dialects now extant, occur in a manner that leads distinctly to the inference that these“Disjecta membra”must have simultaneously belonged to the language of the old Celts. Dr. Prichard, who has examined these vestiges of the[pg 061]ancient Celtic Populations of Europe with much ability and success, leans to the opinion that the Cymraeg or Welsh Dialects predominate in these names. But the following examples, which comprise many names derived from the Irish or Gaelic that have not been noticed by Dr. Prichard or by previous writers on this subject, will serve to render it manifest that the ancient Names in Europœa Celtica did, in fact, include all the various living Celtic dialects very equally and harmoniously blended.How luminous and distinct these proofs of the identity of the ancient with the modern Celtic nations are, will be better understood by a preliminary statement of certain rules, which will serve to give greater precision and perspicuity to the illustrations selected:1. There can be no doubt that the Romans, in the Celtic, as in other countries conquered by them, modified the native terms by the addition of their own peculiar grammatical inflections, as in“Judæ-i, Britann-i, Sen-ones,”&c. Now it is obvious that in identifying the Celtic terms we must reject these mere Roman inflections.542. In many cases the Roman Names cannot be supposed to involve complete transcripts of the Celtic Names; frequently they were doubtless convenient abbreviations of the original names—names consisting of descriptive terms to them unintelligible. According to Mr. Reynolds, the Saxons generally adopted the first syllable only of the Roman or British names they found in this island. According to Bullet,“Vic,”a word of Roman origin for a Village or Town, has, from similar causes, become common as a Proper name in Dauphiné; in modern times we have numerous Villages called“Thorpe,”the name for a Village in Anglo-Saxon and[pg 062]German. In instances of this kind, there can be no doubt that originally the names were descriptive, such as“Long-town,”“Old-town,”&c. Tre or Trev is the common Welsh word for a Town, Village, or residence; it had the same meaning in Cornwall:“By Tre, Tres, and Tren,You shall know the Cornish men.”A consequence of the names of the gentry of the county having been derived from those of their residences, into which this word commonly entered!In Wales we have numerous examples of“Tre,”as in“Tre-llwng,”“The Town”of the“Pool,”(i.e. Welshpool,) from an adjoining“Llyn,”or Pool, near Powis Castle;“Tre-lydan,”the Broad Village, or Residence near Welshpool;“Trev-alyn,”near Chester, the Residence on the Stream; the“Alyn,”&c. &c.Now according to the Roman mode, such a term as Trev-alyn would have been changed into Trev-iri, the designation actually given to the Celts of“Treves,”&c.The following are analogous examples:There is a tribe of Brig-antes in Yorkshire, another in Ireland, and a third in the North-east of Spain. Many unsuccessful attempts have been made to show that these distant Celtic tribes must have been scions of the same tribe. A much simpler explanation may be given.By referring to the Roman maps the reader will find a word,“Briga,”in such general use as part of the names of towns as to leave no reasonable doubt that it must have been, like Tre, a Celtic name for a town—now obsolete. Thus in Spain we have, Laco-briga, Meido-briga, Ara-briga, Tala-brica, Augusto-briga, &c. Now the analogous instances already noticed suffice to point out that the occurrence of[pg 063]Brig-antes as a Roman name of Tribes in three Celtic countries, is a natural result of the frequent occurrence of Briga as the first part of the names of Celtic places.The“Allo-bryg-es.”The name of this warlike tribe, the Celtic inhabitants of Savoy, has also been the source of perplexity, which may be removed in the same manner. This tribe had a town, called by the Romans“Brig-icum,”which was said to be“the only one they had.”55Now Allo-Bryga may reasonably be identified with Alpo-Briga, the Town of the Alps (Briga being clearly the common base of“Allo-bryg-es,”and“Brig-icum.”)The names of Celtic communities, as they appear on the Roman Maps, may, I conceive, be proved to have been descriptive of the most prominent natural features of the regions they inhabited, and not of their lineage or descent, as seems to have been often supposed. Thus we have the Mor-ini in Belgium, and the Ar-mor-ici in Gaul on the Sea; we have the Sen-ones on the Seine, the Tamar-ici on the Tamar-is, in Hispania, &c. In the Mountainous regions it will be observed that the names of tribes are derived from the Mountains. In the flat countries they take their names from Rivers or the confluence of Rivers. In the same manner it is highly deserving of remark, that the names of the different French Departments have been derived from precisely the same natural features. Thus in the Hilly countries we have the Departments of the High Alps,“Hautes Alpes;”of the Low Alps,“Basses Alpes;”in the Champaign countries the Departments are named from the Rivers; such as the Seine, the Marne, and the Somme, &c. Many of these French names are literally equivalent to translations of the ancient Gaulish names, as interpreted by means of the Welsh and Irish languages. It is impossible to conceive a[pg 064]more perfect verification of the accuracy of these interpretations!I may here observe, that as far as we can perceive, the various independent communities of Britain and Gaul mentioned by Cæsar, such as the Edui, the Venetes, &c., did not consist ofoneclan or sept, they seem rather to have been a combination of several contiguous septs, to whom no appropriate common namecould have beengiven, except one derived from the natural features of the district they occupied.The durability of local names has been already noticed. Of this truth we possess remarkable proofs in those of localities in France, as preserved by the modern French to the present day. I do not doubt that the present French names are, in many instances, much more faithful transcripts of the original Celtic appellations than those which occur in the Roman Maps are. Thus, for example, Bonomia, a name conferred by the Romans upon Boulogne, and of which the origin has perplexed Antiquarics, may easily be explained as a Roman abbreviation of the word Boulogne itself, of which the Celtic meaning will be shown hereafter to be appropriate and unequivocal. Here it may be noticed, that the Celtic language did not become extinct in Gaul until many centuries after the termination of the Roman sway and the establishment of the Franks in that country. The use of the old Gaulish or Celtic continued until the eighth century, nearly until the time of Charlemagne.56Now we know that the modern Welsh and Irish, for the most part, continue to use their own primitive names of localities in those cases in which abbreviations or translations have been substituted by the English. There can be no reasonable doubt that the ancient Gauls did the same, and that these names were in use among the inhabitants[pg 065]of each locality at the time of the final subjugation of Gaul by the Franks, by whom, in many instances, these names are more likely to have been adopted than those used by the Romans.It will also be observed in the course of the following examples, that names of the class about to be noticed, viz., Topographical names of which the elements are not extant in the existing Celtic dialects, but occur in Oriental words, &c., are remarkably well preserved by the modern French. Thus the“Aube,”as pronounced by the French, is identical in sound with the Asiatic terms for Water, and names of Rivers, to which it is allied.3. By many, perhaps by all those Celtic scholars who have investigated this subject, it has been assumed that the living Celtic dialects may be expected to furnish a complete clue to all the Local Names of ancient Celtic regions. This conclusion, like the theory of Lhuyd above discussed, is founded on an exaggerated idea of the stability of Human Tongues! Neither the Irish nor the Welsh, nor a combination of all the Celtic dialects, will be found to afford a complete solution of the Topographical nomenclature of the ancient Celtic regions of Europe. Names undoubtedly occur in these countries which have been preserved in none of the Celtic tongues, names which I shall indisputably show to be positive transcripts, in many instances, of appropriate terms occurring in the Hebrew and other languages, with which, in other parts of this work, the original Celtic dialects will be proved to have been originally identical. These facts lead to the conclusion that the ancient nomenclature of Celtic countries forms in reality a connecting link between the existing dialects of the Celts and the language of the Oriental stock from which they are descended.This conclusion, though at variance with the views of many estimable writers, is nevertheless in unison with those anticipations[pg 066]which historical facts legitimately suggest. It is only reasonable to infer that since the period of their first arrival in Europe, the era at which many of these names must have been conferred (see page10), the Celtic tribes must have lost many words which none of the modern Celtic nations have preserved. The Celts were settled about the sources of the“Ister, and the city,”(perhaps the mountains)“of Pyrene,”even in the time of Herodotus, and how many ages had elapsed since their first arrival is unknown!57There is a certain Class of terms of which the meaning can reasonably be inferred from their extensive use in combination with other terms, of which the meaning may be considered as ascertained. To this class may be referred the terms immediately following.Catti, Cassii, Casses, or Cad, seem to have meant a People, Tribe, &c., as in the following examples of the names of Celtic Tribes:The Abr-in-Catui, in Normandy. The Catti-euch-lani, the people of Cambridgeshire and the adjoining counties. The Cassii, in Hertfordshire. The Bidu-casses, in Normandy. The Tri-casses, a people in Champagne. The Cad-ur-ci, on the Garonne.The above words seem clearly derivable from the following Welsh words, which are allied to the Hebrew:Welsh.Hebrew.From Kiw-dod (Kiw-dod-æ,plur.) a Clan, a Nation.Gow, a Body of Men, a Society or Association.Kiw-ed, a Multitude, a Tribe.Gowee, a Nation.Kyf, a Body or Trunk, a Pedigree.Gow, Gowe, Goweeth, the Body of a Man or Animal.[pg 067]Tre, Trev, a Village, Town, or Residence, (Welsh,) a Tribe, (Irish.)58Trev-iri, the people of Treves. A-Treb-ates, the people about Arras. (For further examples see Dr. Prichard's work.) Trev is a common element in names of places in Wales, as Tre-vecca, Tre-gynnon.Trigo, to reside, dwell, (Welsh.)Duro-trig-es, the dwellers on the Water or Sea, the people of Dorsetshire. (Camden.)Catt uriges. (See Dour.)Dun-um, a Hill, a Fort or Town, generally on a Hill, (occurs inWelshandIrish.)Oxell-dunum, a Hill-fort in Gaul, described by Cæsar. (See numerous instances in Dr. Prichard's work.)“Castell Din-as Bran,”on a lofty eminence in the Vale of Llangollen, Wales.Dur, Duvr, Awethur (Welsh), Dour (Cornish), Dur (Armorican), Dovar (Irish, obsolete, but occurs in ancient MSS.)“Water.”This word, and Ydōr or Hudōr (Greek), and Tschur (Armenian),“Water,”have an obvious affinity. These forms may be traced in the names of Celtic Localities.“Dour”occurs in the following names of Rivers: Dur, (Hibernia,) Dur-ia Major,“The Doria,”and Duria Minor, (Gallia Cisalpina,) Dur-ius,“The Douro,”and“Dero,”(Hispania,) Dur-anius,“The Dordogne,”(Gallia). InBucharianDeriâ means“The Sea.”Ydōr or Hudōr (Greek), Awethur (Welsh), occur in the[pg 068]Rivers“The Adour,”59Atur-is (Gallia),“The Adder”(Britain),“The Adare”(Ireland.)“Tschur”(Armenian), occurs in“Stura,”(Gallia Cis.),“The Stour”(Britain),“The Suir”(Britain & Ireland),“The Souro”(Spain, a branch of the Tagus.)From the frequent recurrence of all these different forms in several Celtic countries thus widely separated, it is plain that they were used conjointly by the early Celts, and represent various transitions of the same word. Thus“Stura”(inGal. Cis.), flows between the neighbouring streams Duria Major and Duria Minor, &c.This word“Dour”enters very largely into the names of tribes; it forms singly a natural clue to a great number of names that hitherto have been referred to a complication of Roots. Thus the Roman name for the people of Dorsetshire, Duro-trig-es, i.e. The dwellers on the Water or the Ocean, has been noticed by Camden.In the preceding, and in several of the following, it will be apparent that the old Celts applied this term to the“Sea or Ocean,”as the Bucharians do, and also to a“River.”At present the Welsh apply the term to Water only, in a restricted sense.In the South-east of England names abound (applied to places on Rivers or the Sea) in which the two slight variations of Dur and Du-v-r (or Do-v-ar,Irish), still preserved in Welsh, are apparent. Duro-vern-um,“Canterbury,”from Duro, Water, and Vern or Veryn, a Hill. (Compare the name of the“Ar-vern-i,”under Beryn, at p. 78.) The Town was on a Hill by the Stour.Portus Du-b-r-is or Dub-r-œ, i.e.“Sea Port,”the modern[pg 069]“Do-v-or,”a word which is an echo of the Irish Dovar and the Welsh Du-v-r.Duro-brivæ, Rochester on the Medway, (Briva or Brivis, the ancient Celtic for a Town.) Duro-levum, Milton on the Thames.Lan-du-b-r-is, a Portuguese Island. Lan, a Bank of a Stream, or the Sea: also an inclosed Space, (Welsh.)Tur-ones, the inhabitants of the country at the junction of several streams with the Loire, the neighbourhood of the modern Tours.Bi-tur-ig-es, from Bi“Two,”Tur or Dour, Water, and trigo, to reside.There are two tribes of this name in Gaul; the Bituriges Cubi, situated between two of the branches of the Loire, and the Bi-turi-ges Vobisci, between the Garonne and the Sea, at the junction of the Dordogne and the Garonne.Cat-ur-iges, from Catti, Tribes or People; Dour, Water, and Trigo, to reside; on the Durentia, South-east of France, about Embrun or Eburo-Dunum, which was their principal town. Cad-ur-ci, from Catti, Tribes, and dur.There is one tribe of this name on the Dordogne, and another contiguously placed on the Garonne.The mutual support that these interpretations give to each other will be obvious.The following Irish word for“Water,”which is not extant in the Welsh, may be traced in Celtic regions in its various modifications: Uisge (Irish),“The Usk”(South Britain)—Eask60(Irish, obsolete),“The Esk”(Scotland),“The Escaut”(North of France), Isca,“The Exe”(South Britain)—Easkong (Irish, obsolete), Axona (Gallia,Belg.),“The Aisne,”Axones, the neighbouring tribe.[pg 070]Names Of Estuaries, Or Mouths Of Streams.The terms of this class, which occur in ancient Gaul, &c., consist either of terms still thus applied in the living Celtic dialects, or of compounds of which the elements may be recognized, unchanged, in those dialects. Moreover it will be highly interesting to observe that these terms, for the most part, consist of Metaphors derived respectively from the same sources as the two English words“Estuary”and“Mouth,”and the two Latin words“Æstuarium”and“Os Fluminis.”One of the principal arguments of those writers who maintain that the separation of the Irish from the other Celtic tribes must have been of remoter date than the first peopling of these islands, is founded on the fact that the Irish use the word In-ver for the Mouth of a Stream, while the Welsh use Ab-ber (spelt Aber); a feeble support for so wide a conclusion, which a correct analysis of these terms, and a comparison of some interesting coincidences in the local names of ancient Gaul will show to be utterly futile! In-ver and Ab-ber are not simple but compound terms, literally corresponding to the Latin expression“Fluminis Æstuarium.”Æstuarium is from Æstuo,“To boil,”a metaphorical term, obviously derived from the agitation of the Waters where two Streams meet, or where a River enters the Sea.In the first syllable“Inver”and“Ab-ber”differ, but they agree in the last. Both“In”and“Ab,”the first syllables of these terms, occur so often in Celtic regions that there can be no doubt they were both in use among the ancient Celts as words for a River, or Water. The last syllable of these words, Ber or Ver, I shall show to mean an“Estuary.”“In”occurs in the name of“The Inn,”in the Tyrol, the“Æn-us”of the Romans, and in other instances previously[pg 071]noticed.“An”is a Gaelic or Irish term for“Water,”which is identical in sound and sense with terms of frequent occurrence among the tribes of the American Continent, as in Aouin (Hurons, N. America), Jin Jin (Kolushians, extreme North-west ofN. America), Ueni (Maipurians, S. America.)“Ab”occurs in“The Aube,”in France, &c., a name of which the pronunciation may be considered identical with Ab,“Water,”(Persian.) Ap in Sanscrit, and Ubu Obe in Affghan, mean“Water.”“Obe”occurs in Siberia as the name of a well-known river. In India also the term has been applied to“Rivers;”thus we have in that country the Punj-âb, (the Province of“The Five Rivers,”) an appellation of which the corresponding Celtic terms“Pump-ab”would be almost an echo!Further it may here be noticed—as an example of the complete identity of the Celtic and Oriental languages when all the“Disjecta Membra”are compared—that this word does not exist in the modern Celtic in the simple form of Ab, but in the derivative form of Avon, which is found in the Roman maps spelt“Abon,”&c. Now this form also occurs in the East. Abinn,“A River,”is given by Klaproth from the language of the inhabitants of the Mountains to the North of Bhagalpur. Apem means“Water,”in Zend, an ancient Persian dialect. Af is“Water,”in Kurdish.

“But the oldest remains of the Danish language are to be found on our Runic stone monuments, and here at length it perfectly coincides with the earliest Swedish, Norwegian, and Icelandic.“The Danish is closely allied to the Swedish, and both, in the earliest times, lapse into the Icelandic, which, according to all ancient records, was formerly universal over all the North, and must therefore be considered as the parent of both the modern Scandinavian dialects.”45On the subject of the differences of dialect in the differentprovincesof the Northern Kingdoms he says that,“In Norway as well as in Denmark one province terminates its verbs[pg 045]ina, another distinguishes allthe three genders, while a third has preserved a vast number ofoldwordsand inflectionswhich to the othersare unintelligible.”We have thus a proof that even in the provinces of the same kingdom there are differences of“words, grammar,and inflections.”The differencein the numberofgendersis a very remarkable one.The researches of Professor Rask will be found distinctly to warrant the following conclusions. These conclusions are in the nature of results that legitimately flow from his researches; they do not represent the inferences which he himself has thence deduced. With regard both to the languages of England and of his native Scandinavia, this learned writer seems evidently to have been perplexed by the extent and variety of the changes he has described. Hence, in both instances, he has shown an inclination to ascribe to the influence of War and Social disturbance changes which his own researches clearly prove to have been the effects neither of transient nor of local influences, but of causes progressively at work through a series of ages, and embracing large groups of nations and languages in their action.1. The differences which now exist between the various Scandinavian Languages extend to all those features in which it is possible that one Language, or one Class of Languages, can differ from another; viz. to Words, Grammar,Inflections,46and to thearrangementof Words in sentences,47orIdioms.2. Not only do differences of this nature present themselves in the various ScandinavianKingdoms—but also in the various[pg 046]Provinces of the sameKingdom, which in many instances are distinguished by the most marked differences in Words, Grammar, &c. Thus the Dialect of Dalecarlia in Sweden is very ancient and distinct, and approaches to the Gothic.483. These characteristic features of the various languages and dialects of Scandinavia have arisen progressively during the course of ages.4. These differences principally consist in the abandonment in one Kingdom or Province of a portion of the Words, Idioms, Grammar, &c. of the Parent Speech—that part of the elements of the Original Tongue which have become obsolete in one dialect having generally been preserved in the dialects of other kingdoms and provinces—which have at the same time generally lost other distinct portions of the Vocabulary, Grammar, &c. of their common Original. In other words, the“Disjecta Membra”of the old Scandinavian, or“Danska Tunge,”when not preserved in the Danish, have been retained for the most part in the Swedish, Icelandic, and Norwegian, or in some of the Provincial dialects of Scandinavia, and vice versâ.In the various provinces in which it was once spoken different portions of the Parent speech have been abandoned or preserved.5. Hence it follows that the Primitive Language of Scandinavia, or“Danska Tunge,”does not exist in any one—but is dispersed inallits derivative dialects. (Compare the motto from Grotius on the title-page.)6. It is a necessary consequence of the third and fourth propositions that the more ancient remains of the derivative dialects approach more nearly to the Parent Speech, and—in the ratio of their superior antiquity—unite a greater proportion of the distinctive peculiarities of all the sister-dialects,[pg 047]which, as previously stated, have arisen in consequence of certain portions of the Parent speech having been abandoned in some provinces and retained in others, and vice versâ.An interesting illustration of this maxim occurs in a passage from Professor Rask's preface already quoted, in which, after giving a specimen of old Danish, which approaches closely to the Icelandic, he adds,“The few deviations from the Icelandic bear for the most parta strong resemblanceto theSwedish.”In other words, the older specimens of the Danishunitethose peculiarities by which the modern collateral Tongues of Iceland, Denmark, and Sweden are distinguished from each other.Let it be borne in mind, that the lapse of one thousand years has produced these changes, and the instructive nature of this example will be fully apparent. Of the accuracy of the data on which the previous deductions rest, all doubt must be removed by reference to one remarkable event. It is historically certain that the Island of Iceland is inhabited by a nation descended from emigrants from the opposite Norwegian coast. It is historically certain, also, that previously to the Ninth Century these warlike adventurers had not established themselves on the Icelandic soil. Anterior to that period, therefore, it is self-evident that, inasmuch as the Icelanders had no existence as a nation, the Icelandic Tongue could not have had a separate existence as a language. Yet it is certain that in the present day the Icelandic deviates at least as widely from the language of the adjoining Norwegian Coasts as that language deviates from the other Scandinavian Tongues.The evidence furnished by Professor Rask and the writers whose views he has combated, will be found, when fairly balanced, distinctly to support a very important Conclusion, contemplated by neither. The facts adduced on both sides conspire to show a rapid approximation of the Teutonic and[pg 048]Scandinavian branches of the Gothic as we ascend into remote ages.Of this approximation, the features of identity between the Anglo-Saxon and the Icelandic, pointed out by the writers whose views Professor Rask combats, furnish a reasonable presumption, which is converted into positive proof by the evidence collected by Professor Rask himself, that the same features occur in all the ancient, though they do not in the modern, specimens of the Languages of the Scandinavian Peninsulas. It is true, this learned writer, of whose researches I have chiefly availed myself in this Section, maintains that there are some features in which all the Scandinavian differ from the Anglo-Saxon and other Teutonic Dialects, a conclusion, however, but feebly supported by the examples he has adduced, and scarcely reconcilable in any way with the resemblance which the primitive Swedish dialect of Dalecarlia is said to bear to the Gothic. But, assuming the occurrence of some features of difference, even in the earliest specimens we possess, this assumption leaves untouched the proposition that these specimens show a rapid rate of approximation, which, if equally rapid prior to their date, implies that at an era not many ages anterior the identity of the languages of Germany and Scandinavia must have been complete.[pg 049]Section V.The Origin of the Irish Nation. The original Language of the British Isles was a Union of Welsh and Irish. Union of the Irish, Welsh, &c. in the ancient Local Names in the Celtic Countries of Gaul, &c. These Names a connecting Link between the existing Celtic Dialects and the Oriental, Greek, and other Languages, &c.The origin of the Irish nation, or Gael, forms—for numerous reasons—a highly interesting and important subject of inquiry. Of this Nation the very same theories have been maintained as those which have been adopted in some quarters with respect to the North American Indians, the Negroes, and other branches of the Human Family; viz., that they are of a stock aboriginally inferior and distinct, by nature incapable of the virtues of civilization. Let the views advocated by Pinkerton with respect to the Gaelic race—views received with no slight degree of favour in his time—be compared with the doctrines of many modern writers on the subject of the native African and American Races, and an instructive lesson will be learnt on the force of prejudice and the uniformity of error!On the other hand, it must be allowed that the opinions which have been generally espoused on the subject of the origin of the Gael by many of the Historians and Scholars of Ireland and of the Highlands of Scotland, can scarcely be said to possess a better claim to the approbation of a calm and dispassionate judgment. Eminently distinguished as the Irish are by Literary genius, there is probably no subject on which their native talent has appeared to less advantage than in the investigation of the early History of their own[pg 050]Country. Fictions the most extravagant, borrowed from the Chronicles of the dark ages, have been credulously adopted by their first Scholars in lieu of those solid truths to which a calm and sober inquiry alone can lead. Thus we find Mr. Moore, at once the Poet and the Historian of Ireland, lending the sanction of his name to the Fable that the Irish are of Spanish origin; and citing, in answer to the more reasonable hypothesis of a British origin, a variety of Irish writers of no mean note, and some Welsh writers also, in favour of the assertions: 1, that the Irish Language is almost totally unlike the Welsh or Ancient British; and 2, that the Welsh is not a Celtic but a Gothic Tongue! There is every reason to conclude that Mr. Moore—unacquainted, probably, with any of the Celtic dialects himself—resorted to those authorities which he might naturally have deemed most deserving of confidence. But this only renders more lamentably conspicuous the credulity, carelessness, and ignorance of those to whose labours he has appealed. The assertions, 1, that the Welsh and Irish are unlike; and 2, that the Welsh is a Gothic dialect, are contradictions of the plainest facts.Influenced by national feelings Gaelic Scholars have also advanced various other theories, calculated to exhibit the antiquity of their language and race in a favorable point of view. The Gaelic has been maintained to be the Parent, at least in part, of the Latin, the Welsh, &c.; while to the first Colonists of Ireland a Carthaginian or Phœnician origin has been assigned.These conclusions cannot be sustained. But it is highly probable, notwithstanding, that the proofs on which they have been based will be found, in many instances, to contain the germs of important truths, though blended with an admixture of error. The traces of affinity between the Irish[pg 051]and other ancient languages which have been collected by Gaelic Scholars, may be open in many cases to the same remark, which is clearly applicable to the examples of affinity pointed out by Mr. Catlin between the dialect of the North American Indian tribe the Mandans and the Welsh; viz., these features may consist of clear and genuine traces of a generic, though they may afford no proofs of a specific, affinity of race. There can be no doubt that the Irish preserves many primitive forms which the kindred Celtic of Wales has lost; there can be no doubt also that the Irish approximates to the Latin, to the Greek, and to the Egyptian,49&c. in many features which the Welsh no longer exhibits. The examples adduced in Appendix A of the connexion of the Irish language with the Hebrew, Egyptian, &c. are sufficient to show that the Irish are a nation of Oriental origin. But on the other hand it must be borne in mind, that inasmuch as the Welsh, Latin, &c., have also preserved primitive forms which the Irish has lost, there is no ground for concluding that the Gaelic is a Parent rather than a Sister of these venerable Tongues; and inasmuch as the evidence of the Eastern origin of the Gael, however unequivocal, is not clearer or closer than the accompanying50evidence with respect to the Welsh, English, and other European nations, there are no peculiar grounds for referring the first colonization of Ireland to a direct migration from the shores of Palestine or Africa, rather than to the gradual diffusion of population from a central point.The following comparison presents examples of features in which the Irish approximates to the Gothic and other Languages, at the same time that it differs more or less from the Welsh.[pg 052]Words in which the Gaelic resembles the Gothic, and other European Languages, more closely than it resembles the Cymraeg or Welsh.English.Gaelic.Illustrations.Cymraeg.1. Father.Ath-air, (Ir.)Atta, (Gothic.), Ayta, Aydia, (Basque.), Attia, (Hung.), Otek, (Russ.), Fader, slightly varied in all the Gothic dialects, except the Gothic properly so called, Pater, (Greek & Latin.)Tad, (W.)2. Mother.Math-air, (Ir.)Mater or Mutter (with some trifling variations) in Latin, Greek, and all the Teuto-Scandinavian dialects except the Gothic—also in the Sclavonic and Bohemian. Ath-ei, (Gothic.)Mymmog, (Manx dialect.Mam, (W.)3. Brother.Brathair, (Ir.)The Irish form,Brathair,occurs in the Latin and Teuto-Scandinav. tongues; the Welsh form,Brawd,in the Sclavonian tongues.Brawd, (W.), Bredar, (Cornish.)Breur, (Manx dialectBreur, (Arm.)4. Sister.|Siur, (Ir.)The Irish form prevails in the Latin, Teuto-Scand. and Sclavonic.Chwaer, (W.)Piur, (Scotch.)Hor, Huyr, (Cornish.)5. A Company.Drong, (Ir.)Drang, a Throng, a Crowd, (German.)Torv.6. Mock.Magom, (Ir.)Mock, (English.)Gwatwor, (W.)7. Evil.Neoid, (Ir.)Naughty, (Eng.)Droug, (W.)Olk, (Ir.)Ill, (Eng.)8. The Bank of a stream.Rang, (Ir.)Rand,51(Germ.)Glan, (W.)9. A Step.Beim, (Ir.)Bēm-a, a Step, (Greek.), Bain-o, to go, Bahn, a Path, (Germ.)Cam.10. To bear.Beir-im, (Ir.)Fero, (Latin.) Ge-Bähr-en, (Germ.)Dwyn.11. Jeering, Delight, A Desire.Fon-amhad (Ir.), Foun, (Ir.)Fun, (Eng.), Vonne, Delight, (Germ.), Vunsch, a Wish, (Germ.)Vynn, or Mynn, a Wish, (W.)12. A Woman.Geon, (Ir.)Cwen, (Ang.-Sax. & Icel.)Gen-eth, a Girl, (W.)13. To know.Fis-ay-im, Fod-am, (Ir.)Viss-en, (Germ.), Vit-an, (Ang.-Sax.),“I wot,”(Eng.)Wys, or Gwys, Wyth, or Gwyth, Knowledge (W.)14. To heat, or warm.Gorm, (Ir.)Warm, (Eng.)Gwresogi, (W.)15. A Shadow.Sgath, (Ir.)Skia, Skiad-on, (Greek.), Schatten, (Germ.)Cysgod, (W.)16. To speak.|Raid-him, (Ir.)Read-en, (Germ.)Siarad, (W.)Some of these examples would furnish a more plausible argument to show that the Irish are a Gothic race than any which have been advanced to prove that the Welsh are of Gothic origin! It is singular, for instance, that the Irish terms expressive of the Domestic relations are so near the English as to excite in the first instance a suspicion that they must have been borrowed from the followers of Strongbow! But this impression must be dispelled by the reflection that terms of this class are never borrowed from its conquerors by a nation that continues to retain its primitive language. Moreover, it will be observed, that the Irish, in the instance of these words, approaches much more nearly to the Gothic,[pg 054]Hungarian, and Russian, &c. than it does to the English. Again, the Irish word“Gorm,”To heat or warm, islikethe English“Warm.”But, on the other hand, its genuineness is rendered indisputable by itsabsolute identitywith the word 'Gorm' in Persian and Egyptian, (See Appendix A, p.21.) Finally, the resemblances manifested above by the Irish to the Greek are quite as close as those which the former language displays to the English and other Gothic Tongues. In these examples, therefore, we may recognize proofs not of any partial results or specific connexions, but of the more complete approximation of the European languages as we enlarge our range of inquiry, and obtain more ample specimens of each Class.But, notwithstanding the occurrence of some features of difference, it is indisputable that there exists a close specific affinity between the Irish and Welsh Languages, which renders the common origin of the nations who speak them evident. The original identity of the Irish and Welsh Languages was established as far back as the commencement of the eighteenth century, by the investigations of the excellent Archæologist, Edward Lhuyd, who spent five years in travelling through the various Celtic regions, and whose comparison of the dialects of Wales, Cornwall, Armorica, the Highlands of Scotland, and the Isle of Man, is not inferior either in soundness of reasoning, or in patient, extensive, and honest research, to the best German works of the present day. But although the writings of Lhuyd may be said to have established the original unity of the Welsh and Irish races, since the publication of his work, a peculiar opinion has been adopted by some learned men with regard to thetimeof their original separation. Of this opinion, Edward Lhuyd was himself the first advocate; his conclusion was that though the Irish and British Celts were both descendants from one stock, they must have been separated into two[pg 055]distinct Tribes before their arrival in the British Islands. The Gaelic or Irish Tribe he supposes to have preceded the Welsh or British Tribe, by whom he conceives them to have been gradually driven to the West, as the Britons were by the Saxons in subsequent ages. Lhuyd's grounds are as follows:The most ancient names of Rivers and Mountains in the Island of Britain are very generally composed of terms still preserved in the Welsh or Ancient British Tongue. But there are some remarkable exceptions, and in these instances it frequently happens that the Names may be clearly identified with Words still preserved in the Irish or Gaelic branch of the Celtic. For example, the names of the British rivers, the Usk and the Esk, are particularly noticed by Lhuyd; these names are identical with“Uisge, Eask,”the Irish term for“Water.”This word, he observes, does not exist in the Welsh, and he had looked for it in vain in the sister dialect of Armorica; but, he adds, it is still retained by the Irish or Gaelic. Hence, he suggests that the Irish or Gaelic branch of the Celts must have colonized the Island of Britain before the arrival of the Cymry or Welsh branch, by whom, as he conceives, they were expelled, after having conferred names on the principal localities.The evidence of language will be found sufficient to show not merely the common origin of the Welsh and Irish, but also to fix a much more recent date for their separation than that which has been assigned by Lhuyd. It will thence appear that the Irish are descendants of Colonists of the Welsh or British race, not of a distinct Celtic sept, and that the commencement of the separate existence of the Irish nation must be referred to a comparatively recent date, propositions of much interest, of which the proofs about to be advanced will probably be deemed to be at once clear and simple.Lhuyd's reasoning in favour of his theory, that the Irish or Gael existed in Britain as a separate Tribe, prior to the arrival[pg 056]of the Britons who fought against Cæsar, the ancestors of the modern Welsh, is founded on a false analogy not unnatural to a first inquirer.The proposition that the most important local names in every country for the most part consist of terms belonging to the language of the very first inhabitants, is one of which I conceive the truth will be evident. For a proof of this principle, I may refer to Chalmers'52admirable analysis of local names in the Lowlands of Scotland, where, in spite of a succession of Conquests, and the utter extinction in that part of Britain of the language of the original inhabitants, viewed as a vernacular dialect, Welsh and other Celtic names are still preserved, after the lapse of ages, for the most prominent features of the country. This result, it may be observed, is one that flows from the very nature of things. Even the most fierce and ruthless invaders are compelled to hold sufficient intercourse with the first population to enable them to learn the proper names of their localities, and these names, from obvious motives of convenience, they almost universally adopt.Now, had Lhuyd shown that the most ancient Local names in Britain areexclusivelyIrish, there can be no doubt that, consistently with the principle just noticed, his theory would have been supported by the facts to which he adverts. But the most ancient local names in Britain are notexclusivelyorprincipallyIrish; in an equal number, perhaps in a majority, of cases they are Welsh.Moreover, it may be observed that the names of localities in this Island furnish highly instructive evidence, not merely with respect to the different races by whom it has been successively peopled, but also of the order in which they arrived. Thus the names of Rivers and Mountains, and other natural[pg 057]objects, at least of the most conspicuous, are Celtic; the names of the most ancient Towns are Latin, or Latin grafted on British words; more modern Towns and Villages have Saxon appellations; those of more recent origin have frequently Norman designations; and last of all come those places which have names derived from our present English. These various classes of names cannot be nicely distinguished in each particular instance. Of the correctness of the general principle, however, there is no doubt.But the terms noticed by Lhuyd as significant in the Irish language do not belong to a different class of appellations from those which are obviously of British or Cymraeg origin. The Irish and Cymraeg terms are both found to predominate most in the names of the most ancient Class, viz. in those of Rivers, Mountains, &c., and to be thus applied in conjunction. Hence the natural inference that flows from his facts is not that these names were conferred by two distinct and successive races, but that they were imposedcontemporaneouslyand by the same People!Further it may be noticed, that if British Topography presents words extant only in the Irish Tongue, Irish Topography also presents names which cannot be explained by means of the Irish, though their meaning is preserved in Welsh; for example: There is a place near the head of a Stream in Roscommon, called“Glan a Modda,”(from Glan,“The bank of a Stream,”Welsh.) There is a place in Wales, called“Glan a Mowdduy.”There is a place called“Glan-gora,”in a Creek at the head of Bantry Bay; and another place in Ireland called“Glan-gort.”“Ben-heder,”the ancient Irish name for“The Hill of Howth,”interpreted by Mr. Moore“The Hill of Birds.”(Adar,“Birds,”Welsh. The word does not exist in Irish.)Arran, A mountainous Island. (Arran, a Mountain,Welsh. This word does not exist in Irish,) &c. &c.[pg 058]Mr. Chalmers in his Caledonia states that the prevalent ancient names of localities in Britain and Ireland are essentially the same.The conclusions to which these facts legitimately and necessarily lead are, that the British Islands were originally colonized by Settlers, who, at the time of the first occupation of Great Britain and Ireland, spoke one uniform language, in which the Welsh, Irish, and other living Celtic Dialects were combined. We may infer, and I conceive most clearly, that these dialects must be viewed in the light of“Disjecta Membra”of the speech of the old British and Irish Celts, just as the Icelandic, Norwegian, &c. are fragments of the ancient“Danska Tunge,”as noticed in the previous section.It has been shown by Dr. Prichard that the population of Islands has been derived from the neighbouring Continents, and that the population of the more distant Islands has been derived in like manner from those which are nearer to the common source of migration. It is highly unreasonable to assume that Ireland has formed an exception to this general rule, considering that the common basis of the Irish and ancient British or Welsh languages are confessedly the same, unless it can be proved that the accompanying differences are such as to require the solution Lhuyd has suggested. Here, then, the question arises, are the features of difference between the Welsh and Irish languages more numerous or more fundamental, in relation to the interval of time that has elapsed since the Roman Invasion of Britain, than the varieties of dialect among the Scandinavian nations are in relation to the period that has elapsed since the colonization of Iceland? They are not! It will thence be seen that Lhuyd's theory, as to the remote date of the separation of the Gaelic or Irish from the British or Cymraeg branch of the Celts, is founded on an exaggerated conception[pg 059]of the stability of Human Tongues; and that the abandonment by various septs of different synonymes used conjointly by their common forefathers will satisfactorily account for the differences between the Welsh and Irish, to which he attaches so much weight. It will be perceived, for example, that in the Icelandic, of which the existence commenced in the ninth century, and the Continental Scandinavian from which it then sprang, totally different terms are used for“Water,”the very instance to which Lhuyd especially adverts, as regards the languages of the Welsh and Irish, whom we know to have existed as separate nations in the time of Cæsar eighteen centuries ago!Another highly instructive test of the correctness of his theory may be derived from the investigations of Lhuyd himself, who, in his comparison of the Welsh and Irish languages, uniformly distinguished the current terms from the obsolete synonymous words that occur only in ancient MSS. This comparison proves distinctly that the Irish and Welsh languages approximate, as we ascend, at a rate which, if as rapid previously as we know it to have been up to the date of the earliest MSS., would imply that these languages must have been identical about the era of the Roman invasion. As the changes which languages undergo in their infancy are more rapid than those which occur at later stages of their growth, it is possible that the unity of these Tongues may be ascribed even to a much later period, an opinion which has been maintained by a very judicious and excellent writer, Mr. Edward Davies, who in his“Claims of Ossian”has published an early specimen of Irish Poetry, which in Language and Style he regards as identical with the most ancient productions of the Welsh Bards. Making every allowance for the irregularity of the changes which occur in Languages, I do not conceive it possible that the Welsh and Irish could have differed very essentially[pg 060]in the time of Cæsar. This leads directly to another conclusion, viz. that the first colonization of Ireland could not have taken place a great many centuries before the Roman invasion. Had such been the case, the differences between the Welsh and Irish Languages must have been proportionately more extensive. In the time of the Romans we learn that an Irish traitor arrived in Britain, who stated that Ireland might be kept in subjection by a single legion, an incident which tends, however slightly, to favour the opinion that the sister Island was at that period but thinly, perhaps because but recently, peopled.Of the extent of the changes which the Celtic languages have undergone since the first arrival of the Celts in Europe, we possess proofs of far more ancient date than the earliest literary specimens of the living dialects of the Celtic in the Local names of Celtic regions, as preserved in Roman Maps, and in the existing languages of the French, English, and other nations, who occupy countries of which the Celts were the first inhabitants. These names I shall show to consist of three elements: A union of 1, Welsh, Cornish, &c.; 2, Irish, Highland Scotch, &c.; and 3, Terms not extant in any Celtic Tongue, but preserved in the Oriental, Greek, and other languages.As regards the Names of the 1st and 2d Classes, it will abundantly appear from the ensuing examples that, in the Topographical Nomenclature of Gaul, Britain, and other Celtic regions of Europe,53words derived from all the various Celtic dialects now extant, occur in a manner that leads distinctly to the inference that these“Disjecta membra”must have simultaneously belonged to the language of the old Celts. Dr. Prichard, who has examined these vestiges of the[pg 061]ancient Celtic Populations of Europe with much ability and success, leans to the opinion that the Cymraeg or Welsh Dialects predominate in these names. But the following examples, which comprise many names derived from the Irish or Gaelic that have not been noticed by Dr. Prichard or by previous writers on this subject, will serve to render it manifest that the ancient Names in Europœa Celtica did, in fact, include all the various living Celtic dialects very equally and harmoniously blended.How luminous and distinct these proofs of the identity of the ancient with the modern Celtic nations are, will be better understood by a preliminary statement of certain rules, which will serve to give greater precision and perspicuity to the illustrations selected:1. There can be no doubt that the Romans, in the Celtic, as in other countries conquered by them, modified the native terms by the addition of their own peculiar grammatical inflections, as in“Judæ-i, Britann-i, Sen-ones,”&c. Now it is obvious that in identifying the Celtic terms we must reject these mere Roman inflections.542. In many cases the Roman Names cannot be supposed to involve complete transcripts of the Celtic Names; frequently they were doubtless convenient abbreviations of the original names—names consisting of descriptive terms to them unintelligible. According to Mr. Reynolds, the Saxons generally adopted the first syllable only of the Roman or British names they found in this island. According to Bullet,“Vic,”a word of Roman origin for a Village or Town, has, from similar causes, become common as a Proper name in Dauphiné; in modern times we have numerous Villages called“Thorpe,”the name for a Village in Anglo-Saxon and[pg 062]German. In instances of this kind, there can be no doubt that originally the names were descriptive, such as“Long-town,”“Old-town,”&c. Tre or Trev is the common Welsh word for a Town, Village, or residence; it had the same meaning in Cornwall:“By Tre, Tres, and Tren,You shall know the Cornish men.”A consequence of the names of the gentry of the county having been derived from those of their residences, into which this word commonly entered!In Wales we have numerous examples of“Tre,”as in“Tre-llwng,”“The Town”of the“Pool,”(i.e. Welshpool,) from an adjoining“Llyn,”or Pool, near Powis Castle;“Tre-lydan,”the Broad Village, or Residence near Welshpool;“Trev-alyn,”near Chester, the Residence on the Stream; the“Alyn,”&c. &c.Now according to the Roman mode, such a term as Trev-alyn would have been changed into Trev-iri, the designation actually given to the Celts of“Treves,”&c.The following are analogous examples:There is a tribe of Brig-antes in Yorkshire, another in Ireland, and a third in the North-east of Spain. Many unsuccessful attempts have been made to show that these distant Celtic tribes must have been scions of the same tribe. A much simpler explanation may be given.By referring to the Roman maps the reader will find a word,“Briga,”in such general use as part of the names of towns as to leave no reasonable doubt that it must have been, like Tre, a Celtic name for a town—now obsolete. Thus in Spain we have, Laco-briga, Meido-briga, Ara-briga, Tala-brica, Augusto-briga, &c. Now the analogous instances already noticed suffice to point out that the occurrence of[pg 063]Brig-antes as a Roman name of Tribes in three Celtic countries, is a natural result of the frequent occurrence of Briga as the first part of the names of Celtic places.The“Allo-bryg-es.”The name of this warlike tribe, the Celtic inhabitants of Savoy, has also been the source of perplexity, which may be removed in the same manner. This tribe had a town, called by the Romans“Brig-icum,”which was said to be“the only one they had.”55Now Allo-Bryga may reasonably be identified with Alpo-Briga, the Town of the Alps (Briga being clearly the common base of“Allo-bryg-es,”and“Brig-icum.”)The names of Celtic communities, as they appear on the Roman Maps, may, I conceive, be proved to have been descriptive of the most prominent natural features of the regions they inhabited, and not of their lineage or descent, as seems to have been often supposed. Thus we have the Mor-ini in Belgium, and the Ar-mor-ici in Gaul on the Sea; we have the Sen-ones on the Seine, the Tamar-ici on the Tamar-is, in Hispania, &c. In the Mountainous regions it will be observed that the names of tribes are derived from the Mountains. In the flat countries they take their names from Rivers or the confluence of Rivers. In the same manner it is highly deserving of remark, that the names of the different French Departments have been derived from precisely the same natural features. Thus in the Hilly countries we have the Departments of the High Alps,“Hautes Alpes;”of the Low Alps,“Basses Alpes;”in the Champaign countries the Departments are named from the Rivers; such as the Seine, the Marne, and the Somme, &c. Many of these French names are literally equivalent to translations of the ancient Gaulish names, as interpreted by means of the Welsh and Irish languages. It is impossible to conceive a[pg 064]more perfect verification of the accuracy of these interpretations!I may here observe, that as far as we can perceive, the various independent communities of Britain and Gaul mentioned by Cæsar, such as the Edui, the Venetes, &c., did not consist ofoneclan or sept, they seem rather to have been a combination of several contiguous septs, to whom no appropriate common namecould have beengiven, except one derived from the natural features of the district they occupied.The durability of local names has been already noticed. Of this truth we possess remarkable proofs in those of localities in France, as preserved by the modern French to the present day. I do not doubt that the present French names are, in many instances, much more faithful transcripts of the original Celtic appellations than those which occur in the Roman Maps are. Thus, for example, Bonomia, a name conferred by the Romans upon Boulogne, and of which the origin has perplexed Antiquarics, may easily be explained as a Roman abbreviation of the word Boulogne itself, of which the Celtic meaning will be shown hereafter to be appropriate and unequivocal. Here it may be noticed, that the Celtic language did not become extinct in Gaul until many centuries after the termination of the Roman sway and the establishment of the Franks in that country. The use of the old Gaulish or Celtic continued until the eighth century, nearly until the time of Charlemagne.56Now we know that the modern Welsh and Irish, for the most part, continue to use their own primitive names of localities in those cases in which abbreviations or translations have been substituted by the English. There can be no reasonable doubt that the ancient Gauls did the same, and that these names were in use among the inhabitants[pg 065]of each locality at the time of the final subjugation of Gaul by the Franks, by whom, in many instances, these names are more likely to have been adopted than those used by the Romans.It will also be observed in the course of the following examples, that names of the class about to be noticed, viz., Topographical names of which the elements are not extant in the existing Celtic dialects, but occur in Oriental words, &c., are remarkably well preserved by the modern French. Thus the“Aube,”as pronounced by the French, is identical in sound with the Asiatic terms for Water, and names of Rivers, to which it is allied.3. By many, perhaps by all those Celtic scholars who have investigated this subject, it has been assumed that the living Celtic dialects may be expected to furnish a complete clue to all the Local Names of ancient Celtic regions. This conclusion, like the theory of Lhuyd above discussed, is founded on an exaggerated idea of the stability of Human Tongues! Neither the Irish nor the Welsh, nor a combination of all the Celtic dialects, will be found to afford a complete solution of the Topographical nomenclature of the ancient Celtic regions of Europe. Names undoubtedly occur in these countries which have been preserved in none of the Celtic tongues, names which I shall indisputably show to be positive transcripts, in many instances, of appropriate terms occurring in the Hebrew and other languages, with which, in other parts of this work, the original Celtic dialects will be proved to have been originally identical. These facts lead to the conclusion that the ancient nomenclature of Celtic countries forms in reality a connecting link between the existing dialects of the Celts and the language of the Oriental stock from which they are descended.This conclusion, though at variance with the views of many estimable writers, is nevertheless in unison with those anticipations[pg 066]which historical facts legitimately suggest. It is only reasonable to infer that since the period of their first arrival in Europe, the era at which many of these names must have been conferred (see page10), the Celtic tribes must have lost many words which none of the modern Celtic nations have preserved. The Celts were settled about the sources of the“Ister, and the city,”(perhaps the mountains)“of Pyrene,”even in the time of Herodotus, and how many ages had elapsed since their first arrival is unknown!57There is a certain Class of terms of which the meaning can reasonably be inferred from their extensive use in combination with other terms, of which the meaning may be considered as ascertained. To this class may be referred the terms immediately following.Catti, Cassii, Casses, or Cad, seem to have meant a People, Tribe, &c., as in the following examples of the names of Celtic Tribes:The Abr-in-Catui, in Normandy. The Catti-euch-lani, the people of Cambridgeshire and the adjoining counties. The Cassii, in Hertfordshire. The Bidu-casses, in Normandy. The Tri-casses, a people in Champagne. The Cad-ur-ci, on the Garonne.The above words seem clearly derivable from the following Welsh words, which are allied to the Hebrew:Welsh.Hebrew.From Kiw-dod (Kiw-dod-æ,plur.) a Clan, a Nation.Gow, a Body of Men, a Society or Association.Kiw-ed, a Multitude, a Tribe.Gowee, a Nation.Kyf, a Body or Trunk, a Pedigree.Gow, Gowe, Goweeth, the Body of a Man or Animal.[pg 067]Tre, Trev, a Village, Town, or Residence, (Welsh,) a Tribe, (Irish.)58Trev-iri, the people of Treves. A-Treb-ates, the people about Arras. (For further examples see Dr. Prichard's work.) Trev is a common element in names of places in Wales, as Tre-vecca, Tre-gynnon.Trigo, to reside, dwell, (Welsh.)Duro-trig-es, the dwellers on the Water or Sea, the people of Dorsetshire. (Camden.)Catt uriges. (See Dour.)Dun-um, a Hill, a Fort or Town, generally on a Hill, (occurs inWelshandIrish.)Oxell-dunum, a Hill-fort in Gaul, described by Cæsar. (See numerous instances in Dr. Prichard's work.)“Castell Din-as Bran,”on a lofty eminence in the Vale of Llangollen, Wales.Dur, Duvr, Awethur (Welsh), Dour (Cornish), Dur (Armorican), Dovar (Irish, obsolete, but occurs in ancient MSS.)“Water.”This word, and Ydōr or Hudōr (Greek), and Tschur (Armenian),“Water,”have an obvious affinity. These forms may be traced in the names of Celtic Localities.“Dour”occurs in the following names of Rivers: Dur, (Hibernia,) Dur-ia Major,“The Doria,”and Duria Minor, (Gallia Cisalpina,) Dur-ius,“The Douro,”and“Dero,”(Hispania,) Dur-anius,“The Dordogne,”(Gallia). InBucharianDeriâ means“The Sea.”Ydōr or Hudōr (Greek), Awethur (Welsh), occur in the[pg 068]Rivers“The Adour,”59Atur-is (Gallia),“The Adder”(Britain),“The Adare”(Ireland.)“Tschur”(Armenian), occurs in“Stura,”(Gallia Cis.),“The Stour”(Britain),“The Suir”(Britain & Ireland),“The Souro”(Spain, a branch of the Tagus.)From the frequent recurrence of all these different forms in several Celtic countries thus widely separated, it is plain that they were used conjointly by the early Celts, and represent various transitions of the same word. Thus“Stura”(inGal. Cis.), flows between the neighbouring streams Duria Major and Duria Minor, &c.This word“Dour”enters very largely into the names of tribes; it forms singly a natural clue to a great number of names that hitherto have been referred to a complication of Roots. Thus the Roman name for the people of Dorsetshire, Duro-trig-es, i.e. The dwellers on the Water or the Ocean, has been noticed by Camden.In the preceding, and in several of the following, it will be apparent that the old Celts applied this term to the“Sea or Ocean,”as the Bucharians do, and also to a“River.”At present the Welsh apply the term to Water only, in a restricted sense.In the South-east of England names abound (applied to places on Rivers or the Sea) in which the two slight variations of Dur and Du-v-r (or Do-v-ar,Irish), still preserved in Welsh, are apparent. Duro-vern-um,“Canterbury,”from Duro, Water, and Vern or Veryn, a Hill. (Compare the name of the“Ar-vern-i,”under Beryn, at p. 78.) The Town was on a Hill by the Stour.Portus Du-b-r-is or Dub-r-œ, i.e.“Sea Port,”the modern[pg 069]“Do-v-or,”a word which is an echo of the Irish Dovar and the Welsh Du-v-r.Duro-brivæ, Rochester on the Medway, (Briva or Brivis, the ancient Celtic for a Town.) Duro-levum, Milton on the Thames.Lan-du-b-r-is, a Portuguese Island. Lan, a Bank of a Stream, or the Sea: also an inclosed Space, (Welsh.)Tur-ones, the inhabitants of the country at the junction of several streams with the Loire, the neighbourhood of the modern Tours.Bi-tur-ig-es, from Bi“Two,”Tur or Dour, Water, and trigo, to reside.There are two tribes of this name in Gaul; the Bituriges Cubi, situated between two of the branches of the Loire, and the Bi-turi-ges Vobisci, between the Garonne and the Sea, at the junction of the Dordogne and the Garonne.Cat-ur-iges, from Catti, Tribes or People; Dour, Water, and Trigo, to reside; on the Durentia, South-east of France, about Embrun or Eburo-Dunum, which was their principal town. Cad-ur-ci, from Catti, Tribes, and dur.There is one tribe of this name on the Dordogne, and another contiguously placed on the Garonne.The mutual support that these interpretations give to each other will be obvious.The following Irish word for“Water,”which is not extant in the Welsh, may be traced in Celtic regions in its various modifications: Uisge (Irish),“The Usk”(South Britain)—Eask60(Irish, obsolete),“The Esk”(Scotland),“The Escaut”(North of France), Isca,“The Exe”(South Britain)—Easkong (Irish, obsolete), Axona (Gallia,Belg.),“The Aisne,”Axones, the neighbouring tribe.[pg 070]Names Of Estuaries, Or Mouths Of Streams.The terms of this class, which occur in ancient Gaul, &c., consist either of terms still thus applied in the living Celtic dialects, or of compounds of which the elements may be recognized, unchanged, in those dialects. Moreover it will be highly interesting to observe that these terms, for the most part, consist of Metaphors derived respectively from the same sources as the two English words“Estuary”and“Mouth,”and the two Latin words“Æstuarium”and“Os Fluminis.”One of the principal arguments of those writers who maintain that the separation of the Irish from the other Celtic tribes must have been of remoter date than the first peopling of these islands, is founded on the fact that the Irish use the word In-ver for the Mouth of a Stream, while the Welsh use Ab-ber (spelt Aber); a feeble support for so wide a conclusion, which a correct analysis of these terms, and a comparison of some interesting coincidences in the local names of ancient Gaul will show to be utterly futile! In-ver and Ab-ber are not simple but compound terms, literally corresponding to the Latin expression“Fluminis Æstuarium.”Æstuarium is from Æstuo,“To boil,”a metaphorical term, obviously derived from the agitation of the Waters where two Streams meet, or where a River enters the Sea.In the first syllable“Inver”and“Ab-ber”differ, but they agree in the last. Both“In”and“Ab,”the first syllables of these terms, occur so often in Celtic regions that there can be no doubt they were both in use among the ancient Celts as words for a River, or Water. The last syllable of these words, Ber or Ver, I shall show to mean an“Estuary.”“In”occurs in the name of“The Inn,”in the Tyrol, the“Æn-us”of the Romans, and in other instances previously[pg 071]noticed.“An”is a Gaelic or Irish term for“Water,”which is identical in sound and sense with terms of frequent occurrence among the tribes of the American Continent, as in Aouin (Hurons, N. America), Jin Jin (Kolushians, extreme North-west ofN. America), Ueni (Maipurians, S. America.)“Ab”occurs in“The Aube,”in France, &c., a name of which the pronunciation may be considered identical with Ab,“Water,”(Persian.) Ap in Sanscrit, and Ubu Obe in Affghan, mean“Water.”“Obe”occurs in Siberia as the name of a well-known river. In India also the term has been applied to“Rivers;”thus we have in that country the Punj-âb, (the Province of“The Five Rivers,”) an appellation of which the corresponding Celtic terms“Pump-ab”would be almost an echo!Further it may here be noticed—as an example of the complete identity of the Celtic and Oriental languages when all the“Disjecta Membra”are compared—that this word does not exist in the modern Celtic in the simple form of Ab, but in the derivative form of Avon, which is found in the Roman maps spelt“Abon,”&c. Now this form also occurs in the East. Abinn,“A River,”is given by Klaproth from the language of the inhabitants of the Mountains to the North of Bhagalpur. Apem means“Water,”in Zend, an ancient Persian dialect. Af is“Water,”in Kurdish.

“But the oldest remains of the Danish language are to be found on our Runic stone monuments, and here at length it perfectly coincides with the earliest Swedish, Norwegian, and Icelandic.“The Danish is closely allied to the Swedish, and both, in the earliest times, lapse into the Icelandic, which, according to all ancient records, was formerly universal over all the North, and must therefore be considered as the parent of both the modern Scandinavian dialects.”45On the subject of the differences of dialect in the differentprovincesof the Northern Kingdoms he says that,“In Norway as well as in Denmark one province terminates its verbs[pg 045]ina, another distinguishes allthe three genders, while a third has preserved a vast number ofoldwordsand inflectionswhich to the othersare unintelligible.”We have thus a proof that even in the provinces of the same kingdom there are differences of“words, grammar,and inflections.”The differencein the numberofgendersis a very remarkable one.The researches of Professor Rask will be found distinctly to warrant the following conclusions. These conclusions are in the nature of results that legitimately flow from his researches; they do not represent the inferences which he himself has thence deduced. With regard both to the languages of England and of his native Scandinavia, this learned writer seems evidently to have been perplexed by the extent and variety of the changes he has described. Hence, in both instances, he has shown an inclination to ascribe to the influence of War and Social disturbance changes which his own researches clearly prove to have been the effects neither of transient nor of local influences, but of causes progressively at work through a series of ages, and embracing large groups of nations and languages in their action.1. The differences which now exist between the various Scandinavian Languages extend to all those features in which it is possible that one Language, or one Class of Languages, can differ from another; viz. to Words, Grammar,Inflections,46and to thearrangementof Words in sentences,47orIdioms.2. Not only do differences of this nature present themselves in the various ScandinavianKingdoms—but also in the various[pg 046]Provinces of the sameKingdom, which in many instances are distinguished by the most marked differences in Words, Grammar, &c. Thus the Dialect of Dalecarlia in Sweden is very ancient and distinct, and approaches to the Gothic.483. These characteristic features of the various languages and dialects of Scandinavia have arisen progressively during the course of ages.4. These differences principally consist in the abandonment in one Kingdom or Province of a portion of the Words, Idioms, Grammar, &c. of the Parent Speech—that part of the elements of the Original Tongue which have become obsolete in one dialect having generally been preserved in the dialects of other kingdoms and provinces—which have at the same time generally lost other distinct portions of the Vocabulary, Grammar, &c. of their common Original. In other words, the“Disjecta Membra”of the old Scandinavian, or“Danska Tunge,”when not preserved in the Danish, have been retained for the most part in the Swedish, Icelandic, and Norwegian, or in some of the Provincial dialects of Scandinavia, and vice versâ.In the various provinces in which it was once spoken different portions of the Parent speech have been abandoned or preserved.5. Hence it follows that the Primitive Language of Scandinavia, or“Danska Tunge,”does not exist in any one—but is dispersed inallits derivative dialects. (Compare the motto from Grotius on the title-page.)6. It is a necessary consequence of the third and fourth propositions that the more ancient remains of the derivative dialects approach more nearly to the Parent Speech, and—in the ratio of their superior antiquity—unite a greater proportion of the distinctive peculiarities of all the sister-dialects,[pg 047]which, as previously stated, have arisen in consequence of certain portions of the Parent speech having been abandoned in some provinces and retained in others, and vice versâ.An interesting illustration of this maxim occurs in a passage from Professor Rask's preface already quoted, in which, after giving a specimen of old Danish, which approaches closely to the Icelandic, he adds,“The few deviations from the Icelandic bear for the most parta strong resemblanceto theSwedish.”In other words, the older specimens of the Danishunitethose peculiarities by which the modern collateral Tongues of Iceland, Denmark, and Sweden are distinguished from each other.Let it be borne in mind, that the lapse of one thousand years has produced these changes, and the instructive nature of this example will be fully apparent. Of the accuracy of the data on which the previous deductions rest, all doubt must be removed by reference to one remarkable event. It is historically certain that the Island of Iceland is inhabited by a nation descended from emigrants from the opposite Norwegian coast. It is historically certain, also, that previously to the Ninth Century these warlike adventurers had not established themselves on the Icelandic soil. Anterior to that period, therefore, it is self-evident that, inasmuch as the Icelanders had no existence as a nation, the Icelandic Tongue could not have had a separate existence as a language. Yet it is certain that in the present day the Icelandic deviates at least as widely from the language of the adjoining Norwegian Coasts as that language deviates from the other Scandinavian Tongues.The evidence furnished by Professor Rask and the writers whose views he has combated, will be found, when fairly balanced, distinctly to support a very important Conclusion, contemplated by neither. The facts adduced on both sides conspire to show a rapid approximation of the Teutonic and[pg 048]Scandinavian branches of the Gothic as we ascend into remote ages.Of this approximation, the features of identity between the Anglo-Saxon and the Icelandic, pointed out by the writers whose views Professor Rask combats, furnish a reasonable presumption, which is converted into positive proof by the evidence collected by Professor Rask himself, that the same features occur in all the ancient, though they do not in the modern, specimens of the Languages of the Scandinavian Peninsulas. It is true, this learned writer, of whose researches I have chiefly availed myself in this Section, maintains that there are some features in which all the Scandinavian differ from the Anglo-Saxon and other Teutonic Dialects, a conclusion, however, but feebly supported by the examples he has adduced, and scarcely reconcilable in any way with the resemblance which the primitive Swedish dialect of Dalecarlia is said to bear to the Gothic. But, assuming the occurrence of some features of difference, even in the earliest specimens we possess, this assumption leaves untouched the proposition that these specimens show a rapid rate of approximation, which, if equally rapid prior to their date, implies that at an era not many ages anterior the identity of the languages of Germany and Scandinavia must have been complete.[pg 049]Section V.The Origin of the Irish Nation. The original Language of the British Isles was a Union of Welsh and Irish. Union of the Irish, Welsh, &c. in the ancient Local Names in the Celtic Countries of Gaul, &c. These Names a connecting Link between the existing Celtic Dialects and the Oriental, Greek, and other Languages, &c.The origin of the Irish nation, or Gael, forms—for numerous reasons—a highly interesting and important subject of inquiry. Of this Nation the very same theories have been maintained as those which have been adopted in some quarters with respect to the North American Indians, the Negroes, and other branches of the Human Family; viz., that they are of a stock aboriginally inferior and distinct, by nature incapable of the virtues of civilization. Let the views advocated by Pinkerton with respect to the Gaelic race—views received with no slight degree of favour in his time—be compared with the doctrines of many modern writers on the subject of the native African and American Races, and an instructive lesson will be learnt on the force of prejudice and the uniformity of error!On the other hand, it must be allowed that the opinions which have been generally espoused on the subject of the origin of the Gael by many of the Historians and Scholars of Ireland and of the Highlands of Scotland, can scarcely be said to possess a better claim to the approbation of a calm and dispassionate judgment. Eminently distinguished as the Irish are by Literary genius, there is probably no subject on which their native talent has appeared to less advantage than in the investigation of the early History of their own[pg 050]Country. Fictions the most extravagant, borrowed from the Chronicles of the dark ages, have been credulously adopted by their first Scholars in lieu of those solid truths to which a calm and sober inquiry alone can lead. Thus we find Mr. Moore, at once the Poet and the Historian of Ireland, lending the sanction of his name to the Fable that the Irish are of Spanish origin; and citing, in answer to the more reasonable hypothesis of a British origin, a variety of Irish writers of no mean note, and some Welsh writers also, in favour of the assertions: 1, that the Irish Language is almost totally unlike the Welsh or Ancient British; and 2, that the Welsh is not a Celtic but a Gothic Tongue! There is every reason to conclude that Mr. Moore—unacquainted, probably, with any of the Celtic dialects himself—resorted to those authorities which he might naturally have deemed most deserving of confidence. But this only renders more lamentably conspicuous the credulity, carelessness, and ignorance of those to whose labours he has appealed. The assertions, 1, that the Welsh and Irish are unlike; and 2, that the Welsh is a Gothic dialect, are contradictions of the plainest facts.Influenced by national feelings Gaelic Scholars have also advanced various other theories, calculated to exhibit the antiquity of their language and race in a favorable point of view. The Gaelic has been maintained to be the Parent, at least in part, of the Latin, the Welsh, &c.; while to the first Colonists of Ireland a Carthaginian or Phœnician origin has been assigned.These conclusions cannot be sustained. But it is highly probable, notwithstanding, that the proofs on which they have been based will be found, in many instances, to contain the germs of important truths, though blended with an admixture of error. The traces of affinity between the Irish[pg 051]and other ancient languages which have been collected by Gaelic Scholars, may be open in many cases to the same remark, which is clearly applicable to the examples of affinity pointed out by Mr. Catlin between the dialect of the North American Indian tribe the Mandans and the Welsh; viz., these features may consist of clear and genuine traces of a generic, though they may afford no proofs of a specific, affinity of race. There can be no doubt that the Irish preserves many primitive forms which the kindred Celtic of Wales has lost; there can be no doubt also that the Irish approximates to the Latin, to the Greek, and to the Egyptian,49&c. in many features which the Welsh no longer exhibits. The examples adduced in Appendix A of the connexion of the Irish language with the Hebrew, Egyptian, &c. are sufficient to show that the Irish are a nation of Oriental origin. But on the other hand it must be borne in mind, that inasmuch as the Welsh, Latin, &c., have also preserved primitive forms which the Irish has lost, there is no ground for concluding that the Gaelic is a Parent rather than a Sister of these venerable Tongues; and inasmuch as the evidence of the Eastern origin of the Gael, however unequivocal, is not clearer or closer than the accompanying50evidence with respect to the Welsh, English, and other European nations, there are no peculiar grounds for referring the first colonization of Ireland to a direct migration from the shores of Palestine or Africa, rather than to the gradual diffusion of population from a central point.The following comparison presents examples of features in which the Irish approximates to the Gothic and other Languages, at the same time that it differs more or less from the Welsh.[pg 052]Words in which the Gaelic resembles the Gothic, and other European Languages, more closely than it resembles the Cymraeg or Welsh.English.Gaelic.Illustrations.Cymraeg.1. Father.Ath-air, (Ir.)Atta, (Gothic.), Ayta, Aydia, (Basque.), Attia, (Hung.), Otek, (Russ.), Fader, slightly varied in all the Gothic dialects, except the Gothic properly so called, Pater, (Greek & Latin.)Tad, (W.)2. Mother.Math-air, (Ir.)Mater or Mutter (with some trifling variations) in Latin, Greek, and all the Teuto-Scandinavian dialects except the Gothic—also in the Sclavonic and Bohemian. Ath-ei, (Gothic.)Mymmog, (Manx dialect.Mam, (W.)3. Brother.Brathair, (Ir.)The Irish form,Brathair,occurs in the Latin and Teuto-Scandinav. tongues; the Welsh form,Brawd,in the Sclavonian tongues.Brawd, (W.), Bredar, (Cornish.)Breur, (Manx dialectBreur, (Arm.)4. Sister.|Siur, (Ir.)The Irish form prevails in the Latin, Teuto-Scand. and Sclavonic.Chwaer, (W.)Piur, (Scotch.)Hor, Huyr, (Cornish.)5. A Company.Drong, (Ir.)Drang, a Throng, a Crowd, (German.)Torv.6. Mock.Magom, (Ir.)Mock, (English.)Gwatwor, (W.)7. Evil.Neoid, (Ir.)Naughty, (Eng.)Droug, (W.)Olk, (Ir.)Ill, (Eng.)8. The Bank of a stream.Rang, (Ir.)Rand,51(Germ.)Glan, (W.)9. A Step.Beim, (Ir.)Bēm-a, a Step, (Greek.), Bain-o, to go, Bahn, a Path, (Germ.)Cam.10. To bear.Beir-im, (Ir.)Fero, (Latin.) Ge-Bähr-en, (Germ.)Dwyn.11. Jeering, Delight, A Desire.Fon-amhad (Ir.), Foun, (Ir.)Fun, (Eng.), Vonne, Delight, (Germ.), Vunsch, a Wish, (Germ.)Vynn, or Mynn, a Wish, (W.)12. A Woman.Geon, (Ir.)Cwen, (Ang.-Sax. & Icel.)Gen-eth, a Girl, (W.)13. To know.Fis-ay-im, Fod-am, (Ir.)Viss-en, (Germ.), Vit-an, (Ang.-Sax.),“I wot,”(Eng.)Wys, or Gwys, Wyth, or Gwyth, Knowledge (W.)14. To heat, or warm.Gorm, (Ir.)Warm, (Eng.)Gwresogi, (W.)15. A Shadow.Sgath, (Ir.)Skia, Skiad-on, (Greek.), Schatten, (Germ.)Cysgod, (W.)16. To speak.|Raid-him, (Ir.)Read-en, (Germ.)Siarad, (W.)Some of these examples would furnish a more plausible argument to show that the Irish are a Gothic race than any which have been advanced to prove that the Welsh are of Gothic origin! It is singular, for instance, that the Irish terms expressive of the Domestic relations are so near the English as to excite in the first instance a suspicion that they must have been borrowed from the followers of Strongbow! But this impression must be dispelled by the reflection that terms of this class are never borrowed from its conquerors by a nation that continues to retain its primitive language. Moreover, it will be observed, that the Irish, in the instance of these words, approaches much more nearly to the Gothic,[pg 054]Hungarian, and Russian, &c. than it does to the English. Again, the Irish word“Gorm,”To heat or warm, islikethe English“Warm.”But, on the other hand, its genuineness is rendered indisputable by itsabsolute identitywith the word 'Gorm' in Persian and Egyptian, (See Appendix A, p.21.) Finally, the resemblances manifested above by the Irish to the Greek are quite as close as those which the former language displays to the English and other Gothic Tongues. In these examples, therefore, we may recognize proofs not of any partial results or specific connexions, but of the more complete approximation of the European languages as we enlarge our range of inquiry, and obtain more ample specimens of each Class.But, notwithstanding the occurrence of some features of difference, it is indisputable that there exists a close specific affinity between the Irish and Welsh Languages, which renders the common origin of the nations who speak them evident. The original identity of the Irish and Welsh Languages was established as far back as the commencement of the eighteenth century, by the investigations of the excellent Archæologist, Edward Lhuyd, who spent five years in travelling through the various Celtic regions, and whose comparison of the dialects of Wales, Cornwall, Armorica, the Highlands of Scotland, and the Isle of Man, is not inferior either in soundness of reasoning, or in patient, extensive, and honest research, to the best German works of the present day. But although the writings of Lhuyd may be said to have established the original unity of the Welsh and Irish races, since the publication of his work, a peculiar opinion has been adopted by some learned men with regard to thetimeof their original separation. Of this opinion, Edward Lhuyd was himself the first advocate; his conclusion was that though the Irish and British Celts were both descendants from one stock, they must have been separated into two[pg 055]distinct Tribes before their arrival in the British Islands. The Gaelic or Irish Tribe he supposes to have preceded the Welsh or British Tribe, by whom he conceives them to have been gradually driven to the West, as the Britons were by the Saxons in subsequent ages. Lhuyd's grounds are as follows:The most ancient names of Rivers and Mountains in the Island of Britain are very generally composed of terms still preserved in the Welsh or Ancient British Tongue. But there are some remarkable exceptions, and in these instances it frequently happens that the Names may be clearly identified with Words still preserved in the Irish or Gaelic branch of the Celtic. For example, the names of the British rivers, the Usk and the Esk, are particularly noticed by Lhuyd; these names are identical with“Uisge, Eask,”the Irish term for“Water.”This word, he observes, does not exist in the Welsh, and he had looked for it in vain in the sister dialect of Armorica; but, he adds, it is still retained by the Irish or Gaelic. Hence, he suggests that the Irish or Gaelic branch of the Celts must have colonized the Island of Britain before the arrival of the Cymry or Welsh branch, by whom, as he conceives, they were expelled, after having conferred names on the principal localities.The evidence of language will be found sufficient to show not merely the common origin of the Welsh and Irish, but also to fix a much more recent date for their separation than that which has been assigned by Lhuyd. It will thence appear that the Irish are descendants of Colonists of the Welsh or British race, not of a distinct Celtic sept, and that the commencement of the separate existence of the Irish nation must be referred to a comparatively recent date, propositions of much interest, of which the proofs about to be advanced will probably be deemed to be at once clear and simple.Lhuyd's reasoning in favour of his theory, that the Irish or Gael existed in Britain as a separate Tribe, prior to the arrival[pg 056]of the Britons who fought against Cæsar, the ancestors of the modern Welsh, is founded on a false analogy not unnatural to a first inquirer.The proposition that the most important local names in every country for the most part consist of terms belonging to the language of the very first inhabitants, is one of which I conceive the truth will be evident. For a proof of this principle, I may refer to Chalmers'52admirable analysis of local names in the Lowlands of Scotland, where, in spite of a succession of Conquests, and the utter extinction in that part of Britain of the language of the original inhabitants, viewed as a vernacular dialect, Welsh and other Celtic names are still preserved, after the lapse of ages, for the most prominent features of the country. This result, it may be observed, is one that flows from the very nature of things. Even the most fierce and ruthless invaders are compelled to hold sufficient intercourse with the first population to enable them to learn the proper names of their localities, and these names, from obvious motives of convenience, they almost universally adopt.Now, had Lhuyd shown that the most ancient Local names in Britain areexclusivelyIrish, there can be no doubt that, consistently with the principle just noticed, his theory would have been supported by the facts to which he adverts. But the most ancient local names in Britain are notexclusivelyorprincipallyIrish; in an equal number, perhaps in a majority, of cases they are Welsh.Moreover, it may be observed that the names of localities in this Island furnish highly instructive evidence, not merely with respect to the different races by whom it has been successively peopled, but also of the order in which they arrived. Thus the names of Rivers and Mountains, and other natural[pg 057]objects, at least of the most conspicuous, are Celtic; the names of the most ancient Towns are Latin, or Latin grafted on British words; more modern Towns and Villages have Saxon appellations; those of more recent origin have frequently Norman designations; and last of all come those places which have names derived from our present English. These various classes of names cannot be nicely distinguished in each particular instance. Of the correctness of the general principle, however, there is no doubt.But the terms noticed by Lhuyd as significant in the Irish language do not belong to a different class of appellations from those which are obviously of British or Cymraeg origin. The Irish and Cymraeg terms are both found to predominate most in the names of the most ancient Class, viz. in those of Rivers, Mountains, &c., and to be thus applied in conjunction. Hence the natural inference that flows from his facts is not that these names were conferred by two distinct and successive races, but that they were imposedcontemporaneouslyand by the same People!Further it may be noticed, that if British Topography presents words extant only in the Irish Tongue, Irish Topography also presents names which cannot be explained by means of the Irish, though their meaning is preserved in Welsh; for example: There is a place near the head of a Stream in Roscommon, called“Glan a Modda,”(from Glan,“The bank of a Stream,”Welsh.) There is a place in Wales, called“Glan a Mowdduy.”There is a place called“Glan-gora,”in a Creek at the head of Bantry Bay; and another place in Ireland called“Glan-gort.”“Ben-heder,”the ancient Irish name for“The Hill of Howth,”interpreted by Mr. Moore“The Hill of Birds.”(Adar,“Birds,”Welsh. The word does not exist in Irish.)Arran, A mountainous Island. (Arran, a Mountain,Welsh. This word does not exist in Irish,) &c. &c.[pg 058]Mr. Chalmers in his Caledonia states that the prevalent ancient names of localities in Britain and Ireland are essentially the same.The conclusions to which these facts legitimately and necessarily lead are, that the British Islands were originally colonized by Settlers, who, at the time of the first occupation of Great Britain and Ireland, spoke one uniform language, in which the Welsh, Irish, and other living Celtic Dialects were combined. We may infer, and I conceive most clearly, that these dialects must be viewed in the light of“Disjecta Membra”of the speech of the old British and Irish Celts, just as the Icelandic, Norwegian, &c. are fragments of the ancient“Danska Tunge,”as noticed in the previous section.It has been shown by Dr. Prichard that the population of Islands has been derived from the neighbouring Continents, and that the population of the more distant Islands has been derived in like manner from those which are nearer to the common source of migration. It is highly unreasonable to assume that Ireland has formed an exception to this general rule, considering that the common basis of the Irish and ancient British or Welsh languages are confessedly the same, unless it can be proved that the accompanying differences are such as to require the solution Lhuyd has suggested. Here, then, the question arises, are the features of difference between the Welsh and Irish languages more numerous or more fundamental, in relation to the interval of time that has elapsed since the Roman Invasion of Britain, than the varieties of dialect among the Scandinavian nations are in relation to the period that has elapsed since the colonization of Iceland? They are not! It will thence be seen that Lhuyd's theory, as to the remote date of the separation of the Gaelic or Irish from the British or Cymraeg branch of the Celts, is founded on an exaggerated conception[pg 059]of the stability of Human Tongues; and that the abandonment by various septs of different synonymes used conjointly by their common forefathers will satisfactorily account for the differences between the Welsh and Irish, to which he attaches so much weight. It will be perceived, for example, that in the Icelandic, of which the existence commenced in the ninth century, and the Continental Scandinavian from which it then sprang, totally different terms are used for“Water,”the very instance to which Lhuyd especially adverts, as regards the languages of the Welsh and Irish, whom we know to have existed as separate nations in the time of Cæsar eighteen centuries ago!Another highly instructive test of the correctness of his theory may be derived from the investigations of Lhuyd himself, who, in his comparison of the Welsh and Irish languages, uniformly distinguished the current terms from the obsolete synonymous words that occur only in ancient MSS. This comparison proves distinctly that the Irish and Welsh languages approximate, as we ascend, at a rate which, if as rapid previously as we know it to have been up to the date of the earliest MSS., would imply that these languages must have been identical about the era of the Roman invasion. As the changes which languages undergo in their infancy are more rapid than those which occur at later stages of their growth, it is possible that the unity of these Tongues may be ascribed even to a much later period, an opinion which has been maintained by a very judicious and excellent writer, Mr. Edward Davies, who in his“Claims of Ossian”has published an early specimen of Irish Poetry, which in Language and Style he regards as identical with the most ancient productions of the Welsh Bards. Making every allowance for the irregularity of the changes which occur in Languages, I do not conceive it possible that the Welsh and Irish could have differed very essentially[pg 060]in the time of Cæsar. This leads directly to another conclusion, viz. that the first colonization of Ireland could not have taken place a great many centuries before the Roman invasion. Had such been the case, the differences between the Welsh and Irish Languages must have been proportionately more extensive. In the time of the Romans we learn that an Irish traitor arrived in Britain, who stated that Ireland might be kept in subjection by a single legion, an incident which tends, however slightly, to favour the opinion that the sister Island was at that period but thinly, perhaps because but recently, peopled.Of the extent of the changes which the Celtic languages have undergone since the first arrival of the Celts in Europe, we possess proofs of far more ancient date than the earliest literary specimens of the living dialects of the Celtic in the Local names of Celtic regions, as preserved in Roman Maps, and in the existing languages of the French, English, and other nations, who occupy countries of which the Celts were the first inhabitants. These names I shall show to consist of three elements: A union of 1, Welsh, Cornish, &c.; 2, Irish, Highland Scotch, &c.; and 3, Terms not extant in any Celtic Tongue, but preserved in the Oriental, Greek, and other languages.As regards the Names of the 1st and 2d Classes, it will abundantly appear from the ensuing examples that, in the Topographical Nomenclature of Gaul, Britain, and other Celtic regions of Europe,53words derived from all the various Celtic dialects now extant, occur in a manner that leads distinctly to the inference that these“Disjecta membra”must have simultaneously belonged to the language of the old Celts. Dr. Prichard, who has examined these vestiges of the[pg 061]ancient Celtic Populations of Europe with much ability and success, leans to the opinion that the Cymraeg or Welsh Dialects predominate in these names. But the following examples, which comprise many names derived from the Irish or Gaelic that have not been noticed by Dr. Prichard or by previous writers on this subject, will serve to render it manifest that the ancient Names in Europœa Celtica did, in fact, include all the various living Celtic dialects very equally and harmoniously blended.How luminous and distinct these proofs of the identity of the ancient with the modern Celtic nations are, will be better understood by a preliminary statement of certain rules, which will serve to give greater precision and perspicuity to the illustrations selected:1. There can be no doubt that the Romans, in the Celtic, as in other countries conquered by them, modified the native terms by the addition of their own peculiar grammatical inflections, as in“Judæ-i, Britann-i, Sen-ones,”&c. Now it is obvious that in identifying the Celtic terms we must reject these mere Roman inflections.542. In many cases the Roman Names cannot be supposed to involve complete transcripts of the Celtic Names; frequently they were doubtless convenient abbreviations of the original names—names consisting of descriptive terms to them unintelligible. According to Mr. Reynolds, the Saxons generally adopted the first syllable only of the Roman or British names they found in this island. According to Bullet,“Vic,”a word of Roman origin for a Village or Town, has, from similar causes, become common as a Proper name in Dauphiné; in modern times we have numerous Villages called“Thorpe,”the name for a Village in Anglo-Saxon and[pg 062]German. In instances of this kind, there can be no doubt that originally the names were descriptive, such as“Long-town,”“Old-town,”&c. Tre or Trev is the common Welsh word for a Town, Village, or residence; it had the same meaning in Cornwall:“By Tre, Tres, and Tren,You shall know the Cornish men.”A consequence of the names of the gentry of the county having been derived from those of their residences, into which this word commonly entered!In Wales we have numerous examples of“Tre,”as in“Tre-llwng,”“The Town”of the“Pool,”(i.e. Welshpool,) from an adjoining“Llyn,”or Pool, near Powis Castle;“Tre-lydan,”the Broad Village, or Residence near Welshpool;“Trev-alyn,”near Chester, the Residence on the Stream; the“Alyn,”&c. &c.Now according to the Roman mode, such a term as Trev-alyn would have been changed into Trev-iri, the designation actually given to the Celts of“Treves,”&c.The following are analogous examples:There is a tribe of Brig-antes in Yorkshire, another in Ireland, and a third in the North-east of Spain. Many unsuccessful attempts have been made to show that these distant Celtic tribes must have been scions of the same tribe. A much simpler explanation may be given.By referring to the Roman maps the reader will find a word,“Briga,”in such general use as part of the names of towns as to leave no reasonable doubt that it must have been, like Tre, a Celtic name for a town—now obsolete. Thus in Spain we have, Laco-briga, Meido-briga, Ara-briga, Tala-brica, Augusto-briga, &c. Now the analogous instances already noticed suffice to point out that the occurrence of[pg 063]Brig-antes as a Roman name of Tribes in three Celtic countries, is a natural result of the frequent occurrence of Briga as the first part of the names of Celtic places.The“Allo-bryg-es.”The name of this warlike tribe, the Celtic inhabitants of Savoy, has also been the source of perplexity, which may be removed in the same manner. This tribe had a town, called by the Romans“Brig-icum,”which was said to be“the only one they had.”55Now Allo-Bryga may reasonably be identified with Alpo-Briga, the Town of the Alps (Briga being clearly the common base of“Allo-bryg-es,”and“Brig-icum.”)The names of Celtic communities, as they appear on the Roman Maps, may, I conceive, be proved to have been descriptive of the most prominent natural features of the regions they inhabited, and not of their lineage or descent, as seems to have been often supposed. Thus we have the Mor-ini in Belgium, and the Ar-mor-ici in Gaul on the Sea; we have the Sen-ones on the Seine, the Tamar-ici on the Tamar-is, in Hispania, &c. In the Mountainous regions it will be observed that the names of tribes are derived from the Mountains. In the flat countries they take their names from Rivers or the confluence of Rivers. In the same manner it is highly deserving of remark, that the names of the different French Departments have been derived from precisely the same natural features. Thus in the Hilly countries we have the Departments of the High Alps,“Hautes Alpes;”of the Low Alps,“Basses Alpes;”in the Champaign countries the Departments are named from the Rivers; such as the Seine, the Marne, and the Somme, &c. Many of these French names are literally equivalent to translations of the ancient Gaulish names, as interpreted by means of the Welsh and Irish languages. It is impossible to conceive a[pg 064]more perfect verification of the accuracy of these interpretations!I may here observe, that as far as we can perceive, the various independent communities of Britain and Gaul mentioned by Cæsar, such as the Edui, the Venetes, &c., did not consist ofoneclan or sept, they seem rather to have been a combination of several contiguous septs, to whom no appropriate common namecould have beengiven, except one derived from the natural features of the district they occupied.The durability of local names has been already noticed. Of this truth we possess remarkable proofs in those of localities in France, as preserved by the modern French to the present day. I do not doubt that the present French names are, in many instances, much more faithful transcripts of the original Celtic appellations than those which occur in the Roman Maps are. Thus, for example, Bonomia, a name conferred by the Romans upon Boulogne, and of which the origin has perplexed Antiquarics, may easily be explained as a Roman abbreviation of the word Boulogne itself, of which the Celtic meaning will be shown hereafter to be appropriate and unequivocal. Here it may be noticed, that the Celtic language did not become extinct in Gaul until many centuries after the termination of the Roman sway and the establishment of the Franks in that country. The use of the old Gaulish or Celtic continued until the eighth century, nearly until the time of Charlemagne.56Now we know that the modern Welsh and Irish, for the most part, continue to use their own primitive names of localities in those cases in which abbreviations or translations have been substituted by the English. There can be no reasonable doubt that the ancient Gauls did the same, and that these names were in use among the inhabitants[pg 065]of each locality at the time of the final subjugation of Gaul by the Franks, by whom, in many instances, these names are more likely to have been adopted than those used by the Romans.It will also be observed in the course of the following examples, that names of the class about to be noticed, viz., Topographical names of which the elements are not extant in the existing Celtic dialects, but occur in Oriental words, &c., are remarkably well preserved by the modern French. Thus the“Aube,”as pronounced by the French, is identical in sound with the Asiatic terms for Water, and names of Rivers, to which it is allied.3. By many, perhaps by all those Celtic scholars who have investigated this subject, it has been assumed that the living Celtic dialects may be expected to furnish a complete clue to all the Local Names of ancient Celtic regions. This conclusion, like the theory of Lhuyd above discussed, is founded on an exaggerated idea of the stability of Human Tongues! Neither the Irish nor the Welsh, nor a combination of all the Celtic dialects, will be found to afford a complete solution of the Topographical nomenclature of the ancient Celtic regions of Europe. Names undoubtedly occur in these countries which have been preserved in none of the Celtic tongues, names which I shall indisputably show to be positive transcripts, in many instances, of appropriate terms occurring in the Hebrew and other languages, with which, in other parts of this work, the original Celtic dialects will be proved to have been originally identical. These facts lead to the conclusion that the ancient nomenclature of Celtic countries forms in reality a connecting link between the existing dialects of the Celts and the language of the Oriental stock from which they are descended.This conclusion, though at variance with the views of many estimable writers, is nevertheless in unison with those anticipations[pg 066]which historical facts legitimately suggest. It is only reasonable to infer that since the period of their first arrival in Europe, the era at which many of these names must have been conferred (see page10), the Celtic tribes must have lost many words which none of the modern Celtic nations have preserved. The Celts were settled about the sources of the“Ister, and the city,”(perhaps the mountains)“of Pyrene,”even in the time of Herodotus, and how many ages had elapsed since their first arrival is unknown!57There is a certain Class of terms of which the meaning can reasonably be inferred from their extensive use in combination with other terms, of which the meaning may be considered as ascertained. To this class may be referred the terms immediately following.Catti, Cassii, Casses, or Cad, seem to have meant a People, Tribe, &c., as in the following examples of the names of Celtic Tribes:The Abr-in-Catui, in Normandy. The Catti-euch-lani, the people of Cambridgeshire and the adjoining counties. The Cassii, in Hertfordshire. The Bidu-casses, in Normandy. The Tri-casses, a people in Champagne. The Cad-ur-ci, on the Garonne.The above words seem clearly derivable from the following Welsh words, which are allied to the Hebrew:Welsh.Hebrew.From Kiw-dod (Kiw-dod-æ,plur.) a Clan, a Nation.Gow, a Body of Men, a Society or Association.Kiw-ed, a Multitude, a Tribe.Gowee, a Nation.Kyf, a Body or Trunk, a Pedigree.Gow, Gowe, Goweeth, the Body of a Man or Animal.[pg 067]Tre, Trev, a Village, Town, or Residence, (Welsh,) a Tribe, (Irish.)58Trev-iri, the people of Treves. A-Treb-ates, the people about Arras. (For further examples see Dr. Prichard's work.) Trev is a common element in names of places in Wales, as Tre-vecca, Tre-gynnon.Trigo, to reside, dwell, (Welsh.)Duro-trig-es, the dwellers on the Water or Sea, the people of Dorsetshire. (Camden.)Catt uriges. (See Dour.)Dun-um, a Hill, a Fort or Town, generally on a Hill, (occurs inWelshandIrish.)Oxell-dunum, a Hill-fort in Gaul, described by Cæsar. (See numerous instances in Dr. Prichard's work.)“Castell Din-as Bran,”on a lofty eminence in the Vale of Llangollen, Wales.Dur, Duvr, Awethur (Welsh), Dour (Cornish), Dur (Armorican), Dovar (Irish, obsolete, but occurs in ancient MSS.)“Water.”This word, and Ydōr or Hudōr (Greek), and Tschur (Armenian),“Water,”have an obvious affinity. These forms may be traced in the names of Celtic Localities.“Dour”occurs in the following names of Rivers: Dur, (Hibernia,) Dur-ia Major,“The Doria,”and Duria Minor, (Gallia Cisalpina,) Dur-ius,“The Douro,”and“Dero,”(Hispania,) Dur-anius,“The Dordogne,”(Gallia). InBucharianDeriâ means“The Sea.”Ydōr or Hudōr (Greek), Awethur (Welsh), occur in the[pg 068]Rivers“The Adour,”59Atur-is (Gallia),“The Adder”(Britain),“The Adare”(Ireland.)“Tschur”(Armenian), occurs in“Stura,”(Gallia Cis.),“The Stour”(Britain),“The Suir”(Britain & Ireland),“The Souro”(Spain, a branch of the Tagus.)From the frequent recurrence of all these different forms in several Celtic countries thus widely separated, it is plain that they were used conjointly by the early Celts, and represent various transitions of the same word. Thus“Stura”(inGal. Cis.), flows between the neighbouring streams Duria Major and Duria Minor, &c.This word“Dour”enters very largely into the names of tribes; it forms singly a natural clue to a great number of names that hitherto have been referred to a complication of Roots. Thus the Roman name for the people of Dorsetshire, Duro-trig-es, i.e. The dwellers on the Water or the Ocean, has been noticed by Camden.In the preceding, and in several of the following, it will be apparent that the old Celts applied this term to the“Sea or Ocean,”as the Bucharians do, and also to a“River.”At present the Welsh apply the term to Water only, in a restricted sense.In the South-east of England names abound (applied to places on Rivers or the Sea) in which the two slight variations of Dur and Du-v-r (or Do-v-ar,Irish), still preserved in Welsh, are apparent. Duro-vern-um,“Canterbury,”from Duro, Water, and Vern or Veryn, a Hill. (Compare the name of the“Ar-vern-i,”under Beryn, at p. 78.) The Town was on a Hill by the Stour.Portus Du-b-r-is or Dub-r-œ, i.e.“Sea Port,”the modern[pg 069]“Do-v-or,”a word which is an echo of the Irish Dovar and the Welsh Du-v-r.Duro-brivæ, Rochester on the Medway, (Briva or Brivis, the ancient Celtic for a Town.) Duro-levum, Milton on the Thames.Lan-du-b-r-is, a Portuguese Island. Lan, a Bank of a Stream, or the Sea: also an inclosed Space, (Welsh.)Tur-ones, the inhabitants of the country at the junction of several streams with the Loire, the neighbourhood of the modern Tours.Bi-tur-ig-es, from Bi“Two,”Tur or Dour, Water, and trigo, to reside.There are two tribes of this name in Gaul; the Bituriges Cubi, situated between two of the branches of the Loire, and the Bi-turi-ges Vobisci, between the Garonne and the Sea, at the junction of the Dordogne and the Garonne.Cat-ur-iges, from Catti, Tribes or People; Dour, Water, and Trigo, to reside; on the Durentia, South-east of France, about Embrun or Eburo-Dunum, which was their principal town. Cad-ur-ci, from Catti, Tribes, and dur.There is one tribe of this name on the Dordogne, and another contiguously placed on the Garonne.The mutual support that these interpretations give to each other will be obvious.The following Irish word for“Water,”which is not extant in the Welsh, may be traced in Celtic regions in its various modifications: Uisge (Irish),“The Usk”(South Britain)—Eask60(Irish, obsolete),“The Esk”(Scotland),“The Escaut”(North of France), Isca,“The Exe”(South Britain)—Easkong (Irish, obsolete), Axona (Gallia,Belg.),“The Aisne,”Axones, the neighbouring tribe.[pg 070]Names Of Estuaries, Or Mouths Of Streams.The terms of this class, which occur in ancient Gaul, &c., consist either of terms still thus applied in the living Celtic dialects, or of compounds of which the elements may be recognized, unchanged, in those dialects. Moreover it will be highly interesting to observe that these terms, for the most part, consist of Metaphors derived respectively from the same sources as the two English words“Estuary”and“Mouth,”and the two Latin words“Æstuarium”and“Os Fluminis.”One of the principal arguments of those writers who maintain that the separation of the Irish from the other Celtic tribes must have been of remoter date than the first peopling of these islands, is founded on the fact that the Irish use the word In-ver for the Mouth of a Stream, while the Welsh use Ab-ber (spelt Aber); a feeble support for so wide a conclusion, which a correct analysis of these terms, and a comparison of some interesting coincidences in the local names of ancient Gaul will show to be utterly futile! In-ver and Ab-ber are not simple but compound terms, literally corresponding to the Latin expression“Fluminis Æstuarium.”Æstuarium is from Æstuo,“To boil,”a metaphorical term, obviously derived from the agitation of the Waters where two Streams meet, or where a River enters the Sea.In the first syllable“Inver”and“Ab-ber”differ, but they agree in the last. Both“In”and“Ab,”the first syllables of these terms, occur so often in Celtic regions that there can be no doubt they were both in use among the ancient Celts as words for a River, or Water. The last syllable of these words, Ber or Ver, I shall show to mean an“Estuary.”“In”occurs in the name of“The Inn,”in the Tyrol, the“Æn-us”of the Romans, and in other instances previously[pg 071]noticed.“An”is a Gaelic or Irish term for“Water,”which is identical in sound and sense with terms of frequent occurrence among the tribes of the American Continent, as in Aouin (Hurons, N. America), Jin Jin (Kolushians, extreme North-west ofN. America), Ueni (Maipurians, S. America.)“Ab”occurs in“The Aube,”in France, &c., a name of which the pronunciation may be considered identical with Ab,“Water,”(Persian.) Ap in Sanscrit, and Ubu Obe in Affghan, mean“Water.”“Obe”occurs in Siberia as the name of a well-known river. In India also the term has been applied to“Rivers;”thus we have in that country the Punj-âb, (the Province of“The Five Rivers,”) an appellation of which the corresponding Celtic terms“Pump-ab”would be almost an echo!Further it may here be noticed—as an example of the complete identity of the Celtic and Oriental languages when all the“Disjecta Membra”are compared—that this word does not exist in the modern Celtic in the simple form of Ab, but in the derivative form of Avon, which is found in the Roman maps spelt“Abon,”&c. Now this form also occurs in the East. Abinn,“A River,”is given by Klaproth from the language of the inhabitants of the Mountains to the North of Bhagalpur. Apem means“Water,”in Zend, an ancient Persian dialect. Af is“Water,”in Kurdish.

“But the oldest remains of the Danish language are to be found on our Runic stone monuments, and here at length it perfectly coincides with the earliest Swedish, Norwegian, and Icelandic.

“The Danish is closely allied to the Swedish, and both, in the earliest times, lapse into the Icelandic, which, according to all ancient records, was formerly universal over all the North, and must therefore be considered as the parent of both the modern Scandinavian dialects.”45

On the subject of the differences of dialect in the differentprovincesof the Northern Kingdoms he says that,“In Norway as well as in Denmark one province terminates its verbs[pg 045]ina, another distinguishes allthe three genders, while a third has preserved a vast number ofoldwordsand inflectionswhich to the othersare unintelligible.”

We have thus a proof that even in the provinces of the same kingdom there are differences of“words, grammar,and inflections.”The differencein the numberofgendersis a very remarkable one.

The researches of Professor Rask will be found distinctly to warrant the following conclusions. These conclusions are in the nature of results that legitimately flow from his researches; they do not represent the inferences which he himself has thence deduced. With regard both to the languages of England and of his native Scandinavia, this learned writer seems evidently to have been perplexed by the extent and variety of the changes he has described. Hence, in both instances, he has shown an inclination to ascribe to the influence of War and Social disturbance changes which his own researches clearly prove to have been the effects neither of transient nor of local influences, but of causes progressively at work through a series of ages, and embracing large groups of nations and languages in their action.

1. The differences which now exist between the various Scandinavian Languages extend to all those features in which it is possible that one Language, or one Class of Languages, can differ from another; viz. to Words, Grammar,Inflections,46and to thearrangementof Words in sentences,47orIdioms.

2. Not only do differences of this nature present themselves in the various ScandinavianKingdoms—but also in the various[pg 046]Provinces of the sameKingdom, which in many instances are distinguished by the most marked differences in Words, Grammar, &c. Thus the Dialect of Dalecarlia in Sweden is very ancient and distinct, and approaches to the Gothic.48

3. These characteristic features of the various languages and dialects of Scandinavia have arisen progressively during the course of ages.

4. These differences principally consist in the abandonment in one Kingdom or Province of a portion of the Words, Idioms, Grammar, &c. of the Parent Speech—that part of the elements of the Original Tongue which have become obsolete in one dialect having generally been preserved in the dialects of other kingdoms and provinces—which have at the same time generally lost other distinct portions of the Vocabulary, Grammar, &c. of their common Original. In other words, the“Disjecta Membra”of the old Scandinavian, or“Danska Tunge,”when not preserved in the Danish, have been retained for the most part in the Swedish, Icelandic, and Norwegian, or in some of the Provincial dialects of Scandinavia, and vice versâ.In the various provinces in which it was once spoken different portions of the Parent speech have been abandoned or preserved.

5. Hence it follows that the Primitive Language of Scandinavia, or“Danska Tunge,”does not exist in any one—but is dispersed inallits derivative dialects. (Compare the motto from Grotius on the title-page.)

6. It is a necessary consequence of the third and fourth propositions that the more ancient remains of the derivative dialects approach more nearly to the Parent Speech, and—in the ratio of their superior antiquity—unite a greater proportion of the distinctive peculiarities of all the sister-dialects,[pg 047]which, as previously stated, have arisen in consequence of certain portions of the Parent speech having been abandoned in some provinces and retained in others, and vice versâ.

An interesting illustration of this maxim occurs in a passage from Professor Rask's preface already quoted, in which, after giving a specimen of old Danish, which approaches closely to the Icelandic, he adds,“The few deviations from the Icelandic bear for the most parta strong resemblanceto theSwedish.”In other words, the older specimens of the Danishunitethose peculiarities by which the modern collateral Tongues of Iceland, Denmark, and Sweden are distinguished from each other.

Let it be borne in mind, that the lapse of one thousand years has produced these changes, and the instructive nature of this example will be fully apparent. Of the accuracy of the data on which the previous deductions rest, all doubt must be removed by reference to one remarkable event. It is historically certain that the Island of Iceland is inhabited by a nation descended from emigrants from the opposite Norwegian coast. It is historically certain, also, that previously to the Ninth Century these warlike adventurers had not established themselves on the Icelandic soil. Anterior to that period, therefore, it is self-evident that, inasmuch as the Icelanders had no existence as a nation, the Icelandic Tongue could not have had a separate existence as a language. Yet it is certain that in the present day the Icelandic deviates at least as widely from the language of the adjoining Norwegian Coasts as that language deviates from the other Scandinavian Tongues.

The evidence furnished by Professor Rask and the writers whose views he has combated, will be found, when fairly balanced, distinctly to support a very important Conclusion, contemplated by neither. The facts adduced on both sides conspire to show a rapid approximation of the Teutonic and[pg 048]Scandinavian branches of the Gothic as we ascend into remote ages.

Of this approximation, the features of identity between the Anglo-Saxon and the Icelandic, pointed out by the writers whose views Professor Rask combats, furnish a reasonable presumption, which is converted into positive proof by the evidence collected by Professor Rask himself, that the same features occur in all the ancient, though they do not in the modern, specimens of the Languages of the Scandinavian Peninsulas. It is true, this learned writer, of whose researches I have chiefly availed myself in this Section, maintains that there are some features in which all the Scandinavian differ from the Anglo-Saxon and other Teutonic Dialects, a conclusion, however, but feebly supported by the examples he has adduced, and scarcely reconcilable in any way with the resemblance which the primitive Swedish dialect of Dalecarlia is said to bear to the Gothic. But, assuming the occurrence of some features of difference, even in the earliest specimens we possess, this assumption leaves untouched the proposition that these specimens show a rapid rate of approximation, which, if equally rapid prior to their date, implies that at an era not many ages anterior the identity of the languages of Germany and Scandinavia must have been complete.

Section V.

The Origin of the Irish Nation. The original Language of the British Isles was a Union of Welsh and Irish. Union of the Irish, Welsh, &c. in the ancient Local Names in the Celtic Countries of Gaul, &c. These Names a connecting Link between the existing Celtic Dialects and the Oriental, Greek, and other Languages, &c.

The origin of the Irish nation, or Gael, forms—for numerous reasons—a highly interesting and important subject of inquiry. Of this Nation the very same theories have been maintained as those which have been adopted in some quarters with respect to the North American Indians, the Negroes, and other branches of the Human Family; viz., that they are of a stock aboriginally inferior and distinct, by nature incapable of the virtues of civilization. Let the views advocated by Pinkerton with respect to the Gaelic race—views received with no slight degree of favour in his time—be compared with the doctrines of many modern writers on the subject of the native African and American Races, and an instructive lesson will be learnt on the force of prejudice and the uniformity of error!

On the other hand, it must be allowed that the opinions which have been generally espoused on the subject of the origin of the Gael by many of the Historians and Scholars of Ireland and of the Highlands of Scotland, can scarcely be said to possess a better claim to the approbation of a calm and dispassionate judgment. Eminently distinguished as the Irish are by Literary genius, there is probably no subject on which their native talent has appeared to less advantage than in the investigation of the early History of their own[pg 050]Country. Fictions the most extravagant, borrowed from the Chronicles of the dark ages, have been credulously adopted by their first Scholars in lieu of those solid truths to which a calm and sober inquiry alone can lead. Thus we find Mr. Moore, at once the Poet and the Historian of Ireland, lending the sanction of his name to the Fable that the Irish are of Spanish origin; and citing, in answer to the more reasonable hypothesis of a British origin, a variety of Irish writers of no mean note, and some Welsh writers also, in favour of the assertions: 1, that the Irish Language is almost totally unlike the Welsh or Ancient British; and 2, that the Welsh is not a Celtic but a Gothic Tongue! There is every reason to conclude that Mr. Moore—unacquainted, probably, with any of the Celtic dialects himself—resorted to those authorities which he might naturally have deemed most deserving of confidence. But this only renders more lamentably conspicuous the credulity, carelessness, and ignorance of those to whose labours he has appealed. The assertions, 1, that the Welsh and Irish are unlike; and 2, that the Welsh is a Gothic dialect, are contradictions of the plainest facts.

Influenced by national feelings Gaelic Scholars have also advanced various other theories, calculated to exhibit the antiquity of their language and race in a favorable point of view. The Gaelic has been maintained to be the Parent, at least in part, of the Latin, the Welsh, &c.; while to the first Colonists of Ireland a Carthaginian or Phœnician origin has been assigned.

These conclusions cannot be sustained. But it is highly probable, notwithstanding, that the proofs on which they have been based will be found, in many instances, to contain the germs of important truths, though blended with an admixture of error. The traces of affinity between the Irish[pg 051]and other ancient languages which have been collected by Gaelic Scholars, may be open in many cases to the same remark, which is clearly applicable to the examples of affinity pointed out by Mr. Catlin between the dialect of the North American Indian tribe the Mandans and the Welsh; viz., these features may consist of clear and genuine traces of a generic, though they may afford no proofs of a specific, affinity of race. There can be no doubt that the Irish preserves many primitive forms which the kindred Celtic of Wales has lost; there can be no doubt also that the Irish approximates to the Latin, to the Greek, and to the Egyptian,49&c. in many features which the Welsh no longer exhibits. The examples adduced in Appendix A of the connexion of the Irish language with the Hebrew, Egyptian, &c. are sufficient to show that the Irish are a nation of Oriental origin. But on the other hand it must be borne in mind, that inasmuch as the Welsh, Latin, &c., have also preserved primitive forms which the Irish has lost, there is no ground for concluding that the Gaelic is a Parent rather than a Sister of these venerable Tongues; and inasmuch as the evidence of the Eastern origin of the Gael, however unequivocal, is not clearer or closer than the accompanying50evidence with respect to the Welsh, English, and other European nations, there are no peculiar grounds for referring the first colonization of Ireland to a direct migration from the shores of Palestine or Africa, rather than to the gradual diffusion of population from a central point.

The following comparison presents examples of features in which the Irish approximates to the Gothic and other Languages, at the same time that it differs more or less from the Welsh.

Words in which the Gaelic resembles the Gothic, and other European Languages, more closely than it resembles the Cymraeg or Welsh.

Some of these examples would furnish a more plausible argument to show that the Irish are a Gothic race than any which have been advanced to prove that the Welsh are of Gothic origin! It is singular, for instance, that the Irish terms expressive of the Domestic relations are so near the English as to excite in the first instance a suspicion that they must have been borrowed from the followers of Strongbow! But this impression must be dispelled by the reflection that terms of this class are never borrowed from its conquerors by a nation that continues to retain its primitive language. Moreover, it will be observed, that the Irish, in the instance of these words, approaches much more nearly to the Gothic,[pg 054]Hungarian, and Russian, &c. than it does to the English. Again, the Irish word“Gorm,”To heat or warm, islikethe English“Warm.”But, on the other hand, its genuineness is rendered indisputable by itsabsolute identitywith the word 'Gorm' in Persian and Egyptian, (See Appendix A, p.21.) Finally, the resemblances manifested above by the Irish to the Greek are quite as close as those which the former language displays to the English and other Gothic Tongues. In these examples, therefore, we may recognize proofs not of any partial results or specific connexions, but of the more complete approximation of the European languages as we enlarge our range of inquiry, and obtain more ample specimens of each Class.

But, notwithstanding the occurrence of some features of difference, it is indisputable that there exists a close specific affinity between the Irish and Welsh Languages, which renders the common origin of the nations who speak them evident. The original identity of the Irish and Welsh Languages was established as far back as the commencement of the eighteenth century, by the investigations of the excellent Archæologist, Edward Lhuyd, who spent five years in travelling through the various Celtic regions, and whose comparison of the dialects of Wales, Cornwall, Armorica, the Highlands of Scotland, and the Isle of Man, is not inferior either in soundness of reasoning, or in patient, extensive, and honest research, to the best German works of the present day. But although the writings of Lhuyd may be said to have established the original unity of the Welsh and Irish races, since the publication of his work, a peculiar opinion has been adopted by some learned men with regard to thetimeof their original separation. Of this opinion, Edward Lhuyd was himself the first advocate; his conclusion was that though the Irish and British Celts were both descendants from one stock, they must have been separated into two[pg 055]distinct Tribes before their arrival in the British Islands. The Gaelic or Irish Tribe he supposes to have preceded the Welsh or British Tribe, by whom he conceives them to have been gradually driven to the West, as the Britons were by the Saxons in subsequent ages. Lhuyd's grounds are as follows:

The most ancient names of Rivers and Mountains in the Island of Britain are very generally composed of terms still preserved in the Welsh or Ancient British Tongue. But there are some remarkable exceptions, and in these instances it frequently happens that the Names may be clearly identified with Words still preserved in the Irish or Gaelic branch of the Celtic. For example, the names of the British rivers, the Usk and the Esk, are particularly noticed by Lhuyd; these names are identical with“Uisge, Eask,”the Irish term for“Water.”This word, he observes, does not exist in the Welsh, and he had looked for it in vain in the sister dialect of Armorica; but, he adds, it is still retained by the Irish or Gaelic. Hence, he suggests that the Irish or Gaelic branch of the Celts must have colonized the Island of Britain before the arrival of the Cymry or Welsh branch, by whom, as he conceives, they were expelled, after having conferred names on the principal localities.

The evidence of language will be found sufficient to show not merely the common origin of the Welsh and Irish, but also to fix a much more recent date for their separation than that which has been assigned by Lhuyd. It will thence appear that the Irish are descendants of Colonists of the Welsh or British race, not of a distinct Celtic sept, and that the commencement of the separate existence of the Irish nation must be referred to a comparatively recent date, propositions of much interest, of which the proofs about to be advanced will probably be deemed to be at once clear and simple.

Lhuyd's reasoning in favour of his theory, that the Irish or Gael existed in Britain as a separate Tribe, prior to the arrival[pg 056]of the Britons who fought against Cæsar, the ancestors of the modern Welsh, is founded on a false analogy not unnatural to a first inquirer.

The proposition that the most important local names in every country for the most part consist of terms belonging to the language of the very first inhabitants, is one of which I conceive the truth will be evident. For a proof of this principle, I may refer to Chalmers'52admirable analysis of local names in the Lowlands of Scotland, where, in spite of a succession of Conquests, and the utter extinction in that part of Britain of the language of the original inhabitants, viewed as a vernacular dialect, Welsh and other Celtic names are still preserved, after the lapse of ages, for the most prominent features of the country. This result, it may be observed, is one that flows from the very nature of things. Even the most fierce and ruthless invaders are compelled to hold sufficient intercourse with the first population to enable them to learn the proper names of their localities, and these names, from obvious motives of convenience, they almost universally adopt.

Now, had Lhuyd shown that the most ancient Local names in Britain areexclusivelyIrish, there can be no doubt that, consistently with the principle just noticed, his theory would have been supported by the facts to which he adverts. But the most ancient local names in Britain are notexclusivelyorprincipallyIrish; in an equal number, perhaps in a majority, of cases they are Welsh.

Moreover, it may be observed that the names of localities in this Island furnish highly instructive evidence, not merely with respect to the different races by whom it has been successively peopled, but also of the order in which they arrived. Thus the names of Rivers and Mountains, and other natural[pg 057]objects, at least of the most conspicuous, are Celtic; the names of the most ancient Towns are Latin, or Latin grafted on British words; more modern Towns and Villages have Saxon appellations; those of more recent origin have frequently Norman designations; and last of all come those places which have names derived from our present English. These various classes of names cannot be nicely distinguished in each particular instance. Of the correctness of the general principle, however, there is no doubt.

But the terms noticed by Lhuyd as significant in the Irish language do not belong to a different class of appellations from those which are obviously of British or Cymraeg origin. The Irish and Cymraeg terms are both found to predominate most in the names of the most ancient Class, viz. in those of Rivers, Mountains, &c., and to be thus applied in conjunction. Hence the natural inference that flows from his facts is not that these names were conferred by two distinct and successive races, but that they were imposedcontemporaneouslyand by the same People!

Further it may be noticed, that if British Topography presents words extant only in the Irish Tongue, Irish Topography also presents names which cannot be explained by means of the Irish, though their meaning is preserved in Welsh; for example: There is a place near the head of a Stream in Roscommon, called“Glan a Modda,”(from Glan,“The bank of a Stream,”Welsh.) There is a place in Wales, called“Glan a Mowdduy.”There is a place called“Glan-gora,”in a Creek at the head of Bantry Bay; and another place in Ireland called“Glan-gort.”

“Ben-heder,”the ancient Irish name for“The Hill of Howth,”interpreted by Mr. Moore“The Hill of Birds.”(Adar,“Birds,”Welsh. The word does not exist in Irish.)

Arran, A mountainous Island. (Arran, a Mountain,Welsh. This word does not exist in Irish,) &c. &c.

Mr. Chalmers in his Caledonia states that the prevalent ancient names of localities in Britain and Ireland are essentially the same.

The conclusions to which these facts legitimately and necessarily lead are, that the British Islands were originally colonized by Settlers, who, at the time of the first occupation of Great Britain and Ireland, spoke one uniform language, in which the Welsh, Irish, and other living Celtic Dialects were combined. We may infer, and I conceive most clearly, that these dialects must be viewed in the light of“Disjecta Membra”of the speech of the old British and Irish Celts, just as the Icelandic, Norwegian, &c. are fragments of the ancient“Danska Tunge,”as noticed in the previous section.

It has been shown by Dr. Prichard that the population of Islands has been derived from the neighbouring Continents, and that the population of the more distant Islands has been derived in like manner from those which are nearer to the common source of migration. It is highly unreasonable to assume that Ireland has formed an exception to this general rule, considering that the common basis of the Irish and ancient British or Welsh languages are confessedly the same, unless it can be proved that the accompanying differences are such as to require the solution Lhuyd has suggested. Here, then, the question arises, are the features of difference between the Welsh and Irish languages more numerous or more fundamental, in relation to the interval of time that has elapsed since the Roman Invasion of Britain, than the varieties of dialect among the Scandinavian nations are in relation to the period that has elapsed since the colonization of Iceland? They are not! It will thence be seen that Lhuyd's theory, as to the remote date of the separation of the Gaelic or Irish from the British or Cymraeg branch of the Celts, is founded on an exaggerated conception[pg 059]of the stability of Human Tongues; and that the abandonment by various septs of different synonymes used conjointly by their common forefathers will satisfactorily account for the differences between the Welsh and Irish, to which he attaches so much weight. It will be perceived, for example, that in the Icelandic, of which the existence commenced in the ninth century, and the Continental Scandinavian from which it then sprang, totally different terms are used for“Water,”the very instance to which Lhuyd especially adverts, as regards the languages of the Welsh and Irish, whom we know to have existed as separate nations in the time of Cæsar eighteen centuries ago!

Another highly instructive test of the correctness of his theory may be derived from the investigations of Lhuyd himself, who, in his comparison of the Welsh and Irish languages, uniformly distinguished the current terms from the obsolete synonymous words that occur only in ancient MSS. This comparison proves distinctly that the Irish and Welsh languages approximate, as we ascend, at a rate which, if as rapid previously as we know it to have been up to the date of the earliest MSS., would imply that these languages must have been identical about the era of the Roman invasion. As the changes which languages undergo in their infancy are more rapid than those which occur at later stages of their growth, it is possible that the unity of these Tongues may be ascribed even to a much later period, an opinion which has been maintained by a very judicious and excellent writer, Mr. Edward Davies, who in his“Claims of Ossian”has published an early specimen of Irish Poetry, which in Language and Style he regards as identical with the most ancient productions of the Welsh Bards. Making every allowance for the irregularity of the changes which occur in Languages, I do not conceive it possible that the Welsh and Irish could have differed very essentially[pg 060]in the time of Cæsar. This leads directly to another conclusion, viz. that the first colonization of Ireland could not have taken place a great many centuries before the Roman invasion. Had such been the case, the differences between the Welsh and Irish Languages must have been proportionately more extensive. In the time of the Romans we learn that an Irish traitor arrived in Britain, who stated that Ireland might be kept in subjection by a single legion, an incident which tends, however slightly, to favour the opinion that the sister Island was at that period but thinly, perhaps because but recently, peopled.

Of the extent of the changes which the Celtic languages have undergone since the first arrival of the Celts in Europe, we possess proofs of far more ancient date than the earliest literary specimens of the living dialects of the Celtic in the Local names of Celtic regions, as preserved in Roman Maps, and in the existing languages of the French, English, and other nations, who occupy countries of which the Celts were the first inhabitants. These names I shall show to consist of three elements: A union of 1, Welsh, Cornish, &c.; 2, Irish, Highland Scotch, &c.; and 3, Terms not extant in any Celtic Tongue, but preserved in the Oriental, Greek, and other languages.

As regards the Names of the 1st and 2d Classes, it will abundantly appear from the ensuing examples that, in the Topographical Nomenclature of Gaul, Britain, and other Celtic regions of Europe,53words derived from all the various Celtic dialects now extant, occur in a manner that leads distinctly to the inference that these“Disjecta membra”must have simultaneously belonged to the language of the old Celts. Dr. Prichard, who has examined these vestiges of the[pg 061]ancient Celtic Populations of Europe with much ability and success, leans to the opinion that the Cymraeg or Welsh Dialects predominate in these names. But the following examples, which comprise many names derived from the Irish or Gaelic that have not been noticed by Dr. Prichard or by previous writers on this subject, will serve to render it manifest that the ancient Names in Europœa Celtica did, in fact, include all the various living Celtic dialects very equally and harmoniously blended.

How luminous and distinct these proofs of the identity of the ancient with the modern Celtic nations are, will be better understood by a preliminary statement of certain rules, which will serve to give greater precision and perspicuity to the illustrations selected:

1. There can be no doubt that the Romans, in the Celtic, as in other countries conquered by them, modified the native terms by the addition of their own peculiar grammatical inflections, as in“Judæ-i, Britann-i, Sen-ones,”&c. Now it is obvious that in identifying the Celtic terms we must reject these mere Roman inflections.54

2. In many cases the Roman Names cannot be supposed to involve complete transcripts of the Celtic Names; frequently they were doubtless convenient abbreviations of the original names—names consisting of descriptive terms to them unintelligible. According to Mr. Reynolds, the Saxons generally adopted the first syllable only of the Roman or British names they found in this island. According to Bullet,“Vic,”a word of Roman origin for a Village or Town, has, from similar causes, become common as a Proper name in Dauphiné; in modern times we have numerous Villages called“Thorpe,”the name for a Village in Anglo-Saxon and[pg 062]German. In instances of this kind, there can be no doubt that originally the names were descriptive, such as“Long-town,”“Old-town,”&c. Tre or Trev is the common Welsh word for a Town, Village, or residence; it had the same meaning in Cornwall:

“By Tre, Tres, and Tren,You shall know the Cornish men.”

“By Tre, Tres, and Tren,You shall know the Cornish men.”

“By Tre, Tres, and Tren,

You shall know the Cornish men.”

A consequence of the names of the gentry of the county having been derived from those of their residences, into which this word commonly entered!

In Wales we have numerous examples of“Tre,”as in“Tre-llwng,”“The Town”of the“Pool,”(i.e. Welshpool,) from an adjoining“Llyn,”or Pool, near Powis Castle;“Tre-lydan,”the Broad Village, or Residence near Welshpool;“Trev-alyn,”near Chester, the Residence on the Stream; the“Alyn,”&c. &c.

Now according to the Roman mode, such a term as Trev-alyn would have been changed into Trev-iri, the designation actually given to the Celts of“Treves,”&c.

The following are analogous examples:

There is a tribe of Brig-antes in Yorkshire, another in Ireland, and a third in the North-east of Spain. Many unsuccessful attempts have been made to show that these distant Celtic tribes must have been scions of the same tribe. A much simpler explanation may be given.

By referring to the Roman maps the reader will find a word,“Briga,”in such general use as part of the names of towns as to leave no reasonable doubt that it must have been, like Tre, a Celtic name for a town—now obsolete. Thus in Spain we have, Laco-briga, Meido-briga, Ara-briga, Tala-brica, Augusto-briga, &c. Now the analogous instances already noticed suffice to point out that the occurrence of[pg 063]Brig-antes as a Roman name of Tribes in three Celtic countries, is a natural result of the frequent occurrence of Briga as the first part of the names of Celtic places.

The“Allo-bryg-es.”The name of this warlike tribe, the Celtic inhabitants of Savoy, has also been the source of perplexity, which may be removed in the same manner. This tribe had a town, called by the Romans“Brig-icum,”which was said to be“the only one they had.”55Now Allo-Bryga may reasonably be identified with Alpo-Briga, the Town of the Alps (Briga being clearly the common base of“Allo-bryg-es,”and“Brig-icum.”)

The names of Celtic communities, as they appear on the Roman Maps, may, I conceive, be proved to have been descriptive of the most prominent natural features of the regions they inhabited, and not of their lineage or descent, as seems to have been often supposed. Thus we have the Mor-ini in Belgium, and the Ar-mor-ici in Gaul on the Sea; we have the Sen-ones on the Seine, the Tamar-ici on the Tamar-is, in Hispania, &c. In the Mountainous regions it will be observed that the names of tribes are derived from the Mountains. In the flat countries they take their names from Rivers or the confluence of Rivers. In the same manner it is highly deserving of remark, that the names of the different French Departments have been derived from precisely the same natural features. Thus in the Hilly countries we have the Departments of the High Alps,“Hautes Alpes;”of the Low Alps,“Basses Alpes;”in the Champaign countries the Departments are named from the Rivers; such as the Seine, the Marne, and the Somme, &c. Many of these French names are literally equivalent to translations of the ancient Gaulish names, as interpreted by means of the Welsh and Irish languages. It is impossible to conceive a[pg 064]more perfect verification of the accuracy of these interpretations!

I may here observe, that as far as we can perceive, the various independent communities of Britain and Gaul mentioned by Cæsar, such as the Edui, the Venetes, &c., did not consist ofoneclan or sept, they seem rather to have been a combination of several contiguous septs, to whom no appropriate common namecould have beengiven, except one derived from the natural features of the district they occupied.

The durability of local names has been already noticed. Of this truth we possess remarkable proofs in those of localities in France, as preserved by the modern French to the present day. I do not doubt that the present French names are, in many instances, much more faithful transcripts of the original Celtic appellations than those which occur in the Roman Maps are. Thus, for example, Bonomia, a name conferred by the Romans upon Boulogne, and of which the origin has perplexed Antiquarics, may easily be explained as a Roman abbreviation of the word Boulogne itself, of which the Celtic meaning will be shown hereafter to be appropriate and unequivocal. Here it may be noticed, that the Celtic language did not become extinct in Gaul until many centuries after the termination of the Roman sway and the establishment of the Franks in that country. The use of the old Gaulish or Celtic continued until the eighth century, nearly until the time of Charlemagne.56Now we know that the modern Welsh and Irish, for the most part, continue to use their own primitive names of localities in those cases in which abbreviations or translations have been substituted by the English. There can be no reasonable doubt that the ancient Gauls did the same, and that these names were in use among the inhabitants[pg 065]of each locality at the time of the final subjugation of Gaul by the Franks, by whom, in many instances, these names are more likely to have been adopted than those used by the Romans.

It will also be observed in the course of the following examples, that names of the class about to be noticed, viz., Topographical names of which the elements are not extant in the existing Celtic dialects, but occur in Oriental words, &c., are remarkably well preserved by the modern French. Thus the“Aube,”as pronounced by the French, is identical in sound with the Asiatic terms for Water, and names of Rivers, to which it is allied.

3. By many, perhaps by all those Celtic scholars who have investigated this subject, it has been assumed that the living Celtic dialects may be expected to furnish a complete clue to all the Local Names of ancient Celtic regions. This conclusion, like the theory of Lhuyd above discussed, is founded on an exaggerated idea of the stability of Human Tongues! Neither the Irish nor the Welsh, nor a combination of all the Celtic dialects, will be found to afford a complete solution of the Topographical nomenclature of the ancient Celtic regions of Europe. Names undoubtedly occur in these countries which have been preserved in none of the Celtic tongues, names which I shall indisputably show to be positive transcripts, in many instances, of appropriate terms occurring in the Hebrew and other languages, with which, in other parts of this work, the original Celtic dialects will be proved to have been originally identical. These facts lead to the conclusion that the ancient nomenclature of Celtic countries forms in reality a connecting link between the existing dialects of the Celts and the language of the Oriental stock from which they are descended.

This conclusion, though at variance with the views of many estimable writers, is nevertheless in unison with those anticipations[pg 066]which historical facts legitimately suggest. It is only reasonable to infer that since the period of their first arrival in Europe, the era at which many of these names must have been conferred (see page10), the Celtic tribes must have lost many words which none of the modern Celtic nations have preserved. The Celts were settled about the sources of the“Ister, and the city,”(perhaps the mountains)“of Pyrene,”even in the time of Herodotus, and how many ages had elapsed since their first arrival is unknown!57

There is a certain Class of terms of which the meaning can reasonably be inferred from their extensive use in combination with other terms, of which the meaning may be considered as ascertained. To this class may be referred the terms immediately following.

Catti, Cassii, Casses, or Cad, seem to have meant a People, Tribe, &c., as in the following examples of the names of Celtic Tribes:

The Abr-in-Catui, in Normandy. The Catti-euch-lani, the people of Cambridgeshire and the adjoining counties. The Cassii, in Hertfordshire. The Bidu-casses, in Normandy. The Tri-casses, a people in Champagne. The Cad-ur-ci, on the Garonne.

The above words seem clearly derivable from the following Welsh words, which are allied to the Hebrew:

Tre, Trev, a Village, Town, or Residence, (Welsh,) a Tribe, (Irish.)58

Trev-iri, the people of Treves. A-Treb-ates, the people about Arras. (For further examples see Dr. Prichard's work.) Trev is a common element in names of places in Wales, as Tre-vecca, Tre-gynnon.

Trigo, to reside, dwell, (Welsh.)

Duro-trig-es, the dwellers on the Water or Sea, the people of Dorsetshire. (Camden.)

Catt uriges. (See Dour.)

Dun-um, a Hill, a Fort or Town, generally on a Hill, (occurs inWelshandIrish.)

Oxell-dunum, a Hill-fort in Gaul, described by Cæsar. (See numerous instances in Dr. Prichard's work.)

“Castell Din-as Bran,”on a lofty eminence in the Vale of Llangollen, Wales.

Dur, Duvr, Awethur (Welsh), Dour (Cornish), Dur (Armorican), Dovar (Irish, obsolete, but occurs in ancient MSS.)“Water.”

This word, and Ydōr or Hudōr (Greek), and Tschur (Armenian),“Water,”have an obvious affinity. These forms may be traced in the names of Celtic Localities.

“Dour”occurs in the following names of Rivers: Dur, (Hibernia,) Dur-ia Major,“The Doria,”and Duria Minor, (Gallia Cisalpina,) Dur-ius,“The Douro,”and“Dero,”(Hispania,) Dur-anius,“The Dordogne,”(Gallia). InBucharianDeriâ means“The Sea.”

Ydōr or Hudōr (Greek), Awethur (Welsh), occur in the[pg 068]Rivers“The Adour,”59Atur-is (Gallia),“The Adder”(Britain),“The Adare”(Ireland.)

“Tschur”(Armenian), occurs in“Stura,”(Gallia Cis.),“The Stour”(Britain),“The Suir”(Britain & Ireland),“The Souro”(Spain, a branch of the Tagus.)

From the frequent recurrence of all these different forms in several Celtic countries thus widely separated, it is plain that they were used conjointly by the early Celts, and represent various transitions of the same word. Thus“Stura”(inGal. Cis.), flows between the neighbouring streams Duria Major and Duria Minor, &c.

This word“Dour”enters very largely into the names of tribes; it forms singly a natural clue to a great number of names that hitherto have been referred to a complication of Roots. Thus the Roman name for the people of Dorsetshire, Duro-trig-es, i.e. The dwellers on the Water or the Ocean, has been noticed by Camden.

In the preceding, and in several of the following, it will be apparent that the old Celts applied this term to the“Sea or Ocean,”as the Bucharians do, and also to a“River.”At present the Welsh apply the term to Water only, in a restricted sense.

In the South-east of England names abound (applied to places on Rivers or the Sea) in which the two slight variations of Dur and Du-v-r (or Do-v-ar,Irish), still preserved in Welsh, are apparent. Duro-vern-um,“Canterbury,”from Duro, Water, and Vern or Veryn, a Hill. (Compare the name of the“Ar-vern-i,”under Beryn, at p. 78.) The Town was on a Hill by the Stour.

Portus Du-b-r-is or Dub-r-œ, i.e.“Sea Port,”the modern[pg 069]“Do-v-or,”a word which is an echo of the Irish Dovar and the Welsh Du-v-r.

Duro-brivæ, Rochester on the Medway, (Briva or Brivis, the ancient Celtic for a Town.) Duro-levum, Milton on the Thames.

Lan-du-b-r-is, a Portuguese Island. Lan, a Bank of a Stream, or the Sea: also an inclosed Space, (Welsh.)

Tur-ones, the inhabitants of the country at the junction of several streams with the Loire, the neighbourhood of the modern Tours.

Bi-tur-ig-es, from Bi“Two,”Tur or Dour, Water, and trigo, to reside.

There are two tribes of this name in Gaul; the Bituriges Cubi, situated between two of the branches of the Loire, and the Bi-turi-ges Vobisci, between the Garonne and the Sea, at the junction of the Dordogne and the Garonne.

Cat-ur-iges, from Catti, Tribes or People; Dour, Water, and Trigo, to reside; on the Durentia, South-east of France, about Embrun or Eburo-Dunum, which was their principal town. Cad-ur-ci, from Catti, Tribes, and dur.

There is one tribe of this name on the Dordogne, and another contiguously placed on the Garonne.

The mutual support that these interpretations give to each other will be obvious.

The following Irish word for“Water,”which is not extant in the Welsh, may be traced in Celtic regions in its various modifications: Uisge (Irish),“The Usk”(South Britain)—Eask60(Irish, obsolete),“The Esk”(Scotland),“The Escaut”(North of France), Isca,“The Exe”(South Britain)—Easkong (Irish, obsolete), Axona (Gallia,Belg.),“The Aisne,”Axones, the neighbouring tribe.

Names Of Estuaries, Or Mouths Of Streams.

The terms of this class, which occur in ancient Gaul, &c., consist either of terms still thus applied in the living Celtic dialects, or of compounds of which the elements may be recognized, unchanged, in those dialects. Moreover it will be highly interesting to observe that these terms, for the most part, consist of Metaphors derived respectively from the same sources as the two English words“Estuary”and“Mouth,”and the two Latin words“Æstuarium”and“Os Fluminis.”

One of the principal arguments of those writers who maintain that the separation of the Irish from the other Celtic tribes must have been of remoter date than the first peopling of these islands, is founded on the fact that the Irish use the word In-ver for the Mouth of a Stream, while the Welsh use Ab-ber (spelt Aber); a feeble support for so wide a conclusion, which a correct analysis of these terms, and a comparison of some interesting coincidences in the local names of ancient Gaul will show to be utterly futile! In-ver and Ab-ber are not simple but compound terms, literally corresponding to the Latin expression“Fluminis Æstuarium.”Æstuarium is from Æstuo,“To boil,”a metaphorical term, obviously derived from the agitation of the Waters where two Streams meet, or where a River enters the Sea.

In the first syllable“Inver”and“Ab-ber”differ, but they agree in the last. Both“In”and“Ab,”the first syllables of these terms, occur so often in Celtic regions that there can be no doubt they were both in use among the ancient Celts as words for a River, or Water. The last syllable of these words, Ber or Ver, I shall show to mean an“Estuary.”

“In”occurs in the name of“The Inn,”in the Tyrol, the“Æn-us”of the Romans, and in other instances previously[pg 071]noticed.“An”is a Gaelic or Irish term for“Water,”which is identical in sound and sense with terms of frequent occurrence among the tribes of the American Continent, as in Aouin (Hurons, N. America), Jin Jin (Kolushians, extreme North-west ofN. America), Ueni (Maipurians, S. America.)

“Ab”occurs in“The Aube,”in France, &c., a name of which the pronunciation may be considered identical with Ab,“Water,”(Persian.) Ap in Sanscrit, and Ubu Obe in Affghan, mean“Water.”“Obe”occurs in Siberia as the name of a well-known river. In India also the term has been applied to“Rivers;”thus we have in that country the Punj-âb, (the Province of“The Five Rivers,”) an appellation of which the corresponding Celtic terms“Pump-ab”would be almost an echo!

Further it may here be noticed—as an example of the complete identity of the Celtic and Oriental languages when all the“Disjecta Membra”are compared—that this word does not exist in the modern Celtic in the simple form of Ab, but in the derivative form of Avon, which is found in the Roman maps spelt“Abon,”&c. Now this form also occurs in the East. Abinn,“A River,”is given by Klaproth from the language of the inhabitants of the Mountains to the North of Bhagalpur. Apem means“Water,”in Zend, an ancient Persian dialect. Af is“Water,”in Kurdish.

Back to IndexNext