CORNELL UNIVERSITY LIBRARY UNDERGRADUATE LIBRARY DATE DUE m^&m- TSiji f*»»™pj ijaumffS; GAYLODD NTED IN U S > Cornell University Library PR 3597.B85 1921a Milton's prosody with a chapter on accen 3 1924 014 156 859 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924014156859 CMILTON'S TROSODT dMILTON'S TROSODT with a chapter on ^Accentual Verse & Notes by ROBERT BRIDGES Revised 4 & Final, * hi Edithn d OXFORD i AT THE CLARENDON PRESS \ First published 1921 Reprinted lithographically in Great Britain at the University Press, Oxford from sheets of the first edition 1965 PREFACE WHEN I had finished the revise of this book, and in some parts rewritten it, the need of a preface confronted me, and, feeling disinclined to write anything, I fell to considering prefaces in general, and I thought of that great treatise-maker, Cicero, who, if I truly remember, kept a store of prefaces on hand, so that when he had perfected any dissertation, he had but to select from his stock the accomplished little lucubration that appeared most suitable, or, as we should say, the one that would do best. But even had this elegant method wholly pleased me, I had no prefaces ready on hand : and then I saw what a thousand pities it is that a book cannot write its own preface. One imagines the growing book as its organization develops and gradually gathers into unifying existence, bursting at completion into personal self-consciousness, open- ing like the flower of a child's mind to the miracle of its being ; then I fancied how it would come slowly to muse on its creation, to feel the discom- forts inseparable from mortal existence, till it arose in the rebellion of artistic dissatisfaction to be critical of its creator. Few indeed are the books which, like the children of the wise woman, would rise up and bless their parent : they would talk rather like iv Preface those who with preposterous intelligence grumble at their fate, complaining that their brains are too de- pendent on their stomachs, or that their knee-joints are clumsily fashioned, and their toes unsightly and useless ; they might even emulate the bold proficiency of the German Helmholtz who asserted that, if he the creature had only been the Creator, he would have supplied mankind with a better eye. Then I took sorrowful compassion on my deaf and dumb child, a poor little grammar, not born to be clothed in gorgeous raiment of morocco or enamelled leather, to lie golden-edged on drawing-room tables or by the king's bedside ; yet surely with some honest faculty of delicate feeling and, alas ! all the inconveniences of an embryonic and embarrassed in- heritance, pains like to those which we ourselves — whether from bestial ancestry or a fall from Paradise — know too well, our Dropsies and asthmas and joint-racking rheums, with all other ills that flesh is heir to : and with these pitiable imperfections of body it would bewail its ignorance, the frailties and baulking limitations of its reasoning powers, and be deeply troubled at soul by unintelligible glimpses of spiritual beauty, those adumbrations of glory, those interrupted strains and broken echoes of poetry, those flashes of Miltonic music that are embedded in it without consequence or correlation. I wish, indeed, that it could relieve itself by utter- ance of vituperation against me its maker. I should Preface v rejoice, not only because my sense of justice and sport would welcome it — nor would I resent unpleasant truths — but because such a prelude would be attrac- tive and useful to my readers, and supply that first utility of a preface, which is to spare critics the labour of examining the book. This cannot be. One service, however, I can render better than the book itself could have done it ; I can tell the story of its creation : but as that is not fit for a preface I shall put it among the notes at the end. CONTENTS PART I ON THE PROSODY OF PAR. LOST Method of Treatise Digression on Quantity I. Exception to the syllables being ten Digression on the midverse extrametrical Digression on Elision II. Variety in number of stresses . III. Inversion of feet . Remarks on the break in the verse PAGE i i 2-4 4 6-8 9-1 8 37 40 43 PART II ON THE PROSODY OF PARADISE REGAINED AND SAMSON AGO- N1STES 46 The relaxation which is found in Paradise Regained and Samson Agonistes of the laws of ' Elision ' in Paradise Lost . . 46 The prosody and rhythm of Samson Agonistes . . . . . .50 PART III ON OBSOLETE MANNERISMS . 67 Recession of accent .... 67 Spelling ...... 77 Pronunciation ..... 79 viii Contents PART IV PAGE ON THE PROSODY OF ACCENTUAL VERSE 85 The rules of the common lighter stress- rhythms, and of the English accentual hexameter . . . . .89 Heavier accentual rhythms . . -103 The accentual hexameter . . .105 NOTES 113 PART I ON THE PROSODY OF PARADISE LOST In this treatise the scheme adopted for the exami- Method, nation of Milton's matured prosody in the blank verse of Paradise Lost is to assume a normal regular line, and tabulate all the variations as exceptions to that norm. For this purpose English blank verse may con- veniently be regarded as a decasyllabic line on a disyllabic basis and in rising rhythm (i. e. with accents or stresses on the alternate even syllables) ; and the disyllabic units may be called feet. Let such lines as the following be taken as normal lines, 1 Of that Forbidden Tree, whose mortal tast. i. 2. Torments him ; round he throws his baleful eyes. 56. A Forrest huge of Spears : and thronging Helms, 547. and we find that these lines have ten syllables with five stresses all on the even places. In the following chapters we will examine the exceptions to these conditions, namely : I Exceptions to the number of syllables being ten, II Exceptions to the number of stresses being five, III Exceptions in the position of the stresses, and this will give all the variations due to prosody ; Con- fer quantity, though a main factor of rhythm, is not sideration considered in the prosody of syllabic verse. ofquantity 1 In accentual blank verse been satisfactorily determined, these would be normal lines, see nor will it be discussed in this p. 38; but the rhythmical basis book; see ch. II, pp. 37 and of syllabic blank verse has never 38, and again on p. 84. Digression What quantity No quan- titive rule for the feet in English syllabic verse. [§ Digression on Quantity. Quantity, which means the relative duration of time which different syllables fill in pronouncing, is an omnipresent efficient factor of rhythm, and as we are not going to deal with it, we shall do well to exhibit exacdy what it is that we are excluding. Take the first of these regular lines quoted above, (i) Of that | Forbid|den Tree, | whose mor|tal tast: this line may be read extremely well with all its five accents at perfectly regular intervals of time : Let it be so read. If this reading be now set out in musical notation, with the isochronous musical bars (as is necessary) before the accents, we shall not get -J-l-J— J- 1 J J | J J | J-J-fJ- Of that ' For - bid'- den Tree', whose mor'-tal tast nor _«LLg J | e t-j-f-J- J | J J | J - but something like this J | J J | J J- J A JJ-J el | J- that is to say the accent in the second foot forbid is on a very short syllable bid, and the unaccented syllable den is held on to fill the bar : and this affects the rhythm very deeply, but it does not concern the prosody ; that is to say, the prosody admits of either long or short syllables in any place of the verse. If now we were to try to write this same line in equal- timed prosodial feet we should get Of that | Forbid'n|n Tree, whose mor-tal tast : and although one very effective and common way of reciting the verse of P. L . is to set up an equal-timed musical beat and keep as nearly to it as possible, yet such a reading will sometimes give five and sometimes on Quantity 3 only four bars to the verse ; and if it serve for a rhythmical interpretation, it will exhibit to the ear, as the notation above exhibits to the eye, the fact that time-value or quantity is not considered in the prosody of English syllabic verse any more than classical prosody concerns itself with the rhythms produced by the incidence of the verbal accents on the prescribed prosodial units, although in each case these rhythms are primal factors in the beauty of the verse. The example of the word forbidden will show what Common it is that English writers on metre confuse, when they confusion call accented syllables ' long ', and take all unaccented accent .... , ' , , b ' andquan- syilables to be ' short . t ; ty- That in syllabic English verse the prosody proper is not concerned with the rhythmical effects caused by 'quantity' (i.e. by the different lengths of the syllables when spoken) may possibly give rise to the idea that there is no such thing as ' quantity ' in English speech : and if a man can persuade himself that he is insensible to the actual different time-length of spoken syllables — as roughly illustrated above — he may possibly consider ' himself at liberty to apply the terms 'long' and 'short' to accented and un- accented syllables as such. The confusion is of course irremediable ; and it is little credit to such exponents of verse that, having deprived these essential terms of their proper meaning, they do not, when they discuss rhythm, seem hampered in their vocabulary by the absence of any terms that distinguish these primary and omnipresent conditions. To clinch the absurdity, note the indisputable fact that they cannot speak without differentiation of the quantities of the syllables, but yet they maintain that they cannot differentiate them. It is fruitless to show colours to the blind. One example may be of use. The Greek word typical reTVfj.fji.iv6s is in quantity and accent similar to such example. line, 4 Number of Syllables English words as scientific, apostolic, unemphatic, disembody, recognition, unambitious, anaesthetic, &c, and may be represented in musical notation thus, ESE T€ - TV/A -fJL€- VOS an - aes - the - tic ; but boys are taught to accent the Greek word as if it were Latin, and the Latin, rule being that short penultimates are unaccented and in polysyllables throw the accent back, they pronounce the word thus : rmnr re - TV/X. - (V - OS in doing this the long unaccented syllable tv/jl,u. (with, its double a) is shortened by being accented, because the pp is treated as if it were merely the English device of marking the short quality of a preceding accented vowel, instead of giving syllabic length by the pro- duction of the liquid /x. Both accent and quantity are thus falsified, the long syllable is pronounced short, and the accented syllable is unaccented ; and thus it is that boys are expensively trained to be incapable of distinguishing between accent and quantity, and to read all Greek and Latin verse wrongly.] I. Exceptions to the syllables being ten. These exceptions may be either of deficiency or excess. ^L Deficiency of syllables. Nine- There is no example in P. L. of a line having less syllable th an ten syllables ; but this is worth noting because it is probable that Milton was acquainted with Chaucer's practice of using a line that omits the first Final extrametrical 5 unaccented syllable, as in these examples from his most perfected style in the Prologue to the Tales : Ginglen in a whistling wind as clere. I 70. For to delen with no swich poraille. 247. Twenty bokes, clad in blak or reed. 294. and one cannot read Chaucer unless one is prepared for this. It must be concluded that Milton rejected this form of the line, though the question may be raised again in considering the rhythms in Samson Agonistes. In the first edition of P. L. there hap- pened to be a nine-syllable line printed at x. 827, With me ; how can they acquitted stand ; this was corrected in the 1674 edition thus, With me ; how can they then acquitted stand. (Beeching, p. 40 1.) d. Supernumerary syllables. These should be separated into two classes. First, those which may be considered as Extrametrical; secondly, those that are to be accounted for by Poetical Elision and speech contraction. And first to define the extrametrical. § Extrametrical syllables. 1. At end of the line. An extra syllable sometimes occurs at the end of Final the line, more rarely in Milton than in most writers, extra- e rr metrical Of Rebel Angels, by whose aid aspiring, i. 38. sometimes there are two such syllables. Imbu'd, bring to thir sweetness no satietie. viii. 216. For solitude somtimes is best societie. ix. 249. P.R. i. 302. It is possible that these words satiety and society are allowed in this place because they admit of ' elision ' [see below] and can be therefore considered as single 811629 B 6 3\did-verse 1 hangers ' : but the elision being optional — (compare the following lines What higher in her | soci|etie | thou findst. viii. 586. Him with her lov'd | soci|etie, | that now. ix. 1007. In solemn troops, and sweet | societies. Lye. 179. ) — all such endings, having two syllables extrametrical at the end, whether theoretically elidible or not, will still have a hexametric effect, and they do not differ from verses intended to have six feet. This ambiguity of prosody is illustrated in Sam. Ag. see p. 61. The matter worth noticing is that Milton is sparing in the use of all such endings in blank verse. (See note B). Mid- 2. In other parts of the line. verse j n Shakespeare it is common to find an analogous metrical syllable in the midst of the line ; and thus Milton, in his earlier work, e.g. in Comus, To quench the drouth of Phoe(bus), which as they taste. 1. 66. And as I past, I wor(shipt) : if those you seek. 302. And earths base built on stub-(ble). But com let 's on. 599. But for that damn'd magi-(cian), let him be girt. 602. Root-bound, that fled Apol-(lo). Fool do not boast. 662. Cramms, and blasphemes his fee-(der). Shall I go on ? 779. In P. L. Milton disallowed the use of this syllable. In lines like the following, Of high collateral glor(ie): him Thrones and Powers, x. 86. where its rhythmical effect is maintained by the colon, the extra syllable is accounted for by l elision '. [§ Digression on the midverse extrametrical syllable. its I suppose there is no doubt that this midverse French extrametrical syllable came from the old French ongin ' practice of regarding their twelve-syllable line as com- posed of two hemistichs divided by a marked caesura : concerning which Littre writes in the preface to his translation of the first book of the Iliad : ' Autrefois l'bimistiche £tait consider! comme une fin de vers. Ainsi dans un poeme du xn e siecle il est dit de Berthe : Sxtrametricals 7 Oncque plus douce chose ni vi, ne n'acointrai ; in twelve- Elle est plus gracieuse que n'est la rose en mai. syllable Et . . . d'un guerrier blesse 1 a mort : verse. Pinabaux trebucha sur 1'herbe ensanglantee Et fors de son poing destre lui echappa l'epee. Cette habitude est constante ; ' but it was discarded in the seventeenth century. The practice also invaded the French ten-syllable in ten- verse, and as this has no middle it divided it unequally, syllable There are two examples in these five lines : Quant vient en mai que l'on dit as Ions jors, Que Franc de France repairent de roi cort, Reynauz repaire devant el premier front ; Si s'en passa lez lo meis Erembor, Ainz n'en dengna le chief drecier amont. 1 This extrametrical syllable being originally attached in to the old caesura of the twelve-syllable line, its place Shake- is properly after the sixth syllable, as in all the examples s P eare " quoted from Comus, but the indeterminate position of the break in ten-syllable verse allowed it to appear in other places as a few quotations from Shakespeare will illustrate. After the fourth place it is common, and this corresponds with the French examples just quoted, Burnt on the wat(er) : the poop was beaten gold. Ant. 6° Cle. ii. 2. From mine own know(ledge). As nearly as I may. Ibid. but in The Tempest, his last play (?), we find So dear the love my people bore (me), — nor set. Temp. i. 2. With all the honours on my broth(er) : whereon. This extrametrical syllable within the line is then a borrowed licence and has no title to admission into English syllabic blank verse, but Shakespeare made a very good use of it in his dialogue. Where a line is divided between two speakers, the second speaker often disregards the last syllable of the first speaker, 1 Quoted from Les Po'etes fratifais. Crepet, vol. i. p. 42. Twelfth century. 8 ^Mid-verse Sxtrametricals and treats it as extrametrical. This avoids the effect of the second speaker having his answer conditioned for him by the first, who being in possession of the line ceded as it were only as much as he chose ; and in drama the value of a reply is actually impaired, if it seems to be led up to and prearranged, so as to fall pat. A stichomythia, as it is called, in which each speaker is bound to fill and not to exceed one line, requires the art to be free from all realism whatever ; a condition not often presented by our drama. The extrametrical syllable in the condition above described is so common in Shakespeare that the ear becomes familiarized with it, and does not resent it in other places : it was freely abused by the Eliza- bethan dramatists : it has probably become con- founded with the true trisyllabic foot and imagined to be a bad attempt at that : some modern writers have thus used it, with a sort of affectation of antique robustiousness.] its effect There is no foot in Milton's line where this effect obtained cannot be obtained by interrupted elision, as by elision. Departed from (thee), and thou resembl'st now. iv. 839". Before (thee); and not repenting, this obtaine. x. 75. and the conditions are sometimes very elaborate, e.g. in the following line the last syllable of condescension appears to be extrametrical, if the prosodial elision of be honour d is neglected ; and as that is optional, it cannot be said that the effect of the extrametrical is not intended : but Milton's rules would not have allowed the line without the elision Thy condescension, and shall be i&onour'd ever. viii. 649. Had the midverse extrametrical syllable been admitted into P. L., the whole prosody would have been thrown into confusion. Digression on Elision 9 § Supernumerary syllables accounted for by Elision. [§ Digression justifying the use of the term Elision. Since the word elision signifies c cutting out ', Elision or there would seem an impropriety in using it to svn ", describe the condition of syllabic vowels which are acep a ' not truly elided or cut out of the pronunciation. The following justification of my use of it is provoked by my critics, to whom I hereby dispense the readers' maledictions. In English verse when there is poetic elision of the terminal terminal vowel of one word before the initial vowel ellslon - of the next word, the sound of it is not lost, the two vowels are glided together, and the conditions may be called synaloepha. For instance the first example of terminal syn- aloepha in P. L. is Above th' Aonian Mount, while it pursues, i. 1 5. where the final vowel of the is glided into the A of Aonian, it is still heard in the glide, though pro- sodially asyllabic. Now since this synalcepha of vowels between two words was historically an imitation of the true Greek elision, that name is convenient and historically correct, and it is commonly used by correct gram- marians, and as a matter of fact the first of two such vowels is theoretically c cut out ' of the prosody or scansion. In Milton's prosody, this terminal c elision ' is not confined to naked vowels, he treats the semivowels /, «, and r as vowels, so that his terminal { elisions ' require different phonetic explanations, and would not all fall under one definite grammatical name. Moreover when these same collisions of vowels and semivowels occur (under the same phonetic i o Digression conditions) within the word, he subjects them to the same prosodial law as when they are terminal, that is between two words. The various phonetic con- ditions are the same in both cases, and it is convenient to have one name to cover all. came from it might be argued that the terminal synalcepha is e *j not the type, but is itself a mere extension of the y " midword synalcepha native to our speech ; but since historically the Chaucerian and Miltonic terminal synalcepha seems to be the true direct descendant, a great-grandchild of the Greek elision, I preferred a term which recognized that kinship ; and if in our prosody all such cases of syllabic loss are to have a general term, then 'elision' is justifiable, and is the better for having no phonetic significance ; it does not describe any one of the conditions, and cannot be mistaken for anything but a label. Greek \ SU pp se that the practice of terminal synalcepha lsion. actually came to our verse something in this way : Firsdy — In Greek when a word ended with a naked vowel, then, if the following word began with a vowel, the naked terminal vowel was cut out ; it was neither spoken nor written : and this was true elision. This condition of things raises some curious questions in Greek prosody. Coales- Greek grammarians are, I believe, agreed that cence of a Q ree fc syllable was essentially and typically com- posed of a consonant (or consonants) followed by a vowel ; and thus in all Greek manuscripts, and sub- sequently in print, where a word is divided between two lines, the division is made on this principle, phonetically, regardless of the philological articulation of the word. For instance, they would have divided a word like disorderly not dis-orderly as we do, but disorderly, the s being annexed to the second syllable, to complete it. However strange this division may look to us, we ourselves observe it in singing, and on Slision 1 1 that shows it to have a phonetic propriety of some kind. We must suppose that the Greeks used the same practice in reading as they did in writing, and that e.g. Mr)viv aeili? was read as Myjvi vaetie. But a word beginning with a vowel if it were preceded by a word that ended with a vowel would have no con- sonant with which to initial itself except by the practice of elision, which seems to have come of an opisthophagic habit by which it ate away the final vowel of the preceding word, if that was short and syllabically unimportant, in order to get at the last consonant and annex it to complete its own first syllable. This I imagine may have been the origin of Greek elision. Now words thus treated renounce in speech their The ver- proper formal unity ; and this coalesced condition bal unit would seem at first sight irreconcilable with the great importance which the Greeks attached to the verbal unit, as that plainly appears in all their laws for caesuras and verbal breaks, in their incommunicable sensibility to the effect of a trisyllable at the end of an iambic line, and their strict disallowance of unequal division of the fourth foot in the hexameter. Such rules imply that the verbal unit had to them an over- ruling force, and it might seem that the recognition of that was incompatible with the phonetic amalgama- tion of the Words. I have never come across any treatment of this question : the facts, I think, show that the phonetic amalgamation of the words cannot in any way have destroyed the force of the verbal unit — which has some special recognition perhaps in the fact that elision rules between words, but not between similar syllables within words — and if that is so, we must conclude that its force did not lie in its formal literary structure, but in the voice- (in voice tone which translated it to the plane of ideas and tone) emotions. 1 2 Digression I suppose that it is something of this sort. The entrance of any important word into the verse is a call of the attention by change of voice-tone to some idea, with its trail of associations ; and this attention implies duration ; the attention is shifted to a certain level and as it were held there until it is released by the new call of the next word ; and in the break between the two, that part of our mind which is attending to the metre is conscious of the place where it has arrived. The coarseness of such an analysis is apparent, but the subtlety and obscurity of our sub- conscious intelligence defy definition, and I only try to suggest the plane on which these effects are alive. If it is somewhat as I suggest, then the value of word- length is well explained, and also the advantage of recognizing enclitics and proclitics. We certainly recognize the force of the verbal unit in English, almost as much as the Greeks did, although the mono- syllabic quality of our language is unfavourable to the full use of its best effects, has no If it should be questioned what place this discus- rules in s i on h as here, I would point out that the rules for the ng ? position of verbal units were an essential matter in Greek prosody, but that we, though bound to recog- nize their almost equal force in our own verse, do not regard them as subject to laws of prosody. Milton, so far as I know, has no rule for the division of feet between words, although his practice is no doubt subtle and careful enough. In English sylla- bic verse the rules for the distribution of syllables as of quantities are absent. Elision in Secondly — when the Latins discarded their native Latin accentual verse and invented their close imitation of verse. fa e G ree i C) they adopted the same rule of excluding these naked terminal vowels from their reckoning of syllables in the scansion ; but, as their habit of speech required, they still wrote them, and did not wholly on Slision 13 cut them out of the pronunciation, but glided the two colliding vowels together, as is done to this day in Italian verse. This gliding together of vowels, called synaloepha, Latin being forced into the reading of Latin verse as a con- sy™ 06 - scious device whereby the prosody might be assimi- j^ 1 "^ lated to the Greek, produced what seems a gross inconsistency in the system of the Latins ; because they applied it only to the naked vowels that collided between separate words, and not to the same vowels colliding within the word : and though they followed the Greeks in this, it must in their own practice have been a distinction without a difference. And even the terminal vowels that had no true phonetic glide were forced into the synalcepha that was forbidden to the most confluent vowel-combinations within the word, such as are more easily glided together than kept syllabically independent. On such questions of speech-condition I have not learning enough to formulate anything above a very modest opinion ; but history seems to show that our European versification has been strongly determined through the Romance languages by Latin practice ; and however that may have been adapted to suit different conditions, it may well have been the cause of logical inconsistencies : and this knowledge should forbid us to assume that the laws which at any time ruled in any school of poetry were the result of free experiment. Convention has been very strong. Force of In versification we know that the manner to which conven- every one is accustomed, even though it be pedantry, tl0n " has a far greater propriety to our ears than that which we should rightly prefer if we were not preju- diced by custom — the ridiculous distortion of sense and speech-rhythm in the chanting of the Psalms by the trained choirs of our Cathedrals is a good example — and one can only say that the Latins trained them- 1 4 Digression selves to regard the collision of vowels between words as a different condition from the same collision within a word, and that this convention was handed down. But when French theorists argue that the laws of hiatus enforced by Malherbe are rubbish because the same forbidden conditions are tolerated within words, it is not thereby proved that the for- bidden conditions are agreeable, or that it is not better to avoid them. Nor have they the right to assume" that the ends of words are in the same conditions as their middles ; for the word as a unit demands special treatment, and it may have been uncon- sciously protected by the very inconsistency that they denounce. Romance Thirdly — when the syllabic poetry of the Romance verse languages arose, its basis was the inheritance of the corrupted quantitive Latin practice ; and we have to explain how quantitive verse became syllabic. I would here again plainly warn the reader that I have never been a serious student in this field, and that my knowledge is most superficial and fragmentary ; scattered examples, however, seem to me to admit of one simple interpretation, concerning which I feel no doubt, and it would be as follows, how it The recognition of only two lengths of syllables, be | : 1 am . e on which the Greek and Latin prosody had depended, ' though convenient and defensible, was artificial, and was bound to perish from practice in the lapse of tradition ; so that the early ecclesiastical Latin verse- writers, in following the old verse-forms, while they satisfied their ear by preserving something of the old superimposed accentual rhythms — for those could not be misread — , neglected the old quantities, and merely counted the syllables as units of the feet, omitting from that reckoning the naked terminal vowels which they no doubt still glided together. Some of the old hymns look as if the tradition of observing the law of on Slision 1 5 terminal elision was lost, 1 and hiatus is common in old French verse ; but this rule of elision eventually survived, and the whole of Romance verse thus became syllabic and subject to the Latin elision. Eventually this terminal synalcepha was consistendy extended to the collision of gliding vowels within the word in the early Italian and French verse. Fourthly — when Chaucer adopted the Romance Chaucer prosody he imitated the French very closely ; and it followed was thus that the Latin practice, as handed down by "" the Romance writers, came into English verse. It would be essential for a thorough treatment of the subject to give the rules which Chaucer elaborated, but unfortunately this is not possible. The difficulty of fully formulating his practice is due to his having extended it gradually, so that it is developed only in his latest writing — and this is the same with Milton — ; and since his later work can in many cases be dated only by the prosody, this uncertainty together with that of the manuscripts and of the syllabic values of the French words makes it impossible to proceed without unwarrantable assumption of some of the questions in dispute. I shall therefore not attempt to describe Milton it here ; and it is more easily investigated when followed Milton's rules are understood, since those would im ' appear to be only a learned systemization of Chaucer's practice. 2 The main point is that the elision (synalcepha) extended which in the Latin tradition ruled between words applica- was now freely extended to the same syllabic con- tl0n- ditions within the word, and that it was in every case 1 Some examples are given Milton whether Spenser were not below on p. 86. his model, and that Milton gladly 2 That Milton took Spenser assented in order to avoid a fruit- for his master we have on Dry- less discussion on prosody, — a den's authority. I always guessed motive which I can well under- that Dryden must have isked stand. i6 Digression Equi- valence or trisylla- bic feet re- cognized in English before Chaucer. optional ] ; and it is no doubt a liberty in my termi- nology to extend the prosodial term ' elision ' to cover all these conditions of syllabic loss. One reason why the historic theory of elision in English prosody is not universally recognized is that these so-called elisions may be all described as making trisyllabic feet, which are agreeable both to accentual and quantitive theorists. Moreover neither Chaucer's nor Milton's rules were ever formulated, so that their practice not being understood was not strictly ob- served by poets, who therefore supply apt illustrations for any theory. If the explanation of these elided syllables is quantitive they can be interpreted by the old classical rule of c equivalence ', two short syllables being equivalent to one long — as we find spondees interchangeable with dactyls, and trisyllabic feet ad- mitted into Greek and Latin disyllabic verse. And there is no doubt that many of the elisions have this effect fully, and must have been agreeable to Milton's ear for that reason, and it is moreover evident that certain of them have their best rhythmical explanation on the equivalence theory. Moreover Anglo-Saxon poets had recognized ' equivalence ' ; at least one ex- pert tells me that they enforced length on their accented places, so that if, in their accentual verse the accent fell on a short syllable, then that syllable needed another short syllable with it to fill up its time. Our native speech-instinct which led to this practice is, I should say, exemplified in such lines as the follow- ing in P. L. I had persisted hSppie|, had not thy pride, x. 874. Of £ne|mie hath beguil'd thee, yet unknown, ix. 905. 1 Optional with exception of larity to French accidence, the those weak endings of words, effects of which he was imitating, which though they were, as ex- Concerning the rationale of op- perts tell us, obsolete in his day, tional elision see the sections Chaucer consistently used, being on hypermonosyllables, pp. 20 probably influenced by their simi- and 26. on Elision 17 though these verses scan by means of their elisions before had and hath, their rhythm seems to be in the equivalence, as I have divided them. Although there can have been no confusion between Tri- synaloepha and true resolution by equivalence in Latin syllabic verse, yet the rhythmic effect of synaloepha must have e " ect of been very plain to the ear. For instance, while Virgil's J," in epic hexameter is much more spondaic than Homer's, Latin, his Latin lines have nearly as many syllables in them as the Greek. In Catullus's two pure iambic poems the elisions in the odd places (where spondees were normal) are twice as frequent as those in the even places. But it would be rash to deduce much from this. Again, there is no doubt that any English poet who in should write accentual trisyllabic verse (that is verse E . n s' lsh on the basis of having two unaccented syllables be- ^c verse tween the normal metric stresses), would freely use these same combinations of vowels and liquids to make his trisyllabic feet ; but he will use many other short syllables also, and the characteristic of the Mil- tonic verse is that those other short syllables are forbidden to come in as couples to fill the single places of the disyllabic scheme. There is one insuperable theoretic objection to Objection explaining Milton's trisyllabic places as examples of t0 e 1 ul " c equivalence ' in the fact that he does not refuse to ^eonTin admit two classically short syllables for a full foot. />. /,. On the theory of equivalence these feet are defective, so that it can be only enforced inconsistently to explain the trisyllabic places. But it matters little how these are imagined, if it be recognized that the ' equivalence ' in P. L. is always assured by some ' elision '. Blank verse which admits into such places any kind of unaccented syllables, whether elidible or not, ceases to be syllabic verse and becomes so far accentual. There is no objection to such accentual verse, and there has been an increasing propensity to allow its forcible 1 8 Digression on Elision effects to override the subtler grace of the purely syl- labic structure : and nowadays all readers of English verse are accustomed to find syllabic and accentual verses alternating in a poem ; and such a poem will please them none the less ; and they read the rhythms easily enough, being familiar with both manners, as the writers were, and probably enough in neither case distinguishing them. The fault of such verse is that syllabic verses coming in among accentual verses must often invite (if they do not compel) a wrong accentuation, and one cannot be forbidden to read them on the same system as one has to use for the others. The practice is artistically indefensible. It is true that the practice of optional elision brings Mil- ton's syllabic verse under a somewhat similar objection : but its speech-rhythm gives its interpretation and leaves the prosody consistent ; while in the other case it reveals the inconsistency. As a matter then of rhythm it may well enough be maintained that Milton's elisions are trisyllabic feet; but historically and in prosody they are ' elisions ', the great-grandchildren of Homeric elision. On the extension of elision to semivowels see p. 26. Dryden's Finally there is some indication that Dryden and prosody his school at one time read Greek elision into Latin un-Mil- verse, and in their own verse intended the ' elided ' vowels — which they represented by apostrophs — to be omitted from pronunciation : but if so, the absurdity perished. This anti-Miltonic school, which I should guess to have been under the influence of Malherbe, objected also to elision being optional, considering naked terminal vowels as blots on the page. Readers of Dryden and Pope are apt to acquire their un-Mil- tonic ear, in spite of the blots. A theory of speech which is not only unnatural but impossible to conform to, does not call for refutation. Dryden's practice is some further justification for the term ' elision '.] tonic. Common-speech Elisions 19 § Vowel-elisions of common speech. In tabulating Milton's rules for poetic elision it Syna- will be convenient to separate off in a distinct class, as l«pha in 'Elisions of common speech', those cases where common contiguous vowels are run together in common speech, " ee ' and universally recognized as making but one syllable, even though they were in earlier times separately pronounced so as to make two syllables, and still in most cases retain something of their double sound. The first line of P. L. gives a particularly good example, Of Man's First Disobedience, and the Fruit, where the ie of disobedience is neither a diphthong nor a disyllable, and the two vowels are both heard, though they make but one syllable. This condition came from the conversion of the first vowel into a y-glide, and the phonetic condition is exactly the same as the glide or synaloepha of the naked vowels in the poetic elision He effected. The difference between the two cases is this. Common speech had already adopted obedience as a trisyllable, and it is admitted into the verse with that value without question or option, whereas in the poetic elision a fuller pronunciation of the vowels is not only possible without archaism, but is more usual in com- mon speech, and the synaloepha is optional. It was in Elizabethan times that such contiguous much of vowels as the ie in obedience finally lost their disyllabic A ** xed value : so that the syllabic values of such words is a fi out one of the most easily recognized distinctions between the earlier and later verse : and Milton in his earlier poems had used the older pronunciation, thus in Comus, With all the greisly legi-ons that troop. 603. Or gastly furies appariti-on. 641. By a strong siding champion Consci-ence. 212. and thus he uses de/usi-on, conditi-on, complexi-on, visi~on, obsolete syllabi- zing still used in verse. Hyper- mono- syllables. Vocalic. 20 Common-speech 8 /mom contemplati-on, &c, which were all old-fashioned and out of date : but when he wrote P. L. he accepted so far the pronunciation of his time, and there is no example of the old pronunciations of such words in that poem. The usage of all poets, with regard to obsolescent pronunciations is conservative and archaic, and in our contemporary poetry it is still common to find such a word as obedience in full syllabic extension at the end of a line. The line last quoted is a good example of the transitional stage, because champion is accepted with its i lost in the glide, whereas conscience is not. Conscience, patience, vision, and all the host of words ending in -ation, are examples of words that have utterly lost the syllabic value of their i in the y-glide, and we are in danger now of losing the glide in these last, for our phoneticians write -ation as eisKn without any glide ; but Milton respected it (p. 33). But of words containing vowel-glides there is another intermediate class in which the words have not established absolutely fixed values in English prosody. It is not our business to treat them historically or philologically, but their prosody should be considered in this place. The typical and common examples of the y-glide are,jire, desire, tire, &c. e.g. That fires the length of Ophiucus huge. ii. 709. and of the w-glide, our, hour, flower, power, poor, fewer, &c. Of these words Milton used power as a disyllable in Penseroso. Whose power hath a true consent. 95. But in P. L. it is always a monosyllable e.g. Powers and Dominions, Deities of Heav'n. ii. 11. His utmost power with adverse power oppos'd. i. 103. Hyper monosyllables 2 1 and though power became recognized as monosyllabic in English poetry it has still partly maintained its right to be used as a disyllable, e.g. Are portions of one power, which is mine. Shelley, Ap. 14. A power from the unknown God. Hellas, 211. Yet not for power (power of herself. Tennyson, (Enone. The following examples will illustrate the uncertain value of these words. Than tir'd (tieFd) eyelids upon tir'd eyes. Tennyson, Lotos. As Desires lightning feet. Shelley, Prom. i. 734. Thou most desired (iel"d) Hour, more loved and lovely, iii. 3. 69. Upon that path where flower's never grew. Triumph, 65. Few flower's grow upon thy wintry way. Fragment (p. 580). With blackest moss the flower-plots. Tennyson. Higher still and higher Shelley, Skylark. where the first higher is a disyllable (as elsewhere) the second rhymes with fire, which he almost always uses as a monosyllable, but To hear the fire roar and hiss. Mariannes Dream, xvii. The lyre's (liel's) voice is lovely everywhere. M. Arnold. Let us take power as the type and analyse the sound, their It contains three vowels and a glide, and may be repre- typical sented to the eye thus, p a u w a (r), giving a strong P ho "^ lc vowel or diphthongal sound au followed by a weak vowel sound 3, the two being connected by a falling w-glide ; nor can this be shortened : the only question that can be made is whether the u and the w are both present, or whether the place of the vowel (u) be not wholly occupied by the glide (w). But since in pro- portion as the sound of the long vowel is disallowed, the strength of the glide is asserted, it does not seem to matter how this is decided. There is no doubt that power is generally received a prece- into English verse as a monosyllable, and thus is ^ en } *° r sometimes written pow'r or powr : but the omission °P t ! onal of the sound indicated by the apostroph is not effected by omitting it from the spelling ; the word is always 811629 c Apparent excep- tions. ('to- ward. ') 2 2 Common-speech Slisions a strong vowel glided into a weak one, and the intro- duction of such units into the verse clears the way for all other c elisions ' of vowel sequences which can be treated in a like manner l : and if such a unit can be regarded as either monosyllabic or disyllabic at will, that is also a full precedent for such elisions being in other cases optional. Again, when considering the semivowel elisions we shall find quite another set of hypermono- syllables, which having the same doubtful syllabic value as power &c. are treated by the prosody in the same way as filling either one or two places in the syllabic verse. Note that the apparently similar words (which are always monosyllabic) air fair prayer bare., &c, and ear, hear, fear, &c, differ from the type power, the first in having almost no tail-glide to their accented vowel ; the second set by having their accented vowel so short, that its glide is able wholly to supplant it. Thus dialectal speech sometimes prints hear or here as hyah, where the y is, I take it, intended for a consonant or short glide, not for a long vowel-. There may be some doubt perhaps how Milton contracted toward and towards. This word is generally monosyllabic in his verse, but there are these exceptions Strait toward Heav'n my wondring TEyes I turn'd. viii. 257. In Serpent, Inmate bad, and toward Eve. ix. 495. Safe t6wards Canaan from the shoar advance, xii. 215. Justification towards God, and peace, xii. 296. Yet t6w5rd these thus dighifi'd, thou oft. S. A. 682. The contraction which in our common speech is now iward used to be toS-rd, and this is the better form ; and since its probability is very much favoured by the fiye lines just quoted, it will be assumed in 1 Power is of course an example of one class only, two glides, next page. See on the Poetic Slisions 23 the classification, and the word will fall under the w tail-glides, with optional elision. § Poetic elision of vowels. When two vowel sounds come together, then if Vocalic the first of the two has a tail-glide, there may be svna ; elision, i.e. the sounds may be glided together so as d " p * ds to make a sound which can be reckoned as one on u ;i_ syllable in the disyllabic verse. glide. 1 Diphthongs ' are included, and h is often considered as no letter. 1 There are two vowels in English which have no tail-glide. They are the a of father, the broad which we write aw {law). (The vowel in air, Mary, &c. — written as a diphthong by phoneticians ed — varies much and may be neglected here.) The tailless vowels may of course be involved in elisions when they follow a gliding vowel, e.g. th' Army, tK almighty, tK Air. He ceas'd ; and th' Archangelic Power prepard. xi. 126. With other notes then to th' Orphean Lyre. P. L. iii. 1 7. They summ'd their Penns, and soaring th' air sublime, vii. 421. For mere prosody, it would be sufficient to say that all other vowel sounds are subject to elision in all conditions, but in giving examples it may be as well to classify them phonetically. They may be classified under their two glides, The two each of which will show three conditions. glides. The two glides are the y-glide and the u- (00) or w-glide. We will take the y-glide first because it has already been illustrated in obedience, and champion, and hear. Its three conditions are these : 1 The value of written h varies again with its collocation in the from complete suppression to the sentence ; so that the questions full force of a consonant ; and it raised would not repay discus- varies also with the speaker and sion. See p. 26. 24 Toetic Blision i. When the first vowel sound is stressed as in riot. i. When the second vowel sound is stressed as in humiliation. 3. When neither vowel is stressed as in Michael. In this third case it may be conceded that one of the two vowels is generally more stressed than the other, so that most cases of this class (and it is of course the same with the w-glide) might be correctly ascribed to one of the previous classes, but it is more convenient to class them separately. The w-glide has the same three conditions viz. 4. When the first vowel is stressed as in ruin. 5. When the second vowel is stressed. Of this there is no example within the word, unless fluctuats, ix 668, may pass. 6. When neither vowel is stressed as in virtuous. v-eHde Examples of all the six classes are here given. 1 . Saying, being, flying, diet, riot, giant, higher (h?er), hierarch, violence, diamond, variety, Deity, piety. Half flying; behoves him now boath Oare and Saile. ii. 942. Be it so, for I submit, his doom is fair. x. 769. Is his wrauth also ? be it, man is not so. 795. 2. Humiliation. Tiresias iii. 36 (hardly any accent: see note C.) Mediator xii. 240, is confirmed by x. 60. Therefore thy Humiliation shall exalt, iii. 313. Timely 'interposes, and her monthly round, iii. 728. From Hell continu'd reaching th' utmost Orbe. ii. 1029. Hypocrisie, th« only evil that walks, iii. 683. His day, which else as th' other Hemisphere, iii. 725. More grateful then harmonious sound to th* tare. viii. 606. and thus the air, the earth, the eye, the hour, note that the hour has four vowels glided as one syllable, iauer. of Vowels 25 3. Michael, Sinai, (but Milton may have pro- nounced this word as a disyllable :) Michael is a tri- syllable sometimes in P. L. Examples of this class are almost all between words, because when within the word they are usually elisions of common speech : Above th' Aonian Mount, while it pursues, i. 15. To set himself in Glory above his Peers, i. 39. Strange horror seise ihee, and pangs unfelt before, ii. 703. Though kept from Man and worthy to be admir'd. ix. 746. He effected ; Man he made, and for him built, ix. 152. Little inferior, by my adventure hard. x. 468. With spattering noise rejected : oft they assayd. x. 567. Life in myself for ever, by thee I live. iii. 244. Forc't I absolve : all my evasions vain. x. 829. and thus the acclaim : the almighty, me and, thee and, the unwary, thee unblamed. Vitiated x. 169, may belong here or to class 2. 4. doing, ruin, toward, see p. 22. Th No ingrateful food : and food alike those pure. v. 407. With noises loud and ruinous, to compare, ii. 921. 5. Examples within the word missing. And rapture so oft beheld? those heav'nly shapes, ix. 1082. Into atter darkness, deep ingulft, his place, v. 614. Of somthing not unseasonable to ask. viii. 201. In another writer this line would have been intended and rightly read unseasonable | to ask ; and Milton has unreasoning with the i elided ' : but his rule and use of adjectives in able decides certainly in favour of the elision here exemplified. 6. followers, Siloa's, bellowing, shadowy, gradual, effluence, influence, extenuate, tumultuous. For God is also in sleep and Dreams advise, xii. 611. Vertae in her shape how lovly, saw, and pin'd, iv. 848. Presaging, since with sorrow and hearts distress xii. 613. Thou didst accept them j wilt thou enjoy the good. x. 758. Damasco, or Marocco, or Trebisond. i. 584. As Lords, a spacious World, to our native Heaven, x. 467. e w-glide. 26 Poetic Elision H treated as non- conso- nantal Elision through H. 7. It is plain that there cannot be a true glide through a consonantal h. The following examples show various conditions. Harp seems to forbid synaloepha, see note D. Horizon may have its clas- sical aspirate, it has none in Italian. Whom seems possible by using the u of to (tit) as a glide and dis- guising the h, = tukum. See p. 23, note. For still they knew, and ought to have still remember'd. x. 12. In Gems and wanton dress ; to the Ifarp they sung. xi. 579. T' whom thus the Portress of Hell Gate reply'd. ii. 746. Two onely •who yet by sov'ran gift possess, v. 366. Had rounded still th' Horizon, and not known, x. 684. To a fell Adversary, his hate or shame, x. 906. And left to her self, if evil thence ensue, ix. 1 1 8 5. Semi- vocalic hyper- mono- syllables. § Poetic Elision of the semivowels. It has been shown that the poetic elision (syna- loepha) of naked vowels between words, is only a natural extension of the similar treatment in com- mon speech of the same vowels within the worcf: and so it can be shown that the elision of the semi- vowels has an exact counterpart in our habitual treat- ment of certain other monosyllables, which I have called hypermonosyllables. The words schism, prism, chasm, spasm, are usually reckoned to be monosyllables, and thus the finals of baptism, abysm, despotism, paroxysm, &c. &c. ; a few examples will illustrate this. It should be noted that in all these words the vowel is short. Thus schism is a much shorter word than size-m would make. Of Baptisms, Sunday-schools and graves. Shelley, Peter Bell, vi. 61 5. Murmur'd this pious baptism — Be thou called. (Ed. i. 360. conditior of the Semivowels 27 *In the dark backward and abysm of time. Tempest, i. z. 50. *Into the abysm of Hell. If he mistake. A. & C. iii. 13. 147. Or sun from many a prism within the cave. Shelley, R. of I. vii. 20. As from a thousand prisms and mirrors, fills. Epip. 1 66. *Of Newton with his prism and silent face. Wordsworth. And filled with frozen light the chasms below. Shelley, Ath. 269. To other lands, leave azure chasms of calm. Shelley, Epip. 466. And where its chasms that flood of glory drank. Laon. xi. (4245). O'er chasms with new-fallen obstacles bestrown. Words., Eccl. Son. ii. 12. Now taking prism for type, and analysing it, we their find' that we do not combine the sm as in small, but typical vocalize the m separately. It may be represented to P h °o^ lc the eye as^>w m : there is some vocalic sound between the s and the m ; and it is thus just like heaven and prison, which are hev'n and pris'n : and heaven is freely used in all verse both as disyllable and monosyllable. Seven is a monosyllable in Chaucer and Milton, and Milton uses prison in the same way ; and we may ask what difference there is between these words and battle, or temple, which are bail and temp' I. If any distinction should be drawn between pris'm and pris'n, then pris'm is the longer. As for /, if that have any greater claim to be recognized as syllabic above m or n in prism and prison, this perhaps is because / is the most difficult sound for children to learn, and its quality may therefore have some natural tendency to be delayed. Milton's practice is this, that the unac- cented vowels of syllables closed by r, /, or n, are elidible before another vowel, as freely as if they were ' naked '. m is not admitted so freely, if at all. * In the starred examples very frequently, and prefers it there is correct Miltonic elision before a vowel, or at end of line, (as he has spasm in xi. 481), not but has many purely monosyl la- in the others. Shelley uses chasm bic examples. 28 Poetic Elision In the semivowel elisions which will now be illus- trated, the syllabic loss within the word is of course much more real (i.e. the shortening is more possible), if the consonant that precedes the /, », or r, can be amalgamated with it, so as to be pronounced together with it and without break, as wandering is very easy, because the dr can be spoken together. But this facility is not required for poetic elision. Glimmering and murmuring are elidible although mr is a very awkward combination. Glimmering is easier than glimring, and therefore is no longer, while the second u in murmuring is so essential to the word, that if it were really cut out by 'elision' it would be bad writing to ' elide ' it. Again, Ir is almost impossible, and arlillry (ii. 715) is in itself a sufficient refuta- tion of the notion that all poetic elisions are com- mon speech contractions ; if it had been so, then some of Milton's elisions would never have been tolerated by him. Syllabic As the inconsistent or uncertain values of the. power value of anc i jj re WO rds were illustrated by examples from other * emi ". poets, so the tendency to give syllabic value to semi- vowels in words where common speech usually treats them as asyllabic, may be seen in the following quo- tations : O how this spring of love resemb-1-eth. Shakespeare. And death's dark chasm, hurrying to and fro. Shelley, Hellas, 203. Filling the abyss with sunlike light-n-ings. Prom. iv. 276. Like hues and harmonies of eve-n-ing. Int. Beauty. Round which Death laughed, sepulchred emb-l-ms. Pr. 294. The dazz-1-ing sunrise ; two sisters sweet. Keats. Turn'd syllab-1-ing thus : Ah Lycius bright. The parts and graces of the wrest(e)ler. Sh., As T. L. ii. 2. 13. While she did call me rascal fidd(e)ler. T. of S. ii. 1. 158. You, the great toe of this assemb(e)ly. Cor. i. 1. 159. That croaks the fatal ent-e-rance of Duncan. Mac. i. 5. 40. A rotten case abides no hand(e)ling. z H.1V. iv. i. 161. of the Semivowels 29 And strength by limping sway disab(e)led. Sonn. 66. But who is man that is not ang(e)ry. T. of A. iii. 5. 56. 1 Rule of R. If two unstressed vowels be separated by r there Syllabic may be ' elision ' ; that is, the two syllables may count loss for one, the syllabic loss falling on the first of the w two. These r-words are very numerous and suffer syllabic contraction in common speech : but though the spelling is often misleading to the eye, Milton did not always wish to mitigate it : it is as if he insisted on the syllable when he prints : Labouring had rais'd, and thus to Eve replied, x. 1012. Examples are the following : Nectarous, weltering, suffering, glimmering, murmuring, labouring, neighbouring, honouring, endeavouring, measur- ing, disfiguring, &c, mineral, general, several, artillery, desperate, deliberate, emperour, amorous, timorous, torturer, savoury, conqueror, &c. Not distant far from thence a murmuring sound, iv. 453. Luxuriant; mean while murmuring waters fall. 260. Murmuring, and with him fled the shades of night. 1015. These are quoted in Milton's spelling, and it is some evidence that he did not even degrade the affected vowel in pronunciation : and thus are A Pillar of State; deep on his Front engraven, ii. 302. Before them in a Cloud, and Pillar of Fire. By day a Cloud, by night a Pillar of Fire. xii. 202. 3. 1 These last examples are from most ragged irregularity (which Abbott's Shakespearian Grammar, his explanations are designed to I would refer the reader to his avoid) to the effect of torturing sections on the lengthening and a word to make the verse scan. shortening of words which he In the face of this experience, finds necessary in order to make no one can deny the advantage some of Shakespeare's lines scan, of a fixed prosody, such as the Quite apart from the disastrous French have observed. Law- uncertainty of rhythm which lessness means uncertainty of many of his examples exhibit, rhythm, and that is the ruin I find that I often prefer the very of the verse. 3