It is the headmasters of schools like Eton and Winchester who tune the preparatory schools; the latter, whose prestige depends on winning entrance scholarships, are compelled to adapt their general course of studies to suit the papers set in these examinations. So long as four-fifths of these scholarships are assigned to classics and mathematics, modern languages, including English, will necessarily take a back seat. We need not have studied the "Soul of the Child," or attended Miss Mason's Ambleside Conference to convince ourselves of the folly of attempting to teach three foreign languages at once to a child of ten or eleven. It is not a question of the relative merits of French and Latin, or of German and Greek, and as the lesser of two evils, we would far sooner see modern languages deferred to a later stage than begun simultaneously with two dead languages, and not taken seriously. But as it is certain that only the few elect will carry on their Greek beyond the beggarly elements, while all secondary pupils may be reasonably expected to reach the stage of reading a French author with ease and profit, it follows that French should take precedence of Greek at least in point of time. If half a dozen of our great headmasters were to put their heads together and agree that henceforward Greek should not be set in entrance scholarship examinations, they could by a stroke of the pen effect a far-reaching and, in our judgment, a most salutary reform in secondary education. Nor do we believe that classical studies would thereby suffer. The Frankfort experiment has conclusively proved that no loss accrues from beginning Latin at à later stage, and the foremost champions of Greek in England-Mr. S. H. Butcher and the late Professor Jebb-have expressed themselves in favour of not beginning Greek before the age of fourteen.

A discussion of methods does not, as we said,

fall within our brief, and all we can attempt is to estimate actual results. In the first stage there can be no dispute that by the Direct Method the pupil's progress is greatly facilitated and quickened. He learns approximately as the infant learns the mother tongue, and he can from the very first apply his acquired learning; he has a sense of new power. There may be some loss of accuracy, but accuracy is not a childish virtue, and in the long run, so those who have made trial of both systems assure us, he will outstrip his scholastically trained competitor. Talking naturally comes before reading, reading comes before writing, and still more before grammar. But there are dangers in the Direct Method, and the zeal of the new reformers sometimes outruns their discretion. It is indeed essential to secure from the outset correct pronunciation, and to convince the pupil that French is a living language which he may turn to practical uses. But, after all, a pure Parisian accent is not the one thing needful, and endless imitation is not the scholar's one vocation. There is a tendency for the teacher to turn himself into a phonograph, or, literally, to make a phonograph take his place. In a recent report of a public examination board it was laid down by the French examiner that conversation was the crown and summit of French teaching, the supreme test of a knowledge of the language. Such a view is, in our judgment, utterly mistaken, and it gives some colour to Professor Kennedy's gibes at "A Courier Tripos." As far as the Medieval and Modern Language Tripos at Cambridge is concerned it has indeed no point, for at starting there was no viva voce test whatsoever, and even now conversation is not obligatory, but it does give a handle against those who are advocating the admission of a modern language as an alternative to Greek in the first University examination. It is, we are afraid, a fact that standard French

authors (and the same holds of German) are less read in schools than they were ten years ago, and it is quite possible for a boy to leave the sixth form of a public school at nineteen without ever having read a play of Molière or Racine, of Lessing or Goethe. Lists of educational publishers show that there is a steady and increasing demand for recent French fiction, and of About's "Roi des Montagnes" there are a dozen rival school editions. The French and Germans from whom we borrowed the method have not fallen into this extravagance, and no boy could go through a Lycée or a Gymnasium without having read a play of Shakespeare.

It is worth while in passing to enter a protest against the absolute neglect of French prosody in English schools. We are not recommending the introduction of a treatise like Kastner as a text book, still less the composition of French verse in imitation of classical methods. Where Mr. Swinburne has failed it is not likely that the schoolboy or even the undergraduate will succeed. But it is monstrous, to give what must be a common experience of inspectors and examiners, for a sixth form to have been occupied for a whole term with a play of Rostand, and not to know at the end that it is written in Alexandrines.

If we must begin by familiarising our pupils with the spoken phrase as the easiest and most effective way of giving them entrance to the foreign tongue, it is none the less true that the end to be kept in view is the power to read and understand, not to converse.

Books are always with us, but not one in ten of our pupils, even if he has the ability, will have the opportunity of conversing with educated foreigners except at rare intervals. Whether he is a scholar or a theologian, or a man of science or of letters, or a leader of industry, he cannot keep abreast with the

thought and discoveries of the times without a knowledge of French and German. Of all the satiric touches that George Eliot put into her portrait of Mr. Casaubon none is sharper than the bare fact that he did not know German.

And this leads us to remark on the lamentable neglect of German in English schools. While French has been progressing, and virtually every boy and girl in a secondary school now learns French in a way, German has actually retrograded. To give a single instance: at the last Matriculation Examination of the University of London while over 90 per cent. of the 3,000 candidates took French, less than seven per cent. took German.

There is a general agreement among educators, though with some distinguished dissentients, such as Mr. H. W. Eve, that French shall be the first foreign language learnt, and if Latin is still retained, as it is on most modern sides, German cannot be begun with profit before the age of 16 or 17. But there are signs that Latin is losing its prerogative, and teachers are beginning to recognise that the first steps in German, as a highly inflected language, present exactly the same difficulties as the first steps in Latin with some difficulties, such as the order of words, that are peculiar to itself. We may at least hope that the arrangement which now commonly prevails by which German is made an alternative with science will be altered. The complaint of Dr. Clifford Allbutt and the professors of science at both Universities, that the schools send up pupils as ignorant of German as they are of Greek, should not be unheeded by our headmasters.

In conclusion we may say that the fortune of modern language teaching in schools lies with the Universities. By the institution of a tripos or schools they have done much to encourage modern languages, though professors are yet to seek. They have to

some extent modified their joint board and local examinations to suit the reformers, though formal grammar still plays too prominent a part. But they have not yet seen their way to meet the resolution of the Modern Language Association, that no examination in modern language can be considered satisfactory which does not include a viva voce test. A yet more important reform is called for, the inclusion of either French or German as an obligagatory subject in the entrance examination. Modern languages would then cease to be looked on as an extra, to be dropped by all boys who are in the running for a scholarship. If all Oxford and Cambridge students are required to present both Latin and Greek it is surely not extravagant to demand that they should also be tested in the rudiments of French or of German.