Still, the extent to which there are true forces, in certain branches of psychology, and the extent to which they are scientific, so far as they are true, is transparently clear; and, on the ground of this, the consideration of formal conditions has been made equally a matter of science with the consideration of forces and causes. Whether it be so or not, depends upon the definition of the word. It is only certain that between such a science as logic, which is purely formal, and physics and mechanics, which are, in common language at least, the typical sciences, there is as wide an interval as there is between logic the art, and logic the science. Now, Language is in the same order, if not in the same genus, with Logic.

Professor Müller, however, insists on its being a physical science. It is and it is not. It is a science more formal than dynamic-though with dynamic elements. It has more dynamic elements than logic, fewer than the physiology of the nervous system. What it has in common with physics is better shown by example than precept. Let philologers, ethnologists, and archæologists write, muta'is mutandis, after the fashion of geologists and naturalists, and the world will discover their science of its own accord. But it is physical rather than historical. We doubt this. At any rate, no such contrast between history and physics as Dr. Müller has drawn is tenable. If comparative philology have its physical aspect, history has its; and (more than this) it has its able advocates, who are just as ready to make it a science as Professor Müller is anxious to make philology.

Philology is a science; formal in the main, but with physical elements which want working out. Language is a growth, not a manufacture, says (after many others) Dr. Müller. Mr. Buckle (after many others, also) says the same of history. Still, both these gentlemen write ordinary history and ordinary philology. That comparative philology, like history, is more scientific than it was, is true; and what it now most wants is a public, not a public of mere readers, but a public of sufficient number to create a healthy atmosphere of criticism, in which not only absurdity and paradox shall be unable to live, but one from which amateurs shall be expelled, and monopoly impossible. When this is the case with any department of knowledge, it is in working order. It is the condition in which geology is now, but in which it was not at the beginning of the century. It is the condition in which comparative philology will probably be in the year 1900, but in which it is not at present. Towards this happy consummation Professor Müller's arguments do much to contribute, but we differ from him in thinking that much of this improvement is due to any notable discoveries. "Whenever a man," said (or is said to have said) Mr. Canning, "talks of figures, put him down for a sophist." The original term is a stronger one-too strong for ears polite. "Whenever a man talks of induction," say we, "expect an illegitimate generalization; and whenever he talks of discoveries, look out for a mare's nest." That some discoveries, properly so called, have been, is not to be denied; but that many have been directly necessary to the advancement of philology is doubtful. The real progress of philology has consisted in the abolition of old errors. It began with a mass of trash. The accumulation of facts, the improvement of criticism, the

promulgation of certain results in the shape of a rough rule-these it is that have helped it on, and not any brilliant discoveries. Dr. Müller speaks of the discovery of the Sanskrit, but take it in which way we will, it was only a new fact of an old kind. "But why," writes Müller, "it may be naturally asked, should the discovery of Sanskrit have wrought so complete a change in the classificatory study of languages? If Sanskrit had been the primitive language of mankind, or at least the parent of Greek, Latin, and German, we might understand that it should have led to quite a new classification of these tongues. But Sanskrit does not stand to Greek, Latin, the Teutonic, Celtic, and Sclavonic languages in the relation of Latin to French, Italian, and Spanish. Sanskrit could not be called their parent, but only their elder sister. It occupies, with regard to the classical languages, a position analogous to that which Provençal occupies with regard to the modern romance dialects. This is perfectly true; but it was exactly this necessity of determining distinctly and accurately the mutual relation of Sanskrit and the other members of the same family of speech, which led to such important results, particularly to the establishment of the laws of phonetic change as the only safe means for measuring the various degrees of relationship of cognate dialects, and thus restoring the genealogical tree of human speech. When Sanskrit had once assumed its right position-when people had once become familiarised with the idea that there must have existed a language more primitive than Greek, Latin, and Sanskrit, and forming the common background of these three, as well as that of the Teutonic, Celtic, and Sclavonic branches of speech, all languages seemed to fall by themselves into their right position. The key of the puzzle was found, and all the rest was merely a work of patience. The same arguments by which Sanskrit and Greek had been proved to hold co-ordinate rank, were perceived to apply with equal strength to Latin and Greek; and after Latin had once been shown to be more primitive on many points than Greek, it was easy to see that the Teutonic, the Celtic, and the Sclavonic languages also, contained each a number of formations which it was impossible to derive from Sanskrit, Greek, or Latin. It was perceived that all had to be treated as co-ordinate members of one and the same class." We submit that this mother-tongue in the background was no result whatever of the so-called discovery of the Sanskrit, and that it was as easily got at through the Greek, Latin, Sclavonic, Lithuanian, and German, minus the Sanskrit, as with it. The fact of its lying behind a language more or less was unimportant. A common grandmother is inferred from two first cousins, just as well as from two hundred.

From another point of view the Sanskrit is interesting. It was an isolated language, lying apart from and in discontinuity with its congeners. But it was no new fact. The Magyar was in the same category. What Sanskrit really was, was this-it was a stimulus. It excited wonder; so much so, that many modern philologues have scarcely got over the shock. What is wanted in philology is the accumulation of facts, order, and common sense. In all these there has been an improvement, and the quiet collection of new details has been great. Half the discoveries, however,

are personal-merely demonstrations that some earlier writer was wrong. Almost all the so-called generalizations, in the way of classification, are merely corrections of some one's ill-considered limitation. Again, most discoveries in comparative philology are joint-stock affairs. A. sends from some outlying locality some new specimen of some new language; B. finds a place for it.

66

According to Professor Müller, the Sanskrit was spoken in India in Solomon's time. Perhaps the names of the articles obtained from Ophir he regards as a proof. They have a meaning in Sanskrit ; "If," he observes, we wished to know from what part of the world gutta percha was first imported into England, we might safely conclude that it came from the place where the name gutta percha formed part of the spoken language." No. We must also know that it was a product of that country. But this exception we waive, and again say, no; and we must also know that there was only one country where it was in use. The Sanskrit is the "language of languages." This is the speech of the beleagured leather-cutter to the municipal Vaubans. In the higher philology, in the deduction of inflections from roots, and of roots from something else, what has it done? Are there five points settled by it which would not have been settled independently of it? We mean by its direct influence, and the new facts it gave. Indirectly, it has done much. When we poke the fire we make a blaze; but the elements of heat are just what they were before. India awoke Schlegel, who encouraged Lassen, who made many disciples. Are there ten men who agree, within ten degrees, as to where Sanskrit was first spoken; as to when it began and ended; as to whether it is free, or not free, from exotic elements; as to whether, in the extreme form in which the grammarians give it, it was ever spoken at all? Is there any proof that it was a language adapted to the common use of common life, a language for a good prose literature, a language in which a historian could write a plain narrative, or an orator deliver an effective speech? For what high question is it a high and undeniable starting-point? What can we infer from it without some undue assumption? Is it not a language to argue about, rather than a language to argue from?

Its locality in India was the fact that arrested attention.

It was like the footstep in Robinson Crusoe's island; nothing elsewhere, but there everything. Had the Sanskrit been found in Hungary it would have been put at what it was worth.

"A primrose on the river's brim,
A yellow primrose was to him;
And it was nothing more."

But it was found in India. and

It was not merely a fly, but a fly in amber;

"We wondered how the mischief it came there."

With as good men as himself, and with better men than the writer, and with ninety-nine hundredths of the waλ, he deduces the Celtic, the

German, the Lithuanic, the Sclavonic, the Latin, and the Greek, from the parts about the present locality of the Sanskrit literature, rather than the present locality of the Sanskrit literature from the existing areas of the German, the Lithuanic, and their congeners; but new friends, the biolo gists, might have taught him that to deduce the order from the locality of the genus, and (à fortiori) of the order, is bad zoology.

We distrust our own judgment, when we differ from Professor Müller as to the merits of the German school of comparative philology. We make no secret of being tired of the encomiums which it is continually passing on itself. When Prior was asked by Louis XIV., whether Kensington Palace could boast of such pictures of such victories as adorned the walls at Versailles, the answer was, 66 No, sire. The memorials of the great things which my master has done, are seen in many places; but not in his own house." What the Germans understand better than the rest of the world is the doctrine of the letter-changes; and those who understand it best have in many cases used it worst. There is nothing in the wildest works of the pre-scientific period which is wilder than parts of Bopp's "Dissertation on the Georgian," the chapter in the "Deutsche Sprache" which treats of the Dacian language, and such an article in Zeuss as his one on the word "Niuthones." We do not press the doctrine, that in philology, as in mechanics, nothing is stronger than its weakest point. Those, however, who do, must demur to praises so liberally bestowed both on the above-named writers and others. On the other hand, to more than one name, more distinguished than many of those mentioned by Dr. Müller, scant justice seems to be done. This, however, is a matter on which we may err. Schlegel may have done all that is claimed for him. He may, possibly, like a greater poet, have promoted philological, as Göethe advanced biological science. We have generally, however, looked upon his work as an advertisment, rather than aught else; and upon his classification of languages, as a mere resumé of the characteristics of their limited groups— that represented by the inflected languages of Europe, that represented by the Arabic and Hebrew, and that represented by the Chinese. However, the credit for the work done by the philologers of other schools awarded by Professor Müller is so irregular that the notice of it had better have been omitted altogether. His doctrine as to the origin of roots is somewhat strange. "Analyse any word you like, and you will find that it expresses a general idea peculiar to the individual to which the name belongs. What is the meaning of moon?-the measurer. What is the meaning of sun?-the begetter. What is the meaning of earth?-the ploughed. The old name given to animals, such as cows and sheep, was pásu, the Latin pecus, which means feeders. Animal itself is a later name, and derived from anima, soul. This anima, again, meant, originally, blowing or breathing, like spirit from spirare, and was derived from a root an, to blow, which gives us anila, wind, in Sanskrit, and anemos, wind in Greek. Ghost, the German geist, is based on the same conception. It is connected with gust, with yeast, and even with the hissing and boiling geysers of Iceland. Soul is the Gothic saivala, and this is clearly related to another Gothic word,

VOL. III.

B

saive, which means the sea. The sea was called saivs, from a root si, or sio, the Greek seio, to shake; it means the tossed about-water, in contradistinction to stagnant or running water. The soul being called saivala, we see that it was originally couceived by Teutonic nations as a sea within, heaving up and down with every breath, and reflecting heaven and earth on the mirrors of the deep." "Analyse any word you like"; well and good; but the words that follow are those that Professor Müller likes—we should like the analysis of I, buzz, and whizz. One of the weak points of the hypothesis suggested by the extract we have given is that it ignores the pronouns, some of which, in spite of their name, are the very oldest words in language. Men pointed to objects, and said this or that (or their equivalents) as soon as they did anything. Another defect is its partial character. The words quoted are simply individual instances, and, instance for instance, "whizz" and "buzz" are, at least, as clearly connected with imitation as soul" is with "sea" and "seiō." The real question is the value of the class that these particular instances represent. That they represent different classes is clear, not one of which is co-extensive with Language. Of Language's vast domain, the class represented by certain undoubted and undeniable imitative words covers one portion; the class represented by certain undoubted and undeniable interjections covers another; and the class represented by "ghost" and "soul" may possibly (for we do not even admit the etymologies as a matter of course) cover another. Which covers the largest space is to be determined by care and counting, not by hard words and bold assertions.

The details of the doctrine are, moreover, hasty-over hasty. If "moon" comes from the root meaning "measure" what does the root meaning "measure," come from? If "sun" is synonymous with "begetter," what is "beget ?" This is the elephant on the back of the tortoise. Are we really asked to believe, that with a sun to shine by day, and a moon to shine by night, the men and women, whose senses were struck by their light, had no names for them until the intellect arrived at such notions as measurement and generation? Had inland Germans no name for soul until they saw the sea, or the sea, in the mouth of an islander who swam and fished in it, no name until men talked of shaking?

Upon the merits of Professor Müller's philological labours it is needless to pass an opinion. No work is so unexceptionable as to preclude differences of opinion on unsettled and unstudied questions. On many of these we unwillingly differ, not only from Professor Müller, but from the whole school, in which, as a German, he takes his pride. When such is the case, a statement of objections is better than indiscriminate eulogy. It would be no compliment to say that his work on the science of language is the best of its kind in our language-perhaps in any. It might be this, and yet be a bad one. We might almost say that it might, at one and the same time, be both the best and worst; so little can its value be tested by its relative merits. It would, however, be a book of mark if the literature of comparative philology were tenfold what it is in mass, and double what it is in value; and Professor Müller will be a philologer of the first class, even