Prosody fascinates me, although it is one of those topics that tends to make students, and others, yawn. Which is fair enough. Still: speaking for myself, at some point I'd like to write something more serious on the subject: for instance to take polite but substantial issue with Derek Attridge's widely-admired The Rhythms of English Poetry (Longman, 1982), which I think is broadly wrong for more-or-less abstruse reasons I won't go into too deeply here.
At any rate, teaching prosody is less fun that it ought to be. English-speakers in my experience generally have a good ear for ictus, or stress, and can pick which syllables in any given line of verse are stressed and which are unstressed. Couple this with the easily taught table of the four most common metrical patterns in stressed verse—the iambic (de DUM de DUM de DUM), the trochaic (DUM de DUM de DUM de), the anapestic (diddy DUM diddy DUM diddy DUM) and the dactylic (DUM diddy DUM diddy DUM diddy) and you've basically got it. There are other metrical patterns, obviously, and you may need the occasional spondee to make your prosodic analysis work out, but that's basically it. Then it's a simple matter tracing the iambic pulse in a Shakespearian line, or contrasting the dactylic
Blow the wind southerly,with the famously anapestic
Blow the wind south o'er the
Bonny blue sea
The Assyrian came down like the wolf on the fold,that used to be Byron's most memorised poem. All well and good. But there's a wrinkle: this later European tradition of scanning verse by stressed and unstressed syllables is fundamentally different to the tradition that obtained in ancient Greek and Latin verse. Their metrical feet are the same: iambs, anapests and so on. But in place of stress, the ancients heard length. Robert Graves described the difference of modern and ancient poetics as being that between the hammer and anvil of the blacksmith—ictus—and the long and short strokes of the boatsman's oar—classical prosody. Maybe that gives you a sense of the distinction.
And his cohorts were gleaming in purple and gold;
And the sheen of their spears was like stars on the sea,
When the blue wave rolls nightly on deep Galilee.
We move into trickier territory. Whether the Ancients also heard stress is a moot point; but their poets, and their grammarians, analysed poetry in terms of the pattern of long and short syllables, not in terms of patterns of ictus. And that's a distinction we can understand too: we hear the difference between the long e (Greek η) in feel and the short e (Greek ε) in fell; between the omega (ω) in dole and the omicron (ο) in doll. But the fact that we can distinguish those sounds in individual words doesn't mean we can all hear the complex patterns of long and short syllables in passages of Homer or Vergil.
I'm going to make a rather shaming confession: I studied this stuff as an undergraduate doing a Classics degree, and went into it in greater detail for my PhD (where it was kind of important) and have kept it up; but I'm still not sure I can properly ‘hear’ the dactyls in Homer or Vergil as I read them aloud; not in the way I can ‘hear’ the stress patterns in English verse. I can see those patterns in Homer and Vergil, when the verse is written down, and can analyse it and so on; but it may be that ictus just strikes me as intuitive and common-sense in a way patterns of long and short vowels don't.
Anyhow, this long preamble is by way of setting up my larger point. For most of its history, the great English poets were trained up on the classics, and taught a prosody of length, before going on actually to write a poetry governed by a prosody of stress. Most of those poets simply wrote metrically in terms of stressed/unstressed syllables, as governed by the particulars of English, and ignored their book learning on how Homer and Horace wrote their verse.
Occasionally a poet of rare ambition would try to write both schemes at once: to write verse in English where the careful patterning of stressed and unstressed syllables matches, or plays creatively against, an equally careful patterning of long and short syllables. But this is very hard to do, and few poets have managed it. I'm prepared to believe that Milton achieves it, though I haven't really looked into his verse on these terms. I know Coleridge did it from time to time, as a particular exercise, but the results (mostly Notebook entries) are a little forced. The one English poet who possessed the needful expertise, whose ear was finely-enough attuned, and whose facility remarkable enough, to write easily and at length in a way that combined both traditions is, I think, Swinburne. And if I had enough time I'd expand upon that point, though it would take a while. Suffice to say I think Swinburne is a much much more interesting poet prosodically than he is on the level of content. Or, indeed, than he has been given credit for.
Now: the reason why I think this dual tradition comes again to have purchase on English poetry is because the main mode of 'poetry' nowadays is music-plus-lyrics. Most pop song lyrics are trash, of course, and many are remarkably clumsy; but the best pops song lyrics work by matching the rhythms of words to the rhythms of music. Harder to do than you might think.
Now as far as that goes, music combines stress and length patternings in a much more organic way than is the case for purely verbal forms of art. Some music (especially pop music, driven as it almost always is by drums) is heavily stressed, rhythmically speaking; other music (Wagner's chromatism for instance) is much more rhythmically defined in terms of pulses of longs and shorts. And for me the ideal song is one where the lyrics match the music both for stress and for length.
It's surprisingly rare. Often pop-sing lyrics will be regularised to match the musical beat in terms of ictus, but will be all over the place in terms of long and short strokes. Then again, quite a few pop lyrics do a crappy job of matching the natural rhythms of words to the rhythms of the music to which the words are attached. And since the whole point of this post is to praise the songwriting of Randy Newman, it's worth noting that sometimes even his best songs wrongfoot their ictus. For instance: ‘Simon Smith and the Amazing Dancing Bear’ is a lovely little song; but its opening two lines
I may go out tomorrow if I can borrow a coat to weartrip up jarringly on ‘sincere’, which needs to be sinCERE but which can only fit the rhythm of the song by being sung SINcere. Ugh.
Oh I'd step out in style with my sincere smile and my dancing bear
Ah, but let us turn to ‘Louisiana 1927’. There it is at the top of this post. It's my favourite Newman song (which is saying a lot) and one of the reasons I love it is that, so far as I can see, it is quite simply flawless in the way it (intuitively, I guess) combines both prosodic paradigms, the modern by stress and the ancient by length, both working together in a way that precisely fits the (beautiful) rhythmic and melodic pattern of the music. It's a kind of perfection, and that's very rare in art. I'm honestly not sure I can think of a better line in modern pop music, prosodically speaking, than Six feet of water in the streets of Evangeline.