Prosody (Latin)

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Latin prosody (from Middle French prosodie, from Latin prosōdia, from Ancient Greek προσῳδία (prosōidía, "song sung to music; pronunciation of syllable")) is the study of Latin poetry and its laws of meter.[1] The following article provides an overview of those laws as practised by Latin poets in the late Roman Republic and early Roman Empire, with verses by Catullus, Horace, Virgil and Ovid as models. Latin poets borrowed their verse forms from the Greeks, despite significant differences in the two languages.

Latin verse: a Greek gift[edit]

A brief history[edit]

The start of Latin literature is usually dated to the first performance of a play by Livius Andronicus in Rome in 240 BC.[2] Livius, a Greek slave, translated Greek New Comedy for Roman audiences. He not only established the genre fabula palliata, but also adapted meters from Greek drama to meet the needs of Latin. He set a precedent followed by all later writers of the genre, notably Plautus and Terence.[3] The principles of scansion observed by Plautus and Terence (i.e. the rules for identifying short and long syllables, the basis of Greek and Latin meter) are mostly the same as for classical Latin verses.[nb 1] Livius, a versatile author, also translated Homer's Odyssey into a rugged native meter known as Saturnian, but it was his near contemporary, Ennius (239–169 BC), who introduced the traditional meter of Greek epic, the dactylic hexameter, into Latin literature. Ennius employed a poetic diction and style well suited to the Greek model, thus providing a foundation for later poets such as Lucretius and Virgil to build on.[4]

The late republic saw the emergence of Neoteric poets. They were rich young men from the Italian provinces, conscious of metropolitan sophistication. They, and especially Catullus, looked to the scholarly Alexandrian poet Callimachus for inspiration.[5] The Alexandrians' preference for short poems influenced Catullus to experiment with a variety of meters borrowed from Greece, including Aeolian forms such as hendecasyllabic verse, the Sapphic stanza and Greater Asclepiad, as well as iambic verses such as the choliamb and the iambic tetrameter catalectic (a dialogue meter borrowed from Old Comedy).[6] Horace, whose career spanned both republic and empire, followed Catullus' lead in employing Greek lyrical forms, though he calls himself the first to bring Aeolic verse to Rome.[7] He identified with, among others, Sappho and Alcaeus of Mytilene, composing Sapphic and Alcaic stanzas, and with Archilochus, composing poetic invectives in the Iambus tradition (in which he adopted the metrical form of the epode or "iambic distich"). Horace also wrote verses in dactylic hexameter, employing a conversational and epistolary style. Virgil, his contemporary, composed dactylic hexameters on light and serious themes and his verses are generally regarded as "the supreme metrical system of Latin literature".[8]

Modern scholars have developed different theories about how Latin prosody was influenced by these adaptations from Greek models.

Two rhythms[edit]

English meter is said to be stress timed: the regular alternation of stressed and unstressed syllables produces an "accentual rhythm." Classical Greek meter is said to be mora timed: the alternation of long and short syllables produces a "quantitative rhythm." Classical Latin meter obeyed rules of syllable length, like the Greek, even though Latin has a strong word accent like English. Modern scholars have had differing opinions about how these different influences affect the way Latin verse was sounded out. Accentual rhythm in Latin has been observed in pre-classical verse (in Saturnian meter) and in some medieval verse,[9] but otherwise the rhythm of Latin verse appears ambivalent and complex.

Latin, like English, was characterized by a stress accent, whereas ancient Greek was characterized by a pitch accent. In English poetry, accent governs the stress-timed rhythm. In ancient Greek poetry, on the other hand, pitch accent rose and fell independently of the mora timed rhythm, just as musical pitch is not governed by the duration of musical notes. Some modern scholars have suggested that the stress accent in Latin turned into a pitch accent under the Greek influence and thus Latin verse could have functioned in the melodic manner of Greek verse,[10] yet most scholars today reject such a theory as unrealistic.[11] Latin poets might instead have recited long and short syllables as if they were stressed and unstressed, or, more probably, they gave words their natural stress, so that the quantitative metrical pattern acted like an orderly undercurrent to natural speech.[12] Here, for example, is dactylic verse from Virgil's Georgics when the words are given their natural stress:

quíd fáciat laétas ségetes, quó sídere térram,

and here is the same verse when the metrical pattern is allowed to determine the stress:

quíd faciát laetás segetés, quo sídere térram.

Possibly the rhythm was held in suspense until stress and meter happened to coincide (as in "sídere térram" above).[13] English-speaking readers of Latin tend to observe the natural word stress, an approach to Latin verse that was also practised in ancient times (a 5th-century AD papyrus shows hexameter verse with accents recorded separately from the meter), yet there is also an ancient precedent for letting the meter produce an artificial stress accent.[14] In the hands of a master poet such as Vergil, however, the natural stress accent may be thought to function as a second rhythm, whose interplay with the quantitative rhythm can even be regarded as a source of unique aesthetic effects.[15]



Generally a syllable in Latin verse is long "by nature" when

  • it has a long vowel or diphthong (scrī-bae) or
  • it ends in two consonants or a compound consonant (dant, dux)

and long "by position" when

  • it ends in a consonant and is followed by a syllable that begins with a consonant (mul-tos; dat sonitum) or
  • it is the final syllable in a line of verse i.e. brevis in longo.

Otherwise syllables are counted as short durations, though a few exceptions to this rule may be found.

However, there are traps here. To give but one example, the u in the oblique cases of dux 'leader' is short, whereas in the oblique cases of lūx 'light' it is long. The length of vowels is quite often governed by rules, for example, in the inflected parts of words; when not, they have to be learnt individually. [16]

Only the nucleus and coda of the syllable are relevant for metrical purposes: the quantity of a syllable is not affected by consonants preceding the vowel. (Long syllables are sometimes called heavy and short ones light, using terminology borrowed from Sanskrit.)

If a final short open vowel stands before a single consonant or a plosive followed by a liquid (see below), in the same line, it remains short, save for some very rare licentious occurrences such as Virgil's "lappaeque tribolique", where the first short final e is treated as long. A short open final vowel may not stand before other double consonants, including x and z, in the same line, again with rare licentious exceptions such as Ovid's "alta Zacynthus", where the short final a remains short.

For the above rules to apply

  • the digraphs ch, th, ph, representing single Greek letters, count as one consonant;
  • h at the beginning of a word is ignored
  • qu counts as one consonant;
  • the consonants x [ks] and z [zz] count as two consonants;
  • A plosive (p, b, t, d, c, g) plus a liquid (r, l) can be divided between two syllables so that the first syllable may become long by position (e.g. ag-rum, pat-ris). Or, if the vowel is short, the syllable may remain short (a-grum, pa-tris): this is at the poet's discretion. This choice is not permitted, as a rule, in compound words (e.g. ab-rumpo begins with a long syllable and will not be resyllabified *a-brumpo), nor is it available for all plosive-liquid combinations.


Verses were divided into "feet" by ancient grammarians and poets, such as Ovid himself, who called the elegiac couplet "eleven-footed poetry" (Amores 1.30). This practice is followed by traditionalists among modern scholars, especially, perhaps, those who compose Latin verses. In foot-based analysis, the "metrically dominant" part of the foot is sometimes called the "rise" and the other is called the "fall," the Greek terms for which are 'arsis' and 'thesis'. In Greek, these terms were applied to the movement of human feet in dancing and/or marching, Arsis signifing the lifting of a foot, and Thesis its placement. In the Greek scheme Thesis was the dominant part of the meter, but the Romans applied the terms to the voice rather than to the feet, so that Arsis came to signify the lifting of the voice and thus the dominant part of the meter (William W. Goodwin, Greek Grammar, MacMillan Education (1894), page 348). This caused confusion, as some authors followed the Greek custom and others the Latin; thus these terms are no longer generally used. Sometimes the dominant part of the foot, in either quantitative or stressed verse, is called the ictus.

Long and short syllables are marked (-) and (u) respectively. The main feet in Latin are:

  • Iamb: 1 short + 1 long syllable (cărō)
  • Trochee: 1 long + 1 short (mēnsă)
  • Dactyl: 1 long + 2 shorts (lītŏră)
  • Anapaest: 2 shorts + 1 long (pătŭlaē)
  • Spondee: 2 longs (fātō)
  • Tribrach: 3 shorts (tĕmĕrĕ)

According to the laws of quantity, 1 long = 2 shorts. Thus a Tribrach, Iamb and Trochee all equate to the same durations or morae: each of them comprises 3 morae. Similarly a Dactyl, an Anapaest and a Spondee are quantitatively equal, each being 4 morae. These equivalences allow for easy substitutions of one foot by another e.g. a spondee can be substituted for a dactyl. In certain circumstances, however, unequal substitutions are also permitted.

It is often more convenient to consider iambics, trochaics and anapests in terms of metra rather than feet; for each of these families, a metron is two feet. Thus the iambic metron is u-u-, the trochaic -u-u and the anapestic uu-uu-.

Cola: a different way to look at it[edit]

The division into feet is a tradition that produces arbitrary metrical rules, because it does not follow the actual metrical structure of the verse (see for example the listed variations in the tables below). In particular, though a long syllable and two short ones have the same number of morae, they are not always interchangeable: some metres permit substitutions where others do not. Thus a more straightforward analysis, favoured by recent scholarship, is by cola, considered to be the actual building blocks of the verse. A colon (from the Greek for "limb") is a unit of (typically) 5 to 10 syllables that can be re-used in various metrical forms.[HOR 1]

Standard cola include the hemiepes, the glyconic, and the lekythion.


A vowel at the end of a word does not count as a syllable if the following word begins with a vowel or h: thus Phyllida amo ante alias reads as Phyllid' am' ant' alias. This is called elision. At the (rare) discretion of the poet, however, the vowel can be retained, and is said to be in Hiatus. An example of this, in Virgil's fémineó ululátú the "o" is not elided.

A word ending in vowel + m is similarly elided (sometimes this is called Ecthlipsis): thus nec durum in pectore ferrum reads as nec dur' in pectore ferrum.[17]

Syllables ending in a vowel are called open syllables, and those ending in a consonant are called closed syllables.


The last syllable of a foot might or might not coincide with the ending of a word. A too frequent repetition of this coincidence apparently grated on the Roman ear, as in the following example of a poorly composed dactylic hexameter, where each of the first four feet consists of a spondaic word (the feet are marked out by vertical lines):

spārsīs | hāstīs | lōngīs| cāmpūs | splēndĕt ĕt | hōrrĕt.[18]

The ending of a word and foot together, as in this line, is termed diaeresis. To avoid such lines , caesuras, or "cuttings", were employed, a caesura being a place where a word ends within a foot, here marked ||. For convenience, in the following two examples from Virgil the foot endings are marked -|- when they occur within words and | when they occur between words.

Tīty̆rĕ | tū || pătŭ-|-laē || rĕcŭ-|-bāns || sūb | tēgmĭnĕ | fāgī
nōn om-|-nīs || ar-|-busta || iu-|-vant, || humi-|-lēsque || myr-|-īcae

There are two kinds of caesura:

  • strong caesura, when the caesura occurs after a long syllable;
  • weak caesura, when the caesura occurs after a short syllable (none in the first line above, two in the second).


The dividing of verse into long and short syllables and analysis of the metrical family or pattern is called 'scanning' or 'scansion.' The names of the metrical families come from the names of the cola or feet in use, such as iambic, trochaic, dactylic and anapaestic meters. Sometimes meter is named after the subject matter (as in epic or heroic meter), sometimes after the musical instrument that accompanied the poetry (such as lyric meter, accompanied by the lyre), and sometimes according to the verse form (such as Sapphic, Alcaic and elegiac meter).

Guide to symbols used[edit]

  • — for long syllable or long element
  • u for short syllable or short element
  • for brevis in longo
  • | for end of foot
  • ‖ main caesura


  • words are hyphenated wherever they include the end of a foot e.g Trō-iae below;
  • long and short vowels are marked with - and u directly above them e.g. Ā, ă, ĭ, ī, ō, ŏ, ŭ, ū (these don't indicate syllable lengths)

There are four basic families of verse: dactylic, iambic (and trochaic), Aeolic, and anapestic. In the dactylic family short syllables come in pairs, and these pairs may be contracted (two short replaced by one long). In the iambic/trochaic family short syllables come one at a time, and some long elements may be resolved (one long replaced by two short). In the anapestic family short syllables come in pairs, and both contraction and resolution are allowed. In the Aeolic family there are both paired and single short syllables, and neither contraction nor resolution is allowed. Other important metres are hendecasyllabics and the Asclepiads, and Catullus composed important poetry in Glyconics. There are individual Wikipedia entries on various metres. A would-be composer in any metre, however, would need a more detailed knowledge than can be found here.

Dactylic meters[edit]

The "dactyl," as a foot, is — u u; the name comes from the Greek for "finger," because it looks like the three bones of a finger, going outward from the palm. The principal colon of dactylic verse is the "hemiepes" or "half-epic" colon, — u u — u u — (sometimes abbreviated D). The two short syllables (called a biceps element) may generally be contracted, but never in the second half of a pentameter, and only rarely in the fifth foot of a hexameter. The long syllable (the princeps element) may never be resolved. Roman poets use two dactylic forms, the hexameter and the elegiac couplet.

Dactylic hexameter[edit]

Dactylic hexameter was used for the most serious Latin verse. Influenced by Homer's Greek epics, it was considered the best meter for weighty and important matters, and long narrative or discursive poems generally. Thus it was used in Ennius's Annals, Lucretius's On The Nature of Things, Virgil's "Aeneid" and Ovid's "Metamorphoses"; also in Juvenal's caustic satires and Horace's genial Talks and Letters.

A dactylic hexameter consists of a hemiepes, a biceps, a second hemiepes, and a final long element, so DuuD—. This is conventionally re-analyzed into six "feet," all dactyls with the last one either catalectic or necessarily contracted. Roman poets rarely contract the fifth foot.[nb 2] Since Latin was richer in long syllables than was Greek, contraction of biceps elements (producing the so-called spondee) was more common among Roman poets. Neoteric poets of the late republic, such as Catullus, sometimes employed a spondee in the fifth foot, a practice Greek poets generally avoided and which became rare among later Roman poets.[19]

Variations 1st 2nd 3rd 4th 5th 6th[nb 3]
dactyls — u u — u u — u u — u u — u u — —
spondees — — — — — — — —

There will be a caesura in the third or fourth foot (or perhaps, of course, in both). If there is a weak caesura, or none, in the third foot, there will usually be a strong one in the fourth, as in these two examples from Virgil:

sī nescīs, meus ille caper fuit, et mihi Dāmōn ...
et nōbīs īdem Alcimedōn duo pōcula fēcit ...

but here is a line from Virgil with only one caesura, a weak one:

frangeret indēprēnsus et irremeābilis error.

Variations are common, and are used to avoid monotony. Their absence would be a definite fault of versification. Various positions for caesura (in the foot-based analysis) have traditional names: the caesura "in the third foot" is called penthemimeral, that in the fourth hephthemimeral, and that in the second trihemimeral. These names refer to the number of half-feet before the position of the caesura.[20] Dactylic hexameter often has a bucolic diaeresis (a diaeresis between the fourth and fifth feet of a line), as in the first of the following lines from the introduction to Virgil's epic poem, the Aeneid.

 -  u  u| -   u  u| -||  -| -    -|  - u  u |- - 
 Ārmă vĭ-rŭmquĕ că-nō, Trō-iae quī prīmŭs ăb ōrīs
 - u u|-   -| - ||  u u| -   -| -  u  u| - -
 Ītălĭ-ǎm fā-tō   prŏfŭ-gŭs Lā-vīniăquĕ vēnĭt
  - u u | -       - |     -   - | - || - | - u  u |-  -
 lītŏră, mŭlt(um) ĭl-l(e) ĕt tĕr-rīs  iăc-tātŭs ĕt ăltō
  -  u u| - ||  - | -   u u| -   -| - u  u |- -
 vī sŭpĕ-rŭm,  sae-vae mĕmŏ-rĕm Iū-nōnĭs ŏb īrăm;

There are two elisions in line 3 and a bucolic diaeresis in line 1 (quī | prīmus ). Venit and iram at the ends of lines 2 and 4 count as spondees by brevis in longo, despite their naturally short second syllables. The 'i' in 'Troiae' and 'iactatus', the first 'i' in 'Iunonis' and the second 'i' in 'Laviniaque' are all treated as consonants. Bucolic diaeresis has this name because it is common in bucolic or pastoral verse. (NB, however, that this term is sometimes, or even usually, reserved for lines where the fourth foot is a dactyl, as in

forte sub argūtā cōnsīdĕrăt īlice Daphnis

Dactylic hexameters regularly end with a dissyllabic or a trisyllabic word. Exceptions tend to be Greek words.

Dactylic pentameter[edit]

The name "pentameter" comes from the fact that it consists of two separate parts, with a word-break between them, with each part, or hemiepes, having two-and-a-half feet, summing to five (thus giving Ovid his count of eleven feet in a couplet). The first hemiepes may have contraction, the second may not. By Ovid's time there was a rule, with very few exceptions, that the last word should be of two syllables, and it was almost always a noun, verb, personal pronoun (mihi, tibi or sibi) or pronominal adjective (meus etc.). The last syllable would either be closed, or a long open vowel or a diphthong: very seldom an open short vowel.

Variations 1st 2nd ½ 3rd 4th ½
— u u — u u — u u — u u
spondees — — — —

There is a strong danger of monotony in this rigid structure, which poets were able to alleviate, up to a point, by keeping the first half of a line out of conformity with the stricter rules governing the second half, and by varying as much as possible the word-pattern of the second half.

Elegiac couplet[edit]

An elegiac couplet is a dactylic hexameter followed by a dactylic pentameter. The sense of the hexameter frequently runs into the pentameter, a flow-on effect known as enjambement, but a pentameter compararively seldom runs on into a following hexameter. The pentameter came into Latin usage later than did the hexameter and therefore it was not always handled with rigour by Catullus, compared for example with the later poets, especially Ovid. Catullus used elisions very freely, and sometimes he even allowed an elision to span the central diaeresis (e.g. Carmina 77.4). The following is from one of his most famous elegies, mourning for a lost brother (Carmina 101).[21]

  -  - | -   - | - ||- | -  u  u | -  u u| -  -
 Mŭltās pĕr gĕn-tēs  ĕt mŭltă pĕr aequŏră vĕctŭs
      -  u u   | -   u u |- ||  - u   u |-  u u|-
      ădvĕnĭ(o) hās mĭsĕr-ās, frātĕr, ăd ĭnfĕrī-ās
 -   -| -   -| -||- |- -  | - u u| -  -
 ŭt tē pŏstrē-mō dōn-ārĕm  mūnĕrĕ mŏrtĭs
      -   -| -   -|  -  ||    -  u  u| -   u u| -
      ĕt mū-tăm nē-quīqu-(am) adlŏquĕ-rĕr cĭnĕ-rĕm,

Note: the diaeresis after the first hemiepes is marked here like a caesura (a conventional practice.)[22] Observe the elisions in line 2 (o) and line 4 (am). The latter elision spans the diaeresis in the last line.

First Archilochian[edit]

If only one hemiepes is employed, instead of a full pentameter, the elegiac couplet takes the form known as the First Archilochian, named after the Greek poet Archilochus. An example is found in the fourth book of Horace's Odes (Carmina 4.7), which A. E. Housman once described as "the most beautiful poem in ancient literature",[23] introduced with these two lines:

 -  -| - u  u|  - || uu| -   - | - u u | - -
Dīffū-gērĕ nĭ-vēs, rĕdĕ-ŭnt iăm grāmĭnă cămpīs
   - u u| -   u  u | -
  ărbŏrĭ-bŭsquĕ cŏm-ae;

Dactylic tetrameter catalectic[edit]

Most extant examples of this meter are found in Lyric poetry, such as Horace's Carmina 1.7 and 1.28, but also in Iambi.

Variations 1st 2nd 3rd 4th
— u u — u u — u u — —
spondees — —

Note: the final syllable in the 4th foot is marked long or short in some schemes to indicate natural syllable length but it is always long by position.

Alcmanian strophe[edit]

A dactylic tetrameter catalectic is sometimes joined to the dactylic hexameter to form a couplet termed the Alcmanian Strophe, named after the lyric poet Alcman (some scholars however refer to the Alcmanian Strophe as the First Archilochian, as indeed there is a strong likeness between the two forms). Examples of the form are found in Horace's Odes (carmina) and Epodes, as here in his Epode 12.[24]

 - u u |  -   - |- ||  - | -  u u |-   u u | -  -
 Ō ĕgŏ | nōn fēl-īx, quăm tū fŭgĭs ŭt păvĕt  ācrīs
  -  u  u| -   u  u| -  u  u|- -
  ăgnă lŭ-pōs căprĕ-aēquĕ lĕ-ōnēs

Note that the plosive + liquid combination pr in 'capreaeque', syllabified, leaves the first open syllable (ca) metrically short.

Iambic meters[edit]

The iambic family has short syllables one at a time, not in pairs like the dactylic family, and it allows resolution of long elements. In this family there may also be anceps positions, that is, positions in which either a long or a short syllable is allowed.

Iambic trimeter and Senarius[edit]

Used for the theatre by poets such as Plautus and Terence, an iambic line of six feet allowed for numerous variations and it is known as an iambic senarius ('senarius' = 'grouped in sixes').

Poets such as Horace and Catullus however employed iambic feet in pairs, each called a metron, in which fewer variations were allowed. Such a line of verse has three metra and the meter in that case is called 'iambic trimeter'. Here is the list of variations found in Horace,[25] which are more numerous than those in Catullus.[26] Some variations are due to resolution (replacing a long with two shorts) and some are due to anceps (when a syllable may be either long or short—see following Note and the article iambic trimeter). Others are the result of outright substitution (prosody), such as the anapaest in the first foot below.

Variations 1a 1b 2a 2b 3a 3b
u u — u u u — u —
spondees — — — — — —
tribrachs uuu uuu uuu uuu
dactyls —uu —uu
anapests uu—


  • The caesura usually follows the first syllable of the third foot (2a), and sometimes the first syllable of the fourth foot (2b).
  • A short syllable can be replaced by a long syllable in certain positions, here marked bold: i.e. the first short syllable in feet 1a, 2a, 3a may be long. These short/long syllables are said to be in anceps.
  • Some variations never appear; thus for example there is no resolution in foot 3a and therefore no tribrach or dactyl in that foot.

The following example of iambic trimeter is from Horace's Epode 5:

 - u u |u   u u| u|| - | u - |u  - |u - |
 Cānĭdĭ-ă brĕvi-bus ĭmp-lĭcāt-ă vīp-ĕrīs

Note that the first syllable Ca- in Canidia is long and it is thus in anceps, while -idi- is the equivalent of a long syllable and it is said to be resolved into two shorts—thus the first foot appears as if a dactyl has been substituted for an iamb. Long syllables are resolved into two shorts (uu) in the first metron (1a-1b), an effect that may have been intended to suggest the quick movement of tiny snakes that Canidia has tied to her hair.[27] The second foot (1b) is read as a tribrach (uuu) since 'br' in 'brevibus' can be taken as a single consonant, making the preceding 'a' short.

Iambic dimeter[edit]

Iambic verse of four feet, paired to make two metra, hence the name 'iambic dimeter'. It is constituted like the first and third metra of the iambic trimeter (1a, 1b, 3a, 3b), with which it is often joined to form couplets known as Iambic Distich.

Variations 1a 1b 2a 2b
u u — u u —
spondees — — — —
tribrachs uuu uuu
dactyls —uu
anapests uu—


  • the short syllables in 1a and 2a (marked bold) are in anceps and can be replaced by long syllables
Iambic distich[edit]

The iambic distich—a couplet comprising an iambic trimeter and iambic dimeter—is the basis of many poems of a genre known as Iambus, in which the poet abuses and censures individuals or even communities, whether real or imaginary. Iambic rhythms were felt to be especially suited to this role. The Greek poet Archilochus was one of the main exponents of the iambic distich and he provided a model for Horace's Epodes 1-10. The following couplet forms the first two lines of Epode 7.

   -   - | u - | -   uu| u  || - | -   - | u - 
 Quo quo sceles-ti  rui-tis?  aut cur dex-teris
    -  - | u  - | -   - | u -  
    aptan-tur en-ses con-diti?

Note that the first syllables in both lines (Quo and ap) are in anceps and thus a spondee appears to have been substituted for an iamb in both the initial feet. Also ti in scelesti is long by nature and it too is in anceps; the third foot would otherwise be a tribrach (uuu) thanks to the resolution of a long syllable into the two shorts in rui. As a result of all these variations, the 3rd foot instead resembles a dactyl (-uu).

Second Archilochian[edit]

An iambic dimeter may be followed by a hemiepes to form the second line of a couplet, in which the first line is dactylic hexameter. This combination is called the second Archilochian. The iambic dimeter functions like an independent line, however, since it keeps the elements of a line-end i.e. it is marked off from the hemiepes by a pause through brevis in longo, or through a hiatus. An example of this system is found in Hoarace's epode 13, lines 9-10

  -  - | - - | -||u  u |-  u u | -  -| - - 
 pěrfŭn-dī nār-dō iuvǎt ět fĭdě Cyllē-naeā 
   -  - | u -| -  -  | u ῡ|| - u u | - uu | -
   lēvār-ĕ dī-rīs pĕc-tǒră  sǒllĭcĭ-tūdĭnĭ-bŭs

Note that the 5th foot in the hexameter line is a spondee—this is rare for Horace and it is meant to evoke the affectation of Neoteric poets like Catullus, thus complementing the sense of being suffused with perfume while listening to the lyre at a drinking party (the Greek word Cyllenaea, which creates the double spondee, adds to the exotic aura).[28] Observe also that the iambic dimeter ends with brevis in longo, the short syllable a in pectora becoming long by the addition of a pause. The hemiepes also ends with brevis in longo with bus short by nature but long by position.

Third Archilochian[edit]

Here an iambic trimeter forms the first line of the couplet, and the positions of the iambic dimeter and hemiepes are reversed to form the second line, the hemiepes now coming before the iambic dimeter. The hemiepes still functions as if it were independent, retaining the pause of a line-end through brevis in longo or hiatus. An example has survived in Horace's Epode 11, as in lines 5-6 here:

  -   - | u- | -  -| u ||- | -  -  |u  -
 hīc těr-tĭŭs Dēcĕm-bĕr, ĕx quō dēs-tĭtī
    - u u |-  u u |ῡ || - -  | u -| -  - | u -
   Iānăchĭ-ā fŭrĕr-ĕ,  sĭlvīs hŏnō-rĕm dē-cŭtĭt.

Another couplet is formed when a line of dactylic hexameter is followed by a line of iambic dimeter, and this is called the First Pythiambic. The Greek poet Archilochus composed in this form but only fragments remain. Two of Horace's epodes (14 and 15) provide complete examples in Latin. The following couplet introduces his Epode 15:

  -  u u |-   - |- || - | -  - | - u u |- -
 Nŏx ĕrăt ĕt cae-lō  fŭl-gē-băt lūnă sĕ-rēnō
    - - | u -| u  - |u ῡ
   ĭntĕr mĭnō-ră sīd-ĕră 

The Second Pythiambic features an iambic trimeter instead of iambic dimeter in the second line. Horace's Epode 16 is an example.

Iambic tetrameter catalectic[edit]

Usually associated with the comic theatre, it consists of seven feet with an extra syllable at the end instead of a full iambic foot. In that case it is called iambic septenarius ('septenarius' means grouped in sevens). Used outside the theatre, the feet are paired and then it is called iambic tetrameter catalectic (catalectic means that the meter is incomplete). The stage allowed many variations but poets were quite strict in their use of it and Catullus allowed variations only in the first and fifth feet:[29]

Variations 1a 1b 2a 2b 3a 3b 4a 4b
u u — u — u — u u — u —
spondees — — — —

An example is found in Catullus' Carmina 25, beginning with these two lines:

 u - | u   - | u   - | u -| u -| u - |u - | -
Cĭnae-dĕ Thăl-lĕ, mŏl-lĭŏr cŭnī-cŭlī căpĭl-lō
 u  - | u - | u - | u -| u  -| u     - |u - | -
vĕl ān-sĕrĭs mĕdŭl-lŭlā vĕl ī-mŭl(ā) ōr-ĭcĭl-lā 

Catullus uses no variations at all here and he employs diminutives (cuniculi, medullula, imula, oricila) contemptuously in a description of the 'soft' Thallus. Note that doubling of the consonant l lengthens several syllables that are naturally short, thus enabling a strict iambic rhythm.


This meter was originated by the Greek iambic poet, Hipponax. The name choliambics means lame iambics and sometimes it is called scazons or limpers. ("Lame trochaics" exist as well, being a trochaic tetrameter catalectic with the same ending as the iambic.) It is intended to be graceless and awkward " order to mirror in symbolically appropriate fashion the vices and crippled perversions of mankind."[30] It was taken up by the neoteric poets Catullus and his friend Calvus but with fewer variations than Hipponax had employed. It is basically an iambic trimeter but with a surprise ending in the last foot, featuring a trochee or spondee that cripples the iambic rhythm. As used by Catullus, the variations are as follows:

Variations 1a 1b 2a 2b 3a 3b
u u — u u u — — —
spondees — — — — — —
tribrachs uuu
dactyls —uu —uu

Caesuras are found after the first syllable either in the third or fourth feet, sometimes in both. Lines 2 and 3 of Catullus' Carmina 59 about the grave-robbing wife of Menenius offer a good example:

 - -| u -|- || - | u       -  | u - | - -
ŭxŏr Mĕnē-nī, sae-pĕ qu(ăm) ĭn sĕpŭl-crētīs
 - - | u  - | -||u u |u - | u -| - -
vīdĭs-tĭs ĭp-sō răpĕr-ĕ dē rŏgō cēnăm

The dactyl in the third foot of the second line reinforces the meaning, as she greedily reaches for food from the funeral pyre without regard for taboos.

Martial used more variations, such as an anapaest in the fourth foot and a tribrach in the third.

Post-classical poetry[edit]

The above versification, based on long and short syllables, was applied only to learned poetry, made by Latin poets of the classical period in imitation of Greek models. The metrics of popular songs, popular poetry, military marches and so on was based on accents.

After the classical period, the pronunciation of Latin changed and the distinction between long and short vowels was lost in the popular language. Some authors continued writing verse in the classical meters, but this way of pronouncing long and short vowels was not natural to them; they used it only in poetry. Popular poetry, including the bulk of Christian Latin poetry, continued to be written in accentual meters (sometimes incorporating rhyme, which was never systematically used in classical verse) just like modern European languages. This accentual Latin verse was called sequentia, especially when used for a Christian sacred subject.

See also[edit]


  1. ^ Two significant differences are that word-final s may not be counted as making a heavy syllable, and mute-plus-liquid combinations never make a syllable heavy. R.H.Martin, Terence: Adelphoe, Cambridge University Press (1976), page 32
  2. ^ According to the stress-timed theory of Latin prosody, there is a strong tendency to harmonize word-stress and verse-ictus in the final two feet of a hexameter. The fifth foot, therefore, is almost always a dactyl whereas the sixth foot always consists of a spondee; this line ending is perhaps the most notable feature of the meter. In classical times, it was the only readily audible metrical feature, and Romans unfamiliar with Greek literature and versification often heard no sound pattern at all save in the stress-pattern of the last two feet (William Sidney Allen, Vox Latina: a Guide to the Pronunciation of Classical Latin, 2nd edition, Cambridge University Press (2003) ISBN 0-521-37936-9, pages 86, 127).
  3. ^ In some schemes, the final syllable in the 6th foot is marked either long or short to reflect the natural syllable length, but it is always long by position and it is therefore only marked long in this table.
  1. ^ Halporn, Ostwald, Rosenmeyer


  1. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), page 201
  2. ^ See Goldberg for details.
  3. ^ R.H.Martin, Terence: Adelphoe, Cambridge University Press (1976), page 1
  4. ^ P.G.McBrown, 'The First Roman Literature' in The Oxford History of the Classical World, J.Boardman, J.Griffin and O.Murray (eds), Oxford University Press (1995) page 450-52
  5. ^ Robin Nisbet, 'The Poets of the Late Republic' in The Oxford History of the Classical World, J.Boardman, J.Griffin and O.Murray (eds), Oxford University Press (1995) page 487-90
  6. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 32-7
  7. ^ "Princeps Aeolium carmen ad Italum," Odes 3.30.13; for Horace's engagement with Catullus see Putnam (2006).
  8. ^ Richard F. Thomas, Virgil: Georgics Vol. I, Cambridge University Press (1988), page 28
  9. ^ For the contrast between stress-based and quantitative verse, and for developments after the classical period, see especially Gasparov. For Saturnians, Halporn et al. say "most Saturnians make some kind of sense if we assume that the natural word accent alone carries the rhythm" (p. 60-61). See also Parsons, Freeman, Cole, Mahoney.
  10. ^ R.G. Kent, The alleged conflict of accents in Latin verse, T.A.P.A. 51 (1920), pages 19-29)
  11. ^ Richard F. Thomas, Virgil: Georgics Vol. I, Cambridge University Press (1988), pages 29
  12. ^ L.P. Wilkinson, Golden Latin Artistry, Cambridge (1963), page 94
  13. ^ William Sidney Allen, Vox Latina: a Guide to the Pronunciation of Classical Latin, 2nd edition, Cambridge University Press (2003) ISBN 0-521-37936-9, pages 83-88
  14. ^ Richard F. Thomas, Virgil: Georgics Vol. I, Cambridge University Press (1988), pages 28-9
  15. ^ R. D. Williams, The Aeneid of Virgil, Books I-VI, MacMillan (1972), Introduction page xxvii; W. F. Jackson Knight, Accentual Symmetry in Virgil, Basil Blackwell (1950)
  16. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), page 201
  17. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), page 203
  18. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), page 204
  19. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 40
  20. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), pages 204-5)
  21. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 40-1
  22. ^ see for example Peter Green, The Poems of Catullus, University of California Press (2005), page 40
  23. ^ William Flesch, The facts on File Companion to British Poetry, 19th Century, Facts on File, Inc. (2010), page 98; see Google preview
  24. ^ David Mankin, Horace: Epodes, Cambridge University Press (1995), pages 20-22
  25. ^ B.H.Kennedy and James Mountford, The Revised Latin Primer, Longman (1962), pages 206)
  26. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 34
  27. ^ David Mankin, Horace: Epodes, Cambridge University Press (1995), note 15 page 114
  28. ^ David Mankin, Horace: Epodes, Cambridge university Press (1995), pages 219-20
  29. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 34-5
  30. ^ Peter Green, The Poems of Catullus, University of California Press (2005), pages 33
  • Allen, William Sidney (2003). Vox Latina — a Guide to the Pronunciation of Classical Latin (2nd ed.). Cambridge University Press. ISBN 0-521-37936-9. 
  • Cole, Thomas (1972). "The Saturnian Verse". Yale Classical Studies. 21: 3–73. 
  • Freeman, Philip (1998). "Saturnian Verse and Early Latin Poetics". Journal of Indo-European Studies. 26: 61–90. 
  • Gasparov, M. L. (1996). A History of European Versification. Oxford University Press. ISBN 0-19-815879-3. 
  • Goldberg, Sander (2005). Constructing Literature in the Roman Republic. Cambridge University Press. ISBN 0-521-85461-X. 
  • Halporn, James W.; Martin Ostwald; Thomas G. Rosenmeyer (1994) [1963]. The Meters of Greek and Latin Poetry. Hackett. ISBN 0-87220-244-5. 
  • Mahoney, Anne (2001). "Alliteration in Saturnians". New England Classical Journal. 28: 78–82. 
  • Parsons, J (1999). "A New Approach to the Saturnian Verse". Transactions of the American Philological Association. 129: 117–137. 
  • Putnam, Michael C. J. (2006). Poetic Interplay: Catullus and Horace. Princeton University Press. ISBN 978-0-691-12537-4. 
  • Watkins, Calvert (1995). How to Kill a Dragon: Aspects of Indo-European Poetics. Oxford University Press. ISBN 0-19-508595-7.