Prosody as it Really Is
A paper given at the Linguistic Circle of Manitoba and
North Dakota, 1980's
Thank you for being here. There are not many of us, but to tell you the
truth, I am surprised that a talk about metrics will turn out anyone at
all. In fact, I often think of myself as the last of the scanners, the
last of that dying breed who, when confronted by poem, instinctively
begins by scanning to see about its meter and who can spend half a hour
with a poem without getting around to asking what it's about. I
offer no explanation other than to say I can't help myself.
Students responses to metrical analysis range from puzzled to
indifferent or hostile a great many poets feel that the same way
about metrical concerns as Katherine Hepburn character feels about
Human nature when she explains to Humphry Bogart that "human nature,
Mr. Oldnose, is what we were put on earth to overcome." To hear some
poets talking about meter, if Homer and Virgil and Dante and
Milton and Shakespeare and Frost hadn't been distracted by
counting syllables, they might have written something really good.
But my own interests stem not from the metrical nature of our greatest
poets, nor even from that fact that most poetry in English,
historically speaking, is in regular meters. No, I think my interest is
quirkier or more mystic than that, and stems from four sources. One is
the frustration I have felt in teaching poetry, where I still feel some
obligation to at least mention the iamb, and my frustration, not with
the students' glazed eyes, but in my own inability to give a clear
account of the matter.
A second is my interest in the metrical issues that arise in my
own writing of verse, in which I take great pleasure, but which I
distinguish from poetry as I distinguish what I do in an exercise room
from what Barishnakov does on a stage. Exercise is good for me, but I
don't expect anyone to pay to watch me do it.
A final source of my addiction is simply the tone of writers on
metrics. There are few things more downright fun than listening to the
tone of wounded outrage one prosodist uses to discuss the work of
another. My favorite metricist, who is, believe it or not, Valdimir
Nabokov, to take but one example, says that he has "not yet come across
a single work that treated the English iambic--particularly the
tetrameter--on a Taxonomical and comparative literature basis, in a way
even remotely acceptable to a student of prosody" (3). And in
criticizing Herbert Grierson's scansion of a line by Donne ("Both years
and days deep midnight is," from a "Nocturnall upon S. Lucies Day")
Harvey Gross (in Sound and Form in Modern Poetry, 1965) throws up his
hands in horror and declares that "we wonder if anyone understands the
nature of English syllable-stress metric" (6).
Fourth in my list of reasons I care about metrics is my own
thoroughly unsubstantiated and probably unsubstantiable certainty that
the central matter of art is not content--whatever that is--but form,
whatever that is.
I will have more--but not much more--to say about that anon, but
for now, let me turn to a brief survey of the literature of scansion,
all of which I hasten to say, I have not read. At one pole we may put a
wonderful observation by Herbert Read. (Certainly as catholic and
sensitive reader as one could name) in The True Voice of Feeling (NY;
Pantheon: 1953) that "blank verse is virtually free verse" (47), a
statement he expands a few pages later where he says that the
apparently random fluctuations of a needle connected to the microphone
of a person reading poetry aloud are "the only accurate scanning that
can be made of English blank verse--perhaps the only accurate scanning
of any kind of verse" ( 50).
From that amiable shrug of the metrical shoulders we pass at the other
extreme to books which proceed with utter certainty and invigorating
confidence to scan dozens of lines of verse, labeling them with those
jaw-jarring names that begin with the iamb and end somewhere on the
other side of cloud-cuckoo land with things like the amphribrach and
amphimacer as in Clement Wood's good-hearted Poet's Handbook
(Cleveland: World Publishing: 1946) or with things like tilted scuds
and false pryyhics in Nabokov's tortured but I think brilliant
analysis.
In between, we have god's plenty though decreasingly so as we
move into free verse. Historically, studies of metric abounded in
the Renaissance when English poets like Wyatt and Tottel and Surrey and
Gascoigne and Barnaby Googe were trying to absorb classical and
continental metrical systems into English verse, a story summarized in
an elegant little book of John Thompson, called The founding of English
Meter. NY: Columbia: 1961). Mr. Thompson does not raise a question I
sometimes raise with my students: Could English have passed
directly to free verse instead of sending us into this 400 year
interlude which produced Shakespeare and Milton and Keats and Browning
and Yeats and Wallace Stevens).
Ruskin wrote a book on prosody in 1880, at about the same time
that Sidney Lanier wrote one in America. Much of 19th century
scholarship is enfolded into Saintsbury's multi-volume History of
English Prosody, published in 1908, remarkable both for the exhaustive
nature of its coverage, but also for the fact that in all its volumes
you can hardly find an actual line actually scanned except in the
footnotes.
Between the wars, the name of T.S. Omund's seems to be the most
respected among prosodists. Robert Bridges's seminal study, Milton's
Prosody appeared in1921. Karl Shapiro's books are readable and short
besides, partly, no doubt, because he was a poet himself. Among
the studies written since the advent of free verse, Paul Fussell
Poetric Meter and Poetic Form (NY: Random House: 1965) and Harvey
Gross's Sound and Form in Modern Poetry (Ann Arbor: U. of Mich: 1965)
are the best known, at least by me.
Though scansion parties are not popular, metrical study is far from
dead. The Princeton Encyclopedia of Poetry and Poetics (which often
seems to me is not as good as it ought to be, given its name) is
out in a new and more formidible edition. Terry Brogan's 1981 study
English Versification 1570-1980 lists the names of more than
7,680 scholars, many of them the author of more than one study.
That study was published by Johns Hopkins which has become the center
for metrical studies. When I searched the on-line version of MLA, which
goes back to 1963, I found 4,000 items under the term "prosody,"
about 25 percent of them in languages other than English. In 1994,
there were over 200.
Some method of classifiying such studies is clearly in order. A useful
classification of metrical studies is given by Dereck Attridge study
The Rhythms of English Poetry (NY: Longman: ____) There we find a
useful distinction between accentual and temporal analytic schemes, ,
that is, between systems which pay attention only to the "force"
of a syllable (presumably its loudness--though even this might be
problematic) of a syllable--usually using a binary system--and those
that identify whether a syllable takes a longer or shorter time to say,
usually using some form of musical notes to convey this information.
Of such temporal studies Nabokov has said, with the charming diffidence
so characteristic of prosodists: "I have of course slammed shout
without further ado any such works on English prosody in which I
glimpsed a crop of musical notes or those ridiculous examples of
strophic arrangements which have nothing to do with English poetry"
(4). The handbooks tell us with great confidence that English poetry is
quantitative, not qualitative like the meters of Homer and Virgil which
counted long and short syllables, not beats and the absence of
beats. Yet we know that it cannot be so easy and that lines like
Eliot's " Stone bronze, stone steel, stone oakleaves, horses
heels," will scan quantitatively as nine consecutive beats
which it would be nonsense to label, but which yield very nicely to a
temporal analysis I show on the overhead.
Still, we must, in our finite time, turn aside from temporal studies
and concentrate on those which study syllabic emphasis. Even here
we must use crude tools to hack through the thicket, and I have mine
which is the distinction between metrical and rhythmical emphasis. That
is, in any line of verse written before the modernist revolution, one
can identify a two patterns of emphasis, one given by the meter and one
given by the "sense" that is, the natural reading rhythm.
When I encounter a discussion which does not acknowledge this simple
fact, I exercise the prerogative of prosodists: I without further ado,
I slam the book shut and move on. Even our best handbooks often
fail us here. Hibbard and Holman draw back from any firm commitment to
whether this distinction is useful and if so, which should be called
stress and which accent. I will at least be
consistent and refer to metrical stress and rhythmical accent, or when
I can get away with it without confusing you, to stress and accent
I will not say at the risk of becoming tedious, but with the certainty
of becoming tedious, let me explain by offering a line of verse:
If music be the food of love, play on.
I assume most of us would scan it as I have it marked on the handout,
as iambic tetramter with emphasis on the MUS, BE, FOOD, ON. But
of course no one would read the line that way. We would probably
read it with emphasis on the syllables MUS, FOOD, and PLAY. If
MUSic be the food of LOVE, play ON. (I would be willing to be accept
another rhythmical accent on LOVE. The point is not worth quibbling
about. ) In Shakespeare's line, we could "get to" the two schemes by
writing out the line twice, capitalizing the different set of syllables
each time:
if MUSic BE the FOOD of LOVE, play ON (metrical
stress)
if MUSic be the food of love, PLAY ON (rhythmical
accent)
But writing out every line twice is awkward. Let us look at simpler
schemes which focus on this distinction between stress and
accent. Only two prosodists I know of, out of 7,680 (though, to
be honest, I have not read them all) have confronted this problem
directly. One is George Stewart in widely cited little book
called The Technique of English Verse (NY; Kennikat: 1930). The other
is Nabokov in Notes on Prosody (Princeton: Bollingen: 1964).
Both of these men observe that once you acknowledge two sets of
emphases, then one of four things must be the case for any given
syllable: 1) It may take both metrical stress and rhythmical accent, or
2) it may take neither, or 3) it make take metrical stress but not
rhythmical accent, or 4) it may take rhythmical accent and not metrical
stress. Stewart and Nabokov have different ways of indicating
each of these four possibilities, which I summarize on the handout.
Of the three, I submit that mine is the simplest and the most
teachable. It calls for us simply to mark the metrical stress in the
conventional way with an acute accent mark (which Nabokov calls an
ictus) and a breve above he line, and then to mark the syllables
which receive rhythmical accent below the line. When I can write the
line out with plenty of space, I like to use the ictus under the line
as well. When I scan lines on the printed page, I use the
underline. Syllables on which the meter and rhythmical stress are
"in conflict" will then immediately jump out and can be marked with an
arrow pointing up under the syllable.
If the meter "wins," I label the arrow with an M; if the Rhythm
"wins," I label the arrow with an R. On the handout I have scanned
Shakespeare's line using first Stewart's system, then Nabokov's,
then my own. (handout) My scheme calls attention to the doubled
accent at the end of the line: play on, which other schemes might try
to call a spondee, but which would fail to acknowledge that the two
emphases come from different sources. My scheme also accounts for the
way the line rushes over the stressed but unaccented "be" in the fourth
syllable in its hurry to get to the fanfare of the last two syllables.
But Shakespeare is too easy, you say. Well, yes. So let us look at line
of Marvell:
"Vaster than empires and more slow," which Nabokov scans with a
trochaic "substitution" in the first foot as indicated on the
handout. Well, yes, but that scansion overlooks
the other more interesting feature of the line, which is the doubled
emphasis at the end: More slow. If we do acknowledge that, we shall
have to say that a spondee has been "substituted" for the last iamb as
well as the trochaic for the first one. Frankly the whole business of
substitutions seems to me simply a morass. So far as I can tell--and I
stand open to correction--most schemes which talk about substitutions
scan lines with some vague mixture of attention to the accent and
stress and then try to impose some kind of ad hoc labeling on it,
moving the foot divisions around to get feet they can pronounce the
names of. When I see these scansions, I want to ask with Mr.
Gross or at least to ask if anyone understands the nature of English
syllablic stress metric.
My scansion is indicated on the handout. I get it by forcing (violently
I admit) the line to fit the metrical stress, with emphases on ER, EM,
AND, and SLOW. The rhythmical stresses are on VAS, EM, MORE, and
SLOW: VASTer than EMpires and MORE SLOW. Conflicts abound, on the
first, second, sixth and seventh syllables, with the rhythmical accent
and the metrical stress each winning twice. The first set of
conflicts, it could be argued, calls attention to the Vastness of the
empires in question, and in the second set, the stressed but unaccented
AND creates a rush which "sets up" the final doubled emphasis with its
assonant "O" sounds in "more slow." But do not get me started on
assonance.
We have been working with tetrameter. How does it work on pentameter?
Well, not bad. I have scanned many, many lines from Shakespeare's
sonnets, and while I have not examined the considerable
literature on Shakespeare's metrics, I think I can account for much of
what is interesting in those poems. Take for example, the first line of
sonnet #118. That line reads:
Let me not to the marriage of true minds
Forcing the metrical stresses to the surface produces something I grant
is bizarre:
let ME not TO the MARriage OF true MINDS.
While the rhythmical accents seem to fall thus:
let me NOT to the MARriage of TRUE MINDS--five stresses, four accents
with irregularities at syllables number 2, 3, 4, 8 and 9, the large
number of irregularities pointing up the particular ironies in the
line: a mind sensing impediments all over the place while saying that
it only wants harmony.
Frost, as you might expect, yields well to this kind of
treatment. The first line of Mending Wall goes: "Something there is
that doesn't love a wall," which scans as I have it on the handout,
with irregularities at syllables 1, 2 and 4. The substitution of a
trochaic in the first foot is something any scheme would have picked
up, but this scheme also picks up the conflict between meter and
rhythm in syllable #4--it takes stress, that's clear. Does it take
accent as well? The very fact that we ask the question calls attention
to the word reversal, and thus to the formality of the poem's opening
line, in a poem about formality.
But you are all thinking of lines this scheme won't work on. April is
the cruelest month. Well, I admit I can't do much with
anything that is not really accentual-syllabic, Beowulf for example or
Eliot's Four Quartets or Coleridge's Christabel, which are in some
verson of accentual or strong-stress verse, not accentual
syllabic. Pentameter lines that don't have ten syllables for
example are troubling. Shakespeare's sonnet 144 has a line "The
worser spirit a woman colored ill." Unless you elide IT A between
the third and fourth words, that line has 11 syllables. I deal with it
by adding a third kind of "irregular" symbol--X or extra syllable, but
I don't know how far I can push that. I have not done any
work with Hopkins, or with Dickinson. Stay tuned. Nor have I done
anything with the so-called triple rhythms: dactylic and anapestic. So
many lines; so little time.
About free verse, I would prefer to reserve judgment. More work with
the rougher rhythms of metrical verse, in, say Donne, might provide
some insights into free verse. Eliot said that no verse is really free
for a poets who want to do their job, and Pound said he composed
in the sequence of the musical phrase, which raises lots of questions,
since even musical notation does not always indicate the nature of the
musical phrase, since such phrases often run across the bar divisions.
But such statements surely point us in the direction of temporal
analytic schemes with their musical notations. It also seems clear to
me that while free verse may mean a poem has no overall metrical
scheme, line by line, verse, whether free or not, cannot help but set
up tensions in which the counterpoint between rhythm and meter is
relevant to the poem's effect. The first two lines of The Wasteland
can, in fact, be scanned as iambic pentameter (APril IS the
CRUElest MONTH, breedING) and the next five as tetrameter, and doing so
might tell us something. It might even be that the next book I read or
the next line I scan or the question you ask after I finish--may
convince me that all I have done is utterly wrongheaded and I must
start again from scratch. In short, all I am offering you here is a
tentative and clumsy baby step and we all have miles to go--and books
to read--before we sleep.
We could go on scanning lines for hours, but it's time to play you bet
your accent. More accurately, it is time to turn, in conclusion, from
individual lines and syllables to a bit of speculation which metrical
analysis raises: to on larger matters of form in art Let me close
then, with eight large and fuzzy statements, offered to clear your
palate after asking you to look so closely at some tedious matters
of meter and rhythm.
1) To answer the question in my own title, No, Prosody does not
exist in the way that, say Chemistry exists--as an agreed-upon method
for investigating a body of material, producing replicable results.
Even prosodists who focus on old-fashioned binary analysis don't even
agree on mechanics, some of them borrowing from classical prosody the
horizontal dash (called the macron) and using it to indicate a light
syllable, some using the short curved line, generally called the breve.
2) In the nature of things, prosody probably never will exist in the
way Chemistry exists, nor should it, since lines of poetry are atoms
but something else, though what else I am not prepared to say.
3) Still, it seems to me we can do better than we do. Schemes which do
not somehow distinguish between rhythm and meter seem unnecessarily
confusing and muddled.
4) So long as poetry consists of lines, we are going to have to
account for those lines. And the line seems to be the one irreducible
fact of poetry. Milton justifying the ways of God to men and a
semiliterate but broken-hearted lover pouring out his heart in a dorm
room agree on one thing: In a poem, the lines do not come out even down
the right-hand edge.
5) I am not sure that current emphasis on free verse and the
de-emphasis on metrics is doing poetry any great service. It sometimes
seems to me that the most profound and moving poetry of our time as
well as the most popular is found in the song lyrics of Paul Simon and
Bob Dylan and Leonard Cohen. And I cannot help but wonder if some of
their power comes from the fact that because they set their poems
to music have to be aware of meter in ways not required of
those who write from the heart, breaking the lines where it strikes
their fancy.
6) We need to remind ourselves of an obvious truth: that when we
scan a line with whatever method, we are not reversing a process the
poet went through in composing the line. Obviously, poets do not stop
to consciously embed in their lines the relationship between
counterpoint and meaning that our anal-retentive analytic schemes might
uncover. When we read, we uncover things that were never covered over
in the first place. Good poets composing good poetry are doing
the same thing the rest of us are as we write--thinking about what they
want to say, trying to find the words that will, as Stevesn says,
"suffice." A poet is trying to say something, but that something has
has more "behind it" that simple prose: propositional truth,
metaphorical associations, emotional states, rhythmical considerations,
and the literary tradition.
7) We obviously have to be very careful in how we introduce matters of
meter to beginning writers and beginning readers. It is right
that we recognize that literature has a spirit and a letter and that
the letter killeth, but the spirit giveth light. But it may be that the
romantic movements of 60's education have led us to distinguish more
sharply between letter and spirit than we really can. There is a
place for literary jargon and metrical analysis, though only sensitive
teaching can say where it is for particular students.
8) It may be, in fact, that letter and spirit come together and even
generate one another in truly mysterious ways. In more cases than
we think, it may be that attention to matters of form (matters that
beginners find most tedious) is just what, under the right
conditions, produces the Sacred Trance in which emotions are
rendered most precisely and truths revealed and the world made
new. Many artists have noticed that when they are
pouring out their hearts we are least likely to succeed in doing so,
whereas when they are trying to solve some technical problem, they are
most likely to be tricked into telling the real truth of their hearts,
the truth they could not even know until they stopped thinking about it
and starting thinking about some mechanical matter like rhyme or
syllable count.
This is a consistent theme in the notebooks of Theordore Roethke,
which I read at the University of Washington. One version of it
copied out a passage from Pater, where Pater says that "the perfection
of poetry seems to depend in part on a certain suppression of mere
subject so that meaning reaches us through ways not distinctly
traceable by the understanding." Now a person who talks about the
"perfection of poetry" and about "mere subjects" is to be taken with
grain of salt, surely. But I think Pater's statement points us in a
fruitful direction. Prosody need not be the part of the course where we
say, Now this is really dull stuff but you have to know it. It can be
one path into the heart of the mystery of art. Prosody is dead. Long
live prosody.