[1] One hears sometimes of 'rhythmic thought' and 'rhythmic feeling.' This is merely a further extension or metaphorical usage of the term. In Othello, for instance, there is a more or less regular alternation of the feelings of purity and jealousy, and of tragedy and comedy. In some of the Dialogues of Plato there is a certain rhythm of thought. This usage is fairly included in the Oxford Dictionary's definition: "movement marked by the regulated succession of strong and weak elements, or of opposite or different conditions."

[2] There is, however, another phenomenon (to be discussed later) called by the same name, 'tone-color,' but having only a metaphorical relation to it. Many words—father, soul, ineluctable, for example—have emotional associations which stand to the literal meaning somewhat like overtones to the fundamental. This tone-quality of language is one of the primary and most significant sources of poetical effect, but it should never be confused with the musical term on which it is patterned.

[3] Walter Pater, "Leonardo da Vinci," in The Renaissance. For an account of scientific experiments on the time and stress rhythm of this sentence, see W. M. Patterson, The Rhythm of Prose, New York, 1916, ch. iv. An idea of the complexity may be obtained from Patterson's attempt to indicate it by musical notation:

[4] Paul Verrier, Essai sur les Principes de la MÉtrique Anglaise (Paris, 1909), Deuxieme Partie, Livre II, ch. x, pp. 56, 57; and cf. p. 90, n. 1.

[5] A simple experiment will illustrate this. Place two persons back to back, so that they cannot see each other, and have them beat time to an audible melody; as soon as the music ceases they will begin to beat differently. (Verrier, II, p. 65.) The difficulty of keeping even a trained orchestra playing together illustrates the same fact.

[6] "If rhythm means anything to the average individual, it means motor response and a sense of organized time." Patterson, p. 14.

[7] Musicians often 'dot' a note for the sake of emphasizing the accent, especially in orchestral music and with such instruments as the flute, where variations of stress are difficult to produce.

[8] Cf. Patterson, p. 3, " ... the possibility of preserving a certain series of time intervals, but of changing in various ways the nature of the motions or sensations that mark the beats." This may be tested by a simple experiment. With the foot or finger tap evenly, regularly, and rather rapidly. Without changing the regularity of the tapping, but merely by a mental readjustment, the beats may be felt as tum-te, tum-te, tum-te (or te-tum, etc.) or as tum-te-te, tum-te-te, tum-te-te (or te-te-tum, etc.), or even as tum-te-te-te, tum-te-te-te (or te-te-te-tum, etc.). It is but a step from this successive perception of various rhythms from the same objective source to a combined and simultaneous perception of them.

[9] Patterson, p. xx, n. 3.

[10] Experiments have shown that with a little practice one can learn to beat five against seven, and thus actually though unconsciously count in thirty-fives. (Patterson, p. 6.)

[11] Those who are interested will find the scientific experiments discussed in Patterson, ch. i and Appendix III.

[12] When no organization of the irregularity is possible, the language is unrhythmical; and such, of course, is often the case in bad prose and bad verse.

[13] Compare the sentence from Disraeli on page 24, above.

[14] Vol. i, pp. 79, 80. The musical notation must be taken as merely approximate.

[15] Experiments have been made to obtain absolute measurements of syllabic quantity, and elaborate rules formulated for determining longs and shorts. Thus far, however, the results have been very variable and unsatisfactory, and should be accepted with great caution.

[16] To adduce Greek in explanation of English pitch would be a clear case of ignotum per ignotius. But interesting parallels have been noted by Mr. Stone (in R. Bridges, Milton's Prosody, 2d ed.). "The ordinary unemphatic English accent," he says, "is exactly a raising of pitch, and nothing more" (p. 143); and there are similar habits in English and Greek of turning the grave accent into acute, as in to gÈt money and to gÉt it. The Greeks recognized three degrees of pitch: the acute (high), and the grave (low), (which, according to Dionysius, differed by about the musical interval of a fifth), and midway, the circumflex. Compare thÁt? (acute, expressing surprise); thÂt? (circumflex, expressing doubt); and thÀt book (grave—'book' and not 'table'). The main difference between the two languages is that so far as we can tell classical Greek had (very much like modern French) a pitch-accent and very little or no stress-accent, whereas English has both (though stress-accent preponderates).

[17] Cf. J. W. Bright, "Proper Names in Old English Verse," Publications of the Modern Language Association, vol. 14 (1899), pp. 347 ff.; especially pp. 363-365.

[18] Verrier, vol. iii, p. 229. A more ambitious attempt, from Pierson, MÉtrique naturelle du langage (Paris, 1884), pp. 226, 227, is given by Verrier, vol. ii, p. 14—a musical transcription of the opening verses of Racine's Athalie.

[19] Coleridge, Biographia Literaria, ch. xviii. Compare the more poetical expression of the same truth in Carlyle's Heroes, Hero-Worship, and the Heroic in History: "Observe too how all passionate language does of itself become musical—with a finer music than the mere accent; the speech of a man even in zealous anger becomes a chant, a song. All deep things are Song. It seems somehow the very central essence of us, Song; as if all the rest were but wrappings and hulls! The primal element of us; of us, and of all things. The Greeks fabled of Sphere-Harmonies: it was the feeling they had of the inner structure of Nature; that the soul of all her voices and utterances was perfect music.... See deep enough, and you see musically; the heart of Nature being everywhere music, if you can only reach it." (The Hero as Poet.)

[20] The first is from a poem in free-verse, Meditation, by Richard Aldington; the second is blank verse, from the Small Sweet Idyl in Tennyson's Princess; the third is from Henley's Margaritae Sorori (also in free-verse); the fourth is from DeQuincey's English Mail-Coach, Dream Fugue IV (prose); the fifth is from Milton's Samson Agonistes, ll. 1092 ff. (blank verse); the sixth is from Browning's The Ring and the Book, Bk. V (blank verse).

[21] Thomas Rudmose-Brown, "English and French Metric," in Modern Language Review, vol. 8 (1913), p. 104.

[22] A convenient collection of extracts from various writers is made by Professor R. M. Alden in Part IV of his English Verse, New York, 1903.

[23] The chief difference perhaps between the foot and the bar is that the latter always begins with a rhythmic stress, whereas the foot may begin with an unstressed element.

[24] Some metrists, holding that every foot should begin with a stress, divide thus:

Such a division can be justified on several grounds, but it remains awkward and obscures the plain fact of rising rhythm. It does not affect the division of word and foot; for compare Shelley's line:

[25] An historical survey of the problems and theories, somewhat colored by the author's own theory, may be found in English Metrists, Oxford, 1921, by T. S. Omond.

[26] I take these figures from the two articles by Professor Ada L. F. Snell in the Publications of the Modern Language Association for September, 1918, pp. 396-408, and September, 1919, pp. 416-435. For the first example I have made an average from the records of three different readers; for the second Miss Snell gives only one set of figures.

[27] The second and fourth lines have two feet each, the alternate lines throughout the rest of the poem have three feet each; but it is noteworthy that the average length of these two short lines (1.61) is only .37 less than the average of the four longer lines (1.98). The first, third, fifth, etc., lines have four feet each.

[28] This statement is based on Miss Snell's computations from analysis of several records for blank verse and several kinds of lyric verse. The short syllables range in blank verse from .02 to .54, in lyrics from .09 to .7; the long syllables range in blank verse from .08 to .84, in lyrics from .11 to .92. The average length of all long syllables is .4, of all short syllables is .21.

[29] In the latter case it is supplemented by a pause in Miss Snell's marking. Many readers would no doubt combine the hold and pause; as was done in fact in l. 5.

[30] It should be noted that the average line length here (including pauses within the line, excluding those at the end of the line) is 2.8, and the first line is therefore only .32 shorter than the average. If additional allowance (omitted in Miss Snell's computation) be made for the theoretical initial the average would be 2.85 and l. 1 would total 2.92. If the end pause is included the average would be 3.38 and l. 1 2.78—a difference of .66; or with the additional allowance the average would be 3.44 and l. 1 3.22. While too much faith is not to be placed in the mere figures, the inference is plain that the rests practically compensate here for the omitted .

[31] Miss Snell, Pause; a Study of its Nature and its Rhythmical Function in Verse (Ann Arbor, 1918), pp. 78, 79.

[32] Modern English verse theory may be dated from Coleridge's famous manifesto in the prefatory note to Christabel in 1816: "I have only to add that the metre of Christabel is not, properly speaking, irregular, though it may seem so from its being founded on a new principle: namely, that of counting in each line the accents, not the syllables. Though the latter may vary from seven to twelve, yet in each line the accents will be found to be only four. Nevertheless, this occasional variation in number of syllables is not introduced wantonly, or for mere ends of convenience, but in correspondence with some transition in the nature of the imagery or passion." Even here there is implied a vague perception of the time unit, but Coleridge was apparently unaware of its significance. See Leigh Hunt's comments in "What is Poetry?" in Imagination and Fancy.

[33] The expression '4-foot line' is too suggestive of fishing or surveying; 'tetrameter' is confusing because of its different usage in classical prosody; '4-stress line' is open to objection because it seems to overlook the temporal quality of the foot. On the whole, however, the last seems preferable.

[34] From the point of view of stanzaic rhythm Could may be said to complete the final trochee of the previous line:

[35] Apparent paeons occur now and then, where the usual contraction would reduce them to triple time. Mr. Omond, Study of Metre, pp. 96, 97, gives among others these examples:

[36] In the last stanza occurs the foot:

[37] See Sidney Lanier's scansion of the first stanza, in his Science of English Verse, p. 228.

[38] An interesting variation of this rhythm (though perhaps to be related to the Middle English descendant of the Anglo-Saxon long line) occurs in Shelley's Prometheus Unbound, Act I,

Compare also Shelley's earlier poem, Stanzas—April, 1814; and for a more recent example:

[39] This metre has been used, e. g., by George Darley (1795-1846) in The Flower of Beauty (four stanzas) and (rather monotonously) by Charles Swain (1803-74) in Tripping down the Field-Path (cf. Stedman's Victorian Anthology, pp. 17, 76); and more recently by Mr. Alfred Noyes.

[40] For a classified collection see Alden, English Verse, pp. 24 ff.

[41] This whole poem abounds in substitutions. See Shelley's The Cloud, above, pages 59 f., which may be regarded as 2-and 3-stress anapestic lines, though two 2-stress lines are printed as one.

[42] Tennyson's To Virgil, though it has nine stresses in each line and is therefore an exception to the statement made above, page 69, is shorter in respect of the number of syllables. There is, moreover, a poem, After Death, by Fanny Parnell, consisting of fourteen 10-stress lines. The cumbrousness of the rhythm is apparent in these two specimens—which are rather better than the others—

The whole of this poem may be found in Sir Edward T. Cook's More Literary Recreations, p. 278.

[43] By a series of experiments C. R. Squire found a natural preference for duple over triple rhythms (though the triple rhythms seemed 'pleasanter'), and for trochaic and dactylic over iambic and anapestic. (Am. Journal of Psychology, vol. 12 (1901), p. 587.)

[44] Wordsworth and Mrs. Browning have written rimed septenaries.

[45] The usual and most convenient way of indicating stanzaic structure is with small italic letters for the rimes and either superior or inferior numbers for the number of stresses in each line. Thus Landor's Rose Aylmer:

is described as a⁴b³a⁴b³. The repetition of a whole line is indicated by a capital letter. When all the lines are of the same length, one exponent figure suffices, as abba⁴ for the In Memoriam stanza.

[46] See above, p. 66, n. 1.

[47] Byron follows now one model, now another. In Parisina he consciously tried the metrical scheme of Christabel.

[48] It is no doubt significant that the rhythmic pulses which come most naturally to us are in twos and threes and their multiples; while even to beat time in fives requires a special effort. In music 5/8 or 5/4 time is extremely rare. There is an example of the latter in Chopin's Sonata I (the larghetto movement).

[49] On the source and origin of the 5-stress couplet in English, authorities are in disagreement. See Alden, English Verse, pp. 177 ff., and the references there given.

[50] Note Professor Woodberry's praise of the heroic couplet for its simple music, its suppleness, its power of forcing brevity: "the best metrical form which intelligence, as distinct from poetical feeling, can employ." (Makers of Literature, p. 104.)

[51] See Pope's own analysis of his system of verse in a letter to Cromwell, November 25, 1710.

[52] For complete lists and examples of all the various stanzaic forms, the larger works of Alden and Schipper should be consulted.

[53] On its origin and the twenty-five poems in it by seventeen different poets, from Ben Jonson to Clough and Rossetti, before the publication of In Memoriam, see E. P. Morton in Modern Language Notes, 24 (1909), pp. 67 ff.

[54] H. Corson, Primer of English Verse, pp. 70 f.

[55] Early examples may be conveniently found in the Oxford Book of English Verse, Nos. 75, 96, 102, 108,172.

[56] For early examples see again the Oxford Book of English Verse, Nos. 74, 140, 182, 184, 187.

[57] There are two ottava rime in Lycidas, one at the close of the Blind Mouths passage and one at the end of the poem.

[58] A. Dobson, in Ward's English Poets, vol. iv, p. 240.

[59] On the Spenserian stanza see especially Corson, pp. 87 ff. Lowell's characterization of Spenser's use of it is interesting: "In the alexandrine, the melody of one stanza seems forever longing and feeling forward after that which is to follow.... In all this there is soothingness, indeed, but no slumberous monotony; for Spenser was no mere metrist, but a great composer. By the variety of his pauses—now at the close of the first or second foot, now of the third, and again of the fourth—he gives spirit and energy to a measure whose tendency it certainly is to become languorous" (Essay on Spenser). See also Mackail's chapter on Spenser in Springs of Helicon; and Shelley's praise in his Preface to the Revolt of Islam: "I have adopted the stanza of Spenser (a measure inexpressibly beautiful), not because I consider it a finer model of poetical harmony than the blank verse of Shakespeare and Milton, but because in the latter there is no shelter for mediocrity; you must either succeed or fail. This perhaps an aspiring spirit should desire. But I was enticed also by the brilliancy and magnificence of sound which a mind that has been nourished upon musical thoughts can produce by a just and harmonious arrangement of the pauses of this measure."

[60] See also the collection of Sonnets on the Sonnet, edited by M. Russell, London and New York, 1898.

[61] On the origin of the sonnet in Italy (Sicily) see the references in Alden's English Verse, p. 267. Still a standard work is C. Tomlinson's The Sonnet, London, 1874.

[62] Elaborate rules for the sonnet are given by William Sharp in the introduction to his Sonnets of the Century, and by Mark Pattison in the introduction to his edition of Milton's sonnets. There is valuable matter in the Introduction of J. S. Smart's The Sonnets of Milton, Glasgow, 1921. Compare also the 'divisioni' of Dante's sonnets in the Vita Nuova.

[63] "In the production of a sonnet of triumphant success, heart, head, and hand must be right." Corson, p. 145.

[64] Two of these are irregular, the 99th, with fifteen lines (ababacdcdefefgg) and the 126th with twelve (aabbccddeeff). Milton's On the Admirable Dramatick Poet, W. Shakespear, still traditionally miscalled a sonnet, resembles the latter, with its aabbccddeeffgghh or eight couplets. The 16-line stanza of Meredith's Modern Love (abbacddceffeghhg) is sometimes loosely called a sonnet.

[65] Quoted by permission of Dodd, Mead & Co., owners of the copy-right.

[66] The rime scheme of the Progress of Poesy is: strophe and antistrophe a⁴b⁵b⁴a⁵cc⁴d⁵d⁴e⁵e⁴f⁴f⁶, epode aabb⁴a³ccdede⁴fgfgh⁵h⁶. The formula is three times repeated. Note the unusual arrangement of parts in Collins' Ode to Liberty and Shelley's Ode to Naples.

[67] J. A. Symonds, Blank Verse, London, 1895, p. 23. (This little volume contains a valuable, though incomplete and somewhat extravagant, summary of the history of English blank verse.)

[68] The main crux of this passage is "fit quantity of syllables." Quantity in such a context suggests syllabic length; and one recalls the sonnet to Lawes—

But, on the other hand, Mr. Robert Bridges has made it almost if not quite certain that Milton counted syllables, and therefore the phrase would mean "ten syllables to a line," proper allowance being made for elision. Since both interpretations agree pretty well with Milton's practice, one cannot be sure which he had in mind.

[69] J. W. Mackail, The Springs of Helicon, pp. 181 ff.

[70] Cf., for example. Paradise Regained, III, 68 ff.

[71] Browning's blank verse, like all his metres, is typically Browningesque; instead of moulding his verse to fit the idea perfectly, he too often effected the compromise between content and form by slighting the latter.

[72] Mr. Frost in some of his later work permits himself such laxness as—

[73] Strongly to be deprecated is the frequent confusion not only of the different varieties of English free-verse, but of the fundamentally distinct phenomena of free-verse as commonly understood and French vers libre. Vers libre itself has many aspects, from the literally freer use of rime and the mute-e than the traditional French prosody allowed and an escape from the old principle of syllabification to what superficially corresponds with English free-verse, that is, a substitution of prose for verse; but only superficially, since the French language is phonetically different from English, and its ordinary prose has a naturally greater song potentiality. Since the phenomena differ they should not be called by the same name. The English term 'free-verse' is wholly adequate.

[74] "In the effort to get rhyme, 'the rack of finest wits,'" says a pseudonymous newspaper writer, "and in the struggle, writhing, and agony of trying to get the wrong words to say the right thing, one sometimes achieves the impossible, or, rather, from the flame of frantic friction (of 'Rhyming Dictionary' leaves) rises, phoenix-like, another idea, somewhat like the first, its illegitimate child, so to say, and thus more beautiful.

"With vers libre one experiences the mortification one sometimes feels in having roared out one's agony in perfectly fit terms. With rhymed poetry one feels the satisfaction of a wit who gives the nuance of his meaning by the raise of an eyebrow, the turn of a word."

[75] See a part of MargaritÆ Sorori, page 43, above.

[76] From Sword Blades and Poppy Seeds, by permission of Houghton Mifflin Co.

[77] The advanced student should of course read carefully the paper on "Classical Metres in English" by W. J. Stone in Bridges' Milton's Prosody (2d ed.), pp. 113 ff. Mr. Stone regards the hexameters of Clough's ActÆon and some specimen verses by Spedding (the biographer of Bacon) as the best he has seen.

[78] According to the commonest American pronunciation.

[79] The unaccented vowel sounds show the usual predominance of the obscure vowel e, with three occurrences of ĭ and ī.

[80] Reference to the text will identify the symbols.

[81] Rime occurs, however, here and there in Greek and Latin poetry, and is more frequent than perhaps we commonly suppose.

[82] In the 3182 lines of Beowulf, for example, there are sixteen exact rimes and many more approximate rimes. There is also in Anglo-Saxon the so-called Riming Poem, of uncertain date, composed probably under Scandinavian influence.

[83] See the whole of Ben Jonson's Fit of Rhyme against Rhyme.

[84] Compare Flaubert's extreme statement: "that a beautiful verse without meaning is superior to one that has meaning but is less beautiful."

[85] Triple rimes are naturally excellent for joco-serious purposes, like the celebrated intellectual: henpecked you all, Timbuctoo: hymn book too, thin sand doubts: ins and outs.

[86] Swinburne, Dedication, 1865.

[87] For an extreme example of mimicry, see Southey's Lodore.

[88] Lines 370 ff. Dr. Johnson's comment on this last line is curious: "The swiftness of Camilla is rather contrasted than exemplified. Why the verse should be lengthened to express speed, will not easily be discovered. In the dactyls, used for that purpose by the ancients, two short syllables were pronounced with such rapidity, as to be equal only to one long; they, therefore, naturally exhibit the act of passing through a long space in a short time. But the alexandrine, by its pause in the midst, is a tardy and stately measure; and the word 'unbending,' one of the most sluggish and slow which our language affords, cannot much accelerate its motion."

[89] It is not to be understood, however, that the higher the pitch the greater the emphasis; for the contrary is often the case.

[90] It is perhaps useless to debate about this line. Whether one divides thus:

and says there is an 'inversion' in the first, third, and fourth feet, or preferably thus:

the rhythm is extraordinary; and the added complexity of 'own' puts it entirely hors concours. Compare with it, however, Milton's

[91] The italics are not Wordsworth's.

[92] Here the italics are the poet's.

[93] Some readers take the line thus:

with emphasis or pitch-accent on 'first'; in which case the above explanation does not hold.

[94] Milton's Prosody, p. 30 (ed.1901).

[95] The pronounced syncopations of ragtime partially illustrate this.

[96] In the specific cases mentioned below, this phenomenon is historically known as "recession of accent"; and it sometimes occurs in non-metrical contexts. It is also very similar to one of the aspects of pitch; see pages 181 f., above.

[97] Note also the spondaic effect in the second line, the rime in the third, and the imitative movement in the fourth.

[98] Here, dividing the lines into parts measured by the number of syllables, the series is: 6+4, 6+4,—, 2+4+4, 6+4, 8+2, 6+4, 6+4, 6+4, 8+2, 8+2, etc.

[99] Verrier, vol. i, p. 134.

[100] Mark H. Liddell, An Introduction to the Scientific Study of English Poetry, New York, 1901, pp. 291 f.