Tuesday, March 01, 2011

The Elegiac Couplet

The elegiac couplet was one of the most common forms of Latin poetry. It consists of two lines, one hexameter followed by a pentameter. An example is Martial's 1.38, here without macrons:

Quem recitas meus est, o Fidentine, libellus;
sed male cum recitas, incipit esse tuus!

Latin poetic forms are adaptations of forms originating in Greece, and thus, like Greek, the Latin forms are based on syllabic length (English poetry, as a contrast, is based on accent strength). To understand this, let's first investigate the basics of Latin meter.

The most common unit (or foot) is the dactyl, which consists of one long syllable (the thesis) followed by two short syllables (the arsis), with long syllables being roughly twice the length of a short syllable.* "Dactyl" is, incidentally, the Greek word for digit --you can figure out the connection between the meanings by looking at the length of the segments of one of your fingers.

The two short syllables may usually be replaced by a single long syllable -- this makes sense, since one long syllable is the same length as two short syllables: the result is called a spondee. This gives us two possibilities (- represents a long syllable, u a short one) for a foot.

Dactyl: - u u

Spondee: - -

We can refer to both of these with the following short hand:

- U

Where U indicates either a long syllable, or two shorts.

When we refer to a hexameter, the prefix tells us how many feet (dactyls or spondees) we find in a given line: six. So the following are some examples of a hexameter line:

- - | - u u | - - | - - | - u u | - -
- u u | - - | - - | - u u | - u u | - u u

And so on.

(Note that u u - (called an anapest), though metrically the same length, does not occur.)

Now we can understand the typical elegiac meter. The first line is a hexameter, with the following pattern:

- U | - U | - U | - U | - u u | - -

So, as you can see, in the first four feet either a spondee or dactyl is possible. The fifth foot is always a dactyl, and the last a spondee: this gives the last two feet the rhythm of "Shave and a haircut."

The latter line is a pentameter with the following pattern:

- U | - U | - || - u u | - u u | -

The double line indicates a pause, called the caesura. In the elegiac couplet, it sets up the punchline. As you can see, the first two feet can be either dactyls or spondees. After those two a half-foot occurs, a single long syllable. The caesura is then followed by two dactyls and another half-foot long syllable.

Now, before we apply this to Martial's couplet, we have to know a few rules of determining the length of a syllable.

1. A long vowel (in Latin, some vowels always occurred long, just as in English some vowels are always diphthongs) or a diphthong is always long (it's long by nature).

2. A short vowel, if it's in a syllable that ends in a consonant, is also long (long by position). Generally consonants attach to the following vowel, if possible. If a consonant cluster occurs the first consonant remains with the preceding vowel, the latter with the second. (It's actually a bit more complicated. See note ** at the bottom.)

Here's the couplet again, with the long vowels marked.

Quem recitās meus est, ō Fīdentīne, libellus;
sed male cum recitās, incipit esse tuus!

We can now tell which vowels are long.

Long by nature: the ā in "recitās" is long, as is the ō, and the two ī's in "Fīdentīne."

Long by position: "Quem" has a long vowel, because the syllable ends in a consonant, as does "est". Similarly the first "e" in "Fidentine" is long because the syllable ends in "n", as does the "e" in libellus, because the syllable ends in "l". The "us" in "libellus" must be long because the syllable ends in "s". "Sed" ends in a consonant, so must be long. Likewise with "cum". The first syllable of "incipit" is long, ending in a consonant. The last syllable is not long however, because as we've noted, the consonant goes with the following vowel if possible (even across word boundaries!), so here it attaches to the first syllable of "esse". The first syllable in "esse" is long, as it ends in a consonant (it's impossible to move both consonants of a double consonant to the next syllable: Martialis serpens non erat). The "us" in "tuus" is long: there's no following vowel for the consonant to move to!

So, knowing that, let's break the couplet into its feet with their length marked:

Quem recit | ās meus | est, ō | Fīden | tīne, li | bellus

- u u | - u u | - - | - - | - u u | - -

sed male | cum recit | ās, || incipit | esse tu | us!

- u u | - u u | - || - u u | - u u | -

Altogether, this give us:

- u u | - u u | - - | - - | - u u | - -
- u u | - u u | - || - u u | - u u | -

Compare this to the prescribed pattern of the elegiac couplet:

- U | - U | - U | - U | - u u | - -
- U | - U | - || - u u | - u u | -

It matches perfectly (Greek poets would frequently break the "rules," the Golden Age Latin poets less so, and the later Latin poets seldom if ever).

I mentioned the caesura being the break before the punchline. Here's the same couplet translated into English with the caesura marked to give you an idea what this means:

What you recite, Fidentinus, is mine
But when you recite badly || it begins to be yours!

Let's try another Martial. This is 1.47:

Nūper erat medicus, nunc est vespillo Dialus.
Quod vespillo facit, fēcerat et medicus.


- u u | - u u | - - | - - | - u u | - -
- - | - u u | - || - u u | - u u | -

Note this meter is identical to the last couplet, save the first foot of the second line was before a dactyl, but here has had a spondee substituted, which we know is acceptable in the meter. Our "Shave and a haircut" ending is: "... illo Diaulus"

And translated (with the flexible Latin word order shuffled a bit to sound more natural), with the caesura marked.

Diaulus was recently a doctor, now he's an undertaker.
What the undertaker does ||

(Wait for it.)

the doctor also did!

One of the benefits of the strict meter is it aids in memorization, which may be one of the reasons for its strict adherence in a time where poetry was written to be recited, not read. These are just couplets, but the Iliad, written in dactylic hexameter, is over 15,000 lines long, and began as an oral tradition, only written down by Homer later on.

In fact, I was surprised a few days ago to find that I had, without intending to, memorized the first Martial couplet just by having spent some time figuring out how it scanned.

* When it comes to dactyls, spondees, anapests, etc., these are all also used in languages that rely on stress rather than length, with the long syllables corresponding to stressed syllables, the short syllables to unstressed ones.

** Nothing's easy. First h's don't count as consonants, so "ch", "th" and "ph" are just one breathy consonant that moves together to the next vowel if possible. "Gn" and "qu" are likewise considered one consonant and move as a group. A stop (b, p, d, t, g, c) plus a liquid (l, r) can move together as well, or they may be broken apart if the poet wants to.

Second, if a word ends in a vowel and the following word begins in a vowel, the first vowel disappears. Thus, "vita est" (3 syllables) is pronounced "vit-ast" (two syllables). That's easy to buy, but somewhat more surprising is that the same rule holds if the first word ends in an "m". So "vitam est" is also pronounced "vit - ast."

Why? Probably because Latin had started to develop a degree of nasalization, whereby the "m" simply nasalized the preceding vowel in certain positions but didn't sound by itself at all. This may be why so many words with "m" endings in Latin ended up with vowel endings in the daughter Romance languages. (After those endings were gone, French and Portuguese underwent a secondary nasalization, using "m" and "n" to nasalize another set of vowels. It may have been something like this: "Bonum" (Classic Latin) --> "Bono~" (Later Classic Latin) --> "Bono" (Proto-Romance) --> "bon" (Old French) --> "bo~ɔ̃" (Modern French). The clear significance is the French will eventually express approval of something by a loud "b" sound.)

Note that the poet could also avoid this elision at his discretion.

No comments: