3331 
 H!>9 
 
THE LIBRARY 
 
 OF 
 
 THE UNIVERSITY 
 
 OF CALIFORNIA 
 
 IN MEMORY OF 
 
 PROFESSOR WILLIAM MERRILL 
 
 AND 
 
 MRS. IMOGENE MERRILL 
 
?f'3^i^OuZ: 
 
 ^ ^ 
 
 I i/'^ CCij^- i^^iU^Cy 
 
 ^^■tJV-^-j'''^) 
 
 The Quantitative 
 Reading of 
 Latin Verse 
 
 Joseph Henry Howard, A. M. 
 
 u/r 
 
 SCOTT, FORESMAN AND 
 COMPANY, PUBLISHERS, 
 378=388 WABASH AVE- 
 NUE, CHICAGO, ILLINOIS 
 
'-'^'^^iM^^^^ ^yC^ OUitli/^ 
 
 THE 
 
 QUANTITATIVE EEADING 
 
 OF 
 
 LATIN VERSE 
 
THE 
 
 QUANTITATIVE READING 
 
 LATIN TERSE 
 
 BY 
 
 JOSEPH HENEY HOWARD, A. M. 
 
 Assistant Professor of Latin in the Indiana University 
 
 CHICAGO 
 SCOTT, FORESMAN AND COMPANY 
 
 1898 
 
Copyright, 1898, by 
 Scott, Fobesman & Company 
 
 GIFT 
 
 ROOBSB & SMITH CO., FRINTERS, CHICAGO 
 
Hir 
 
 PEEFAOE. 
 
 THIS little book has been written in the belief that there 
 exist to-day an urgent need and a demand for some more 
 specific and more satisfactory information to guide the teachers 
 and pupils of secondary schools in their study of Latin verse. 
 It is the author's hope that this need will be found to have been 
 met, in part at least, by the contents of the following pages. 
 For an aroused interest in this subject and for the general 
 theory herein presented, he is indebted to Professor William 
 Gardner Hale of the University of Chicago. Several years' 
 practical application of this theory, while teaching classes in the 
 Indiana University, has served to strengthen the author's belief 
 in its correctness and its absolute feasibility. 
 
 It is not intended that this shall be a manual on prosody. 
 The aim has been to omit everything not essential to the prepa- 
 ration for reading poetry. Much, therefore, has been left to be 
 added by the teacher, who will, of course, make use of such 
 helps as are needed for that purpose. The book will have ful- 
 filled its mission if it shall have pointed out the road to be 
 traveled. 
 
 Professor Hale very kindly looked over the manuscript for 
 the following pages and offered many valuable suggestions. 
 
 ^; 
 
 ?Of;rr? 
 
THE QUANTITATIVE EEADING OF 
 LATIN VERSE. 
 
 IT is said by those who have given much time to the study of 
 Latin poetry and the laws of Latin versification, that the 
 Roman poet attached not less importance to the form in which 
 his thoughts were cast than to the thoughts themselves, that he 
 believed it to be just as necessary for the dress to be beautiful 
 and appropriate as for the language it clothed to be polished 
 and select It is known, too, that great care in the choice of 
 vocabulary was made necessary by the limitations imposed by 
 the demands of the verse-rhythm, which absolutely precluded 
 the use of all words that would not harmonize quantitatively 
 with the metrical structure of the verse, and that, as a result of 
 this dependence of the language on the form and the consequent 
 difficulty of selection, higher finish and greater accuracy were 
 attained. 
 
 In view of these facts, it is evident that a thorough under- 
 standing and genuine appreciation of the work of the Roman 
 poets can never be the possession of those who choose to ignore 
 the importance of the form-side. 
 
 It is worth while to seek to have this intimate knowledge of 
 the poet's work ; the time and labor one must give to such a 
 study are profitably employed. 
 
 Great pleasure is experienced in the study of art, and much 
 benefit is derived from it. It is thought to be highly profitable 
 to acquaint one's self with even the most minute details of the 
 works of the master sculptors and painters and architects. 
 Such knowledge is considered an important part of a broad 
 and finished education. It should not be called a waste of 
 
8 THE QUANTITATIVE READING OF LATIN VERSE. 
 
 valuable time and energy to apply one's self, with equal zeal and 
 care, to the study of these other, certainly not less noble, crea- 
 tions of art. What, in truth, are they? Masterpieces from one 
 of the greatest literatures the world has known, embodiments 
 of the thoughts of some of the ablest and best representatives 
 of a people whose influence has scarcely been surpassed by that 
 of any other race. 
 
 Preparation for the true interpretation of a Latin poem 
 involves some labor, just as does the preparation for any work 
 worth doing well. Some, not caring to take this trouble, choose 
 to treat poetry as prose, and thus not only fail to arrive at any 
 just appreciation of the poet's work as a whole (for this is like 
 standing before some beautiful building and selecting for inspec- 
 tion the various parts of which it is composed, with never a 
 glance at the entire structure which reveals the builder's design 
 and the harmony of material and form) but also get, for their 
 pains, poorer results than if they had given the same time to a 
 piece of real prose. For not only does the vocabulary of poetry, 
 as we have already seen, differ much from that of prose, but it 
 also has a word-order peculiarly its own. Some who follow the 
 plan of studying poetry as prose have their pupils read it aloud 
 (or, to be more exact, pronounce it) as if it were prose, but with 
 utter disregard of the quantity of syllables, a factor which can 
 not be overlooked even in the correct reading of Latin prose. 
 Imagine the effect of pronouncing, in this way, some finished 
 English poem of delicate rhythm ! And English poetry does 
 not depend upon quantity, for its rhythm, but upon stress- 
 accent. Others do better who omit the oral reading and make 
 merely a silent interpretation of the author's thought, rather 
 than be guilty of so manifest a violation of the spirit of Latin 
 poetry. 
 
 Will it suffice, then, to seek to apprehend, so far as possible, 
 the beauties of the form and the adaptation of form to language, 
 as one makes this silent search for the meaning of the words, 
 and wholly dispense with the oral reading, if, as asserted, it is 
 folly to pronounce merely, and the reading as poetry is attended 
 
THE QUANTITATIVE BEADING OF LATIN VERSE. 9 
 
 with more or less difficulty? With just as much reason one may 
 hope to understand a song, by reading the words without know- 
 ing the melody, or, to make the resemblance more nearly exact, 
 without a knowledge of melody. Is it, therefore, not possible 
 for one to arrive at a thorough understanding of a Latin poem 
 without reading it orally and as poetry? Silent interpretation 
 is, perhaps, just as effective and affords the reader as great 
 pleasure as oral reading, provided that the music of the verse is 
 as vividly present. It seems that there can be no question, 
 though, that the music must first be heard actually, in so far as 
 the voice can reproduce it, before it can be real to the silent 
 reader. Careful practice in oral reading, continued for some 
 time by one who has previously acquainted himself with the 
 essential facts of Latin versification, is, therefore, indispensable 
 to perfect silent reading, and must precede it. 
 
 What, then, does one need to know before he is ready to read 
 Latin poetry intelligently? First of all, he must know how to 
 pronounce Latin words correctly. No one can do this who is 
 not acquainted with the sounds of the letters, the quantities of 
 syllables and the word-accent. The requirements thus far seem 
 severe, perhaps. They are not, in fact. If the teacher has been 
 properly trained and has conscientiously performed his duty, 
 pupils will come to the study of Latin poetry with the greater 
 part of the necessary preparation already made. For they will 
 have been taught, from the very first day of their work in prose 
 Latin, to pronounce Latin words correctly. Prose texts with all 
 long vowel-quantities marked are now available and should be 
 used. For those who, in their previous study, have paid no 
 attention to the quantities of vowels and syllables, the task is 
 not so easy. But they can afford to make up for this neglect, 
 even at so late a date in their study. Belated learning of quan- 
 tities, therefore, is to be strongly recommended in such cases. 
 Indeed, it seems that one ought not approach the study of Latin 
 poetry at all, if he is not willing to perform this necessary pre- 
 liminary labor. Nothing could be more simple than the law 
 governing the placing of the Latin accent, and it requires no 
 great labor to become acquainted with the sounds of the letters. 
 
10 THE QUANTITATIVE BEADING OF LATIN VERSE. 
 
 Besides the correct pronunciation of Latin words, one needs 
 to know the more important facts of Latin versification. At 
 first, one need not know, in fact, should not burden his memory 
 with, the details of prosody. These should be learned one at a 
 time, as they come up in actual reading. Many of them are 
 met but seldom. The knowledge of the essential facts pertain- 
 ing to Latin verse-structure in general should be supplemented 
 by an acquaintance with the features which are peculiar to the 
 kind of verse to be taken up first, and the learner is then ready 
 to begin reading. 
 
 It would be desirable if, at this point, we could turn at once 
 to the consideration of some particular kind of Latin verse and 
 make clear, by means of an illustration, just what any one start- 
 ing out to read Latin poetry is to do. But there are certain 
 things that ought to be said, certain erroneous views respecting 
 the reading of Latin poetry that need to be referred to and, if 
 possible, shown to be wrong, before we can feel sure that the 
 learner will be in position to distinguish the true method, when 
 it is set before him, from the false one, still much in vogue. To 
 introduce this discussion, let us take up first Latin word-accent. 
 
 It is not certainly known whether the Latin word-accent was 
 one of pitch, that is, a musical accent, or one of stress. The 
 probabilities, however, strongly favor the belief that it was a 
 stress-accent, and the other view has been almost entirely given 
 up by scholars. It is, therefore, assumed in this discussion that 
 it was a stress-accent. When the syllables of the Avords in a lan- 
 guage are pronounced, some at a higher, some at a lower, note 
 that language is said to have pitch-accent, or musical-accent; 
 when they are uttered, some with greater, others with less energy, 
 the language is said to have stress-accent. The English lan- 
 guage, though decidedly one of stress-accent, nevertheless makes 
 much use of musical-accent. One can easily note that, in pro- 
 nouncing such a word as unconstitutio7ial, as an isolated word, 
 successive syllables are not uttered in the same tone, but that 
 the range is from highest on the syllable -fti to lowest on the 
 syllable -tion. Perhaps there has been no language in which 
 
THE QUAls^TITATIYE READIIh'G OF LATIN" VERSE. 11 
 
 both accents have not been present. But we speak, of course, 
 of that one which is predominant, when we speak of the accent. 
 The ancient Greek language had the musical-accent. Care 
 should, therefore, be taken to avoid confusing Greek and Latin 
 word-accent. With the additional statement that the stress- 
 accent of Latin words was not so strong as that of English 
 words, we can leave this topic. 
 
 We have now arrived at the point where we can state defi- 
 nitely what is involved in the correct reading of Latin poetry. 
 One reads Latin poetry correctly when he gives to each word its 
 ordinary pronunciation (which, be sure to remember, includes 
 the proper placing of the word-accent and the preservation of 
 the quantities of all syllables) and, with due regard for natural 
 pauses, utters the successive words of the verse connectedly 
 in such wise that the rhythm of the verse will be everywhere 
 distinctly maintained as an undertone. 
 
 The method generally employed differs from the one here 
 proposed in one important particular. This leads to a consid- 
 eration of this older, long-used method of reading Latin poetry, 
 believed by the author to be faulty. This subject is introduced 
 here, even at the risk of confusing the reader, only because it is 
 hoped that, by making clear the difference, the error of taking 
 up or continuing to cling to the old plan may be avoided. 
 
 Briefly stated, then, it is the common practice, in reading 
 Latin verse metrically, to make the natural word-accent subor- 
 dinate to the verse-beat (which the Romans called ictios). It is 
 unfortunate that any special mention of ictus has to be made in 
 this book, for we do not need to pay particular attention to a 
 verse-beat, do not need to make a conscious effort, at least, to 
 place the verse-beat, in reading Latin poetry. But here again it 
 seems best to risk causing some confusion in order to obviate far 
 greater confusion. The reader must know, then, that the verse- 
 beat, the ictus (Latin for Uow, heat), is thought by some of 
 those who champion the older theory to be, in fact, the pitch- 
 accent (utterance at a higher note) which belongs to a certain 
 part of every verse-foot, but, in practice, to be best represented 
 
12 THE QUANTITATIVE EEADIN^G OF LATIN VERSE. 
 
 by a strong stress-accent, for the reason that we, who speak 
 English, can not reproduce this pitch-accent, it being almost 
 entirely foreign to our language. 
 
 Now the ictus belongs usually, though not always, to a long 
 syllable of the verse-foot. In practice, therefore, it would be 
 doubly difficult for us to keep in mind to give this same syllable 
 pitch-accent and its proper time (quantity), too, for both are 
 almost strangers to our language, the latter practically unknown 
 to it. So it is thought best, in their opinion, to place a strong 
 stress-accent on every syllable that has the ictus and have this 
 stress serve as a substitute for the Latin musical or pitch-accent 
 and long quantity. Some who advocate the old theory believe 
 that the Latin ictus is really a voice stress but make the mistake 
 of having this stress-accent the only one for the verse-foot. 
 From one of our standard Latin Grammars (Allen & Green- 
 ough% p. 406), the following, given under the caption. The 
 Musical Accent, is quoted: "That part of the measure which 
 receives the stress of voice (the musical-accent) is called the 
 Thesis. The stress of voice laid upon the Thesis is called the 
 ictus (beat).'^ The implied meaning is that no other syllable 
 in a verse measure except the ictus syllable receives a stress- 
 accent. Placing the stress on that syllable alone, though, 
 necessitates the violent change of the natural word-accent from 
 one to several times in almost every verse of Latin poetry. A 
 recent view, quite at variance with these already given, is pre- 
 sented in two. lately published Latin Grammars. Bennett's 
 Latin Grammar (p. 243, foot-note) makes this statement : " Ic- 
 tus was not accent — neither stress-accent nor musical-accent, — 
 but was simply the quantitative prominence inherent in a long 
 syllable.'^ Mooneyes Latin Grammar (p. 231) says : "Ictus is 
 the prominence given the long syllable in every foot/' But 
 clearly this can not be correct, for some verse-feet have more 
 than one long syllable, and some have not even one. The ictus 
 must therefore fall on a syllable short in quantity in some verse- 
 feet. 
 
 Those who feel moved to take exceptions to all this hold to 
 
THE QUAI^TITATIVE EEADIKG OF LATIN TERSE. 13 
 
 the belief that the ictus is neither pitch-accent, nor strong stress 
 nor quantitative prominence. In their opinion, it is a stress, 
 but a stress not so strong as that of the normal word-accent, 
 and therefore not destructive of but subordinate to the word- 
 accent. It is the pulse-beat of the verse, which is rather a swell 
 than a staccato accent. 
 
 The fact has been noted that ictus usually belongs to a long 
 syllable. It is easy to see how, in reading, the proper giving of 
 ictus, which is a gradual increase in volume and intensity of 
 sound rather than staccato stress, is greatly favored by this 
 quantity length. 
 
 The statement was made that, in reading Latin verse met- 
 rically, it is the common practice to make the natural word- 
 accent subordinate to the verse-beat (ictus). This subordination 
 is strong enough (so the advocates of the theory claim) to cause 
 the word-accent to be thrown out of its usual place whenever it 
 conflicts with the metrical ictus, that is, whenever the word- 
 accent naturally falls on some other syllable in the verse-meas- 
 ure than the one having the ictus. The stress-accent in every 
 such case of conflict has to be transferred, they think, to the 
 syllable having the ictus, or, if one choose to put it so, all other 
 stress-accents must disappear before the one on the ictus-syllable. 
 The result is that, in almost every verse oi Latin poetry, one or 
 more (oftener more) words have their stress-accent placed on a 
 syllable which does not normally have it. This seems unnatural 
 and unwarranted. In truth, no parallel to such a practice can 
 be found in any known language. Such license is allowed in 
 poetry occasionally. We see it in our own, sometimes. But 
 nowhere can one find support for such radical changes as are 
 demanded by this older method. Now the explanation is that 
 the practice rests entirely on the untenable hypothesis that Latin 
 accent was one of musical-pitch and not one of stress, and that 
 the best method of representing this pitch-accent is to stress the 
 ictus-syllable heavily, thereby not merely obscuring the natural 
 stress-accent in many words, but entirely removing it.* In 
 
 ♦See Professor Hale's discussion of "Did verse-ictus destroy word -accent in 
 Latin poetry ? " (Ginn & Co.) 
 
14 THE qua:n'titative reading of latin yerse. 
 
 what, then, does the music of Latin verse consist ? On what 
 does it depend ? Does it depend on the word-accent ? Cer- 
 tainly not, for the word-accent for poetry is precisely the same 
 as that for prose. * Is it due to the vocabulary, to the character 
 of the words so carefully chosen by the poet ? Unquestionably 
 this is an important factor in it, but not the all-important one. 
 Shall we say it is quantity ? We are near the truth. The 
 rhythm that results from giving to every syllable in a Latin 
 verse its exact time (quantity) as we read, and from placing 
 properly the regularly recurring ictus-stress constitutes the chief 
 element of the music. The placing of the ictus-stress requires, 
 however, no conscious effort after the reader is once familiar 
 with the metrical structure of the verse to be read. So soon as 
 one comes to feel the rhythm, the ictus will place itself. Now 
 some claim that this verse-rhythm is greatly marred if one pro- 
 nounces each word precisely as it is correctly pronounced when 
 standing in a prose sentence, for the reason that the word- 
 accent, in almost every verse of Latin poetry, as before stated, 
 is thus made to conflict, in one or more verse-measures, with the 
 ictus. From this opinion the author strongly dissents, and 
 holds that, on the contrary, additional charm is given the verse 
 by virtue of this very conflict, that serves to break the monot- 
 ony which would otherwise result. This monotony must have 
 been scrupulously avoided by the poets themselves, for we find 
 very few verses in all Latin poetry in which the word-accent 
 falls, in every verse-measure, on the syllable that has the ictus. 
 Three such verses, written in the same meter as that of the 
 Aeneid of Vergil, are here given : 
 
 Eor. Epist. I, 9, 4— 
 Dignum | mente dojmoque lelgentis ho|nesta Nelronis. 
 
 Eor. Epist. I, 14, 30— 
 Multa I mole do|cendus a|prico | parcere | prato. 
 
 Ennius — 
 
 Sparsis | hastis | longis | campus | splendet et I horret. 
 
THE QUANTITATIVE BEADING OF LATIN VEKSE. 15 
 
 Each of these verses has six feet. The vertical line has been 
 used to separate these feet so that the reader can distinguish 
 them at a glance. A dot is placed beneath each syllable having 
 the ictus (which, in this kind of verse, is always the first in the 
 verse-foot). The syllable having the word-accent is given in 
 heavy type.* Now read these verses and imagine the effect of 
 page after page read in the same monotonous, sing-song fashion. 
 It is precisely this effect that is produced by forcing the word- 
 accent to fall, in every measure, on the syllable having the ictus. 
 To make this clear, let us take verses 8 to 11 inclusive from the 
 first book of the Aeneid : 
 
 Musa mijhi cau|sas memo|ra quo | numine | laeso, 
 duidve do|lens re|gina delum tot | volvere | casus 
 Insig|nem pie|tate vi|rum tot ad|ire la|bores 
 Impule|rit. Tanjtaene ani|mjs cae|lestibus | irae ? 
 
 Read first without paying any attention to the heavy type, 
 and give stress-accent to the syllables under which the dots are 
 placed, that is, to those syllables only which have the ictus. 
 What is the result ? We get a good idea of what the sing-song 
 style is, a taste of what it would be, if continued through many 
 verses, and we have, besides, the satisfaction of knowing that in 
 four verses of poetry, we have had the audacity to do what no 
 Eoman would have dared to do, — place the word-accent nine 
 times on syllables that should not have had it. Look at the 
 verses again, this time noting what syllables are in heavy type 
 (for they have the normal word-accent). The reader will see, at 
 a glance, that, in the first verse, the words mihi, causae and 
 memora all happen to have final syllables which have ictus, and 
 that the natural word-accent falls, in all three, on the first 
 syllable. How ridiculous, then, to stress heavily these final syl- 
 lables and give no stress at all to the syllables having the 
 word-acc ent naturally ! 
 
 * The dot and the heavy type will be employed throughout the remainder of the 
 book in this way. 
 
16 THE QUANTITATIVE KEADI2?^G OF LATIN TERSE. 
 
 We can now turn from this negative side of the subject to 
 the positive side. But before taking up extracts from Latin 
 poems representing different meters, with a view to explaining 
 what the author thinks one should do in reading verse quanti- 
 tatively, it will be necessary to define certain terms with which 
 the learner may not be familiar. A few of these terms, such as 
 verse-foot, quantity, were unavoidably used in the preceding 
 pages.* 
 
 QUANTITY. 
 
 Let us first speak of quantity. Quantity is the time required 
 for utterance. We speak of the quantity of a Latin consonant, 
 quantity of a vowel, quantity of a syllable. Certain Latin vowels 
 require twice as much time as some others, certain syllables re- 
 quire twice the time needed by certain others. We speak, there- 
 fore, of long quantity and short quantity. Latin verse demands 
 that there shall be such a choice and arrangement of words as 
 will result in a fixed succession of long and short syllables, f To 
 indicate long and short quantity, we employ the marks - and u 
 respectively. J 
 
 VEESE-FOOT. 
 
 A verse-foot (called also verse-measure) is the verse's unit of 
 measure. This unit repeated a number of times makes up the 
 verse. For instance, the unit in every verse of poetry written 
 in the meter in which VergiFs Aeneid is written is made up of 
 one long syllable followed by two short ones. Such a unit (i. e. 
 foot) is indicated thus : _ u o. For the verse-unit the poet may 
 (and often does) substitute an equivalent. 
 
 The scheme for the whole of the first verse of the Aeneid is : 
 
 _UO _OC/ — uo 
 
 Observe that two long syllables, instead of one long and two 
 short ones, are found in the third, fourth and sixth feet of this 
 
 * It will be found advantageous for the beginner in Latin poetry to go back and 
 re-read the first part of the book in the light of the information gotten from the pages 
 following. 
 
 tOur language knows practically nothing about long and short syllables, long and 
 short vowels. 
 
 Inconsonant quantity and vowel quantity are treated further on. 
 
THE QUAKTITATIVE KEADING OF LATIN VERSE. 17 
 
 verse. In all verse written in this meter the first syllable of 
 every foot has the ictus. 
 
 For a second illustration, let us take the first verse of Hor- 
 ace's second Epode : 
 
 Bea|tus il|le qui | procul | nego|tii8. 
 
 In verse of this kind the unit consists of one short syllable 
 followed by one long one, and is indicated thus : u -. The 
 scheme of the whole verse is : 
 
 u_| o- I v-l w-l o-l u- 
 
 Here the last syllable in every verse-measure has the ictus, 
 but not strong stress-accent, unless it happens that the regular 
 word-accent falls on that syllable. A glance at the verse above 
 shows that the word-accent does fall on the last syllable in three 
 of the six measures. But in the third and last measures there is 
 no word-accent on any syllable. There are many other kinds 
 of verse-feet employed by Eoman poets. Some of these are con- 
 sidered on a later page. 
 
 SCANSION. 
 
 Scansion is the name given to the process of dividing a verse 
 into verse-feet. We scan a verse of poetry when we pronounce 
 the syllables it contains in such a way as to indicate the verse- 
 feet, thus : 
 
 Arma vi — rumque ca — no Tro — iae qui — primus ab — oris. 
 
 Very little of this scanning need be done by the beginner. 
 For one to continue it long would not only be sheer waste of 
 time, but a positive detriment. The practice destroys one's 
 power to see and appreciate the rhythm, for it leads him to 
 form the habit of looking always for the verse-units as separate 
 groups, and this results in the utter confusion, if not virtual 
 destruction, of the individual words. Do not scan, but read. 
 
 LAW OF LATIN ACCENT. 
 
 The law governing the placing of the stress-accent in Latin 
 
18 THE QUANTITATIVE EEADING OF LATIN VERSE. 
 
 words is so simple that it needs only to be stated to be under- 
 stood. It is this : The stress-accent falls on the syllable next 
 the last if it (^. e. the syllable next the last) has long quantity, 
 otherwise on the syllable next to the left. That is all there is 
 of it. A few illustrations will make this doubly clear. The 
 word amahiUs has the stress-accent on the syllable -ma-, because 
 the syllable -hi- is short. The verb-form amahamus has the 
 accent on the syllable -ha-, because that syllable is long. Of 
 course, in the case of words of two syllables, there is no question 
 — the accent falls always on the initial syllable. A few apparent 
 exceptions to this law are found in those words that have lost a 
 former final syllable, We stress the final syllable of istic, istitc, 
 illic, illuc. These words once ended in e and were accented on 
 the syllable next the last. The e was dropped, the accent re- 
 mained as before. 
 
 CAESURA. 
 
 Caesura (Latin for a cutting) is the name given a natural 
 pause made in the course of the verse. Many verses have two, 
 many only one. It is not necessary, however, that a verse have 
 even one. The beauty and finish of a poem is much affected, 
 and the likill of the poet shown, by the number and the placing 
 of these pauses. A poem soon grows dull and monotonous in 
 which the pauses are regularly put in the same places in succes- 
 sive verses. 
 
 The following passage is quoted from the Aeneid for the pur- 
 pose of pointing out the instances of caesura. The position of 
 each caesura is indicated by parallel vertical lines : 
 
 Aen. Ij, vv. 223-253— 
 
 Et iam finis erat,|| cum luppiter aethere summo 
 Despiciens mare velivolum|| terrasque iacentis 
 220 Litoraque et latos populos, || sic vertice caeli 
 Constitit, II et Libyae defixit lumina regnis. 
 At que ilium talis iactantem pectore curas 
 Tristior et lacrimis oculos suffusa nitentis 
 Adloquitur Venus : || ' qui res hominumque deumque 
 
THE QUANTITATIVE READING OF LATIN VERSE. 19 
 
 230 Aeternis regis imperils, || et f ulmine terres. 
 Quid mens Aeneas || in te committere tantum. 
 Quid Troes potuere, || quibus, tot f unera passis, 
 Cunctus ob Italiam|| terrarum clauditur orbis? 
 Certe hinc Romanes olim, || volventibus annis, 
 
 235 Hinc fore ductores, || revocato a sanguine Teucri, 
 Qui mare, qui terras] omni dicione tenerent, 
 Pollicitus : II quae te, genitor,|| sententia vertit? 
 Hoc equidem occasum Troiae|| tristisque ruinas 
 Solabar, || f atis contraria fata rependens ; 
 
 240 Nunc eadem fortuna viros|| tot casibus actos 
 
 Insequitur.|| Quem das finem,|| rex magne, laborum? 
 Antenor potuit, || mediis elapsus Achivis, 
 Illyricos penetrare sinus, || atque intima tutus 
 Regna Liburnorum, et|| fontem superare Timavi, 
 
 245 Unde per ora novem|| vasto cum murmure montis 
 It mare proruptum et| pelago premit arva sonanti. 
 Hie tamen ille urbem Patavi|| sedesque locavit 
 Teucrorum, et|| genti nomen dedit,|| armaque fixit 
 Troia ; || nunc placida compostus pace quiescit : 
 
 250 Nos, tua progenies, || caeli quibus adnuis arcem, 
 Navibus (infandum !) amissis,|| unius ob iram 
 Prodimur atque Italis longe disiungimur oris. 
 Hie pietatis honos? | Sic nos in sceptra reponis?^ 
 
 It will be seen, on examination of this passage, that the 
 caesura occurs after the first long syllable in the third foot of 
 verses 1, 9, 11, 13, 14, 20, 22, 23, 24, 28 and 31 ; after the first 
 long syllable in the fourth foot in verses 2, 3, 8, 12, 16, 18, 19, 
 21, 25 and 29 ; at the end of the first verse-foot in verses 4 and 
 27 ; at the end of the second verse-foot in verse 7; at the end of 
 the fourth verse-foot in verse 26; after the first long syllable in 
 the second foot in verses 15, 17 and 19. The two caesuras most 
 frequent in dactylic hexameter verse are those after the first 
 long syllable in the third and fourth verse-feet. 
 
20 THE QUANTITATIVE EEADIKG OF LATIN VERSE. 
 
 VOWEL-QUANTITY DISTINGUISHED FROM SYLLABLE-QUANTITY. 
 
 It is of vital importance that the learner understand the 
 difference between syllable-length and vowel-length. Every 
 syllable containing a long vowel is long, but a syllable may be 
 long without having a long vowel. Many syllables containing 
 short vowels are long. How can one know when a syllable con- 
 taining a short vowel is long? The rule to be followed is simple: 
 A syllable containing a short vowel followed by a consonant 
 which is obstructed in pronunciation by a following consonant is 
 long in quantity.* 
 
 The explanation is this : An obstructed consonant requires 
 the time of a short vowel — that is its quantity. The short vowel 
 and obstructed consonant together require the time of a long 
 vowel. One of these two consonants may be the initial letter of 
 a following word, but both may not be in the following word. In 
 the first word of Ille re\git dic\tis ani\mos (part of Aen. I, 153), 
 it will be seen that the syllable II- is long, even though the 
 vowel is short, because the first I is obstructed. The syllable -git 
 is long because the short vowel i is followed by the consonant t 
 obstructed by d, initial consonant of the following word. A 
 syllable containing a long vowel followed by an obstructed con- 
 sonant is doubly long, but, in reading poetry, is compressed a 
 little. Let us illustrate. 
 
 Aen. I, 53— 
 
 Luctan|tes ven|tos tem|pesta|tesque so|noras. 
 
 The first syllable of the first word, luc-, is doubly long. The 
 vowel is long by nature and it is also followed by an obstructed 
 consonant. The same is true of -tes, -tos and -tes in the first, 
 second and third words. The syllables -ta- and -no- in the third 
 and fourth words are long because the vowels in the two 
 syllables are naturally long. Additional examples of syllables 
 long in quantity, even though their vowels are short, are -tan-, 
 ven-, tern-, -pes-. 
 
 *The expression, "consonant obstructed In pronunciation," is Mr. Hale's. See 
 his entire discussion, " Syllabification in Roman Speech." 
 
THE QUANTITATIVE READING OF LATIN VERSE. 21 
 
 There is one kind of consonant combination in which the 
 first consonant is not obstructed by the following one, viz.: 
 mute followed by liquid, as tr, ply pr, cl, hr, etc. In every such 
 case, the syllable containing the mute may be either long or 
 short, as the poet pleases.* 
 
 In Aen. II, 663, Vergil treats the first syllable of pair is as a 
 short syllable and the first syllable of pair em as long, dividing 
 . first pa-tris, then pat -rem. 
 
 Natum ante | ora pa|tris, pat(rem qui ob|truncat ad | aras. 
 
 This is not due to mere whim. The explanation is that the 
 sounds of mute and liquid glide together so readily that they are 
 ordinarily given in the time of one consonant. Yet it is pos- 
 sible, if it suits the poet^s convenience, to pronounce the mute 
 with the preceding vowel as if it were obstructed. 
 
 VOWELS LONG BY NATURE. 
 
 We call those vowels long by nature (sometimes we say by 
 authority of the poets) which we find used in poetry in positions 
 where only long vowels can stand. In the verse quoted just 
 above (Aen. II, 663) the o of ora and the a of aras are instances 
 of such vowels. The best Latin Dictionaries have these natu- 
 rally long vowels all marked by means of a horizontal mark 
 placed just over them, thus: dla, Ira, mdturus, sacerdos. Vow- 
 els not marked are short. 
 
 OBSTRUCTED CONSONANTS. 
 
 When a consonant is obstructed in pronunciation by another 
 consonant, it must be given the time of a short vowel. Unless 
 every such consonant is given its proper time, the rhythm of the 
 verse can not be maintained. Do not pronounce cella as if cela 
 but cel-la, only there must not be a break between the two syl- 
 lables. Pronounce propter as prop-ter, not as if pt were but a 
 single consonant. If a word ending in a consonant is followed 
 by one beginning with a consonant (as in occurrit tellus, 
 
 ♦For many instances of this varying quantity before mutes and liquids in the 
 works 01 Vergil, see Johnston's " The Metrical Licenses of Vergil," §§ 17-19. 
 
22 THE QUANTITATIVE READING OF LATIN VERSE. 
 
 Aen. V, 9), the final consonant, because obstructed, must have 
 the time of a short vowel. Of course, in reading, one should 
 not make a complete break between syllables, even in such a 
 word 2^^ propter, where there is a change made in the position of 
 the vocal organs in passing from p to t. In words like cella, 
 occurrit, tellus, and between words in such groups as occurrit 
 tellies, where the same sound is repeated, a break would be even 
 more unnatural and more difficult to make. The vocal organs 
 are merely held in the same position and a new voice-impulse 
 serves to mark the beginning of the next syllable. Perhaps a 
 better graphical representation than cel-la, etc., would be ceTla, 
 oc cur rit tel lus. This is quite important, and yet likely to be 
 disregarded, because it is entirely alien to our language. Con- 
 sonants that are not obstructed do not take appreciable addi- 
 tional time. So I in alia has practically no quantity. The 
 contact of tongue and palate in passing from a to i is but 
 momentary. But I in alter has quantity that must be taken 
 into account. 
 
 SLURRING. 
 
 Whenever a word, in a verse of poetry, ends in a vowel, a 
 diphthong, or m preceded by a vowel and this word is followed by 
 one beginning with a vowel, a diphthong or h, the final and 
 initial syllables are slurred, blended together, pronounced as 
 one syllable and in the time of one syllable. The quantity of 
 the new syllable will be the same as that of the initial syllable 
 of the second word. Let us take some verses of the Aeneid for 
 illustration : [Quotations have thus far been almost exclusively 
 from the Aeneid. The explanation is this: The Aeneid is the 
 poem usually taken up first in our schools, and, besides, the 
 meter, in which it is written, is one of the simplest and most 
 easily understood. In this instance, a passage from any other 
 poem would have served as well.] 
 
 Aen. II, 1 — 
 
 Oonticuere omnes, intentique ora tenebant. 
 Aen. II, 12— 
 
 Quamquam animus meminisse horret, luctuque refugit. 
 
THE QUANTITATIVE READING OF LATIN VERSE. 23 
 
 In each verse here given, all the letters covered by the hori- 
 zontal line are to be run together, in reading, into one syllable. 
 The quantity of the new syllable reom- will be long because om- 
 is long ; queo- will be long because o- is long ; quama- will be 
 short because a- is short; sehor- will be long because hor- is 
 long. What is actually pronounced for quam+a and se-i-hor may 
 be more accurately represented by qua and seor. The explana- 
 tion is that the A-sound drops out entirely (as it always has a 
 tendency to do when between vowels) and the w-sound prac- 
 tically disappears when followed by a vowel, diphthong or h 
 plus vowel or diphthong.* 
 
 Perhaps the most important thing to remember in this con- 
 nection is this, that the new syllable resulting from the slurring 
 is to be so pronounced that the quality (not the quantity) of the 
 vowel of the final syllable shall be preserved. For instance, in 
 the verses just quoted, pronounce re+om as reom, not as rom, 
 que-\-o as queo, not as quo, se-\-hor as seor, not sor. Some slurs 
 are very difficult to make. Take, for example, these : 
 
 Aen. Ill, 296— 
 
 Coniugio Aeacidae Pyrrhi sceptrisque potitum. 
 
 Aen. Ill, 225— 
 
 At subitae horrifico lapsu de montibus adsunt. 
 
 Aen. I, 322— 
 
 Vidistis si quam hie errantem forte sororum. 
 
 When more than two occur in the same verse, the reading is 
 difficult. Very unusual is such a verse as this : 
 
 Catullus, 2 2, 4 — 
 
 Puto esse ego illi milia aut decem aut plura. 
 
 The judicious use of slurring, made by a poet of taste and 
 skill, adds much to the beauty of a Latin poem. 
 
 HIATUS. 
 
 Hiatus (Latin for cleft, opening) violates the law for slurring. 
 An instance of this is seen between Samo and Mc in 
 
 * For the facts concerning the weak force of m, consult some authoritative treat- 
 ment of the sounds of the Latin letters. The most satisfactory is that published in 
 the Report of the Committee of Ten, p. 6. 
 
24 THE QUANTITATIVE HEADING OF LATIN VERSE. 
 
 Aen. I, 16— 
 
 Posthabi|ta colujisse Sa|mo ; hie | illius | arma. 
 
 The final syllable of Samo and the monosyllable hie do not 
 slur. The reasons can be learned from the grammars.* 
 
 RULES OF QUANTITY. 
 
 Only those rules of quantity thought indispensable to the 
 beginner are given here. Others will have to be gotten from 
 the grammars. But it is a great mistake to attempt to learn 
 these rules faster than they are needed in actual reading of 
 poetry. All exceptions are omitted. By far the best plan is for 
 the teacher to supply additional information to pupils respecting 
 the rules for quantity and the exceptions, taking care always to 
 avoid crowding this upon them too rapidly. One new rule at a 
 time is enough. Do not refer beginners to the mass of details 
 in the grammars for their information. This serves only to 
 confuse them. 
 
 A vowel is regularly short before another vowel or h. 
 
 Every diphthong is long. 
 
 Before 7is and nf, a vowel is always long. 
 
 Before nd and 7it, a vowel is regularly short. 
 
 Final -as, -es, -os, -is, in plural cases are long. 
 
 Final -us of the 4th decl. is short in the nom. sg., long in 
 the other cases in which it is found. 
 
 Final a 1st decl. abl., and final e 5th decl. abl. are long. 
 
 Final e of most adverbs made from adjectives of the 2nd decl. 
 is long. 
 
 Final i is long. In mihi, uU, ibi, sihi and tibi, it is long or 
 short, as the poet pleases. 
 
 Final o of the dat. and abl. sg. is long. 
 
 A syllable whose vowel is followed by a consonant, obstructed 
 in pronunciation by a following consonant, is long. 
 
 All particles attached to the ends of words are short, as, -que, 
 -ve, -ne, 
 
 *A11 the instances of hiatus found in Vergil's poems are given in Johnston's 
 "The Metrical Licenses of Vergil," gg 23-27. 
 
THE QUANTITATIVE BEADING OF LATIN VERSE. 25 
 
 We are now ready to make application of what has thus far 
 been said and to say what remains to be told to make the exposi- 
 tion complete. For the reasons already given on p. 16, the verse 
 in which the Aeneid is written will be considered first. 
 
 The verse in which Vergil wrote is called^ Dactylic Hexameter. 
 The word hexameter represents the two Greek words meaning 
 six and measure. A hexameter verse contains six measures (or 
 feet). A dactyl is a measure having a long syllable followed by 
 two short ones. It is indicated thus :«*/«. 
 
 In Dactylic Hexameter verse, each measure is regulai:ly a 
 
 dactyl or its equivalent. A spondee (two long syllables, ) is 
 
 an equivalent and maybe used in any measure. In fact, how- 
 ever, it is very rarely used in the fifth. That is regularly a 
 dactyl.* A trochee (a long and a short, . J) is very often used 
 in the last foot, never in any other. It is not really an equiva- 
 lent, but is treated as one. The last foot is never a dactyl, but 
 either spondee or trochee. .... 
 
 Let us select for our purpose Aen. I, 1 to 8 : 
 
 Arma vi|ruinque ca|no, Tro|iae qui |, primus ab|oris 
 Itali|am fa|to profu|gus La|viniaque [ vehit 
 
 Litora | mUltum il|le et ter|ris iac|tatus et | alto 
 Vi supe|rum sae|vae niemo|rem Iu|nonis ob|iram; 
 
 Multa quolque et bel|lo pas|sus, dum | conderet | urbem, 
 Infer|retque de|os Lati|o genus | unde Lajtinum 
 Alba]nique pajtres, at|que ^altae | moenia | Romae. 
 
 The reader will note that, in this kind of verse, the normal 
 word-accent in the last two measures of every verse, regularly 
 falls on the first syllable of the measure, the syllable which has 
 the ictus. An examination of the first third of successive 
 verses will show that there, also, the two often coincide. In the 
 
 * Johnston's "The Metrical Licenses of Vergil," §§21, 22, giv&s thirty-two in- 
 stances of spondee in the fifth fpot in the poems of Vergil. 
 
26 THE QUANTITATIVE READING OF LATIN VERSE. 
 
 middle of the verse there is more often conflict. The effect of 
 this return to a smooth rhythm toward the close of each verse, 
 after the roughness caused by the conflict in the middle of the 
 verse, is exceedingly agreeable, and may be likened to the 
 pleasurable sensation produced by the return to harmony, in 
 instrumental or vocal music, after an intentional discord. 
 
 Since the verse-unit here is a dactyl, the first syllable in each 
 measure has the ictus. 
 
 To read these verses (and all others written in this meter) 
 correctly, we must give to each the time represented by the 
 scheme : 
 
 Try, now, to read the verses quoted, placing a stress-accent on 
 every syllable given in heavy type (for they have the word- 
 accent), and, at the same time, preserving to each syllable its 
 exact time. At first, it is not easy to do. For those who have 
 not naturally a good sense of time, as that term is used in 
 speaking of time in music, the task is a difficult one. One can 
 help himself much, and teachers can help pupils materially, at 
 first, by keeping time by a downward and upward movement 
 of the hand as in keeping time in music. The downward 
 movement should represent (in this kind of verse) the time 
 of the first half of each measure, the upward movement the 
 time of the second half. This plan of beating time would 
 be very easy to execute if we could place a strong stress of 
 voice on the first syllable in each measure so that the downward 
 beat of the hand and the word-accent would coincide. But the 
 fact is that the normal word-accent often falls on a syllable that 
 constitutes only half of the last part of the measure, as, for 
 instance, on ca- of cano, in the first verse, and pro- of profugus, 
 in the second. At first, one will have to give his entire atten- 
 tion to the words, as he tries to read, and will find it impossible 
 to regard the thought, too. Practice, though, will soon enable 
 him to carry both along together. As soon as this proficiency 
 has been acquired, there should no longer be a complete stop at 
 the end of every verse, as before. The reader should pass with 
 
THE QUANTITATIVE READIIS^G OF LATIN VERSE. 27 
 
 only a very slight pause from verse close to verse beginning, if 
 the thought runs on. The placing of the ictus, the verse^s 
 pulse-beat, which in Dactylic Hexameter belongs to the long 
 syllable constituting the first half of every verse-foot, will 
 require no special, no conscious effort on the part of the reader. 
 Kemember that the ictus-stress is not so strong as Latin word- 
 accent, which is less strong than the stress-accent of English 
 words. 
 
 Partly for the reason that in some secondary schools the Odes 
 of Horace are read and partly because it seems appropriate to 
 show here how the method of reading Latin poetry herein set 
 forth is precisely as applicable and satisfactory when followed 
 with verse written in other meters, verses from two of the Odes 
 are here given and commented on. Two of the meters most 
 employed by Horace in his Odes are the Sapphic and the Alcaic. 
 One stanza from a poem written in the Sapphic meter is given 
 first. The first three verses have the same metrical scheme, the 
 fourth is different. For each of the first three, it is : 
 
 For the fourth, it is : 
 
 — O w -r ^ 
 
 Eor. Odes I, 32, vv. 1 to 5— 
 
 Intelger vi|tae scele|risque | purus 
 
 Non elget Maulris iacu|lis nelque^arcu 
 Nee ve|nena|tis gravi|da sa|gittis, 
 Fusee pha|retra. 
 
 [As a rule, the monosyllabic particles have not stress-accent, 
 non and nee here are emphatic] 
 
 In verse of this kind, the syllable having the ictus is the first 
 in each measure. Here, precisely as in the Dactylic Hexameter, 
 the rhythm is produced by giving to every syllable its due time. 
 Note how frequently, in the first three verses, the word-accent 
 
28 THE QUANTITATIVE READING OF LATIN VERSE. 
 
 fails to coincide with the ictus. Stress all the syllables in heavy 
 type, for they have the word-accent, and note that the swing of 
 the verse, when the poetry is read quantitatively, places the 
 ictus where it belongs. 
 
 Let us now take one stanza written in the Alcaic meter. In 
 an Alcaic stanza of four verses, three different meters are em- 
 ployed. The first two verses have the same. It is : 
 
 ^:r-|7--|T---rTo|r 
 
 The scheme for the third is : 
 
 ^ : r ^ I :r ^ I 7- - I - « 
 
 The scheme for the fourth is : 
 
 ^OUl ^ \j %J I —vjl — W 
 
 mr. Odes II, 3, vv. 1 to 5— . 
 
 Aequam me|niento | rebus in | ardu|is 
 
 Servare ] mentem | non secus | in bo|nis 
 
 Ab inso|lenti | tempejratam 
 
 Laetiti|a mori|ture Delli. 
 
 The entirely new feature in the verse here is the extra syllable 
 which precedes each of the first three verses, indicated in the 
 schemes by the three dots placed after. Each of these verses is 
 to be regarded as beginning, so far as metrical structure is con- 
 cerned, with the second syllable. That is why the dot is placed 
 beneath the second syllable, although it is the first syllable in 
 every measure that has the ictus. Note that the last measure in 
 the first and second verses is not complete. Word-accent and 
 ictus conflict here about as often as they fall together. The 
 reading of this kind of verse brings out the rhythm as indicated 
 by the schemes given. The points to be observed are precisely 
 the same as for the two kinds of verse already discussed. And, 
 in short, for Latin verse of every kind, one may follow satis- 
 factorily the directions implied in the statement given on page 
 9, in which the author states what he believes to be involved in 
 
THE QUAJifTITATIVE READING OF LATIN VERSE. 29 
 
 the correct reading of Latin poetry. For fear that the reader 
 may erroneously imagine that, in every Latin poem, the 
 syllable having ictus is the first in every measure (since it has 
 happened that such is the case in three kinds of verse shown in 
 the illustrations given), he is referred to the scheme of the first 
 verse of Horace's second Epode, p. 17. It is there made clear 
 that, in verse of that kind, the last syllable in each measure has 
 the ictus. 
 
 In conclusion, it should be said that no written directions for 
 reading Latin poetry quantitatively can help the learner so 
 readily and so much as the actual reading by one who has 
 already, by practice, acquired some skill in it. The teacher 
 ought, therefore, to show his pupils how it is to be done. 
 
 BOOKS OF REFERENCE. 
 
 Every high-school library should possess the following books, 
 which all pupils should be encouraged to use : 
 
 "The Quantitative Pronunciation of Latin'' — William Gard- 
 ner Hale (School Eeview, June, 1898). 
 
 '* Syllabification in Koman Speech" — William Gardner Hale 
 (Ginn & Co.). 
 
 "The Metrical Licenses of Vergil"— H. W. Johnston (Scott, 
 Foresman & Co.). 
 
 " Report of the Committee of Ten," (for the sounds of the 
 letters) — (American Book Co.). 
 
 Every pupil should have a copy of Lewis' "Elementary Latin 
 Dictionary" (Harper & Bros.). It is the most trustworthy 
 authority on quantities of vowels. 
 
UNIVERSITY OF CALIFORNIA LIBRARY 
 BERKELEY 
 
 Return to desk from which borrowed. 
 This book is DUE on the last date stamped below. 
 
 15Apr52R» 
 20/».pr52l« 
 
 2iJan'53!>S 
 
 JAN 9 1953 Lii 
 i4Sep'55VL| 
 
 SEPl 1955 LU 
 
 ,..„.. RiiC'D L3 
 
 ■^ '- *rf ui -;.. .g 
 
 m . 
 
 u 
 
 -hS 
 
 ortK 7 
 
 .o>?5«^ -: 
 
 4 19B' 
 
 I iT' O 
 
 Rec'D J 
 
 FE8 1 7 '64 
 
 REC'D LD 
 
 M4fi 9 - J959 
 
 ffr 
 
 5J/1N'60DF2 
 
 LD 21-95m-ll,'50(2877sl6)476 
 
 MAR ^ i 1 9 73 
 
 MURl 2-68-9/ 
 
 LOAN DCPr 
 
 N0V12 jMl 
 
 NOV 1 1 •'* 
 
 * LD 2lA-38Tn-5,'68 
 ^ (J401slO)476B 
 
 General Library 
 
 University of California 
 
 Berkeley