nothing more, and I never held myself in any way responsible for it.

Nor did I wish to attach any mysterious meaning to the purely preliminary definition which I gave of roots, by calling them 'phonetic types.' I might have called them phonetic moulds, or typical sounds, as well as phonetic types; and all that I wished to convey by this expression was that roots are like firm moulds in which all words are cast; that they are like sharply cut types of which numerous impressions have been taken; that, in fact, every consonant and every vowel in them is settled, and that therefore no etymology is admissible which does not account for every link in that long chain of changes which connects, for instance, the Sanskrit root vid, to know, with the English adverb historically. It is the definiteness of these roots which alone has imparted definiteness to etymological research, and it was this important characteristic, their definiteness, which I wished to impress on my hearers by using the name of phonetic types. In etymological researches it matters little what opinion we hold on the origin of roots, as long as we agree that, with the exception of a number of purely mimetic expressions, all words, such as we find them, whether in English or in Sanskrit, encumbered with prefixes and suffixes, and mouldering away under the action of phonetic corruption, must in the last instance be traced back, by means of definite phonetic laws, to these definite primary forms which we are accustomed to call roots. These roots stand like barriers between the chaos and the cosmos of human speech, and they alone prevent that 'ugly rush' which

would follow, and which has followed, wherever words have been derived straight from imitations of the sounds of nature or from interjections.

There is, no doubt, a higher interest which leads the philosopher to inquire into the nature of these phonetic types, and tempts him to transcend the narrow limits of the purely positive science of language. I value as much as any one the labours of Mr. Wedgwood and the Rev. F. W. Farrar in their endeavours to trace the origin of roots back to interjections, imitations, or so-called vocal gestures. I believe that both have thrown much light on a very difficult problem, and as long as such researches are confined to the genesis of roots, without trenching on etymology in the ordinary sense of that term, I mean, on the formation and the history of words, Mr. Farrar is quite right in counting me not as an opponent, but as a neutral, if not an ally.

ST. IVES, CORNWALL:

20th Sept. 1866.

M. M.

PREFACE TO THE SIXTH EDITION.

IN revising once more the two volumes of my

Lectures on the Science of Language, I have fully availed myself of the help and counsel of my numerous reviewers and correspondents. As my Lectures were reprinted in America, and translated into German, French, Italian, Hungarian, and Russian, the number

of reviews, essays, and even independent books which they have elicited has become considerable, and the task of examining them all was not an easy, nor always a grateful one. Yet I have but seldom read a review, whether friendly or unfriendly, without being able to correct a mistake, or without feeling called upon to improve a sentence that had been misunderstood, to soften an expression that had given offence, to insert a new fact, or to allude to a new theory. Although my general views on the Science of Language have remained unchanged, the mere number of pages will show how many additions have been made, while a careful reader will easily discover how much has been changed, and, I hope, improved in my Lectures since they were first delivered at the Royal Institution in 1861 and 1863.

Though I have protested before, I must protest once more against the supposition that the theory on the origin of language which I explained at the end of my first course, and which I distinctly described as that of Professor Heyse of Berlin, was ever held by myself. It is a theory which, if properly understood, contains some truth, but it offers an illustration only, and in no way a real solution of the problem. I have abstained in my Lectures from propounding any theory on the origin of language, first, because I believe that the Science of Language may safely begin with roots as its ultimate facts, leaving what lies beyond to the psychologist and metaphysician; secondly, because I hold that a theory on the origin of language can only be thoroughly treated in close connection with the theory on the origin of thought, i. e. with the funda

mental principles of mental philosophy. Although in treating of the history of the Science of Language I found it necessary in my Lectures to examine some of the former theories on the origin of language, and to show their insufficiency in the present state of our science, I carefully abstained from going beyond the limits which I had traced for myself. Much has been written during the last ten years on the origin of language, but the only writer who seems to me to have approached the problem in an independent, and at the same time a truly scientific spirit, is Dr. Bleek, in his essay Über den Ursprung der Sprache, published at the Cape in 1867. I am not surprised that his essay should have been received with marked favour by the most eminent physiologists, but I think, nevertheless, that in the minds of philosophical readers it will leave a strong conviction that researches into the origin of language transcend the domain of the physiologist as well as of the philologist, and require for their solution a complete mastery of the problems of psychology. At all events it seems now generally admitted that a mere revival of the mimetic or onomatopoeic theory on the origin of words would be an anachronism in the history of our science. That Mr. Darwin in his fascinating work 'On the Descent of Man' should incline towards the mimetic theory is but natural, though it seems to me that even if it were possible to revive the theories of Demokritos and Epikuros, language, articulate and definite language, language derived, as it has been proved to be, not from shrieks, but from roots, i. e. from general ideas, would still remain what I called it in my first course of Lectures,

our Rubicon which no brute will dare to cross (vol. i. p. 403).

On other points I think that those who have done me the honour of carefully examining and freely criticising my Lectures will find that not one of their remarks has been neglected; and I can honestly say that, where I have retained my own opinions against the arguments of other scholars, it has not been done without careful consideration. In some cases my critics will see that I have given up positions which they had proved to be no longer tenable; in others, I have indicated, by a few additional words, that I was prepared for their objections, and able to meet them; in others, again, the fact that I have left what I had written without any change must show that I consider their objections futile. It would have been easy to answer some of my rather over-confident critics, and I confess it was sometimes difficult to resist the temptation, particularly when one finds oneself blamed, as happens not unfrequently, for having followed Copernicus rather than Ptolemaeus. Οψιμαθεῖς quan sint insolentes non ignoras. But controversy, particularly in public, is always barren of good results. I can now look back on five and twenty years of literary work, and whatever disappointment I may feel in seeing how little has been done and how much more remains to be done, and probably never will be done, I have at least this satisfaction, that I have never wasted one hour in personal controversy. I have grappled with opinions, but never with their propounders; and, though I have carefully weighed what has been proved against me, I have never