mere appanage, but an integral part of the whole composition. Bagehot's well-known summary of Enoch Arden 1 is an excellent example of how a summary may be subordinated to the central idea. Another common way to produce a scattering effect is to use the term "some" as a qualifying adjective to the title: out of a complete and possible ten, say, topics connected with the subject, you may use at random numbers, 5, 3, and 8 a thing which happens in many themes.

The only possible motive for mentioning these and other typical faults which will occur to every experienced teacher, is to aid in the avoidance of them, to help the student to think more clearly. The only safe assumption in the teaching of composition is that the young writer has something to say which he wishes to say to somebody. To train him to express his idea and to express it in a way that somebody else will understand and be interested in is, of course, the only end of instruction in composition, that is, after the most elementary training is done.

A word, therefore, of a more positive kind may be added. In single themes of a critical sort, it is well to pin the student down to definite answers to the three immemorial questions of Coleridge: What has the author tried to do? How has he done it? Is it worth doing? The answers will involve a good deal of thinking, and considerable additional skill will have to be employed to make them compose into a fluent and solid piece of work. They admirably serve to put a writer into leading strings and to give him his structure. They are also sound, in that they take into account the author's point of view in criticising his work.

A more extended program may be offered to advanced students. It is not a bad plan subject, of course, to many modifications of detail to make the study of one author for each student the basis of a term's writing. The author should naturally be one for whom the student has some previous liking, and he should be of medium size. Shakespeare, Dante, Milton, are altogether too large and too much has been said about them. On the other hand it is doubtful if luminaries of the magnitude of Mrs. Hemans, "Barry Cornwall," Allan Ramsay, Eugene Field, E. R. Sill, even Holmes, are sufficiently bright to lighten the way of most students over the trackless path of a term of months. DeQuincey, Lowell,

1 Literary Studies, Vol. II; Wordsworth, Tennyson, and Browning; or Pure, Ornate, and Grotesque Art in Poetry. Cf. G. R. Carpenter and W. T. Brewster, Modern English Prose.

George Eliot, and such writers, where they are not too much talked about, are more ponderable. It is a wholesome practice, by way of introduction, to ask each student, without referring to any book of comment, to set down, in a preliminary theme, what he knows or deems it essential to say about the author he has chosen. There should properly follow a compact biography of the author, a plain matter of ascertainable fact, well arranged and divided, without criticism. This is no easy task; for biographies by young writers are likely to be top-heavy and lumpy. A third essay might properly be a classification of the author's works, with a view to bringing out the forms that he uses, their relative importance, and the range of his ideas. It may be remarked in passing that literary classification is a stumbling-block to many writers. It seems easy, but to find, in practice, some fit scheme for bringing out the ideas and forms of an author is no such matter. To name a type, properly to characterize and illustrate it, and to list the specific writings that fall under the class - the essentials of good classification in literature is often very baffling. Such classification may be based on the author's life, as with Lamb and Addison, whose careers were experiments in various literary forms, of which one was eminently successful; it may be based on the occasion of his writing, as with Swift, who was very nearly uniformly successful in all that he did after he was once started on his literary way; it may be a matter of substance, as with the somewhat elaborate classification of DeQuincey's writings in this volume. There are other appropriate ways.

With a good classification as a basis, a variety of possibilities offers itself. A fourth theme may be written on a man's ideas, if the intellectual side is the stronger, or on his quality if it is his literary feeling that predominates. That which distinguishes him from other writers of his class, intellectually and spiritually, is surely a thing worth exposition. Another important source of material for a theme is found in the author's literary art, his method of approaching his task, his style, considered as a combination of phenomena. What things are characteristic and constant in the writings of Arnold, or Keats, or Landor? Naturally discussion of these points tends to run off into questions of quality, but the two may approximately be kept apart. Any criticism that the student has to offer, either by way of personal impression or impersonal discussion, is a good subject for another essay. Here, experience shows, students are likely to forget what

they have been talking about in their preceding themes: in biography, classification, and account of quality, a student may have shown George Eliot, say, to be a great moralist; and yet the criticism may have nothing to do with the ethics of George Eliot but may deal with the irrelevant question of the mechanics of her verse. In short, one should criticise along the lines indicated by the classification and not abjure all preceding labour and knowledge. With regard to another theme, it is most important of all that a student should learn to state, just as a plain matter of fact, what is the vogue, the estimation, the place, etc., in which his author is held. Such "collective" criticism requires considerable research, but is a most necessary check to one's own judgment.

Any special program is, of course, merely by way of illustration and suggestion; the main point is that young writers will avoid the production of rubbish in criticism, only by following sound expository and argumentative methods. The good critic, like other good men, is doubtless more born than made; but there is no real reason why any painstaking student may not learn clearly, adequately, and in an interesting way, to express the faith that is in him. If the foregoing argument is sound, the fact that criticism is largely nothing more than the expression of personal, often temperamental, opinion, checked, for the better part, by historical and rational tests, this fact should make the young critic more confident of his own views and, at the same time, more willing to modify them and to test them.

I

LESLIE STEPHEN

(1832-1904)

WOOD'S HALFPENCE

[Chapter VII. of the Life of Swift in the English Men of Letters Series]

In one of Scott's finest novels the old Cameronian preacher, who had been left for dead by Claverhouse's troopers, suddenly rises to confront his conquerors, and spends his last breath in denouncing the oppressors of the saints. Even such an apparition was Jonathan Swift to comfortable Whigs who were flourishing in the place of Harley and St. John, when, after ten years' quiescence, he suddenly stepped into the political arena. After the first crushing fall he had abandoned partial hope, and contented himself with establishing supremacy in his chapter. But undying wrath smouldered in his breast till time came for an outburst.

No man had ever learnt more thoroughly the lesson, "Put not your faith in princes;" or had been impressed with a lower estimate of the wisdom displayed by the rulers of the world. He had been behind the scenes, and knew that the wisdom of great ministers meant just enough cunning to court the ruin which a little common sense would have avoided. Corruption was at the prow and folly at the helm. The selfish ring which he had denounced so fiercely had triumphed. It had triumphed, as he held, by flattering the new dynasty, hoodwinking the nation, and maligning its antagonists. The cynical theory of politics was not for him, as for some comfortable cynics, an abstract proposition, which mattered very little to a sensible man, but was embodied in the bitter wrath with which he regarded his triumphant adversaries. Pessimism is perfectly compatible with bland enjoyment of the good things in a bad world; but Swift's pessimism was not of this