Tuesday, 28 February 2023

Homer's metre 1. Dactylic hexameter

Were you taught that the hexameter is a line consisting of six feet, each of which is either a dactyl or a spondee? If so, that was ... well, it wasn’t wrong, not in the sense of being an actual falsehood. But it certainly gave you the wrong mental model.

‘Feet’ are to Homeric poetry what roads are to geology. Handy for finding your way around, but nothing to do with the actual thing.

Let’s try a different route. Start from the basics. Dactylic hexameter is a poetic rhythm. A line of hexameter is built out of phrases of regular rhythmic shapes, which combine together to form a twelve beat rhythm. It is regularly used in ancient Greek and Latin epic poetry, such as Homer and Vergil.

1. Structure of the hexameter | 2. Hermann’s Bridge | 3. Catalogue and glossary

The wrong concepts: feet and caesuras

Yes, it’s true that ‘hexameter’ means ‘a line of six feet’ (hex- ‘six’), and ‘dactylic’ means it’s got a dum-diddy rhythm. That’s the literal meaning of the name.

They’re real words. But they’re the wrong words.

They create a mental model of ‘a sausage-string of dactyls’, as M. L. West put it fifty years ago (1973: 188). It’s a weirdly inorganic, artificial way of thinking of the hexameter. In reality it’s a very organic structure. It’s built out of phrases, not feet.

Here are some things that are traditionally taught to beginners reading ancient poetry for the first time:

  • long and short syllables
  • foot, dactyl (— ⏑⏑), spondee (— —)
  • caesura

The one about long and short syllables is fine. But the other two items are back-to-front. Feet and caesuras aren’t important for their own sake: they’re symptoms, side-effects, analytic tools. Poets invented rhythm; grammarians invented feet. And what on earth is the point of identifying caesuras if you don’t know why caesuras exist?

Here are some much more central concepts.

  • hemipes
  • colon
  • bridge

Oh, sorry, did you want to know what these are? Good luck finding an encyclopaedia that covers them! Students are traditionally taught to regard them as ‘advanced’. You won’t find an entry for any of them in The Homer encyclopedia, let alone in generalist encyclopaedias like Wikipedia. (Admittedly The Homer encyclopedia doesn’t have entries for ‘foot’ or ‘dactyl’ either. But both it and Wikipedia do have entries for ‘caesura’.)

Note. Some metricians would add ‘contraction’ to the list of central concepts: that is, the use of — to replace ⏑⏑ (or if you absolutely insist, spondees replacing dactyls). However, contraction assumes a particular set of theories for how the hexameter came into being, and there are problems. I think that genuinely is too advanced for an introduction.

I’d like to see an encyclopaedia that has no entry for ‘caesura’. Because caesuras don’t bloody matter. I want to see an encyclopaedia that does have entries for ‘colon’ and ‘Hermann’s Bridge’. Because, not only are these things integral, but also, as things stand it can be hard to find out what they actually are without resorting to some very esoteric books.

The right concepts (1): rhythm

Let’s start with just the bare rhythm, and move on to structure afterwards.

In early Greek poetry and classical Latin poetry, rhythms can be expressed in modern musical notation as crotchets (quarter notes) and quavers (eighth notes). ‘Dactylic hexameter’ is this twelve beat rhythm:

Note that we’re talking about note lengths, not stressed and unstressed syllables. Ancient grammarians did talk of epic having a downbeat and upbeat (thésis ‘putting down’, ársis ‘lifting’), but it’s specifically a musical beat, not a pattern of word stresses.

In metrical notation, the rhythm looks like this:

— ⏑⏑ — ⏑⏑ — ⏑⏑ — ⏑⏑ — ⏑⏑ — ×

where — means ‘crotchet’, ⏑ means ‘quaver’, and × means ‘indifferent, no really I mean it, please stop marking this long or short’.

Note. The last note is neither long nor short: it is just the last note. In hexameter (though not in all metres) it’s simply wrong to pin down an anceps as long or short. See further West 1987: 5.

Just like in modern double-time music, you can usually have a single long note in place of a pair of short notes. (The reverse doesn’t happen: the long notes always stay long notes.) So the actual rhythm of the hexameter is

— ⏕ — ⏕ — ⏕ — ⏕ — ⏑⏑ — ×

Beginners usually start by working out how to know when you’re looking at a quaver or a crotchet. That is a useful skill, which applies to all ancient poetic rhythms. Here, though, we’re going to focus on the structure of the hexameter.

Some technical terms, in case you decide to look into more formal publications on the subject:

  • hemipes = the technical name for a beat. That is, the hexameter is a twelve-hemipes rhythm. (‘Hemipes’ comes from hemi- ‘half’ + pes ‘foot’, which is obviously missing the point: that’s why I’m using ‘beats’ here.)
  • mora = in metre, the technical name for the time occupied by a short syllable; that is, equivalent to a short note. (You can safely regard a long note as two morae, no matter what M. L. West says.)

No poet ever sat down and thought, ‘Hey, maybe I’ll use 2/4 time with a dum-diddy riff repeated for six measures.’ Rather, it seems what happened is that at some point, early Greek poets took a traditional three-beat unit that was very productive in early Greek music:

— ⏑⏑ —

and expanded it. This rhythm is called a choriamb. The name isn’t essential: I mention it only because it’s the reason that the — ⏑⏑ — × at the end of the line always stays the same. It’s a fossilised choriamb. Some theories are based on the idea that there’s another fossilised choriamb in the third to the fifth beats of the line — but if so, that one is better disguised.

It’s not clear how we got from choriambs to the twelve beat rhythm. The theories that have been proposed all have problems and they’re all incompatible with each other. At the end, below, you’ll find a summary of some of them.

The right concepts (2): prosodic unit, colon, flow

The building block of Homeric poetry is the phrase. Specifically, the prosodic unit, or intonational phrase. A prosodic unit is a chunk of a spoken utterance. It is distinguished from surrounding chunks by syntactical, semantic, tonal, and rhythmic contours.

In Homer, prosodic units combine to become poetry if they have appropriate rhythms. It’s a bit like the phrases you find in modern rap, except rap also uses rhyme: the strongest rhymes are at phrase-end, so it’s a bit more obvious how it’s built out of prosodic units of variegated length and rhythm. Homer’s range of rhythm shapes is more constrained.

People studying Homeric metre regularly refer to a prosodic unit as a colon, a term borrowed from rhythmic units in melic (sung) poetry. It’s been adopted for Homer because Homer’s prosodic units have regular rhythms, but ‘colon’ has a slightly different meaning in melic poetry, so here I’ll carry on using ‘prosodic unit’.

Note. On the sense of ‘colon’ and other constituent rhythmic elements see West 1987: 4–5.

Homeric poetry is built out of prosodic units with well-defined rhythmic shapes. When they’re combined into the twelve beat rhythm we looked at above, you get a hexameter verse. Here are some common combinations, expressed as a number of beats:

5 + 7
5½ + 6½
5 + 2 + 5
3 + 2½ + 6½
3 + 5 + 4
5 + 4 + 3
2 + 3½ + 6½

and so on. Homer gets flow by varying these combinations from one line to the next.

Prosodic units are frequently re-used: these combinations can be built out of identical rhythmical Lego blocks. When this happens, the re-used prosodic unit is called a metrical formula. There’s a huge amount of modern scholarship on metrical formulas and how they’re adapted and varied in different contexts. (Usually without much attention to flow.)

Here are some sample combinations of formulas, using prosodic units with the rhythms 5 + 2 + 5:

5 beats
(— ⏕ — ⏕ —)
2 beats
(⏕ —)
5 beats
(⏕ — ⏑⏑ — ×)
τὴν/τὸν δ’ ἀπαμειβόμενος
ἀγχοῦ δ‘ ἱσταμένη/-ος
τὴν/τὸν δὲ μέγ’ ὀχθήσας
προσέφη γλαυκῶπις Ἀθήνη
λευκώλενος Ἥρη
πολύμητις Ὀδυσσεύς
πόδας ὠκὺς Ἀχιλλεύς
κρατερὸς Διομήδης
ξανθὸς Μενέλαος
answering them
standing near
very angry at them
spoke grey-eyed Athena
white-armed Hera
cunning Odysseus
swift-footed Achilleus
strong Diomedes
brown Menelaos

Mix and match these however you want, and you’ll come up with a perfectly formed hexameter. Here are some combinations using the 5 + 7 rhythm:

5 beats
(— ⏕ — ⏕ —)
7 beats
(⏕ — ⏕ — ⏑⏑ — ×)
ἀγχοῦ δ‘ ἱσταμένη/-ος
καὶ μιν φώνησας
καὶ ῥ’ ὀλοφυρόμενη/-ος
οἴκτρ’ ὀλοφυρόμεναι/-ους
βαίνομεν/ἕσταμεν/κτλ. ἀχνύμενοι
ἔπεα πτερόεντα προσηύδα
θαλερὸν κατὰ δάκρυ χέουσαι/-οντες
standing near
and speaking to them
and then grieving
grieving piteously
we walked/stood/etc. lamenting
they addressed winged words
shedding copious tears

Not all of these combinations work together: the first four phrases in the left column work with either of the phrases on the right, but the last one on the left only works with the last one on the right.

Homer scholars can get very preoccupied with whether a given prosodic unit should be regarded as formulaic or not. That does matter. But it matters much more whether a string of words is a prosodic unit or not. Because every Homeric line is a combination of prosodic units.

A good introduction to the topic of prosodic units and flow is in G. S. Kirk’s introduction to volume 1 of the Cambridge Iliad commentary (Kirk 1985: 17–24). Just note that he uses the term ‘colon’, not ‘prosodic unit’.

Side effects: caesura, bridge

When you’re starting to read Greek poetry and you learn to scan hexameter, you’ll generally be taught to find a caesura — literally a ‘cutting, incision’.

What you’re actually finding is the rhythms of the prosodic units in the line.

A caesura isn’t a pause. It is a word break, but that’s the least interesting thing about it. What matters is the prosodic units on either side.

You know the proverb ‘They can’t see the wood for the trees’? Well, if you’re taught to pay attention to caesuras, that’s like looking for the gaps between the trees. It’s back-to-front.

The characteristic mid-line caesura — the one in the ‘third foot’ — is simply what happens when a line is built out of a 5 + 7 combination or a 5½ + 6½ combination. 5 + 7 produces a ‘penthemimeral’ caesura — so called because it comes after the fifth (pent-) beat — and 5½ + 6½ produces a ‘tritotrochaic’ caesura. Even more confusingly, these are also traditionally known as ‘masculine’ and ‘feminine’ caesuras.

If you find these names strange and hard to learn, that’s partly because it isn’t obvious why you’d want to learn them. They’re names for gaps between the trees. They aren’t the thing that actually matters.

A bridge is a slightly more useful concept. A bridge is a place in the twelve-beat rhythm that’s usually partway through a prosodic unit. A bridge isn’t an arbitrary rule, any more than a caesura is: it’s a side effect of prosodic units. Bridges exist because there are some rhythms that prosodic units just don’t use.

The most important bridge is in the middle of the eighth beat, called ‘Hermann’s Bridge’. Hermann’s Bridge is much stricter than the mid-line caesura. 98% of Homeric lines have a mid-line caesura; somewhere north of 99.93% of lines observe Hermann’s Bridge.

In part 2 we’ll take a dedicated look at Hermann’s Bridge. Partly to see the result of thinking in terms of prosodic units, instead of caesuras; partly because it is by far the strictest feature of Homeric prosody. Part 2 will also include a glossary.

Postlude: proposed theories for the origin of the hexameter

These are theories on the original prosodic units that combined and evolved to produce the hexameter.

  1. Witte 1913. Dactylic tetrameter + adoneus
    • — ⏕ — ⏕ — ⏕ — ⏕ ⫶ — ⏑⏑ — ×
    • pros: accounts for bucolic caesura, Hermann’s Bridge, Wernicke’s Law
    • cons: doesn’t explain caesuras/bridges in the first three feet
  2. Fränkel 1926, Porter 1951. Four cola of variable length
    • A-caesura within first two feet, B-caesura = within third foot, C-caesura in fourth foot or at end of fourth foot
    • pros: versatile; good at incorporating attested colon-shapes
    • cons: too loose to have much explanatory power; doesn’t explain why some colon-shapes don’t appear
  3. West 1973: 187–192. Hemiepes + paroemiac
    • — ⏑⏑ — ⏑⏑ × ⫶ ⏑ — ⏑⏑ — ⏑⏑ — ×
    • explains both forms of mid-line caesura (penthemimeral and tritotrochaic)
    • doesn’t explain bridges or alternative colometries
  4. Nagy 1974: 49–102. Pherecratic with triple expansion
    • × × — ⏑⏑ — ⏑⏑ — ⏑⏑ — ⏑⏑ — × (= expanded form of × × — ⏑⏑ — ×)
    • pros: can explain a range of caesuras/bridges as caused by adaptations from melic cola
    • cons: doesn’t explain why these caesuras/bridges don’t appear internally in analogous melic metres


  • Fränkel, H. 1926. ‘Der kallimachische und der homerische Hexameter.’ Nachrichten der Gesellschaft der Wissenschaften zu Göttingen, philologisch-historische Klasse 1926: 1–33 (197–229).
  • Kirk, G. S. 1985. The Iliad. A commentary, vol. 1. Cambridge.
  • Nagy, G. 1974. Comparative studies in Greek and Indic meter. Cambridge, MA. [CHS]
  • Porter, H. N. 1951. ‘The early Greek hexameter.’ Yale classical studies 12: 3–63.
  • West, M. L. 1973. ‘Greek poetry 2000–700 B.C.’ Classical quarterly 23: 179–192. [JSTOR]
  • Witte, K. 1913. ‘Homerische Sprach‑ und Versgeschichte.’ Glotta 4: 1–21. [JSTOR]

No comments:

Post a Comment