borrowed by the Goths from the Celtic language, and vice versá; or perhaps adopted by each of them from some third language radically different from them both. Thus, from the Welsh Tád, our vulgar have got the common English word Dad and Daddy. And, from the French Delivre, are derived both the English Deliver, and the Armoric Diluir, whence the Cornish Dilver.

Before I conclude these slight remarks, I must beg leave to observe, that as the great subject of this present book is Gothic antiquities, which I apprehend to be totally distinct from the Celtic, I do not take upon me to decide on any of the points which relate either to the Celtic antiquities or Celtic tongues. For this reason I avoid entering into the dispute, which has of late so much interested our countrymen in North Britain: viz. whether the Erse language was first spoken in Scotland or Ireland. Before the inquisitive reader adopts either opinion, he would do well to consider many curious hints, which are scattered up and down in Lluyd's most excellent Archæologia Britannica, 1707, fol., and especially in his Welsh and Irish prefaces, translated in the appendix to Nicholson's Irish Historical Library, &c., 1736, folio.

In reply to those who contend that the true name of the Erse language is Gaelic or Galic, and that this word is the same with Gallic, the name of the ancient language of Gaul, I will merely observe, without deciding the question as to the origin of the Erse language itself, that the ancient name of Gallic does not seem to have been used by the natives of Gaul themselves, but to have been given them by foreigners. They called themselves Celta, and their language Celtic *; in like manner as the inhabitants of Wales, though called Welsh by us, term themselves Cymru, and their own language Cymraeg; who at the same time call us Saissons, and our tongue Saissonaeg, thus reminding us of our Saxon origin.

Qui ipsorum lingua Celta, nostra Galli appellantur. Cæsar de Bell. Gal. L. 1. "Celta, the Gauls, Cadil, Cadil, or Keill, and in the plural, according to our dialect, Keiliet, or Keilt, (now Guidhelod) Irishmen. The word Keilt could not be otherwise written by the Romans, than Ceilte or Celta." See Lluyd's Irish Preface, p. 107, in Nicholson's Irish Historian.

In the same place the reader will find many of the ancient names of offices, persons, &c., mentioned by Caesar as prevailing in Gaul, explained from the modern Irish language, as, Allobrox, Divitiacus, Vercingétorix, Vergasillaunus, Vergobretus, &c,

22

REMARKS ON BISHOP PERCY'S
PREFACE.

BY THE EDITOR.

PROFESSOR RASK, in the Introduction to his Icelandic Grammar, observes that, "after Bishop Percy's most excellent Preface to Mallet's Northern Antiquities, the Teutonic and Celtic languages can no longer be confounded, nor comprised under the vague and unmeaning appellation of Scythian, Sarmatian," &c. Since the publication of the learned Danish Professor's work thirty-five years have elapsed, during which period the study of glossology, or comparative philology, has made as rapid a progress as that of biology, or comparative anatomy and physiology. By the latter we have become acquainted with the organization and affinities of animals; by the former, with the construction and relationship of languages. At the present day a zoologist, by the mere inspection of a few fossil bones, will seldom be at a loss in ascertaining to what description of animal they belonged, and, if the animal be of an unknown species, what place he ought to assign it in the reticulated chain of organic existence. A glossologist, in like manner, by subjecting to a critical examination the few literary remains of some ancient idiom, which a lucky accident may have preserved from oblivion, will not fail to reconstruct them into a language more or less perfect in all its parts, and point out its near or remote affinity to well-known cognate tongues. When a writer of the 17th or 18th century attempted to do any thing of the kind, he was sure to be led astray by some vague theory or other which we should now deem unworthy of serious consideration, or, like a person threading the mazes of a sylvan labyrinth, would frequently turn aside when on the point of entering the right path and wind

*Fublished at Copenhagen in 1811.

round the object he was in quest of without ever being able to at tain it. Cluverius and his followers thus perceived that there was a certain affinity between the Celtic and Teutonic languages, but, in order to account for this affinity, had recourse to a gratuitous assumption on which they founded arguments that were necessarily erroneous and inconclusive. No one could have refuted these arguments with greater perspicuity and force of reasoning than Bishop Percy has done in the preceding admirable dissertation; but, while he completely demolishes the crude theory of Pelloutier, he falls himself into the error of denying that any, even the most distant, resemblance" exists between the languages of the Celtic and Teutonic nations. Upon what grounds then," he adds, " can it be retended that the ancient languages of Gaul and Germany flowed from one common source? or who will believe so improbable a fact?"+

Now this fact, improbable as it appeared to Bishop Percy, has been admitted by the greatest philologists of the present age. No one certainly will any longer hesitate to regard the Celtic and Teutonic languages as forming two distinct linguistic families; but we think sufficient evidence has been adduced by those who have thoroughly investigated the subject, to warrant the conclusion that these families are remotely cognate, and, with six other linguistic families, do really

*These writers could manage well enough to draw up a tolerable grammar from the remains of an ancient tongue, but they generally thrust the language thus reconstructed into a wrong place. This was the case with the Moso-Gothic. The reader will find, by referring to one of Bishop Percy's notes to chap. xi., that the fragments of a translation of the Gospels made by the Gothic Bishop Ulphilas in the fourth century, were discovered in the sixteenth century, in the library of a Westphalian convent. Now Hickes, Lye, and other celebrated English philologists of the last century, confounded this Gothic idiom with the Anglo-Saxon. The first part of Hickes's "Thesaurus," published in 1705, consists of an Anglo-Saxon and Meso-Gothic Grammar," a work," as Rask observes, (see the Preface to his " AngloSaxon Grammar," Cop. 1830,) "far from faultless, as well by reason of the unfortunate idea of treating the two most dissimilar of the Teutonic tongues together, as in the execution of its respective parts." In 1772 Lye published his "Dictionarium Saxonico et Gothico-Latinum," in 2 vols. fol., in which we observe, to use Rask's words, "the same unfortunate blending of Anglo-Saxen and Meso-Gothic, languages which no more admit of being treated together than Hebrew and Arabic, or Greek and Latin."

+ See page 18.

flow from one common source." We shall not reproduce the arguments brought forward in support of this opinion, as we should be obliged to enter into a philological discussion, which for most of our readers would be totally devoid of interest; but as frequent mention will be made in this work of Scandinavians, Saxons, Goths, and other Teutonic nations, we shall attempt to point out, as briefly as possible, the precise relationship they stand in to each other, and conjointly, to other nations or races sprung from the same 66 common source," referring the reader to the works of Rask, Schlegel, Grimm, Klaproth, Bopp, Arndt, and other eminent writers of the German school of philology, for more ample information on the subject, which he will find, on closer examination, to be well deserving of his attention.

Among the numerous sciences which the researches of the present age have given rise to, is one which, for want of a better name, has been called ethnology. This science falls into two branches-a physiological branch, which might appropriately be termed anthropology; and a philological branch, which forms the science of glossology. Anthropology shows the organic distinctions that constitute the varieties of the human species, inquires how these varieties have originated, whether they be reducible to one common type, or to several distinct types, strives to trace the affinities that connect them, and form a systematic classification of the various races that have hitherto appeared on the face of the globe. Glossology, on the other hand, investigates the construction and affinities of the various languages spoken by mankind, from the earliest that have left any vestiges of their existence down to those of the present day, assumes that a certain number may be regarded as primitive, from which all the others are derived, points out clearly this derivation, and then strives to connect the primitive languages themselves by tracing any philological affinities that may exist between them, with a view both of co-ordinating them into a systematic arrangement, and of ascertaining whether they have all sprung from a common source-from a primordial tongue-or constitute a number of glottic groups totally unconnected. It will thus be seen, that although each of these sciences establishes a separate class of facts, they have essentially the same object in view, and may therefore be subordinated to a higher science, which by

investigating and comparing the facts they respectively furnish, will ascertain how far physiological and philological affinities may be made to coincide-whether, for instance, several nations speaking cognate languages, but remotely connected with any other known tongue, are also distinguished by physiological traits sufficiently prominent to constitute a distinct race. A higher class of facts will thus be elicited, from which we may obtain by induction all the truths ethnological science can make us acquainted with.

It is obvious, however, that this science cannot be cultivated with success until the subordinate sciences of anthropology and glossology have attained a certain degree of perfection. Anthropology, unfortunately, is still in its infancy. Man has certainly been made the object of a special study in every quarter of the globe, but it is only since a very recent period that this study has been conducted on scientific principles. The facts elicited are therefore not sufficiently numerous to form the groundwork of a well constructed system or warrant any general conclusion that might be drawn from them. We may, however, take for granted that man-the genus homohas only one species-which zoologists have been pleased to designate by the somewhat equivocal appellation of homo sapiens; or, in other words, that no specific difference exists among mankind. But this fact, even if it were established with a more rigorous certainty than it is at present, does not by any means solve the problem of the origin of the human race; for similar causes operating on two or more points of the globe, under similar circumstances, would necessarily produce similar results. Be this as it may, the varieties of the human species are endless, a circumstance which renders an accurate classification of them as difficult a task as a scientific arrangement of the animal kingdom itself.

When zoologists fancied that the scale of being ascended in a straight line from the lowest radiary animals to man, classifications were comparatively easy, but when they at length discovered that instead of a chain it formed a most intricate network, in some places ravelled, in others rent, the task became much more difficult. They found that their classes and orders were connected by intermediate links, often too imperceptible to be seized, and that nature disregarded these artificial distinctions invented to assist the memory and enable