What’s this essay about?
In the course of writing another post, I stumbled across Dan Schneider’s essay Robinson Jeffers, & The Metric Fallacy. The essay itself was so stuffed with ludicrous and fallacious arguments that I can’t help responding.
Ostensibly, Schneider’s essay is a defense of Robinson Jeffers, but for the first 1867 words (counted by WordPerfect) Schneider only mentions Jeffers’ name twice, and this is in the last 75 words. So Schneider’s essay is really two essays, the first being nothing less than a diatribe. And even after reading the essay as a whole, one isn’t quite sure how the diatribe is a defense of Jeffers. But it’s the first part of the essay I’ll be responding to.
Here’s how Schneider begins his essay:
What if someone actually said to you that all music was composed of just 2 notes? Or if someone claimed that there were just 2 colors in creation? Now, ponder if such a thing were true. Imagine the clunkiness & mechanicality of such music. Think of the visual arts devoid of not just color, but sepia tones, & even shades of gray.
Schneider immediately frames his argument as a choice between absurdity (metrists) and reason (Schneider). How absurd to think that music would be composed of two notes? Wouldn’t it be a fool’s errand to disagree with him? Schneider would like you to think so. This is called framing an argument. Politicians do it all the time. The Republicans are especially good at it. Who could argue with a Clear Skies Initiative or who could argue with the No Child Left Behind act?
But what metrist has ever asserted that meter is composed of just two discrete stresses and that, furthermore, these two stresses are precisely the same no matter the context? None, unless there’s some fusty nineteenth century pedant I’ve never heard of. Even the fustiest recognized that the markings used in scanning poetry were a relative indication – symbols and nothing more.
But let’s examine how Schneider frames his diatribe by altering a couple of words:
What if someone actually said to you that all music was composed [of just 2 beats]?
Well, on the face of it, why not? Any piece of music in 2/4 time is written with 2 beats – a downbeat and an off-beat. Within that binary ground there is an infinite variety of shading. But, when it’s time to dance, everybody knows where the downbeat is. Period. If this weren’t the case, then there would be no such thing as dancing. What if Schneider said the following: No two beats of music are the same (no two are played with the same volume or emphasis), even in 2/4 time, therefore the whole concept of a time signature is absurd. To quote Schneider, to suggest such things to a musician or painter or photographer would most likely engender- if not outright laughter- some strange looks, indeed.
But this is precisely how he defines meter. He writes:
For the uninitiated meter is the theory (claiming origin by several cultures) that spoken language consists of 2 primary vocalizations of a sound- i.e.- stressed & unstressed.
First, Schneider’s definition of meter is wrong. Period. Latin meter does not consist of stressed or unstressed “sounds”. Neither does meter in Chinese or Greek. Meter, to the initiated, is the means by which poets organize their respective languages into predictable patterns. But anyway, the absurdity of Schneider’s assertion makes for an easy straw man . He then states, as though he were the first one to realize it, that “in the context of spoken words, as well as those internal voicings, an absolute plenitude of stress levels ensnares one.”
Yes, and there are an absolute plenitude of stress levels when performing rhythm in music. This doesn’t mean that a time signature doesn’t exist. Meter is nothing more than a kind of time signature. Any Iambic meter could be thought of as a 2/4, 4/4, etc. time signature. It is a binary, Iambic, time signature. A Dactylic meter could be treated as having a 3/4, 6/8, 9/4, etc. time signature.
Think of meter as a time signature and you will avoid the same cognitive trap that Schneider falls into.
In fact the dualistic notion of mere stressed & unstressed sounds is- in practice by its many proponents- almost always so loose as to be meaningless anyway, as metrics should really redefine its definitions as greater & lower stress(es) (with a plenum of in-betweens), since (obviously) a truly unstressed syllable would be silent. But even that is far too inadequate, for even if you would read this essay aloud to this point- you would, if to tape it on any recorder, really hear at least a dozen stress levels- if not several dozen, were your ear fine-tuned enough.
Having set up and burned down his straw man, Schneider then engages in nothing less than reductio ad absurdum.His argument? Since stressed and unstressed syllables can be read with almost infinite gradation, there is no such thing as stressed and unstressed. But this argument is a red herring. No one denies that stress is subject to infinite variation. No metrist is arguing for fixed stress, but only relative stress. Try auditioning for Hamlet speaking your lines like this:
To be or not to be: that is the question.
You would get some strange looks from all those musicians, artists and photographers. Or try declaiming the line with no stress at all.
But if you read the line the way Shakespeare meant it:
To be or not to be, that is the question
It makes sense. The symbols don’t mean that each stress is of a fixed pitch. No, this line could be spoken in an infinite variety of ways – but in all those ways, the syllables would always be stressed and unstressed in relation to each other. This is the way the English language works. What if the actor wanted to emphasize that instead of is in the fourth foot? Then he would speak it this way:
To be or not to be, |that is| the question
If Schneider is to be believed, such a feat wouldn’t be possible. But it is. And it is because English speakers use relative stress (strongly stressing one syllable while weakly stressing another) to indicate focus.
In English, all syllables are relatively stressed or unstressed in relation to each other. Setting aside poetry and metrists, there is a whole field called Stress Linguistics dedicated to the very real and serious understanding of this facet of language. Maybe Schneider would like to argue that Stress Linguistics is a sham? Try teaching a computer how to speak English. The study of stress in language allows the computer to corrrectly identify which words receive stress and which don’t, otherwise they sound monotone and even incomprehensible.
In English, stress is most dramatically realized on focused or accented words. For instance, consider the dialog
“Is it brunch tomorrow?”
“No, it’s dinner tomorrow.”
In it, the stress-related acoustic differences between the syllables of “tomorrow” would be small compared to the differences between the syllables of “dinner“, the emphasized word. In these emphasized words, stressed syllables such as “din” in “dinner” are louder and longer. They may also have a different fundamental frequency, or other properties. Unstressed syllables typically have a vowel, which is closer to a neutral position (the schwa), while stressed vowels are more fully realized. [Wikipedia]
What’s important to understand is that these stresses (unstressed and stressed) can be objectively measured and that the stress is relative. All that meter does is to organize these relative stresses into a pattern: Iambic Pentameter, Tetrameter and Trimeter; Anapestic Pentameter or Tetrameter; Dactylic Trimeter. This is what nursery rhymes do.
From this unpromising beginning, Schneider delves into some real gobbledygook. He places the invention of meter at the feet of “Classical Society’s need for reductivist explanations to fit into their simplistic cosmological view”. The absurdity of this statement is in the implication that “classical society” was responsible for the invention of meter. But meter isn’t limited to the English language. It’s also a feature of ancient Latin and Greek epic poetry. It’s also a feature of recently discovered ancient Egyptian poetry – poetry written long before such a thing as “classical society”. Maybe Schneider wants to argue that meter is a product of a “classical society” within ancient Greece, Rome and Egypt too?
Schneider then compares the English language to the gallop of a horse which, when caught on film, turns out to be far more complex that artists previously thought. But Schneider is merely dismantling an argument no one but he, himself, has made. No metrist or Stress Linguist would assert that English is ruled by two discrete stresses. All stress is relative. A true analogy would be to assert that scanning metrical poetry is, in fact, more like the horse’s known gait while Schneider’s characterization of meter is more like the naive artist’s before film was invented.
Scheider’s fundamental ignorance of the artistry behind the writing of meter is revealed when he touts variant feet as proof that meter is invalid.
Therefore iambic pentameter- easily the most common meter is 5 iambs- or 10 syllables- although an elided final stressed syllable or unvoiced extra syllable often occur, & are allowed. Yet these oft-recurrent exceptions (& tacit admissions of the theory’s invalidity) have not seemed to vitiate the metric rules its adherents cling to. Similarly anapestic trimeter is 3 anapests or 9 syllables. Yet with the same exceptions allowed as in iambs one might confuse the 2 meters quite often- especially when they often flout their own prerogatives.
Why a variant foot is a “tacit admission of the theory’s invalidity” goes unexplained. What does Schneider mean by theory? He doesn’t tell us. And that’s because…
There is no theory. Poets, during different periods, might have prided themselves in their ability to produce strict, unvarying metrical lines, but all the greatest poets recognized that metrical inflexibility produced monotony. The effect was unnatural. Schneider seems to interpret this as meaning that meter is therefore unnatural and invalid, but the interpretation is absurd. Just because unvarying meter produces monotony doesn’t mean that relative stress doesn’t exist. On the contrary, it proves that it does! And it proves that the best way to reduce the monotony of a too-regular metrical pattern is to introduce variant feet. Also, whether or not one “confuses” meters has no bearing on whether relative stress can produce meter.
Schneider then introduces what, to him, is his killer piece of evidence – the word generation.
A nice 4-syllabled word. The dictionary had it diacritically marked as ge¢ ne ra¢ tions, or (stressed-unstressed-stressed-unstressed). But really listen: if you say the word over & over; just this word, mind you, free of context & naked upon the table- generations, generations, generations….it should become clear that there are 4 distinct stress levels.
Once again, Schneider seems to have discovered (all on his own) what the rest of the poetical world has known for thousands of years. Stress is relative. Regardless of whether there are 4 distinct stress levels, the second and fourth syllables will always receive less stress than the first and second syllable. This is what scansion indicates – relative stress. (And I can’t stress that enough.)
But Schneider has more to say:
But here is another truism, long hidden by the Classicist bent- music in language, or poetry, has almost nothing to do with the individual stresses of syllables.
And, actually, I’m not troubled by this assertion. Schneider says “almost nothing”. Depending on how he defines “music in language”, he could be right, somewhere in between, or maybe wrong. But he doesn’t tell us what it means to him. He only tells us that this “music” has “almost nothing to do” with individual stresses. OK, but what does almost mean? Again, he doesn’t tell us. All we know is that stress does have something to do with the music in language but not everything.
From this point Schneider offers an alternative to understanding the “music of words” in lines. They rest on “rime (in all its varied form & types), alliteration, assonance, enjambment, & in the overall tropes of lines in accordance with the lines directly before & after it.” But what is striking is that Schneider essentially rediscovers (perhaps without realizing it) Robert Frost’s principle of the “sound of sense“. Schneider goes on to described “sharply-sounded lines” and states that “the digression- say, to some warm memory of love- may necessitate or facilitate the abrupt switch in line sound”. In a nutshell:
Simply put, music in verse- or language- depends on the congruence of syllable with syllable, word with word, line with line, stanza with stanza, etc.- as well as each of those congruent units’ emotional/intellectual congress with sound & meaning.
And here is Robert Frost (From a Letter to John Bartlett, 4 July 1913):
(…) I am possibly the only person going who works on any but a worn out theory (principle I had better say) of versification. You see the great successes in recent poetry have been made on the assumption that the music of words was a matter of harmonised vowels and consonants. Both Swinburne and Tennyson arrived largely at effects in assonation. But they were on the wrong track or at any rate on a short track…
(…) Those sounds are summoned by the audial imagination and they must be positive, strong, and definitely and unmistakeably indicated by the context. The reader must be at no loss to give his voice the posture proper to the sentence.
The parallels and the extent to which Schneider reinvents Frost’s sense of sound is striking. This whole principle of music in sentence sounds, first elucidated by Frost, is profound, but it’s not the either/or situation that Schneider makes it. It’s not either the “sound of sense” or meter. This is an absurd dichotomy dispelled by none other than Frost himself – who was able to combine sentence sounds and meter. The two facets of language work at different levels. Schneider is right to the extent that meter is not, in and of itself, the source of “music” in poetry. When it is used, it is only a part of the whole.
Where Schneider really veers into the demonstrably nonsensical is with the following:
The answer returns us to the very root of why metrics is such a ballocksed concept. Human beings simply DO NOT think, speak, or even hear things in a rhythmic fashion. There is no innate rhythm to human speech.
Humans do think and hear things in a rhythmic fashion. The unmitigated absurdity of Schneider’s comment is demonstrated by Rap, the Beatles, Schubert and Bach. It’s been demonstrated for thousands of years by every composer who has put music to words. Schneider tells us that the whole idea of a strong and week stress is rubbish, and yet musicians unfailingly match the strongly stressed syllables with the downbeat and the weakly stressed syllables with the off-beat. German is a stressed language like English and Bach was known to rewrite the lyrics of his Cantatas so that the stress patterns (the meter) coincided with the time signature of his music.
But what does Schneider make of rappers? Is he going to tell us that they, of all poets, musicians and artists, “don’t think, speak, or even hear things in a rhythmic fashion”? Rap is nothing if not a thumping example of accentual and accentual syllabic verse. Anyone who has seen 8 Mile by eminem knows that rappers have no trouble finding the “innate rhythm” in human speech. And what about all those children who can pick out the rhythms in Mother Goose – repeating its nursery rhymes over and over and over?
Moreover, if you don’t want to take my word for it, then consider the science. In an essay called “Perceptual biases for rhythm: The Mismatch Negativity latency indexes the privileged status of binary vs non-binary interval ratios” scientists have demonstrated “that the privileged perceptual status of binary rhythmical intervals is already present in the sensory representations found in echoic memory at an early, automatic, pre-perceptual and pre-motor level.” And here’s another study which refutes Schneider’s baseless assertion.
But what brings me to the science is an article I read several years ago. Humans are predisposed to hear rhythm even where there is none! For instance, humans will hear the tick-tock of clocks as being composed of a strong and weak ticking even when scientists calibrate the ticking mechanism to produce exactly the same tick! Humans, contrary to Schneider’s ludicrous pronouncement, have been programmed by nature to hear rhythm even where rhythm doesn’t exist!
What is Schneider’s alternative. He writes: “Humans speak & even think punctually.”
This may or may not be true. However, this has nothing to do with humanity’s ability to find rhythm in language. Schneider creates another false dichotomy.
But around this time, Schneider begins writing the second part of his essay, his defense of Jeffers.
But if the manifest denuding of this fallacy (& the others detailed) by this essay is not enough to get the Established doggerelists to finally drop their theoretically based & politically motivated vendetta against him & re-recognize Jeffers’ towering mastery, then surely the greatness of his verse should be allowed to stake that claim.
But what Schneider’s diatribe against meter has to do with defending Jeffers is muddled. If, as Schneider seems to imply, Jeffers reputation is being discounted because he didn’t use or couldn’t master meter, then the better approach is to question the legitimacy of his critics’ aesthetics. In other words, is a poem or poet’s greatness truly contingent on his or her use of meter? I say no. Great poetry can be and is written without meter or rhyme; but I say so without making the ridiculous counterclaim that metrical poetry is a sham.
In the one, perhaps, revealing portion of Schneider’s essay, he writes:
Jeffers was originally a mediocre formalist. His earliest published poems & books are forced, clunky, & melodramatic- in the worst sense. (…) I, too, have gone through that process- although every artist of excellence progresses at their own pace, & in their own way. Like Jeffers I too studied hard (he in college, me on my own) syntax, grammar, metrics, etc. to such a degree that I became an incredibly proficient mediocrity.
Is this the reason for Schneider’s hostility toward and denial of rhythm in language? – because he was a mediocre formalist? When I sit down to draw or paint a landscape, I am a phenomenally mediocre artist, but I don’t launch into a diatribe against perspective, denying its existence and railing against any artist who thinks he or she can perceive it.
The practice of writing meter and the symbols poets use to scan it represent relative stresses and nothing more. Other languages have produced their own meters – such as the quantitative meters of classical Latin and Greek. It symbolizes the fundamental human desire for rhythm even at its most subtle. While it’s not the only source for “music” in language, it is very real and plays a very real part.
Dan Schneider responds:
To be honest, my first reaction is to be flattered.
That said, I still find his initial essay ludicrous and stuffed with fallacious arguments. He made many points in response to my own assertions, but most of them are tangential to a definition of meter. For example, he points out that I got the title of his essay wrong, true, and that there are typos in my posts, also true. He accuses me of writing him a possibly virus ridden hate-E-Mail which I don’t remember and which he, conveniently, can’t produce. (I’m calling that one, false.) He also takes issue with how I characterized his arguments. I don’t blame him, but I stand by those characterizations. However, none of this has anything to do with meter itself.
On to his assertions concerning meter.
In the entirety of his response, he provides only two (2) specimens to support his arguments.
In answer to my rhetorical question, ‘…what metrist has ever asserted that meter is composed of just two discrete stresses and that, furthermore, these two stresses are precisely the same no matter the context?’, Dan writes the following:
I will now disprove such by using two definitive texts. The first is from Webster’s Universal Unabridged Dictionary (1964). In reference to meter (meaning poetic metrics, no other usages of the term):
1. (a) rhythm in verse; measured, patterned arrangement of syllables, primarily according to stress and length; (b) the specific rhythm as determined by the prevailing foot and the number of feet in the line; as iambic meter; (c) the specific rhythmic pattern of a stanza as determined by the kind and number of lines.
I don’t see how Webster’s helps Dan’s case. Notice that Webster’s does not assert that meter is composed of two discrete stresses or that they are the same no matter the context. Dan’s original assertion was that:
“meter is the theory (claiming origin by several cultures) that spoken language consists of 2 primary vocalizations of a sound- i.e.- stressed & unstressed.”
And this definition, as a very general one, isn’t necessarily wrong. But he then calls that definition into question by writing that:
In fact the dualistic notion of mere stressed & unstressed sounds is- in practice by its many proponents- almost always so loose as to be meaningless anyway, as metrics should really redefine its definitions as greater & lower stress(es) (with a plenum of in-betweens), since (obviously) a truly unstressed syllable would be silent.
In other words, (according to Dan) the “plenum” of stresses available in an accentual language contradicts the notion of “2 primary vocalizations”. But it only contradicts if one assumes that the “2 primary vocalizations” can’t be relative (or widely vary in relation to each other). Schneider’s argument only holds water if the “2 primary vocalizations” are discrete and always the same. But, as I wrote, no metrist, to my knowledge, has ever asserted the same (only, ironically, Dan Schneider). All “theories” of meter recognize that stress is relative and therefore recognize a “plenum” of stresses. They recognize that English is an accentual language, and that within the language’s “plenum” of stresses, one stress will always be relatively strong and one will always be relatively weak.
Webster’s definition in no way bolsters Dan’s contention that meter doesn’t exist. Nowhere does Webster’s definition limit meter to two discrete stresses which are always the same. The Webster’s definition rightly asserts that meter is a pattern of stresses (English for example) or lengths (Latin for example).
What is especially curious about Dan’s example is that Webster’s defines meter the way I do(!) and, most importantly, doesn’t question its very existence.
On to Schneider’s next example:
The oldest and most important device of Verse form, meter selects one phonological feature of lang. (stress, pitch, length) and reduces it several levels or degrees in ordinary speech (3 or 4 levels of stress; high, mid, and low pitch; various durations) to a simple binary opposition (‘stress’ vs. ‘unstress’; ‘level’ vs. ‘inflected’ pitch; ‘long’ vs. ‘short’) which may be generalized as ‘marked’ vs. ‘unmarked’.
This is from the New Princeton Encyclopedia of Poetry and Poetics. Dan rightly mentions that Princeton’s overview covers several pages. However, he glosses over the implications of this concession by writing:
This is very important to note, because from the start of my essay through its end, I am the person arguing that meter is a reductio ad absurdum, it is not real, and it reduces human speech to a false binary opposition. Princeton proves I’m right on that score, and says so in black and white.
(Never mind that Dan’s own example from Webster’s contradicts his claim that meter is “a reductio ad absurdum” – which is to say, it doesn’t exist.) Well, OK Dan, but, as you intimated, Princeton says a lot more in black and white. It also writes:
The traditional view had always been that m. is an arbitrary pattern imposed on words — that, as Gurney put it, “metrical rhythm is imposed upon, not latent in, sppech” (1880). It seems indubitable that meter is in some sense a filter or grid superimposed on langauge. But 20th century linguistics has shown convincingly that many aspects of poetic form are merely extensions of natural processes already at work in language itself.
One page later, and after much exegesis to support this contention, Princeton closes the section by writing:
But modern metrics also holds that strong syllables outside ictus are “demoted” and weaker syllables under ictus “promoted” under the influence of the meter. Promotion of weak syllables under non-ictus weights and slows the line, adding power. Demotion of stresses under ictus gives a quicker and lighter line. This is not a purely metrical mechanism, it shadows normal phonological process by which alternation of weak and stress, and strong and stronger, is effected atomically in polysyllables.
Apparently Dan either couldn’t be bothered to read this far or conveniently chose to ignore this portion. Princeton, in fact, not only disagrees with Dan but recognizes the binary stress pattern of the English language as a “normal phonological process”. And, by the way, did I mention it does so in black and white? Not only that, but Princeton rightly points out, as I have, that 20th Century linguistics has shown convincingly that many aspects of poetry are “extensions of natural processes already at work in language itself.” The next time Dan claims to be a man of science, take it with a grain of salt.
Dan then goes on, at some length, railing at my characterizations of his argument. None of which, curiously, supports his claim that meter doesn’t exist. He repeatedly refers back to Websters and Princeton, neither of which support his argument. Among other things, he writes:
This is really amazing. First, VP spends the bulk of his essay claiming that my claim that meter is a fallacy is wrong, then he cites a study (naturally, the links do not work)…
I just checked the links. They work.
Without, apparently, reading them, he both dismisses and reinterprets the science (which, did I mention, he didn’t read).
More importantly, Dan never counters the example of an artist like Eminem. As I wrote above, Rap is “thumping example” of accentual and accentual syllabic verse.
Dan quotes Princeton out of context, ignores science, and glosses over 8 Mile. He then closes:
As I implied in the piece VP quotes, I was a mediocre formalist. Note the past tense. I am a great poet, formally and in free verse. There are poems of mine that scan perfectly, according to metric nonsense, but not because I was following metric dictates, but because any well musicked poem will, given the reductive aims of meter, scan well. It’s what is in them that matters.
So, according to Dan, meter doesn’t exist but, by gosh(!), when he wants to, he writes meter with genius!
Not that all meter isn’t “nonsense” (but his poem scans perfectly). He’s not following metrical dictates (it’s just that a “well musicked poem” does the same thing), and not that it’s not nonsense (but it scans well). Never in the annals of “seminal” essays has a more self-contradictory paragraph been written.
I guess that’s what happens when you try to have your cake and eat it too. At the very least, readers shouldn’t be taking advice from a man who claims meter doesn’t exist, then hurriedly, as an afterthought, asserts that he nevertheless writes meter with genius. Makes you wonder who the idiot really is, doesn’t it?
By the way Dan, I prefer – Fool.
In a play like King Lear, he’s the only one who lives.