five-beat line the points considered are: hovering accent, accent transposition, headless lines, double upbeat and dissyllabic thesis, slurring, elision, the cesura.

Schipper subjects the verse of Chaucer to a most mechanical and obdurate analysis. True, he repeats now and then that the variety of Chaucer's verseflow is beautiful and marvelous; but the appreciation is so formula-like in character and so surrounded by masses of unrelenting fruitless line-dissection that its effect is lost. The term "fruitless" is applied with intent. It appears to me that no more is gained for understanding of Chaucer's prosody by classing his "cesuras" as (1) masculine after the second foot, (2) feminine after the second foot, (3) lyric within the third foot, (4) masculine after the third foot, (5) feminine after the third foot, etc., than would be gained by classing the poet's lines as those beginning with a noun, those ending with a noun; those containing a noun in the second foot, those containing two nouns, etc. Indeed, the entire basis of Schipper's analysis is questionable, assuming and asserting as it does that Chaucer is to be treated line by line like an Anglo-Saxon versewriter, and that every one of his lines is divided by a cesura into two portions, each with one principal stress. To such excess of archaism does Schipper push this theory that he even treats lines like

He sette nat his benefice to hyre
That I was of hir felawshipe anon
By forward and by composicioun

Prol. 507

Prol. 32
Prol. 848

as showing an almost imperceptible ("verwischte") cesura, which according to Schipper falls where I have drawn the bar.1 In the face of the long recognition given to Schipper's bulky treatise, I would insist that for Chaucer at least the "cesura" and the "hovering accent" are fabrications of a mind which must name before it can see or feel, a mind which when confronted with a group of lines like:

would find in the rime-divergence ever: river not an instance of sound-variety in sound-unity, peculiarly pleasing to the musical ear, but rather material for a theory of Gaelic influence upon Lowland Scottish short e, and the literary transfer of the resultant hybrid to the poetic dialect of the South of England a generation later.

This same anxious analytic insistence has led Schipper to dicta like the following: In the line

And held after the newe world the space

Prol. 176

it is uncertain whether after is to be considered a substituted trochee or a case of hovering accent; but in line 195,

And for to festne his hood under his chyn

1 Ten Brink, Sprache und Verskunst, § 313, will not accept this.

the word under is possibly a trochee, because after a cesura such a reversal of accent is not disturbing.

Schipper would probably refuse the suggestion that in Chaucer, as in modern poets, there are two sorts of rhythmic variation: (1) that which is deliberate and emphatic, arranged by the poet to attract the ear of the reader and thus enhance the force or the beauty of the line-content for the mind. An example of one sort of such emphasis is

That if gold ruste, what shall iren do?

(2) And secondly, that which is unconscious on the poet's part, merely the expression of the inevitable tendency to vary which must be at work in rhythmic unity if monotony is to be avoided, and of the equally inevitable conflict between word-accent and verse-accent. Through analysis of deliberate variation we may better appreciate the verse and the poet; it is a misfortune, and a barrier to the development of prosodic study, that scholars have addressed themselves as much or more to the examination of

unconscious variations.

Perhaps it is the mysterious power exerted by Schipper's array of categories which has prevented the placing of Chaucer's fivebeat verse in comparison with its nearest of kin, the elevensyllabled line of Boccaccio and Dante.

1883. A. Baret. Etude sur la langue anglaise au xivme siècle. Paris. Reviewed Littblatt 5:358, 6: 330 as valueless.

1884. Bernhard ten Brink. Chaucers Sprache und Verskunst. Strassburg. Reviewed Anglia 7: Anz. 141-43 (Wülker); Littblatt 1885 p. 187 (Einenkel); DLZ 1885, p. 607 ff. (Zupitza); Engl. Stud. 10: 114-117 (Koch). Second ed. by Kluge, Leipzig 1899; rev. Anglia Beibl. 12: 237-240 (Holthausen). Transl. into English by M. B. Smith, London 1901, as The Language and Versification of Chaucer.

See Smith, Some Remarks on Chapter III of ten Brink's Sprache und Verskunst, in Mod. Lang. Quart. 5:13-20 (1902), reviewed by Koch, Engl. Stud. 36: 131-133, somewhat curtly; Koch says the point was already settled by Bischoff, see below under 1897.

1887. E. Einenkel. Streifzüge durch die mittelenglische Syntax. Reviewed Archiv 82: 226 (Trautmann), Engl. Stud. 12: 283-96 (Bülbring), Littblatt 1889 pp. 12-15, Acad. 1889 I: 327. Einenkel's references are to the Aldine Chaucer, by volume and page.

1888. A. Graef. Das perfektum bei Chaucer. diss. Kiel 1887, pp. 96, pubd. Frankenhausen 1888, pp. 102. Reviewed Anglia II : 326 (Wülker).

1890. C. W. Kent. On the Use of the Negative by Chaucer, with particular reference to the particle ne. Publ. Mod. Lang. Assn. 5 109-147.

1889. M. Freudenberger. Ueber das Fehlen des Auftakts in Chau

cer's heroischem Verse. Leipzig. Reviewed Anglia Beibl. 1:88-90 (Koeppel). See under E (4) below.

1891. A Graef. Die präsentischen Tempora bei Chaucer. Anglia 12:532-577.

1891. E. Gasner. Beiträge zum Entwicklungsgang der neuengl. Schriftsprache. diss. Göttingen. pubd. Nürnberg, pp. 144.

1892. W. W. Skeat. Chaucer's Use of the Kentish Dialect. Chaucer Society Essays, part VI.

1893. G. Graef. Das futurum und die entwicklung von schal und wil zu futurischen tempusbildnern bei Chaucer. Flensburg, pp. 52. In Jahresbericht der Flensburger Handelsschule.

1893. In Paul's Grundriss der germanischen Philologie, the articles on Englische Metrik are by K. Luick (Heimische Metra) and J. Schipper (Fremde Metra).

1893. George Hempl. Chaucer's Pronunciation and the Spelling of the Ellesmere Manuscript. Boston.

1893. John M. Manly. Observations on the Language of Chaucer's Legend of Good Women. Harvard Studies, II : 1-120.

1894. G. L. Kittredge. Observations on the Language of Chaucer's Troilus. Chaucer Society; and Harvard Studies, vol. III.

1894 ff. W. W. Skeat, in his ed. of Chaucer, vol. VI : xxiii ff., discusses the dialect of Chaucer, his pronunciations, his rimes, Lounsbury's views on Chaucer's rimes, metres and forms of verse, grammar, and versification. Skeat's dicta upon pronunciation, rime, and grammar are those generally accepted in the present state of our textual knowledge; they are based upon the text constructed by himself. His treatment of the versification, similarly based, is very unsatisfying; the scheme of scansion which he adopts is more inadequate than those which he rejects.

1896. L. Morsbach. Mittelenglische Grammatik. Part I. Halle.

1897. O. Bischoff. Ueber zweisilbige Senkung und epische Cäsur bei Chaucer. diss. Königsberg. Pubd. Engl. Stud. 24:353 ff., and 25: 339-398. (See below, p. 499.)

1898. E. Hampel. Die silbenmessung in Chaucer's fünftaktigem Verse. Teil I. diss Halle, pp. 45.

1899. H. C. Ford. Observations on the Language of Chaucer's House of Fame. diss. Univ. of Virginia, pp. 90.

1900. M. Kaluza. Historische Grammatik der englischen Sprache. Part I. Berlin.

1903. H. Remus. Untersuchungen über den romanischen Wortschatz Chaucers. diss. Göttingen, pubd. Halle, pp. 38.

1904. R. D. Miller. Secondary Accent in Modern English Verse (Chaucer to Dryden). diss. Johns Hopkins Univ.

1905. A. G. Kennedy. On the Substantivation of Adjectives in Chaucer. Univ. of Nebraska Studies, vol. 5, pp. 251-269.

1906. George Saintsbury. History of English Prosody. Part I. Lond. and N. Y. Reviewed Mod. Lang. Review 2:65-70 (McKerrow); Nation 1906 II: 189; Mod. Lang. Notes 22: 122-124 (W. Hand Browne). See Omond, English Metrists, 1907, p. 239.

Chapter IV of Book II is devoted to Chaucer. Saintsbury takes the stand that we do not know what Chaucer wrote, and accordingly are in no position to discuss line-structure. His remarks are therefore general; but the student would have been more materially aided by a larger body of examples and a more thorough treatment of Chaucer as a poet, at the expense of the autobiographical comment, excess of literary allusion, and somewhat flippant gibe which appear in these pages. In this essay we have the complete opposite of Schipper's laborious technical analysis.

1906. Louis Round Wilson. Chaucer's Relative Constructions. diss. Univ. of North Carolina, pp. 60.

1906. H. Remus. Die kirchlichen und speziell-wissenschaftlichen romanischen Lehnworte Chaucers. Halle, pp. xii, 154. See Remus' other work, 1903 above.

C. On the Language and Verse of Chaucer

Several considerations have created, and still keep alive, the feeling that Chaucer's verse, despite the imperfect copies which have come down to us, was in its original form smooth and beautiful. These considerations are: our knowledge that these copies are often careless "mismetred" renderings, marred as Chaucer himself deplored in his Words to Adam and at the close of his Troilus. Secondly, the fact that, in spite of this frequent injury, so large a proportion of his lines, as pointed out by Tyrwhitt, Essay part II, § xii, "sound complete to our ears." Thirdly, the superiority of any really critical text to even the best existing MS copy. And lastly, the instinctive inference of his readers that a man so greatly the master of structure, of character drawing, of selective art, of narrative power, must have been master also of verse-technique.

These arguments still hold. The danger in using them is one of detail rather than one of general principle. Pollard's remark, Athenaeum 1901 II : 631, that it is not likely that future editions of the Canterbury Tales will materially alter the reading of any twenty lines in that work, is in all probability true. But the advocate of the critical text, as defined by Liddell, ibid. p. 598, while reasonably assured of the poet's verse-command, desires in the first place to arrive at that assurance by sound methods, and in the second place to learn what licences or liberties Chaucer, the first master of the decasyllabic line in English, may have permitted himself in that form. Many critics, convinced of Chaucer's impeccability, attempt to force upon him verse-perfection according to modern notions, without regard to any metrical licences of his own which he may have originated. Thus, ten Brink disavowed the headless line now accepted as Chaucerian by all scholars, see below; Lowell also declined to credit the theory that Chaucer could have used it; and most authorities still refuse to accept the existence in Chaucer of the so-called "Lydgate line”, i. e., the line in which a pause falls between two accented syllables. But we have not yet made that thorough and unbiassed examination of the texts which alone will. enable us to say what variants Chaucer permitted himself on the analogy of the substituted anapest in Tennyson or the wrenched accent so frequent in Swinburne. Even among modern students there is found the tendency to reason in a circle, to start from the assumption that Chaucer is "impeccable", and, after constructing a body of texts on that hypothesis, by eliminating or altering whatever seems incompatible, to deduce from them the original assumption of the impeccability of Chaucer. The circle is then complete.

As the smoothness of Chaucer's verse is dependent primarily upon the number of its syllables and the position of the accents, the partial establishment of fourteenth-century syntax is a necessary prelude to the discussion of Chaucerian lines. Upon the sounding