[The following text is taken from The Quantitative Reading of Latin Poetry by Charles E. Bennett, Allyn and Bacon, Boston, 1899. Except for corrections of mere typographical errors, or faults such as erronous hidden quantities (usually in cases where new conclusions have been made after the publication), the text is that of Bennett. Additions and commentaries, such as this, have been set within square brackets.]

Part I. The Dactylic Hexameter

§ 1. General character of Latin poetry.

English poetry, as a rule, is based on stress, i.e. on a regular succession of accented and unaccented syllables. The versification of

This is the forest primeval, the murmuring pines and the hemlocks,

depends entirely upon this alternation of accented and unaccented syllables, and the same thing is true of all ordinary English verse. This basis of English poetry, moreover, is a result of the very nature of the English language. Like all languages of the Teutonic group our English speech is characterized by a strong word-accent.

Latin verse, on the other hand, was based on quantity; a line of Latin poetry consisted of a regular succession of long and short syllables, i.e. of syllables which it took a long or short time to pronounce. This basis of Latin poetry, as in the case of English poetry, is stricly in conformity with the character of the spoken language. For classical Latin was not a language in which there was a strong word-accent. The word-accent, in fact, must have been extremely weak. Different languages differ very greatly in this respect, and we ought to bear this fact in mind in thinking of Latin. In Latin, word-accent was so weak that it could not be made the basis of versification as it is in English, while, on the other hand, quantity was a strongly marked feature of the spoken language. Thus we see how it came about that quantity was made the basis of Latin verse, and why accent was not.

[It should be noted that the Latin poetry, as we know it, was artfully adapted from Greek models, so it was possibly not so much a matter of Latin being inherently unsuited for stress based poetry, as of a strong affection for Greek culture amongst the Romans. It is thus not at all necessarily the case that the Latin accent was “extremely weak”. What Bennett most likely had in mind, is simply that the Latin accent was such that it did not affect the quantity of the syllables, as may be the case in English. If you find that playing down the stress helps you to not inadvertently disturb the syllable quantities, then by all means do so, but a weak accent is not a goal in itself.]

We are, then, to conceive of a line of Latin poetry as consisting simply of a regular arrangement of long and short syllables,—nothing else. To read Latin poetry, therefore, it is necessary simply to pronounce the words with the proper quantity. This takes some patience and practice, but it is easily within the power of every pupil of Latin who can read Latin prose with quantitative accurace. It is in Latin as in English: Any one who can read prose with accuracy and fluency has no difficulty in reading poetry. The poet arranges the words in such wise that they make poetry of themselves, if they are only properly pronounced. No other kind of poetry was ever known in any language. No other is conceivable.

Of course it necessarily takes time for the student’s ear to become sensitive to quantitative differences and to acquire a feeling for the quantitative swing of Latin verse. Yet, with patience and abundant practice in careful pronunciation, the quantitative sense is bound to develop. At all events, no other method of reading should be attempted, for no other is right.

§ 2. Ictus.

In Latin there are four chief kinds of feet:

¯˘ Trochee,¯˘˘ Dactyl,
˘¯ Iambus,˘˘¯ Anapaest.

In every one of these feet the long syllable naturally stands out with greater prominence. This quantitative prominence is called ictus.1 Thus in the Dactyl the ictus or quantitative prominence rests upon the first syllable. Whenever a Spondee is substituted for the Dactyl, as is often the case, the ictus or quantitative prominence is naturally felt as resting upon the first long of the Spondee.

Let the pupil, then, remember that ictus is not an accent, as sometimes taught, but is simply the quantitative prominence of a certain syllable of every foot. It may take time for the pupil to appreciate the full force of this definition; but just so soon as a sense is acquired for the quantitative character of Latin poetry, the naturalness of the above conception of ictus will be sufficiently apparent.

§ 3. Word-accent.

In reading Latin poetry, the ordinary accent of the words should not be neglected. But, as we have already seen above, in § 1, the word-accent in Latin was exceedingly slight. We almost invariably accent Latin words altogether too strongly. As a result we destroy the quantity of the remaining syllables of a word. Thus, in a word like ēvītābātur, we are inclined to stress the penultimate syllable with such energy as to reduce the quantity of the vowel in each of the three preceding syllables. In this way, the pupil says ĕ-vĭ-tă-bā-tur. Such a pronunciation is a fatal defect in reading. What we ought to do is to make the quantity prominent and the accent very slight. Where this is done, the accent will be felt to be subordinate to the quantity, as it ought to be, and as it must be if one is ever to acquire a feeling for the quantitative character of Latin poetry. If the quantity is not made more prominent than the accent, the accent is bound to be more prominent than the quantity, which will be fatal to the acquisition of a quantitative sense for the verse.

§ 4. Special cautions to be observed in order to secure correct syllabic quantity in reading.

Inasmuch as Latin poetry was based on the quantity of syllables, it is obvious that the greatest care must be taken in the pronunciation of the words with a view to securing an absolutely correct syllabic quantity. Otherwise the metrical (i.e. quantitative) character of the verse is violated, and the effect intended by the poet is lost. To ignore the proper quantity of the syllables is as disastrous in a line of Latin poetry as it would be in English poetry to misplace the word-accent. If one were to read the opening line of Longfellow’s Evangeline, for example, as follows:

This ís the forést prímeval,

the result would be no more disastrous than to read a line of Latin poetry with neglect of the quantity.

In reading Latin verse, there are two classes of errors to which the student is particularly liable, either one of which results in giving a wrong syllabic quantity.

Class First. In ‘Open’2 Syllables.

Here the quantity of the syllable is always the same as the quantity of the vowel. Thus, in mā-ter the first syllable is long; in pă-ter, the first syllable is short.

This being so, it is imperative that the pupil should in ‘open’ syllables scrupulously observe the quantity of the vowel. If he pronounces a short vowel long, or a long vowel short, he thereby gives a false quantity to the syllable, and thus wrecks the line completely. The pupil, therefore, must know the quantity of every vowel, and must pronounce in the light of his knowledge. He must not say gērō, tērō, sērō (for gĕrō, tĕrō, sĕrō); nor must he say pāter, āger, nīsī, quōd, quībus, ingēnium, ēs (‘thou art’), etc. One such error in a verse is fatal to its metrical structure, and the pupil who habitually commits such errors in reading is simply wasting valuable time.

Class Second. In ‘Closed’3 Syllables.

It is a fundamental fact that a ‘closed’ syllable is long. But in order to be long it must be actually closed in pronunciation. Right there is where the pupil is apt to err. He fails to make the syllable ‘closed,’ i.e. he does not join the first of the two or more consonants to the preceding vowel, but joins all of the consonants with the following vowel. He thus leaves the preceding syllable ‘open.’ Hence, if the vowel itself is short, the syllable by this incorrect pronunciation is made short, where it ought to be made long. Thus the student is apt to say tem-pe-stā-ti-bus where he ought to say tem-pes-tā-ti-bus, i.e. he joins both the s and the t with the following vowel, where he ought to join the s with the preceding vowel (thus making a ‘closed’ syllable), and only the t with the following vowel.4

Errors of the kind referred to are so liable to occur that it seems best to classify them by groups:

(a) The commonest class consists of those words which contains a short vowel followed by doubled consonants (pp, cc, tt, etc.),—words of the type of ap-parābat, ac-cipiēbam, at-tigerant, ges-sērunt, ter-rārum, an-nōrum, ad-diderat, flam-mārum, excel-lentia, ag-gerimus, etc. In Latin, both of the doubled consonants were pronounced, one being combined with the previous vowel (thus closing the syllable and making it long), one with the following vowel. But in English we practically never have doubled consonants. We write them and print them, but we do not pronounce them. Thus, we write and print kit-ty, fer-ry, etc., but we do not pronounce two t’s or two r’s in these words any more than in pity, which we write with one t, or in very, which we write with one r. Now, in pronouncing Latin the pupil is very apt to pronounce the doubled consonants of that language as single consonants, just as he does in English. Thus, he naturally pronounces the words above given, not ap-pa-rā-bat, etc., but ă-parābat, ă-cipiēbam, ă-tigerant, gĕ-sērunt, tĕ-rārum, ă-nōrum, ă-diderat, flă-mārum, excĕ-lentia, ă-gerimus. In other words, the pupil pronounces only one consonant, where he ought to pronounce two, and that one consonant he joins with the following vowel. He thus leaves the preceding syllable ‘open,’ i.e. he makes it short where it ought to be long.

The effects of this pronunciation are disastrous in reading Latin poetry, for these doubled consonants occur on an average in every other line of Latin poetry.

(b) The second groups consists of words in which a short vowel is followed by sp, sc, st; also by scl, scr, str. In English, when the vowel following these combinations is accented, we usually combine the consonants with the following vowel. Thus, we say a-scríbe, a-stoúnding, etc. Now, the Latin pupil is almost certain to do the same thing in pronouncing Latin, unless he is on his guard, i.e. he is likely to say a-spérsus, i-stṓrum, tempe-stī́vus, coru-scā́bat, mi-scúerat, magi-strṓrum, a-scrī́psit, etc. What he ought to do is to join the s with the preceding vowel (thus making the syllable closed, and long), pronouncing as-persus, is-tōrum, tempes-tīvus, corus-cābat, mis-cuerat, magis-trōrum, as-crīpsit, etc. By joining all the consonants to the following vowel he leaves the preceding syllable open. Hence, when the preceding vowel is short, the syllable also becomes short. This destroys the metre of the line.

(c) The third group consists of words containing a short vowel followed by r and some consonant. In our common English utterance we are very apt to neglect the r. This tendency is all but universal in New England, and is widely prevalent in the Middle States. As a result, the pupil is apt to pronounce Latin with the same neglect of the r as he habitually practises in the vernacular. This omission occurs particularly where the preceding vowel is unaccented, e.g. in portā́rum, terminṓrum, etc. The pupil is likely to say po(r)-tārum, te(r)-minōrum, i.e. he makes the preceding syllable ‘open’ and short, where it ought to be ‘closed’ and long. In order to close the syllable, a distinct articulation of the r is necessary. When this is overlooked, the quantity of the syllable is lost and the metrical character of the line is destroyed.

(d) The fourth group of words consists of those ending in s, preceded by a short vowel and followed by words beginning with c, p, t, v, m, n, f. In English we are very apt to join the final s to the initial consonant of the following word. Thus we habitually say grievou stale for grievous tale; Lewi sTaylor for Lewis Taylor, etc. There is great danger of doing the same thing in Latin. Experience teaches that pupils often say urbĭ sportās for urbis portās; capĭ scanem for capis canem; even urbĭ svīcī for urbis vīcī, etc. Care must be taken to join the final s clearly with the preceding vowel. Otherwise the preceding syllable will be left ‘open’ and short where it ought to be ‘closed’ and long.

The foregoing cautions are not mere theoretical inventions. They are vital, and are based on experience of the errors which we as English-speaking people naturally commit when we pronounce Latin. It is only by a conscientious observance of the principle above laid down that any one can read Latin poetry quantitatively; and unless we do so read it, we necessarily fail to reproduce its true character.

§ 5. Common syllables.

As is well known, when a short vowel is followed by a mute with l or r (pl, cl, tl; pr, cr, tr; etc.), the syllable is common, i.e. it may be either long or short in verse at the option of the poet. The explanation of this peculiarity is as follows:

In a word like pătrem, for example, it was recognized as legitimate to pronounce in two ways; either to combine the tr with the following vowel (pa-trem), thus leaving the preceding syllable ‘open’ and short, or to join the t with the preceding vowel (pat-rem), thus closing the preceding syllable and making it long. Hence, in the case of common syllables, the quantity in each individual instance depends upon the mode of pronunciation, i.e. the mode in which we divide the syllable. In reading Latin poetry, therefore, it will be necessary for the pupil to observe how the poet treats each common syllable, and to pronounce accordingly.

§ 6. Elision.

The rule for Elision, as stated in our Latin grammars, is in substance as follows: ‘‘A final vowel, a final diphthong, or m with a preceding vowel,5 is regularly elided before a word beginning with a vowel or h.’’

The exact naure of Elision, as observed by the ancients in reading Latin verse, is still very uncertain. The Romans may have slurred the words together in some way, or they may have omitted the elided part entirely. In practice, the latter procedure is probably the wiser one to follow.6

§ 7. Verses for practice.

In arranging the following examples for practice, the aim has been to proceed from the easier to the more difficult types of verse. The pupil is advises to bear in mind the four following fundamental principles:

  1. Observe the quantity of each syllable scrupulously, taking care to observe the division of the syllables as indicated by the hyphens, joining the consonant before the hyphen with the preceding vowel, and so closing the syllable.
  2. Make the word-accent light; subordinate it carefully to quantity.
  3. Endeavor to cultivate the quantitative sense, i.e. to feel the verse as consisting of a succession of long and short intervals.
  4. Do not attempt to give special expression to the ictus in any way. The ictus will care for itself if the syllables are properly pronounced.

A. Verses containing no elisions.

I. The ‘ictus’ falls upon an accented syllable in all the feet.7

II. The ‘ictus’ falls upon an accented syllable in all the feet but one (usually the third).

III. The ‘ictus’ falls upon an accented syllable in all the feet but two.

IV. The ‘ictus’ fails in three feet to fall upon an accented syllable.

V. The ‘ictus’ fails in four feet to fall upon an accented syllable.

[To be continued, possibly.]

[Footnotes]

1 The justification of this definition may be found in an article by the writer, published in the American Journal of Philology, Vol. XIX, No. 76.

2 An ‘open’ syllable is one whose vowel is followed by a single consonant (or by a mute with l or r). This single consonant (or the mute with l or r) is joined with the vowel of the following syllable, thus leaving the previous syllable ‘open.’

3 A ‘closed’ syllable is one whose vowel is followed by two or more consonants (except a mute with l or r). The first of the two (or more) consonants is regularly joined in pronunciation with the preceding vowel, thus closing the preceding syllable. This is the real significance of the common rule that a syllable is long when a short vowel is followed by two consonants. It is because one of the consonants is joined to the preceding vowel, thus closing the syllable.

4 This doctrine, to be sure, contradicts the rules given in grammars for division of words into syllables; but those rules apply only to writing, not to actual utterance. See Bennet, Appendix to Lat. Gr., § 35.

5 The elision of final m with a preceding vowel is sometimes called Ecthlipsis.

6 The writer of this little pamphlet has frequently been favored by prominent advocates of ‘slurring,’ with practical illustrations of the method of reading recommended by them; but these experiments have invariably seemed to result in producing more syllables than the verse demands. For example, in a verse like

Vix a-de|ō ag-nō|vit,

the second foot has inevitably taken the form ˘ ¯ ¯, where slurring was attempted, while in a verse like

Tan-dem | cor-ri-pu|it sē|sē atque i-ni|mī-ca re|fū-git,

the fourth foot, by slurring, has become ˘ ¯ ˘ ˘ ˘.

7 Verses of this type are extremely rare.

8 [Words like Troja, majus, etc., which are usually spelled with one j (i), are properly separated into syllables as Trŏj-ja (or Troi-ja, if we analyze the first syllable as forming a diphthong – the pronunciation would be virtually the same), rather than Trō-ja. Indeed, both Cicero and Caesar, we are told, used to spell such words with two i, in order to more correctly reflect the pronunciation.]

9 [I.e. dixerat – the letter x is divided into two syllables.]

[Back to the index.]