1559 056 Cornell University Library PE 1559.056 Metrical rhythm: being an examinatlOT 3 1924 027 420 870 Cornell University Library The original of tliis book is in tine Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924027420870 METRICAL RHYTHM: BEING AN EXAMINATION OF A RECENT ATTEMPT TO DETERMINE THE BASIS OF ENGLISH RHYTHM IN VERSE AND PROSE. 'A I,-): BY T. S. OMOND, Author of " A Study of Metre," etc. ©unbri&gE Mtlls: PRINTED AND PUBLISHED BY R. PELTON, THE BROADWAY, 1905- J) A % ' \(»^o METRICAL RHYTHM. (The basis of English rhythm. By William Thomson. 1904. Paper covers, price i/- net. Published by W. and R. Holmes, Dunlop St., Glasgow.) npHIS essay makes a striking contribution to English prosody. Assuming complete identity between the methods of metre and music, it represents syllables by musical notes, and does this with a precision, fulness, and subtlety which I have never seen equalled. I say so, remembering the work of Joshua Steele in the Eighteenth, Sidney Lanier and his pupils in the Nineteenth, Century. Nowhere before, I feel confident, has such representa- tion been made so elaborately and (as regards prose speech) so convincingly ; for this reason alone the pamphlet is worth its modest price to any student. As a practical guide to what he calls the syllabic movement of English speech, Mr. Thomson recommends tapping with the fingers on a table. A tune, thus tapped out, can be recognised by anyone acquainted with it ; the " movement " of a sentence can be similarly depicted. True, this gives only a skeleton ; the beginnings only of syllables aie recorded by taps, not their whole length and content. Yet even this is great gain, and 1 find the device distinctly helpful. By its aid I have convinced sceptics of the structural difference between such phrases as " mother mine " and " Molly mine." Formerly content to regard mother and Molly as equally " trochees," they now admit that the r sound m mother comes perceptibly later than the y of Molly — even when, as is usually the case with South Country speakers, the r sound is a mere vocalic murmur. Quaint as the notion of tapping out the rhythm of syllables may at first appear, a single serious trial should show that it is worth adopting. The essayist next goes on to show (§ 2) what elements of speech are retained in syllabic movement [or rather, in the skeleton representation of it?], and which are excluded. Here occurs some very precise definition or description. Some- of it may seem undue refining. Is there any real difference between the length of a syllable and its " syllabic burden " ? For ray part, however, I desire more rather than less definition. I miss, in these opening paragraphs, definitions on first use of terms like rhythm, time, accent, metre, foot. Mr. Thomson may say that these are words which everybody understands, and whose mean- ing is to be found in any dictionary. But it is precisely in such words that ambiguity loves to lurk. Rhythm is defined later (§ 7, first sentence) as "the regular recurrence of accent"; a definition which is really a postulate.* Accent I do not find ever defined, but it is apparently identified with force. Time is used in two senses, according as it is spelt with or without a capital letter. The latter seems to connote simple duration ; the former means time m the specialized musical sense, " what the musician calls . . . Time" (p. 12, line 7). This distinction might be missed by some readers, especially since-- perhaps through default of the printer — it is not observed with entire consistency through- out. t The word foot occurs first in § 7, and seems afterwards to be replaced by measure; wliile metre specifies the number of measures in a given group (§ 22, near end). As metre and measure are etymologically the same, this terminology may perplex till explained ; and metre seems to bear a somewhat wider mean- ing the first time it occurs (in last sentence of § 7). I think in common usage it bears a wider meaning, covering all that pertains to the structure of verse as distinguished from prose ; it is in this wider sense that I myself use it. These initial paragraphs are not free from a fault which I find pervading the whole essay, namely, a tendency to undue dogmatism. In § 3 Mr. Thomson has no doubt about how to read " Take your own time, Annie.'' " The matter is very simple to a trained or an attentive ear. The higher pitch is normally Cf. a subsequent much more elaborate version in the long middle sentence of p. 38. t E.g. p. 46, third line from foot, " those who feel the time unduly compressed " ; p. 53, fourth line from foot, "a presumed suspension of time." Occasionally Time has inverted commas, e.g. p. 32, line 14. Bpt as a rule the distinction as above is clearly maintained. s [sic] on 'own,' the stronger accent on 'time.'" Were these reversed, he says, the sentence would clearly mean "that Annie is now cautioned against spending, on the work she has in hand, the time, let us say, of her mistress." Surely it might also mean that Annie is advised to do her work at the pace she finds to suit her. What I should like to have seen recognised is that there are at least a dozen different shades of meaning which may be expressed by these words, and that each requires a slightly different accentuation and modulation. I do not see how any one of these is more " normal " than another. What commends itself to Mr. Thomson's judgment may not commend itself to his neighbour's. My habitual reading of Tennyson's line containing this phrase differs from that here laid down, and the twice re- peated words own time are always accented differently by me on the first and on the second occasion of their occurrence. Such unnecessary dogmatism mars what are otherwise valuable paragraphs, full of criticism very suggestive if not always quite so novel as the writer professes. That the quantitative value of syllables in isolation is often no guide to their value in a sentence has been repeatedly pointed out. It is hardly true that few people distinguish long consonants as well as long vowels ; our phonetic scholars have taught this for years. If Mr. Thomson merely means that our school grammars ignore such matters, he is only too right. I doubt whether in the phrases "if thou shouldst mark" and "to shut windows" it is correct to say that shut occupies longer time than shouldst (§ 3} ; nor am I clear that the first vowel in unanimous is pronounced in the same time as the first of inanimate (ibid). The paragraph about Latin "quantities" (§ 4) admirably shows how partial and limited is our ordinary recognition of these, amounting really to mere accentuation of the antipenult when the penult is short ; but its illustrations from English speech scarcely do justice to the differ- ent parts played by quantity in ancient Latin and in modern English. The word long is said (§ 5) to have "three different lengths '' in the phrases long dress, longer dress, longer address ; which reminds one of the three degrees of shortness found by Dionysius in the syllables rod-, trop-, and stroph-, all of which were accounted equally short in Greek verse. That all speech tends to be rhythmical (§ 6) is quite true, but is compatible with more than one explanation of the difference between prose and verse. These paragraphs should therefore be read with some caution, but the general conclusion is excellently put in the following sentence (foot of p. lo) : — "Within certain limits, the insertion or omission of unaccented syllables does not affect the total duration of a phrase, and the length of accented syllables varies according to the character and number of unaccented syllables intervening before the next accent." As a general rule, and " within certain limits," the principle here formulated, as it had previously been adumbrated by Whewell and others, seems unmistakeably true. Having thus cleared the ground, Mr. Thomson next (§§ 7 to 12) applies his method to the phenomena of English speech. He claims that while the distribution of time over the syllables between the accents might conceivably be quite irregular, as a matter of fact it is regular, and usually conforms to " what the musician calls Triple Time." This " Time " he accordingly considers the real basis of English rhythm, in prose as well as verse, while admitting exceptions both through irregularities and through " the occasional intru- sion of Common Time," in either its duple or quadruple form. The difference between prose and verse he describes shortly thus (end of § 7). " In poetry {scilicet verse], variation within the foot is limited ; in prose it is free. Metre [this word here used for the first time] may occur in prose, but it must not be perceptible, as it is in verse." With both these conclusions I disagree widely, but the grounds of disagreement would take too long to state here, so must be reserved for a later page. Sections 8 to 12 give copious illustrations, first of intrusive Common Time, then of normal Triple Time. These are always interesting, even 7 when one cannot accept them. At the beginning of § 9 the author says — "The notations here given are, it must be under- stood, not to be regarded as absolutely correct, but only as possible. What is maintained is that those who cannot accept them as normal will still find in,, the musical notation a means of expressing the rhythm they prefer.'' Both these sentences seem to me to postulate acceptance of general principles, leaving room only for divergence in details. Certainly, the confidence with which assertions are made about rhythm on p. 14, or about the structure of hexameters on pp. 18-19, does not seem to contemplate the possibility of two opinions. I pass these by for the present. Nor shall I try to follow the essayist in his somewhat hazardous excursions into modern French and ancient Latin verse (pp. 20-22 and 27). I so entirely agree with his conviction of the difficulty, if not impossibility, of thoroughly understanding rhythm in a dead or foreign language, that I only wonder he permitted himself these excursions. In his French scansions 1 seem to see the same flaw which I believe exists in his English. As to Sapphic verse, the schoolboy's reading of it is doubtless irrational. If Mr. Thomson cares to consult the Classical Review for 1903, he will there find three different attempts to state the rhythmical as distinct from the metrical structure of this verse. All three differ in detail, but agree in seeking to rhythmize it into isochronous feet, and are expressed in the musical notation which he himself prefers. It was only while his pages were passing through the press, and then apparently at second hand, that Mr. Thomson became aware how largely his view had been anticipated by Joshua Steele a hundred and thirty years ago. One gathers from this what I had otherwise inferred, that Mr. Thomson has no very wide acquaintance with what has been written about English metre. I am far from thinking this wholly a misfortune. He would have gained but little by such reading, find might have lost the freshness and originality which now characterize his teaching. That he has worked out his view independently is much to his credit. Yet I think it reacts on his exposition and his criticism. In expounding his own view, he labours points which are commonly accepted, and deals slightly with points tha^are much disputed. As a critic of others, he is sometimes wide of the mark through mis- apprehending their ideas. The Sections which come next in- order show him turning aside to pass cursory judgments on some fellow prosodists. In forming these judgments, he seems unable to do justice to any view which does not square with his own, and is not expressed in such terms as he himself uses. Whether such views are right or wrong, they should be fairly and adequately presented ; and this I cannot think is always done. His first shaft is aimed at Sidney Lanier (p. 26, cf. 36-8). With Lanier Mr. Thomson has naturally much sympathy. He thinks him a " real enquirer,'' who hears Triple Time in verse, and uses musical notation. But he rejects Lanier's view that rhythm exists before it is signalized by accent. Triumphantly he produces seven quavers, devoid of barring, and asks which of nine duly illustrated rhythms the quavers contain in themselves ; adding " There is no answer." Now, Lanier was a skilled musician as well as a poet, and this point must have been familiar to him. How he would have answered I do not know.* To me, who am no musician, the answer seems to be that all nine rhythms exist potentially in the notes (or rather in the sounds represented by notes), and that accent merely chooses one of these and makes it actual, directing attention on it to the exclusion of the others. I know, of course, that this clashes with the ordinary rule of thumb which makes musical time depend on regular recurrence of accents. But metaphysical questions are not decided by * Yet his own words may be quoted. " The primordial material of rhythm " is " a se.ies of sounds having among themselves definite relations of time or duration." "Accent can effect nothing, except in arranging materials already rhythmical through some temporal recurrence." (Science of English Vcse, Part I., Chap, II., pp. 64 and 65 in 1898 edition). rules of thumb, and this is -really a question of metaphysics and not of practical music. As such alone have I any right to speak on it. Those who make accent necessary to our recognition of rhythmical periods seem to me to mistake occasion for cause, and confuse the indicator with the thing indicated. Mr. Thomson says (p. 25) that " a succession of accentless quantities, in tones or in syllables, however pro- ducible by an automatic machine, are [is?] to the human tongue, governed as it is by the human ear, simply im- possible." I deny the impossibility ; but at any rate such a succession is not unthinkable. And therefore in thought at least, and I believe in practice also, accentuation can be separated from quantity, and rhytlim be perceived even when not signalized by accent. What has prosody to do with such high speculation ? Very little indeed. But Lanier has departed from musical custom, besides annihilating the work-a-day definition of rhythm. He dares divide heroic verse into old-fashioned feet, with accent on the last syllable. To him, as to me, it seems more natural to scan — The Weight | of all | the hopes | of half | the world, than to scan — The I weight of | all the | ho'pes of | ha'lf the Jjvorld. Rhythm is equally -perceptible either way. '^ This upsets Mr. Thomson's ideas, and breaks what he deems a natural law. To me the question is of little importance. I am coming to think there may be something to say on Mr. Thomson's side, not from the point of theory but from that of fact. Certainly it is inconvenient to use a notation which musically trained people seem unable to follow. In these pages, at any rate, I shall adopt the method which alone is intelligible to Mr. Thomson. Wherever I enclose syllables within bar-marks, it is to be understood that accent falls on the first beat of the bar, though that bar need not always correspond to an actual speech-stress, Turning to simpler matters, the essayist presently (§§ 13, 14, is) criticises the systems of Prof. Mayor, Latham, and the anonymous author of Accent and rhythm determined by the law of monopressures, with the last of whom he might have coupled Prof. Skeat. He has Httle difficulty in showing that these systems are inadequate, as all systems must be which ignore time; and his precise musical notation contrasts favourably with their vague and inefificient symbols. Even here, however, I find him assuming that by " accent " Prof Mayor means exactly what he himself means, which I greatly douht ; and laying down as indisputable what is an arguable question, when he says that the so-called trochaic beginning of an iambic line is really part of a three-syllable foot.* In the main, however, I find nothing to quarrel with here. Next (§ i6),t he comes to my Study of Metre, and I am compelled to say that he totally misrepresents me. Be- cause I make duple time the basis of heroic verse, he thinks that I read the syllables to equal beats, which I certainly do not. He is aware that I predicate duple time of the under- lying rhythm, not of the syllables, but he interprets these words in his own way. To him, " underlying rhythm " means merely the actual time of a passage, disguised as it often is by breaking up of beats (p. 34) ; to me, it means a fixed predetermined rhythm, to ■ which the rhythm of syllables is sometimes violently opposed. He misses my meaning, because to my words he attaches a sense which is not mine but his. He has difficulty in understanding my treatment of Tennyson's line — The sun, the moon, the stars, the seas, the hills, and the plains, through similar misconception. He himself says later (p. 38) that "Sometimes more than one line of verse must be read in order to determine whether trisyllabic feet are in Triple or Quadruple Time." And also, surely, to determine whether dissyllabic feet are set to Duple or to Triple Time. This is * ^f- '"f''"' P- =3. t Cf. also references on pp. 34 and 36. II very much what I said of Tennyson's lines, yet Mr, Thomson sees "nothing analogous'' (p. 32) between my view and his. I said, indeed (Study, p. 54), that the one word "and," coming where it does, tells us that the time is triple. Not that it makes the time triple, or even infallibly demonstrates this, but that it " tells us " a fact which becomes certain as we go on, and to which our ear responds at once. On the strength of this misconception, Mr. Thomson brackets me with writers who think a triple-time foot must contain three syllables, though this is entirely contrary to all I have said on the subject. I might with much more justice retort on Mr. Thomson the charge which he brings against me. I might say that it is he, not I, who make syllables constitute rhythm. He uses terms which point that way. He speaks (e.g. p. 38, second line from foot) of dissyllabic and trisyllabic rhythms ; which, to my mind, is like speaking of three-quaver-rhythms and two-crotchet-rhythms. But I do not judge him by such isolated expressions. I read his essay as a whole, and know that he distinguishes syllables from time-beats — though I think, and shall presently contend, that he mistakes the metrical relations between these two. Had he read my book with a similarly open mind, he must have seen that I make rhythm depend always on time-beats, never on syllables. As to Tennyson's line, let me quote a parallel case. The other day I read for the first time a poem beginning — Where the grey West slopes to a greyer sea. I read this in four feet, as if I had come on it in Coleridge's " Christabel." Next moment, glancing on, I saw that I was mis- taken ; that it was meant to be read in five feet, like a line from Paradise Lost. I at once readjusted, mentally, the music of the first line. This is precisely the converse of what I stated about Tennyson's line. Any one, reading it for the first time, would infallibly imagine the first ten words to be set to the time of heroic verse (whatever that time be); but with the word "and" conviction flashes upon him that this is not so, that the rhythm which governs the poem is other than he had supposed. How this can be supposed tantamount to saying that the trisyllabic foot makes the rhythm triple, I wholly fail to perceive. From myself he passes to Messrs. Bridges and Stone (p. 33). In the latest edition of his Miltoris Prosody (p. 108 ; the passage, indeed the appendix containing it, is not in previous editions) Mr. Bridges thus dissects a line of Longfellow's— And they | rode slowly ] along | through the wood | conversing | together. Mr. Thomson omits the dividing lines, and then complains that "it is impossible to gather from this notation how the line is read." As we have already been told the general structure of the line, I don't see where the impossibility comes in. But, besides this, the above dissection is not meant to show time (more the pity !), nor to show how to read the line ; that is assumed. It is meant to show stress-groups, and that the halting effect of the line is due to clashing between these and metrical rhythm. From this point of view Mr. Bridges analyses the structure. Particularly he specifies the word "through," and though Mr. Thomson finds him " weighted by preconceptions of length as based on the quantity of syllables considered apart from their setting," yet here Mr. Bridges expressly says that the clumsy effect is due not to the inherent heaviness of " through " but to its grammatical collocation. Mr. Thomson's own scansion reproduces this clumsiness by making " through " a short syllable. He also transfers the main stress from "they" to "rode," as if there could be no doubt about the matter; I do not myself believe that I>ongfellow meant us to lay stress on either word. 1 do not defend Mr. Bridges' view of verse ; like Mr. Thomson, I condemn it because it ignores time. A full statement of the reasons why its system of scansion is insufficient would have been well in place, and need not have taken up much room. To dismiss the subject in this way, resting the whole case on one misinterpreted instance, does justice neither to himself nor to Mr. Bridges. 13 The late Mr. Stone is declared to be "in even a worse plight." I fear it is the criticism which deserves this verdict. Mr. Thomson actually asserts that, according to Stone, English has no accents at all. Any one who understands Stone's essay (now published in the volume last referred to) knows that this is quite wrong. Stone did, indeed, contend that our speech-accent con- sists mainly of tone ; but he never denied its existence. On the contrary, his whole scheme involves perception of accent as well as quantity. In regard to quantity, Mr. Thomson adds, Stone " pro- ceeds wholly upon the basis of the lengths of vowels in isolated words." This is equally incorrect. The difference between '■'river in" and "rivers shall," on Stone's showing, is in no sense a difference of vowel-quantity ; it is a difference of what Mr. Thomson calls syllabic burden. One is driven to doubt whether the critic understands Stone's contention ; a contention, let me add, which I reject as heartily as he. And the doubt gathers strength when one notices his treatment of a casual line (p. 19, first example), which is thus divided — First the | rivers shall | cease to re- | pay their | floods to the | ocean. I can hardly be wrong in supposing this a "quantitative hexa- meter " written on principles similar to Stone's. If so, it was meant to be read — First the ri | vers shall | cease, etc. Does Mr. Thomson say this is an unnatural way of reading the line ? I quite agree. Does he say it implies ignoring or altering the natural accentuation ? On Stone's principles it does neither one nor other. In any case, might not the alternative scansion have been at least mentioned ? I do not share or wish to defend Stone's view ; but it should be understood before it is rejected. Some other short criticisms need not detain us. The most adequate exposes Ruskin's wild metrical vagaries. One or two incidental remarks may be referred to again. But I hasten to pass from Mr. Thomson as critic to what is of far greater conse- quence, his general conception of verse. This is set forth in the remainder of his essay (§22 to the end) with great clearness and ' 14 admirable wealth of example. Section 22 itself summarizes re- sults already reached, adding some further definition and explana- tion of terms. 1 read with surprise, in its second paragraph, that " nearly all writers think it immaterial whether ... a trisyllabic foot is represented with the accent at the beginning, in the middle, or at the end " ; nor do I see why accenting " and the plain " on the last word implies dividing that word between two feet. On almost every page some such remark arrests one, and tempts to discussion ; but the larger question claims precedence. We should notice that Mr. Thomson does not pretend to give a complete picture of verse. His survey is confined to its rhythmical form. He well knows that aesthetic effects depend partly on nop-rhythmical factors ; on warmth and colour, the qualities of vowels and consonants, the instinctive modulations of tone. These lie outside his task. He restricts himself to re- cording the beat and movement of syllables, as might be done even with inarticulate sounds (cf. last sentence of § 20). 1 agree as to the immense value of doing this. To our ordinary proso- /dists it is folly, since they as a rule wholly ignore time in_verse. Such ignoring (or shall we say ignorance ?) I protest against with all my heart. Like Mr. Thomson, I consider rhythm the most essential and indispensable feature of verse. The music of a line is due to its time-relations as well as its stresses. But I cannot think either that musical notation rightly represents these time- relations, or that Mr. Thomson's scansions are based on a true theory of the nature of verse. I must try to justify my increduUty. This can be best done by going at once to some of Mr. Thomson's final results. On p. 56 he pronounces five lines wholly irregular, and capable of being read only as prose ; " no other reading seems possible, without doing violence to the English language." These five lines are as follow. Blackened about us, bats wheeled, and owls whooped. Beat ; merrily blowing shrilled the martial fife. One calls the square round, t'other the round square. To stop song, loosen flower, and leave path : Law . . Lies to God, lies to man, every way lies. IS These are lines by no unskilled singers. I doubt if any ordinary reader ever thought there was anything wrong with them, unless the last perhaps troubled him a little. The third line is prosaic, as Browning not seldom chose to be, but its metrical structure is not more irregular than, for example, this of Milton's — Hail, Son of the most High, heir of both worlds. More difficult lines could easily be found ; these are all the fairer as "test cases. They are lines which might be paralleled by scores of others, not identical in cadence, but equally far removed from the normal rhythm. Let us consider what is involved in judging such lines unmetrical, to be read only as prose. It cannot be too often repeated that in prosody we start from no fixed rules. We start with a large body of practice, the work of our best poets. We examine this, trying to discover the principles on which they worked. That such principles exist, we are sure. No artistic work can be lawless ; " it is all triumphant art, but art in obedience to laws." If we can discover and truly state such principles, they may be fairly called laws of verse. We know, further, that poets sometimes make mistakes, write bad verse. One or two infractions of a supposed rule do not necessarily disprove its existence. But if, after formulating a law, we find it habitually infringed by our chief poets, no middle course is open. We must withdraw our formula, confess it erroneous, probably because modified by some other principle which we have omitted to take into account. If we arg slow to do this for ourselves, others will do it for us. The history of English prosody is full of such experiences. Time and again, our critics have laid down rules, by which our poets have. refused to be bound. Time and again, they have declared impossible things which our poets have cheerfully pro- ceeded to do. It is exceedingly rash to dogmatize about details, to lay down hard and fast rules tethering a poet's freedom. As a mere matter of fact, such hard and fast rules have generally been falsified by the event. Nine times out of ten, I shall venture to i6 say, in cases of dispute between critics and poets, the critics have been proved wrong and the poets right. The tenth case remains our consolation and encouragement amid our many mistakes. Accordingly, when Mr. Thomson pronounces lines Hke these incapable of accordance with his rules, I feel certain that his rules are too narrow. He may say that these lines contain dehberate irregularities, used as a musician uses discords to enhance his harmonies. To some extent this is probably true, but it is not the whole truth. When good poets write such lines, and intelligent readers enjoy them, it is difficult to think they hear them merely as discords, as sections of prose. I believe myself that a juster perception of what verse is will remove any difficulty we have in calling these lines metrical, and guide to wider and truer notions of prosody. The flaw in Mr. Thomson's method, to my mind, appears when he handles the syllables of verse. Determining their natural prose rhythm, he regards this as giving their metrical value. Here, for instance, is how (p. 6i) he divides the first line of Milton's L'Allegro— Hence, | loathed | melancholy. I'he last word he would have us read in one foot, just as we read the quadrisyllabic feet in this line of Mr. Gilbert's — A I cheap and chippy | chopper on a | big black | block. This is doubtless how we read it in prose, but assuredly does not represent its place in Milton's fine. There, beyond all doubt, we must somehow spread this word over two feet; two main rhythmical beats come on the word. How this happens may be matter of argument ; that it does happen admits of none. The rhyme with " unholy " is enough to prove it. And the whole line must be read in three feet, not in four, as succeeding lines make clear. This implies some reduction of stress on the word " hence." The metrical structure of the line, in short, is quite other than the prose division of the words would lead us to suppose. 1) I do not forget what, on pp. 49-52, Mr. Thomson says, and illustrates by examples, about what he calls " Heel rhythm, Head rhythm, and Heart rhythm." He recognises that there must be " a compromise between pure prose reading and strict adherence to the verse scheme.'' He is on safe ground in censuring alike those who read prose as verse, and those who sacrifice sense to rhythm by sing-song; I do not doubt that his own ideal is a happy mean between these. But I find, with sorrow, that his scansions continually set aside the natural verse-rhythm in favour either ot mere prose, or of some supposed musical precedent. (The scan- sion of " Take your own time, Annie," on p. 43, illustrates both peculiarities.) Every " heroic " line he strives to show possessed of five main stresses. Yet Milton has lines which seem clearly to carry but four, e.g. — And made him bow to the gods of his wives ; and others which carry more than five, like that already quoted — Hail, Son of the Most High, heir of both worlds. No theory of verse, in my belief, can possibly be adequate which regards only syllables and their natural speech-accents. Such regard perceives only the non-metrical elements of verse, and is blind to what really differentiates it from prose. The view of verse illustrated in this essay seems to rest on a theory thus limited, and the extent of limitation is revealed both by its reject- ing lines which ought to be admitted and by its treatment of those which are admitted. I must try to set forth briefly what I think a truer as well as broader theory, one which will find room for what is valuable in Mr. Thomson's analysis while rejecting conclusions too hastily deduced from it. I hold that verse is essentially a setting of syllables to time. There are several ways of doing this. The old Greeks, and the Romans who followed them, did it by assigning a relative temporal value to each syllable. With negligible exceptions, every syllable in their verse was accounted of either single or double value, " short " or " long," occupying one time-beat or two— as we might i8 say, a quaver or a crotchet. The same ratios held good in prose. Therefore the temporal fabric of any Greek sentence, in prose or in verse, can be determined certainly by beating time to its syllables. If we neglect to do this, — if, by substituting accentua- tion for quantity, we attribute wrong rhythms to Greek verse,— the loss is ours. Facts are not changed by being misread. No such canon exists in our language. No English syllable has a fixed time-value, whenever and wherever it occurs. Some syllables take longer to pronounce than others, and it would be strange if our poets made no use of this fact. But we pay little attention to it in our speech. Our dominating stress-accent reduces quantitative distinctions to insignificance, and makes them an insufficient basis of metre. Wherever stress-accent is weak, we find verse resort to other means of signalizing structure, attach importance to the bulk or the number of syllables ; where it is strong, these latter are of small account. One would not expect to find the same basis of prosody in ancient Greek, or in modern French, as in English or German. Human ears, presum- ably, do not differ; but speech-habits do, and verse differs accordingly. r But, just as it does not follow that there are no quantitative differences among our syllables, so it does not follow that our verse bears no relation to time. Because one method is not used, it does not follow that none is. Verse unrelated to time would be an inconceivable monstrosity. If time cannot be predicated of syllables, there must be something else of which it can be predicated. That something I take to be the ''period" or time- unit (usually styled z.foot), which contains the syllables. Contains them, not consists of them. Historically as well as logically, the "period" precedes and is independent of syllables. It may contain one, two, or more syllables, or sometimes none at all; but its duration is not affected thereby. Stress-accent on syllables is the normal and natural instrument used by our poets to mark their periods, but it does not create the periods. Nor do syllables create or wholly define the period; it is prior to and separate '9 from them, with a character of its own. The time-beats of a period need not correspond with the prose accentuation of the syllables contained in it. To musicians this will seem heresy. A musical bar which is in one "time," while its notes are in another, would be indeed unthinkable. But verse is not music, nor a department of music; it is a sister art, with methods of its own, and may employ the immutable laws of rhythm in ways which have no musical pre- cedent. That there is some essential difference of method seems argued by the fact that excellence in one art rarely accompanies excellence in the other. Great poets are seldom musicians, great musicians seldom good writers of verse. So far from being an objection to my view, it may rather be deemed an argument in its favour that it recognises — as will presently appear — a wide and essential difference between the methods of verse and of music. Whether the time of a metrical period bears any relation to musical " times '' may even seem doubtful. It is tempting to think that the time-unit of verse is always one and the same, differing merely in the number of syllables (and pauses) it contains. But this view seems to me opposed both to reason and to the testimony of our senses. The difference between "common time" and "triple time" is so palpable, so universally felt, that I cannot conceive our poets failing to profit by it. And, as a matter of fact, I find this difference underlying our verse. I recognise it, not by listening exclusively to syllables, but by perceiving time-beats which are often at variance with syllables. I thus reach conclusions very different from those taught in this essay. To explain the difference, I must state precisely how I conceive an English poet to work. I take it that he starts knowing his hearers will adjust his syllables to either common or triple rhythm. His first business is to show which of these is intended. A very slight indication suffices for this. Our ears, as I said in my Study, are quick to catch any direction affecting time. It is not the mere prose value of the syllables which decides it ; the very same words, with the very same accents, will be set now to one time, now to another. Both quantity and accent are subjugated by time-beats ; they become as clay in the hands of a potter. Syllables are but counters in the game; yet they are counters. Two syllables ^suggest duple time, three triple, four quadruple. In unfamiliar measures, syllables and time-beats must correspond closely, else we fail to catch the swing; in familiar ones there is no such necessity. It is a crude view of verse which supposes that word- accents must always synchronize with rhythm-accents, or syllables with time-beats. In all our ordinary verse the time-measure is known beforehand, which enables poets to deal very freely with their often stubborn and recalcitrant material. For the poet has much more to do than merely to reproduce time-beats. On the top of his temporal pattern he weaves an accentual. This latter, in our highly developed literature, attracts chief attention. Most of our metrists regard nothing else. Some of them even talk of " stress-rhythms " — as if stresses, in and by themselves, ever did or could create rhythm. That function belongs to time alone. Accentual prosodies, from the "innocent" system of Latham to Mr. Thomson's elaborate notation, err by supposing that verse is to be scanned merely by its speech-accents. Childish " sing-song " seeks to reinforce mental perception of periods by assigning a stress to each, but adult ears need no such hammer-beat. They can perceive recurrent periods without thumping at each, just as a competent musician does not beat time aloud. They do, in fact, perceive periods not merely when these are not emphasized by stress, but even when the accentual pattern varies from and conflicts very considerably with the temporal. It is precisely in this contest between temporal and accentual that I find the charm and secret of good English verse to reside. Our speech-habit prompts us to throw weight on every alternate (or sometimes every third or even fourth) syllable ; our rhythm- instinct urges us to keep the time-beats equal. By playing ofif one tendency against the other our poets secure richness, variety, and 21 elaboration. They do not treat syllables as musicians treat sounds; that would be .suicidal. Music has its intricate and elaborate notation, its symbols created expressly to record minute fractions of time. Its ideal is absolute faithfulness to time. Syllables exist before verse handles them, and are not wholly amenable to its handling. They cannot be coaxed to keep exact time, and of course cannot be chopped or carved into fragments. From this very inability, poets in their unconscious inspiration draw beauty. They delight us by maintaining a continual slight conflict between syllables and time. It must not go too far, or the sense of rhythm perishes, and the line becomes heavy, inert, prosy. But within limits the contest is unceasing. Mr. Thomson's very first example (p. 15) represents Will you walk into my parlour, said the spider to the fly, as a uniform row of fourteen quavers followed by one final crotchet. Does this in the least represent the syllabic variety which obtains in even so trivial a line ? 1 do not know how far musicians are at liberty to depart from precise time in playing similar rows of quavers ; strictly speaking, I imagine, not at all. Anyhow, such mere difference of tempo is quite another thing from the structural differences between syllables. Perfect regu-\ larity, exact recurrence of syllables, is about the last thing EngUsh poets strive after. Variety is their aim, monotony their abhorrence. Contrast between the absolute uniformity of time- beats and the inherent non-uniformity of syllables secures perpetual variety, and gives good verse an orchestral effect not unworthy of comparison with the grandest harmonies of music* * A lately republished paper by R. L. Stevenson ("On some technical elements of style in literature;" Essays in the art of writing, Chatto and Windus, 1905; this paper re- printed from Contemporary Review, April 1885) deals most suggestively with this "double pattern " in verse. I do not hold with all the notions borrowed by Stevenson from his friend Fleeming Jenkin. It is not true that a heroic line must not divide into five syllable-groups, or cannot consist of only two, nor begin with a "double trochee ; " instances of all these are found in good writers. But when he says that the laws of prosody "have one common purpose, to keep alive the opposition of two schemes simultaneously followed ; " and when he compares verse- writing to juggling with several oranges at once ; he shows himself master of the main principle. Eventually he makes out that the verse-writer keeps Jive oranges going at once ! Needless to say the paper charms by manner as well as interests by matter ; I would fain quote many sentences. One I must reproduce for Mr. Thomson's benefit, since it throws at least a side-light on our question. " No verse can ever sound otherwise than trivial when uttered with the delivery of prose." Instead of making triple rhythm the basis of our verse, I restrict this — so far as time-beats are concerned — to what are wrongly called our dactylic and anapaestic measures (wrongly, since true dactyls and anapaests imply quadruple time) ; and I claim that our ordinary so-called iambic verse moves to duple time. It is unbelievable that we read Paradise Lost to waltz- rhythm. The mistake comes from attending only to syllables. These tend to form iambs (or, if the musical order be followed, trochees), but this tendency does not prevent our mental re- tention of the underlying spondaic beat. If any one will honestly tap out triple rhythm on a table, and then think whether this represents the impression left on him by weighty blank verse I am persuaded that he will feel a difference. The "common time" of music is, I am convinced, also the common time of verse, however disguised by syllabic variation. Milton's majestic music is not based on the "jig'' of triple rhythm. What I have elsewhere styled the " higher criticism " of verse seeks to unravel this double thread, and show precisely how syllabic structure is related to temporal. When this work is entered upon, Mr. Thomson's close analysis should be of great service. He untwists one thread with admirable skill; but his whole analysis assumes that this is the only thread, that the rhythm of its syllables constitutes the rhythm of a line. This idea once dismissed, the way would be clear for him to advance the said higher criticism. It is work that much wants doing. The field is almost virgin, and there is room for many labourers. But uniformity must be made clear before we consider diversity. Variations are only inteUigible when we know from what they vary.* As a single illustration, take "inverted feet." What, pre- cisely, does a poet mean when he writes such a line as— He left the upland lawns and serene air ? The common explanation makes " serene " a foot turned back- V. ; ''^'"''"''""'^^'^^"^io" °f tl>= groups is an adornment; but as soon as the original beat has been forgotten, they cease implicitly to be eccentric. "-R. L. Stevenson, as befor,. 23 foremost. Accentual scansionists, on the other hand, would divide the line thus — He I left the | upland | lawns and se- | rene | air, possibly indicating a pause of some sort to fill up the fourth foot. Which view accords best with what seems the practice of our poets ? It would take a paper as long as my present one to discuss that question properly. The short answer, that " serene " must be accented on the first syllable, may be dismissed as childish. Yet the accentualist's view seems hardly nearer truth. It is a possible reading, certainly ; but is it the way we really hear the line ? Is it the way we can conceive a poet writing it ? That triplet in the middle — does it exist to our ear ? Does it not seem a departure from the line's true structure, involving a change just as much as " serene ? " Both seem artificial scansions. Shelley's Epipsychidion contains the remarkable line — Harmonizing silence without a sound. Is it possible to think that "silence" does not repeat the cadence of " Harmo- | nizing?" To make the third foot contain an additional syllable (" silence with ") surely is false scansion. I assume, of course, temporal equality in the feet or periods ; Mr. Thomson and I hold that equally strongly.* The accepted ex- planation, that these are real cases of inverted feet, seems to me on the whole preferable to any other. Such inversions are most common in the first foot of a line, though by no means rare in other feet, even the last. In all such cases Mr. Thomson and others find a trisyllabic foot, as in one quoted by him (p. 31 ; cf p. 37, last sentence of first paragraph ; ax\A passim) — Seemed to have | known a 1 better | day. This must be pronounced at least open to doubt. The "higher criticism " of verse will consider all such questions, also what is implied in such " setting " of words to time as this of Keats — The enchantment that afterwards befell. And it will have much to say about lines which have less or more * It is, however, somewhat amusing to find the phrase "equal periodicity" spoken of as an accepted principle. To most of my reviewers this idea seemed novel. I do not claim to have invented, or to have property rights in, the phrase ; and I quite agree that it ought to be the veriest commonplace of prosody. But I fear that as yet it is far from being so. than the normal number of stresses, rejecting utterly the device of postulating imaginary accents. But such criticism cannot begin till a real foundation is laid, and will be useless while it restricts its view to one side only of the complex tissue of verse- structure. Not till it distinguishes temporal and syllabic, in their friendly yet antagonistic co-operation, shall we reach any satis- factory analysis of the phenomena of English verse. Accentual scansionists nearly always minimize the difference between verse and prose. For, taking English syllables by them- selves, there is really no difference. The difference — a real and true one — lies in the set ting. Verse sets syllables to equal time- l measures, prose to unequal. When either poaches on the other's Ipresetve, we are apt to resent it. One heroic line in prose may escape notice, but hardly a second. That the difference does not lie in the syllables themselves appears from the fact that the same sentence may sometimes be read as prose and sometimes as verse. When we first read " And the doors shall be shut in the streets when the sound of the grinding is low,'' we probably hear it as prose ; but once let it be compared with — 1 I am out of humanity's reach, \ I must finish my journey alone, and it will be difficult ever after not to receive an impression of verse. The difference is subjective. The same words will be prose to one man, verse to another. Whether we find prose or verse in Webster's line — Cover her face ; mine eyes dazzle ; she died young, depends on whether we can or cannot relate these words to the five equal time-measures of heroic verse. So with the five lines rejected by Mr. Thomson ; so with Milton's And made him bow to the gods of his wives ; so with every difficult or unusual line that can be cited. How, precisely, the relation of syllables to time in each case can best be explained and determined is for the " higher criticism " to ex- pound ; but each of us has to solve the question for himself. If through laziness, inattention, or ignorance we fail to solve it, 25 the words remain prose to us. That is why help in solving it is so eminently desirable. Just as the difference between prose and verse is one of setting, so is the difference between duple and triple metre. It depends on how we hear the time-beats. Mr. Thomson says (foot of p. 36) — " Had Mr. Lanier or Mr. Omond met ' Who would believe ' or ' Seenied to have known ' in Browning's ' Kentish Sir Byng stood for the [his] king,' they would have had no doubt at all of its triple character." I should have had no doubt that the words were then set to triple rhythm, because to my mind that is clearly the time of Browning's poem ; but when I meet these phrases in heroic or octosyllabic verse, I read them to a different time. In themselves the syllables are not metrical, but they can be set to either rhythm. The poem gives rhythm to the syllables, not the syllables to the poem. " For poets do not adjust times to syllables, but syllables to times.'' Present limits prevent giving more than this bare outline of my view, which I commend to Mr. Thomson's candid considera- tion, as supplementing rather than contradicting his. It is true that I cannot admit that musical notes adequately represent syllables, since they fail to reproduce the double function just described. Yet the examples in this essay convince me that they can give a closer approximation than I had deemed possible. I have always maintained (e.g. Study, p. 15) that they are the natural means of teaching rudiments of verse ; that a child will learn more from them than from half-understood Greek names which are wrongly applied by our grammatical pundits. I now see that the representation of verse in terms of music can be carried further than I had previously imagined. But I still say that the representation is imperfect, and that the methods of one art are not necessarily those of another. It seems to me al- together certain that verse is not merely a department of music. Many characteristic remarks occur in these later pages. " Normally — though no else notices it [the italics are mine] — there seems to be a gentle crescendo in nearly every unbroken line of 26 verse" (p. 52). I think other writers have noted "the strong finish in which English verse delights," and this seems a truer description of the phenomenon, since the effect can be equally well obtained by a diminuendo, instances of which are not far to seek. That "grave difficulties attend the disentanglement of varying degrees of accent" (p, 53) is an exceedingly true remark, which I should like to have seen borne in mind throughout the essay. The notation on p. 48 of the metre of Shelley's Away ! the moor is dark beneath the moon, admirably brings out the " rests " in the first line, and the identical length of all the irregular-seeming lines in the stanza. But I must not be tempted to criticise scansions in detail, or attempt the ungracious task of showing where and why I diifer from Mr. Thomson on almost every page. The difference is one of general view, which naturally entails divergence in particulars. Even on his own principles, however, it seems to me that pentametric periodicity is violated in the third example on p. 44 ; nor can I see why, in the sixth on p. 60, " no better " is treated as if it were " No ! better, etc." In the fifth on p. 59, surely " fire " has dis- syllabic value. And I must protest against the dictum (p. 62) that "sense, rhyme, and rhythm are all at odds " in this couplet of Pope's — Reason's whole pleasure, all the joys of sense, Lie in three words — Health, Peace, and Competence. If objection be really taken to competence as a rhyme for sense, it is a startling instance of how theories mislead. Such rhymes are sanctioned by all our great poets, and enjoyed by their readers ; a critic's business is to find out why this is so, not presumptuously to pronounce it wrong. The rest of the dictum I do not under- stand, seeing nothing particularly amiss with either sense or rhythm ; the objection on score of rhyme is certainly inadmis- sible. But I must end, as I began, with praise of the excellent work which is at least inaugurated in this essay. Its insistence on time, its unflinching belief in periodicity, are merits that would outweigh 27 many defects. Its notation is remarkably clear, enabling one readily to detect points of difference. I trust it will be widely read, but read with attentive criticism by students not carried away with its semblance of finality, and the apparent rigour of its deductions. As yet I have seen no adequate notice of it, which is much what might have been expected. Any tenth or twentieth re-hash of aesthetic criticism finds easy access to our reviews ; but enquiry into the structure of verse is caviare to most editors. The general public, however, is by no means loath to consider ques- tions which obtrude themselves on every intelligent reader of our literature ; this my own experience shows. To its attention I recommend Mr. Thomson's essay, than which no more vivid pronouncement on matters of prosody has appeared for many a day. 14 Calverley Park, Tun BRIDGE Wells, October igo^. i. ^^ ^A '^^fM>^