I am a beginner poet, and feel I have gotten rid of my awful tin ear only recently. That is: I feel capable of writing regular, metric poetry.

But regular can become relentless, even when the regularity is Steele-style, like a line that would scan 2-4-1-2-3-4-2-4-1-3, which some poets would scan as x/ xx // x/ x/ - as a line with pyrrhic and spondaic substitution.

So I wonder when other substitutions may be permissible, or when substitution with pyrrhics and spondees whose stress curves are 2-1 or 4-3 are permissible in iambic verse.

Trochaic substitution, for instance: I see it used for emphasis sometimes. But do you need to emphasize to make a trochaic substitution? And will trochaic substitution always work well when you want to emphasise something?

I hope to develop "ear" for substitution in time, but think rules-of-thumb will be a good help not only while I work on developing that ear, but also as a means of developing that same ear.


------------------

Svein Olav

.. another life

Eeeeeeeeeeeooooooooouuuuuuuuuyyyyyyyooooooooouuuuu uuu!!!

A three-page lecture about the importance of writing what sounds good to you is about to pop out of me!

On the other hand, if you admit to having a tin ear, perhaps you do need to write by the numbers. I don't usually write by the metrical rules, but let me see if I understand them well enough to list them:

(These rules are for iambic verse.)

1. Most variations are best placed towards the beginning of the line, especially trochees.

2. If you employ a variation in one line, best to do it in the others also, as in Frost's occasional practice of throwing one or two anapests into each line. (Being consistent with anapests is more important than being consistent with trochees, as anapests lengthen the line.)

3. Best not to have more than one anapest in a row; trochees, however, can be doubled-up if you're feeling feisty.

4. A common substitution for two iambs is the pyrrhic/spondee: x x / X X /

5. Be cautious when mixing two different kinds of variations in one line, such as both a trochee and an anapest.

6. Beginning-of-line variations should have some consistency -- e.g., don't have a headless iamb starting one line and an anapest starting the next (I'm guilty of that).

7. Be damn sure to have 5 beats per line, or you'll be executed by the Metrical Police! (A famous formalist poet, who shall remain nameless, once told me that you could have as few as 3 stresses and as many as 7 stresses per pentameter line, but in actual practice he never has more nor less than 5.)

8. Well-placed caesurae (periods or commas within the lines) will do a lot to prevent monotony.

9. Enjambments should occur at logical places in the line -- e.g., at the beginnings or ends of phrases. Startling enjambments ("I escaped my / mother's womb") are best left to the free-versers.

Alan, Tim, what have I left out?

[Damn, there I go again, answering questions for the Lariat! I keep forgetting what board I'm on.]

Caleb

P.S. It really works! I said something a little too irreverent (read: nasty), and I said to myself, Would Rhina approve of that? and then I immediately deleted it! Rhina's presence may have solved the board's Caleb problem!


[This message has been edited by Caleb Murdock (edited August 16, 2001).]

I would modify a few points in Caleb's list.

First a caveat. While we can describe meter objectively (most of the time), metrical preferences are subjective and endlessly debatable.

Stricter metrists will not admit consecutive substitutions in IP. Auden formulated this concept as a rule. I prefer to think of it as a general principle.

A caesura is not necessarily punctuated. It is a pause dictated by phrasing. Phrases in English naturally run in units of roughly four to six syllables.

The success or failure of metrical variation is strongly dependent on the larger metrical context. For example, an IP sonnet with half a dozen strategically-scattered anapests and only one or two other substitutions would have a distinct rhythm, different from a more regular IP, and also different from an irregular IP where most substitutions were trochees.

For this reason, I would not necessarily object to Caleb's example of a headless line followed by an initial anapest. It would depend on the context.

I would, however, almost always object to consecutive headless lines in IP. Variation should be, well, varied. This is another general principle for metrical practice.

Perhaps I should elaborate a bit on this last point. When I speak of variation, I am referring to the placement of substitutions, not their type.

Absent compelling rhetorical need, it is a bad idea to employ identical meter in consecutive lines. A strictly regular IP, if written well, uses secondary stress to break its lockstep, relaxing some strong stresses and raising weak ones to create the feeling of substitution without actually employing other feet. Such subtleties are difficult to analyse unless one admits three or four degrees of stress in scansion. This is why I like Steele's system.

A.S

I've read these points with interest and wish I had time to test each one. I'm sure I could find exceptions, just because all generalizations are false, including this one. Alan clarified some of Caleb's points, and I'd like to clarify a couple more.

Rule #7 about 5 beats per line refers to pentameter lines only. The Metrical Police don't require you to write in pentameter in the first place. I write in tetrameter most of the time. Even sonnet variations are branching out to other meters. But if you ARE writing in pentameter, make damn sure you have five beats to the line. I prefer to use the term "beats" to denote metrical stress and "stresses" to denote vocal stress, in order to avoid such confusion over how many stresses is too many. Thus you might have more or fewer than five vocal stresses in a five-beat line, employing relative promotion to keep the beat.

On rule #2, some lines without variation are desirable for setting up a pattern. Variations should be, as Alan says, variations, and too many may undermine the pattern or rhythm. But I don't think I agree with the 20% (or any other %) rule of thumb for substitutions that I've heard some people cite, or that you can't mix them skillfully, or that they should necessarily occur near the beginning of a line.

In rule #9 about enjambments, they aren't true enjambments if they occur at the end of the phrase or thought. An enjambment occurs when the phrase or thought carries over into the next line. What Caleb says about awkward line breaks is quite true, though. There should be some logic used, even moreso in free verse, which doesn't have rhyme to tell the reader where the line is supposed to end. I've heard some of the more radical writers (Superfluous Poets?) say they put line breaks in odd places to "startle" the reader or "short-circuit his brain" or "throw a monkey wrench into his mental processes," but I don't read their kind of stuff if I can help it.

Solan, one way to develop your ear is by reading lots of poetry. Another is by having native speakers of English (not necessarily other poets, but good readers) read your work cold and point out any places where they stumble. Then try to analyze why they had the problem, working back from the effect to the cause, and making your own cautious generalizations.

Thank you, Rhina, wherever you are!


Carol

Read, yes, and declaim as well.

Like Carol, I write more tetrameter than IP, and a lot of trimeter too. I find longer measures unwieldy and never use them, though I work sometimes with extended tetrameter that might run to eleven or twelve syllables.

Solan's questions about 2,1 pyrrhics or 4,3 spondees demonstrate the difficulty that Steele's system can cause for beginners. Assignment of stress is a tricky business, and use of the word "permissible" troubles me. It is wiser, I think, to speak of what sounds awkward and why, using specific examples, and bearing in mind that awkwardness can sometimes be a virtue.

A.S.

Alan has made many very good points. Steele's system, for one thing, is useful to scholars, and to analysis of poetry, but muddies the waters for novices, and is far too complex to think about consciously when composing. (Also, it is rather arbitrary, I think, in assigning 4 gradations of stresses.) Personally, I dread to see people just learning to write metrical verse using such terms as phyrrics (and amphibrachs and paeons and minor ionics). Eeek! It is best, I think, when starting off to think of number of stresses or beats, rather than fancy Greek misnomers (and to count off on your fingers, for instance, rather than marking it on paper). Meter should be visceral in its composition, not intellectual, though of course it is open to intellectual analysis and interpretation.

Above all, rules are gleaned from what works in great poetry. NOT the other way around. Go with your ear. But make sure that ear is well-schooled through lots of reading. Anything is permissible IF you can carry it off. Poets should learn "prosody" from poetry, not from books on prosody (though books may be helpful used in addition).

(Having said all that, a tried and true subsitution, where you cannot go wrong, and one I so WISH I saw more of on the metrical boards is the trochaic inversion in the first foot of ip -- or you could say it sounds like a dactyl followed by trochees. It is elegant and very, very koscher. Take a look in Shakespeare's sonnets. They are all over the place!)

The advice to read is always a good one. But the problem with good meter is that it is almost always unobtrusive in reading. I needed to have it pointed out to me to even see that there was such a thing as meter. (Same went for rhythm in dancing, actually.) Most of the time, I notice the meter only when it jingles. Declaiming it might work better, though it honestly feels quite awkward to go declaiming to oneself. But maybe that's what "ear" is all about: Sound?

---

Svein Olav

.. another life

[This message has been edited by Solan (edited August 16, 2001).]

Yes, sound. A non-technical book on it, Robert Pinsky's The Sounds of Poetry (FSG, 1998), is a masterpiece on the topic. But his first advice, as said above, is read, read, read. And, yes, yes, yes, read aloud, honoring the syntax not the lineation.

------------------
Ralph



[This message has been edited by RCL (edited August 16, 2001).]

... or in my case, maybe: Declaim, declaim, declaim.

But while learning by rote, repetition and quantity has its virtues, I think many are too focused on that as a method. You can look at the stars for ages without discovering anything about them, whereas a two-day course in astronomy can teach you a lot. If I were to choose, I'd do the astronomy course first and gaze at the stars later. They look so much better when you know what you are looking at. Poetry is like star-gazing: Beautiful even when you know nothing, but gives you so much more when you know what you are looking at.

---

Svein Olav

.. another life

[This message has been edited by Solan (edited August 17, 2001).]

I think I'll be covering most of these matters
on the Scansion thread---too confusing to have
the same thing coming up in different places.
But: Solan, you can have a subsitution anywhere
in the line, even a trochee in the 5th foot. It all
depends, as Alicia says, on whether you can pull it
off. You want to get to the point where you're
adjusting your rhythms entirely by ear and not thinking
about "emphasizing points" with trochees or whatever.
One of the continual mistakes in Perrine and Fussell
(and others) is the peculiar notion that a subsitution
always "means" something; it's nonsense. For example,
beginning an iambic pentameter with a trochee is so
common as not to be felt as a subsitution at all; and
that trochee almost always "means" or "emphasizes"
exactly nothing.
I don't think a sequence that Tim Steele would mark as
1-2-3-4 would ever be xx// (if I understand Solan's
notation). Or almost never. Over on Scansion, I'm
going to try to teach another method of notation, one
that I think works much better, especially for students
and beginners.
I think most of Caleb's rules are generally wrong, or
require so much qualification as not to need saying.
Enough. Let's ride on over to Scansion.