Free Verse: an essay on Prosody by Carles O. Hartman
March 9, 2012
First Things First: What is Prosody?
I remember, way back when, I knew a poet who favored free verse. As his writing developed, he struggled with a question that confronts many writers of free verse. Where does the poet break his or her lines? You can find this same question frequently posed on the internet. In traditional poetry, the line ends where the iambic pentameter ends, basta; but, as far as my friend knew, there was no such rule pertaining to line lengths in free verse. Me friend declared that he was going to systematize lineation in free verse. I never heard back from him. What he was really saying was that he wanted to develop a Prosody, one that he and his readers could mutually understand.
So, when I discovered Hartman’s book, I was excited. Here was a book that tried to answer the question: Is there a prosody of free verse and, if so, what is it?
The definition of prosody (or at least the one with which this book is concerned) is as follows: A system of versification. This is problematic when applied to free verse. What this means (and what Hartman must argue) is that free verse isn’t free, but is a systematic form as rigorous as traditional verse. He must argue that once a reader understands that prosody, he can apply that knowledge to any free verse poem in the same way that a reader of traditional poetry applies the prosody of meter and rhyme to her reading of traditional poetry. Take Shakespeare’s famous line, To be or not to be, that is the question, as an example. If we read the line according to the prosody of traditional poetry, the verb is receives the stress, rather than that. That is, knowing the rules of meter, the line reads: To be or not to be, that is the question. We can assert the likelihood of this reading because the prosody of meter (and blank verse in particular) suggests it.
The first thing Hartman had to do, given that he’s writing a book on the prosody of free verse, is to re-define the word free. Clearly, if something is free, implying in this case that it lacks form (is not formal) then there can’t be, by definition, a prosody. One can’t go writing a book on the form of free verse when the free in free verse is understood as implying a lack of form! Prosody implies a regular and recognizable system that is applicable to all poems (and that is more than a niggling problem for Hartman). He wastes no time explaining what we really mean by free in free verse (or prosody for that matter). First he defines prosody:
The prosody of a poem is the poet’s method of controlling the reader’s temporal experience of the poem, especially his attention to that experience. But how can the poet control the reader’s experience? How does the reader know what to pay attention to, among the many linguistic events the poem comprises? The prosody, to function as a prosody, must be shared. [p. 13]
The italics are the author’s. The underlining is mine. While Hartman emphasizes the intentions of the author with italics (which will be extremely important to his later arguments) the thing that makes prosody, well, a Prosody, is that it “must be shared” – a contrivance understood by both reader and writer. The poet can tell himself that he is controlling the “reader’s temporal experience” until the moon grows grass, but unless the poet’s methods are understood by the reader (unless it is shared) he might as well be writing letters to his dog.
The difference between Poetry & Prose
Anybody who follows my blog knows my opinion on free verse. I do think it’s much easier to write than traditional verse (which has led to its near total dominance), but that doesn’t mean free verse can’t be written with a greatness equal to traditional verse. The qualities of greatness are the same whether the poem is free verse or traditional. That said, traditional verse offers the poet effects that free verse doesn’t and never will. And so begins my many disagreement with Hartman.
At the outset, Hartman states what I have stated many times:
“…it has often been shown that any mode of organization found in any poem (except lineation) will also occur in some passages of prose – usually many, though rhyme, for instance, had a short and relatively disastrous career in English prose.”
The problem is that Hartman means this to include traditional verse, which I don’t. As Hartman himself states (curiously) rhyme’s appearance in prose was “short and relatively disastrous” (meaning that it didn’t work). That’s because internal rhyme isn’t the same as end rhyme. In other words, one can’t separate end rhyme (as it is practiced in the traditional poem) from lineation. And the combination of end rhyme, combined with meter, is also not the same as end-rhyme alone (and is not something that appears in prose). My point is that there is a continuum. As regards free verse, Hartman’s statement holds water. The only feature that separates free verse from prose is lineation. As regards traditional verse, Hartman’s statement doesn’t (as he himself unintentionally admits). Traditional verse adds extra layers to lineation. The metrical line and end-rhyme don’t and have never appeared in prose. For example, regular metrical feet may appear in prose, but a regular metrical line never has and never will.
It’s a curious facet of Hartman’s dialectic that he eagerly (and rightly I think) emphasizes the importance of lineation in free verse, but consistently downplays or fails to recognize the compounding effect of lineation when combined with meter and rhyme. He can’t have it both ways (though he tries) and that’s part of the problem. Hartman wants to establish a prosody of free verse that is equal to traditional verse. That’s a mistake. He can’t do so without altogether disregarding the compounding effect of meter and end-rhyme.
He knows that. It’s the only direction his thesis can take him. That’s why, at the end of chapter three, he triumphantly announces that “what rhyme and meter can do, lineation alone can also do”. We’ll return to that. Yes, we will. By the time Hartman makes this pronouncement, a fairly simple word like rhythm has been turned inside out and upside down and many a reader (to judge by other reviews) becomes lost in the maze of his baroque re-definitions.
Of Rabbit Holes and Rhythm
Unfortunately for Hartman, the one word he fails to accurately define is rhythm. I’ve had this discussion elsewhere (on this blog) with readers who style themselves defenders or proponents of free verse. Most of us use rhythm in a literal and a figurative sense – but mixing these two uses in a book which professes to establish a prosody (and which takes great care to carefully define words like free and prosody) is a considerable oversight that undercuts the entire argument.
We regularly refer to random events or objects as having a rhythm. We can watch the wind on a wheat field and describe the rhythms of the wind – but these are random events. They’re not rhythmic. The human brain, as science has amply demonstrated, is designed, by default, to find rhythm and pattern where none exist. This is important because we also commonly refer to the rhythms of language when, in reality, we’re describing not the rhythms, but the arrhythmia of language. Likewise, listeners and poets will frequently refer to the rhythm of this or that free verse poem when what they’re really describing is the arryhthmia of the poem’s language (which isn’t to say that a free verse poet isn’t making conscious choices — only that the choices result in an irregularity that is unique to the poem).
Strictly speaking, arrhythmia is a medical term, but in this context it’s useful. Here’s how it’s defined by the Farlex Free Dictionary:
1. arrhythmic – lacking a steady rhythm; “an arrhythmic heartbeat”
unsteady – subject to change or variation; “her unsteady walk”; “his hand was unsteady as he poured the wine”; “an unsteady voice”
2. arrhythmic – without regard for rhythm
unrhythmic, unrhythmical – not rhythmic; irregular in beat or accent
The most useful meaning for our purpose is the idea that language is “irregular in beat” (though different languages are obviously irregular in unique ways). Language has no rhythm (in the literal sense of the word) because the rhythm of any given language isn’t regular. However, each language has a unique rhythm in the figurative sense if we understand that to mean arrhythmical — uniquely irregular. Rhythm, on the other hand, means something regular, recurring, having a beat or pattern: “of, relating to, or characterized by rhythm, as in movement or sound; metrical, periodic, or regularly recurring“. Wikipedia makes the link between rhythm and pattern explicit:
Rhythm (from Greek ῥυθμός—rhythmos, “any regular recurring motion, symmetry“) may be generally defined as a “movement marked by the regulated succession of strong and weak elements, or of opposite or different conditions.” This general meaning of regular recurrence or pattern in time may be applied to a wide variety of cyclical natural phenomena having a periodicity or frequency of anything from microseconds to millions of years. [March 3rd, 2012]
So, when readers, and Hartman himself, refers to a free verse poem as having a rhythm, he’s using the term incorrectly. If a free verse poem has a “regularly, periodic and recurring pattern,” then it’s not free verse. As I’ve written many times before, if the verse isn’t free, then it’s not free verse (unless we change the definition of free).
Hartman’s failure to adequately define rhythm (or his misunderstanding of the word) sets him on the wrong course from the get-go.
What I have already said about the temporality of poems suggests that prosodic organization is rhythmic. Rhythm, in poetry, is the temporal distribution of the elements of language. According to this definition, all language unavoidably has rhythm. [p. 14]
The italics are Hartman’s. Temporal means “of or relating to or limited by time”, but not necessarily recurring or periodic. This is a nice dodge. Hartman himself realizes that this definition won’t do, but he fails, utterly, to acknowledge the importance of “pattern” to rhythm because he doesn’t want to. To do so would be to undercut his dependence on the word rhythm as it describes free verse later in the book. Instead he offers up an object lesson in tortured, baroque avoidance that leaves him right where he started. (Remember, he refuses to use or acknowledge the word pattern or recurrence.) He settles on the neutral word: organization. (You might object that he includes the word rhythmic in his definition, but remember that he’s just defined “rhythm in poetry” not as recurring or regular but as temporal.
This suggests a form of my definition of prosody that approximates and includes the traditional one: It is the system of rhythmic organization that governs the construction and reading of a poem. [ibid]
Now that he’s settled on the generic organization, he needs to define it:
“Organization” implies elements to be organized, and prosodic organization will employ the elements of speech: (1) timbre (in recurrences such as alliteration, assonance, and rhyme); (2) duration (which, when applied as it commonly is to syllables, is called quantity); (3) pitch or intonation; (4) intensity or volume (these two being distinguishable acoustically but not psychologically, and so not prosodically); and (5) boundary. [ibid]
Nowhere does Hartman acknowledge the one thing that is part and parcel of rhythm — regularity, recurrence and pattern! In fact, nothing in Hartman’s further definition of prosody distinguishes it, in any way, from his first definition. That is, Rhythm, in poetry, is the temporal distribution of the elements of language. And, like he said, his definition of “rhythm in poetry” does not distinguish it in any way from “all language”. So why make it? All of the 5 elements he lists (as elements implied by “organization”) are occurring, right now, in this paragraph. Without the stipulation that rhythm implies a regular and recurring pattern, the word becomes figurative at best and meaningless and worst (to be applied to anything). That’s going to cause problems for Hartman, problems from which his argument can’t recover.
The first problem is that his definition of “rhythm in poetry” cannot account for meter. This is intentional but it’s still a problem. Since Hartman still has to account for meter, and since he rejects the word rhythm, he has to come up with an alternate that avoids being conflated with rhythm. He does so in spades:
The linguistic elements a poet organizes prosodically are largely chosen from him by the conventions of his language, but each poetic tradition also dictates, by establishing more-specific conventions of verse, what he does with those elements. In almost every case, this traditional mode of organization is or depends on a numerical rule. When this is the case, we call the prosody metrical. A meter is prosody whose mode of organization is numerical. [p. 17 - Hartman's italics]
Numerical? This is so generic and bland as to be cynical. Hartman’s definition completely ignores the aural effect of meter and treats it (and poetry in general) like something that only happens on the silence of the page. (This, in fact, will be a tendency that appears elsewhere.) Hartman’s definition of meter fails on such a grand scale that refuting it is as simple as the humble limerick. I’ll explain what I mean shortly but first, we continue on our tour of Hartman’s logic.
Now that Hartman thinks he has firmly excised rhythm from meter, he goes for the kill in one of the most confused and nonsensical paragraphs I have ever read. I’ll print it in full:
Crude as it is, scansion — the simple diagrammatic indication of stresses and slacks — tells us all we have to know about a poem’s meter. The meter itself, like the scansion, is an abstraction. It is the rule to which a line more or less conforms, and not the line itself. It is not rhythm, but a pattern imposed on rhythm. Not only the unmetered elements of language (such as timbre and quantity), but also the actual instances of the metered elements, the particular stresses and syllables of the line, continue in some sense to occupy the more general area of rhythm. “Rhythm is not metre,” Own Barfield remarks. “It is not another name for metre, but something far subtler. Rhythm is variable about its underlying regularity, whereas metre is invariable” (12,793). Meter is the “underlying regularity” played against by rhytyhm. These two maintain a continual and fructifying tension, like any actuality and the abstraction that shape it. [p. 22]
Where do I start? Let’s begin with the underlined sentence. This sentence is precisely where the previous 21 pages collapse: It is not rhythm, but a pattern imposed on rhythm. Hartman has so separated rhythm from its central meaning of recurrence, regularity and pattern, that he fails to see the absurdity of his statement. I’ll be blunt: A rhythm implies, by definition, a pattern! You cannot, quote-unquote, “impose” a pattern on a rhythm because a rhythm already implies a pattern! This is his attempt, I think, to fully separate rhythm from pattern (essential if he wants to divorce rhythm from meter and leave it nothing but the dry numerical) but the effort defies simple logic.
He follows this with a quote from Own Barfield meant to drive home his point: “Rhythm is not metre.” Barfield then explains this by implying that metre is something separate from language. He too makes the bizarre assertion, counter to every definition of rhythm, that rhythm is distinct from meter’s “underlying regularity”. I just popped up Artha (my linux system’s dictionary) to drive home the point I’m making.
- the basic rhythmic unit in a piece of music
- recurring at regular intervals
- an interval during which a recurring sequence of events occurs
- the arrangement of spoken words alternating stressed and unstressed elements
So, if rhythm is defined as something that recurs “at regular intervals”, how on earth is this distinct from meter’s “underlying regularity”? Between Hartman and Barfield, if there was ever a textbook example of a distinction without a difference, this is it. The whole mess could easily have been avoided if Hartman had simply conceded that meter is, in fact, rhythm; that the link between music and meter is not isochrony but a recurring and regular pattern; and that if there is a distinction to be made, it is not between meter and rhythm but between the rhythm of metrical verse and the arrhythmia of non-metrical verse.
But the proof is in the pudding, and that brings me back to the humble limerick.
A flea and a fly in a flue
Were caught, so what could they do?
Said the fly, “Let us flee.”
“Let us fly,” said the flea.
So they flew through a flaw in the flue.
The meter of all limericks essentially alternates between anapestic trimeter and anapestic dimeter. It’s that simple. The whole reason limericks are so catchy is because of their rhythm, yes rhythm; and their end-rhymes. To call a limerick numeric is to be obtuse. Yes, it’s numeric, but limericks work not because you can count the stresses in their lines but because you can hear the rhythm created by the anapests.
What is Hartman’s answer to the limerick (or any of the hundreds of nursery rhymes)? He wants us to know that a limerick’s effects are “not a meter in the poetic sense”!
This prosody originates in music. It depends on a beat or pulse–not counting the accents, but equalizing the time between them: isochrony, it is called… But though it is a prosody–it controls the audience’s temporal experience more directly than most–it is not a meter in the poetic sense. It organizes rhythm not numerically but temporally. [p. 32]
Poems like limerick’s “organize rhythm not numerically but temporally“. This is the rabbit hole Hartman must navigate because of his refusal to recognize the standard definition of rhythm. Remember page 14? Here’s what Hartman wrote: “Rhythm, in poetry, is the temporal distribution of the elements of language.” He then adds that “according to this definition, all language unavoidably has rhythm.” When you consider that Hartman is defining the effect of poems like limericks as temporally organized rhythm (and remember, he has already defined rhythm as “a temporal distribution”), then you must conclude (based on Hartman’s earlier definition of rhythm as, from what I can tell, the same thing) that there is no difference between a limerick and “all language” (since all language unavoidably has rhythm). Did you get that? I’ll make it clear. According to Hartman:
- Rhythm is “the temporal distribution of the elements of language“.
- Limericks and Nursery rhymes organize “rhythm not numerically but temporally“.
Now, since he’s already told us what rhythm is (according to him) let’s replace the word rhythm with the definition he provided:
- Poems like limericks organize the “temporal distribution of the elements of language not numerically but temporally“.
Right, I’m not making this up. These are his own words. This is where Hartman’s baroque definition of rhythm (absent it’s dictionary definition of recurrence and pattern) lands him – in sheer tautological absurdity. He just can’t bring himself to admit to the rhythmic effect of meter. He can’t because he wants to reserve the word for his prosody of free verse. However, his gymnastics just don’t work. For all intents and purposes, he seems to deny that poems like limericks are written in meter or even exist! This is what allows him to say that line breaks can do anything that meter and rhyme can do. He has written off the very things that meter and rhyme do!
But enough argumentation. Let Hartman write a limerick that doesn’t use meter or rhyme — only line breaks.
Right. I didn’t think so.
Anyway, Hartman is now forced to distinguish between meter “in the poetic sense” and meter in the “temporal” sense. Does that mean, then, that any time one begins to hear the rhythm in meter that it’s not really “a meter in the poetic sense”? Then what does he make of the entirety of Spenser’s Fairy Queen? Here are just two of the hundreds of stanzas:
Now when that idle dream was to him brought,
Unto that Elfin knight he bad him fly,
Where he slept soundly void of evil thought,
And with false shows abuse his fantasy,
In sort as he him schoolèd privily:
And that new creature, borne without her dew,°
Full of the makers guile, with usage sly
He taught to imitate that Lady true,
Whose semblance she did carry under feignèd hew.
Thus well instructed to their work they haste,
And coming where the knight in slumber lay,
The one upon his hardy head him plac’d
And made him dream of loves and lustful play,
That nigh his manly hart did melt away,
Bathed in wanton bliss and wicked joy:
Then seemèd him his Lady by him lay,
And to him ‘plain’d, how that false wingèd boy,
Her chaste hart had subdued, to learn Dame Pleasure’s toy.
Anyone, and I do mean anyone, who actually sits down to read Spenser’s Fairy Queen, cannot fail to hear the steady, near incessant tum-te-tum-te-tum of Spenser’s iambic pentameter. According to Hartman (since it’s obviously rhythmic in every sense but his), this doesn’t count as meter “in the poetic sense”. How about Shakespeare’s blank verse? The meter’s rhythm is subtler, but it’s there. Ask anyone who can hear the difference between Shakespeare’s blank verse and his prose passages. They won’t answer that “the blank verse sounds organized numerically.” That’s just nonsense. They will answer that there’s a rhythm to the blank verse that isn’t heard in the prose passages.
Hartman’s description of meter, at the close of chapter one, comes as no surprise. He describes it as:
…an abstract pattern [the reader] can transfer in detail from poem to poem and codify in a formally closed, quasi-mathematical system that bears only incidentally on the experience of poetry. [p. 28]
Any reader who states that meter is a system “that bears only incidentally on the experience of poetry” doesn’t know how to read it. To Hartman’s credit, some 26 pages later however, he essentially contradicts himself when he writes that “in traditional verse the metrical determination of accent helps to control the interpretation of meaning…” How can meter help to determine the poem’s meaning and yet bear “only incidentally on the experience of poetry”? Hartman’s book is full of contradictions like these, but then again, maybe he has re-defined the meaning of “experience” vis-a-vis poetry.
- It’s worth mentioning, I think, that Hartman dismisses the one word that could have gotten him out of this whole quagmire – cadence.
All these theorists and theories tended to converge on the word cadence. The convergence was more lexical than semantic, since the word came to mean whatever a writer liked. [p. 46-47]
Ironic that he would write that, since this is precisely what he does to the word rhythm. Also ironic in that, by the time he’s done re-defining what he thinks rhythm should mean he’s all but re-defined it as cadence!
Rhythm, Symmetry, Counterpoint and the Free in Free Verse
Like I wrote earlier, one can’t go writing a book on the form of free verse when the free in free verse is understood as implying a lack of form. Hartman has to change the meaning of free. His first stab at this is to argue that we only think it’s free because we’re ignorant of its conventions.
In some sense any verse form is “free” with respect to any other, as the rhapalic line I invented is free if measured by the rules of iambic pentameter. It is “free” until its prosody is discovered. The reader easily discovers the prosody of a poem that belongs to his own tradition. But when the prosodic conventions on which a poem depends are alien to his experience, the poem will puzzle or completely mystify him. [p. 18]
At first glance this seems like a reasonable argument, but the argument is weirdly self-defeating. Hartman’s reasoning would seem to go like this: Poem X only looks like a free verse poem, but it’s not. In fact, once you closely examine it, you realize that it has a form. By way of example, Hartman gives us Marrianne Moore’s “Bird-Witted”. He points out that each stanza is “flawlessly divided” into syllables that count: 9,8,6,4,7,3,6,4,7,4. But what is he saying? Is he saying that Moore’s poem is still free verse, or is he saying that it’s not? If he’s saying it’s not, then what is the point of his argument? Is he saying that some free verse poems are free and some are not? Then what does free mean? What do we call these other “free verse poems”? He does, at least, have an answer to this last question. He divides free verse into vers libre and vers libéré.
On comparing French and English theories of verse, “we discover at once that French distinguish between vers libre and vers libéré — verse which is born free and verse, so to say, which has been liberated from some pre-existing chains. We have not this distinction in English — party I suppose because the neat verbal antithesis between libre and libéré is not available in English language. J.V. Cunningham helpfully provides a full idea of the resources of what Hough calls verse libere: “in general, the lines of a poem [of this kind] will be partly in standard meter, at times parts of what would be a standard line, or they are felt to be equivalent in some aspect of sound or feeling to a standard line, or they exhibit some marked variation of a standard line, or some other principle of meter is used intermittently and supported and given authority by the presence and recurrence of standard lines.” [p. 113]
Even so, these distinctions seem tangential and unhelpful. What exactly is Hartman discussing? For that, we go back a hundred pages:
“Free” is properly a synonym for “nonmetrical,” and it follows that the prosody of free verse is rhythmic organization by other than numerical modes. [p. 24]
Ultimately, “free verse” is “free” only in a special sense. Poems are written in verse so that the rhythms of language can contribute to the whole meaning of the poem; and it is prosody of one kind or another that turns rhythm into meaning. [p. 27]
And in these definitions are the whole reason he tries to excise rhythm from meter. He wants, in effect, to co-opt rhythm as a free verse effect and not a metrical effect. He makes rhythm essential to his notion of a free verse that isn’t “free”. Curiously though, and typically, his definition implies that meter’s “numerical modes” are a species of rhythmic organization! Why else would he write “rhythmic organization by other than numerical modes? (It seems that Hartman is, himself, either forgetful of, or confused by his own rhetoric of rhythm.) At this point, Hartman describes two “rhythmic” modes of organization that are nonmetrical — counterpoint and symmetry. The rest of the book, however, will primarily be concerned with counterpoint.
“I have implied that multiple rhythmic patterns–not all of them metrical and perhaps none–can coexist within a given passage of verse. These multiple patterns may reinforce each other, or they may stand in conflict. In the latter case, we can generally expect to perceive conflict on one level as meaning on another, as any paradox ultimately disproves (but does not deny) itself. This kind of significant conflict I will call counterpoint.” [p. 25]
“A second mode is symmetry. Free verse rarely uses a symmetrical prosody in a primary way. It would give the poem too tedious a stability. But when such elements as accent function at all prosodically in free verse (as they usually do, because of the nature of the langauge), they often adopt a symmetry that seems to arise out of the actual line, unlike an imposed numerical quota.” [ibid]
- Once again you’ll notice that Hartman slips up by stating that meter is one of many “rhythmic patterns” – this after insisting that meter isn’t rhythm!
Chapter Four is called Counterpoint. Chapter 5 is called the Discovery of Form ( touches on Symmetry) and Chapter 6 is called the Discovery of Meter (this is where he makes the distinction between vers libre and vers libéré .
These three chapters are the heart of Hartman’s book, the chapters where he actually tries to establish and demonstrate a workable prosody. Of the three, the fourth chapter is the most interesting and the most useful to anyone who is writing free verse. In my opinion, the book would have been much better if he had started with Chapter 4, resisting the Aristotelian reinvention of the wheel in the first three. I think I can briefly summarize the gist of the three chapters.
Lineation allows the free verse poet to emphasize not just words (by choosing their placement at the ends of lines) but allows the poet to counterpoint linebreak with syntax. Where Hartman fails is in establishing counterpoint as a prosody. Remember that a prosody is something that “must be shared”. The principle error in these three chapters is a categorical one. Meter is numerical in the sense that one can objectively scan it and objectively observe where words are demoted or promoted. The meaning of a promoted word doesn’t necessary change from one poem to the next because syllabic emphasis is a part of our language. For example:
How did you do that?
How did you do that?
Depending on how these words appear in a metrical poem (one way or the other) their meaning subtly changes. Because meter is a prosody and because we all speak the same language, we will know which way to read the line based on its appearance within the metrical line. We can safely assume that the poet means us to read it one way or the other – and our interpretations of these lines will be more alike, than not. Hartman would have us believe that line breaks are no different. So, by contrast, here are the first lines of a poem he analyzes:
Shadows cast by the street light
·······under the stars
··············the head is tilted back,
the long shadow of the legs
·······presumes a world
··············taken for granted
on which the cricket trills.
Now what is the reader to make of these line breaks and indents? Hartman has an explanation and analysis for all of it, but all of it flirts too dangerously (when it doesn’t cross the line) with Intention Fallacy (in a limited sense) and Enactment Fallacy. Hartman must presume to know what the author intended when he used X number of syllables in a line, broke the line at this or that word or phrase, and indented. There’s no way around this. There just isn’t. If Hartman can’t speak to the author’s intentions, then there’s no prosody, there’s no certainty that the poet and reader are sharing a common interpretation of the techniques used. (That is, there’s no common interpretation of the techniques being used.) If Hartman denies this, then his interpretations may or may not represent the intentions of the poet. That, in fact, is precisely what happens. Although Hartman’s interpretation of lineation in this or that poem is interesting, he offers no reason to think the poet actually intended any of it (other than Hartman’s say so).
Likewise, any poet who writes free verse may have her reasons for breaking a line where she does, but how is the reader to know whether to give all line breaks equal weight, some less, or some more. How is the reader to guess at the poet’s meaning? It’s extremely doubtful that any two readers would ever give the same weight or the same meaning to a given line break, let alone a poem. By contrast, the majority of readers will similarly interpret: How did you do that?
In this respect, Hartman’s “prosody” does not withstand comparison to the prosody of traditional poetry. In fact, Hartman’s arguments and assertions can become so diffuse as to be a kind of proto-academese:
“If one distinguishes the constant interval of time measuring each line from the variable pace within it, the relation between them itself appears as a kind of counterpoint. That relation, incidentally, resembles the one between meter and rhythm in accentual-syllabic verse, suggesting that the traditional meter, too, inherently involves counterpoint. But Williams’s poem derives much of its rhythmic interest from a more complex counterpoint, changing the relation between its isochronous lineation–comprising both interval and pace–and its syntax.” [p. 68]
But, as with the limerick, rather than speculate, there’s a concrete test for Hartman’s claims. Prosody has more than one definition. Here’s Wikipedia:
“Many traditional verse forms prescribe a specific verse metre, or a certain set of metres alternating in a particular order. The study of metres and forms of versification is known as prosody. (Within linguistics, “prosody” is used in a more general sense that includes not only poetical metre but also the rhythmic aspects of prose, whether formal or informal, which vary from language to language, and sometimes between poetic traditions.)” [March 9, 2012]
We’re going to use the first definition of prosody – the study of poetic meter. This is fair. Hartman himself invites the comparison. Throughout the book he compares his prosody the that of rhyme and meter (rather than to prose). At the end of Chapter 4, Hartman tells us the following:
“For an introductory course in modern poetry, I had typed out and mimeographed a set of free-verse poems as if they were prose. My purpose, of course, was not to pretend that the two forms are equivalent, but to broach the fundamental question of how free-verse lines are divided or determined. I asked the students to mark the line breaks. The only additional information I gave them about Auden’s poem [Museé Des Beaux Arts] was that it contains two stanzas of unequal length.”[p. 75]
The results, predictably enough, were nothing like Auden’s. What this proved, other than that the students aren’t as good at Auden’s poetry as Auden, is unclear, but Hartman has a method. He means to demonstrate that if the poem is written in any other way, the poem won’t carry the same meaning as that implied by Auden’s original lineation.
In several cases, Auden’s lineation generates quite specific effects which one might call semantic. Some of these depend on what I described in the last chapter as a principle of antithesis, that changing elements take stress and constant ones do not. The reverse applies as well: Where a word is unexpectedly stressed, it suggests the alternatives from among which it has been chosen. [p. 77]
So says Hartman. He may or may not be right in his interpretation of these typographical features. But then what does right mean in this context? There’s no way for him to know whether his thinking reflects Auden’s and what does it matter? It only matters if you’re claiming that your interpretation is based on a “shared” prosody of free verse. With that in mind, the larger purpose of his classroom experiment was, I think, to suggest that Auden’s poem obeyed certain recognizable principles. While I enjoyed Hartman’s detailed examination, it hardly added up to a prosody comparable to that of traditional poetry. I’m willing to offer the following test. If Hartman is so sure that his prosody is the equivalent of meter and rhyme, then I will remove the line breaks in any of Auden’s other poems (presumably one he’s not as familiar with) and let’s see if he can reconstruct it? Or how about William Carlos Williams or any of the other major free verse poets? I’ll bet he can’t reconstruct a single one of them. This tells me that his claims to a “shared” prosody of free verse don’t hold water.
On the other hand, he’s welcome to pick any passage from Milton, any Sonnet, any poem by Donne or even a passage from Shakespeare, and based on the prosody of traditional poetry I, the carpenter from up in Vermont, will reconstruct them exactly as the poet wrote them. Not only that, but give me a sonnet (one that I’m unfamiliar with) remove the line breaks and mix up the order of the lines. I’ll still reassemble the sonnet exactly as the poet wrote it.
I defy any one, using Hartman’s prosody, to reconstruct the randomly scrambled and de-lineated free verse poem. Again, what this tells us is that Hartman’s prosody fails the standard he, himself, set for it, that a prosody “must be shared”.
Hartman’s repeated claims to a free verse prosody (equivalent to that of traditional poetry) are baseless. He does his argument no favors by making such comparisons. Besides that, there’s no reason to. I just don’t see why he feels the need to constantly compare free verse to traditional poetry? Is the book nothing more than sibling rivalry? Why can’t a prosody of free verse be like a prosody of prose?
Hartman describes how the subject of a poem, in this instance at the hands of William Carlos Williams, helps shape the poem itself (its counterpoint) — phrasing and lineation. The poet “discovers” the form of the poem as he writes it. This chapter is probably the weakest and least convincing of the three. Hartman, at the chapter’s outset, seems to anticipate this weakness:
Using free verse did not simply mean discarding metrical principles but substituting new ones. Often the conventions on which these new principles rest, such as lineation itself and its relation to syntactical rhythms, are at once less obvious (less explicitly systematic) and more fundamental that the special conventions of meters. [p. 81]
And then later:
Meaning arises not from what the poems says, but from what it does and the doing that it represents. It cannot be reduced to either a content (a set of propositions) or a form, in the sense in which that word complements “content” — an achieved product, a static stature. Nor, indeed, can meaning be reduced to an accomplished combination or unity of form and content. We comprehend the poem only as a process, not as an object. [p. 85]
This gets to be so rarefied, and the air so thin, that some readers may need oxygen. One begins to notice with Hartman that there’s an inverse relationship between the thinness of the sand under his castle and the academese of his argument. By the time we get to the middle of the chapter, his attempt to describe anything like a prosody of “discovered form” has become so generic, general and diffuse as to be meaningless:
When rhythm renounces the support of abstract or independent systems — meter or isochrony — the basic principle of the line emerges and takes absolute control: Not time alone, nor accent alone, but a combination from among all the elements of sound and of sense must give the line some special twist to justify its individual existence. The details of its rhythm are discovered (by poet and reader) with what it says; they are “organically” united. [p. 92]
So, a combination of all the elements give the line some special twist? Hartman follows this up with the two short poems by William Carlos Williams. In both instances, Hartman’s observations are so specific to his own interpretation that it’s hard to see how any general prosodic conclusion can be drawn. Yes, we can go so far as to impute meaning in the symmetry or lack of symmetry between two scanned lines (which the poet may or may not have been aware of), but that’s nothing two readers are likely to agree on or even recognize.
This chapter is stronger. As with the previous chapter, he argues that the poem’s subject can shape the use or absence of meter. (This is the chapter where Hartman makes the useful distinction between vers libre and vers libéré.) Hartman focuses on vers libéré — the way a free verse poet can fuse elements of traditional poetry with free verse. To me, the most interesting passages in the chapter are not those by Hartman (who like a nervous Putzfrau spends his time fussily admonishing, correcting and revising the words and intent of deceased poets) but those of the deceased poets themselves, like T.S. Eliot:
The most interesting verse which has yet been written in our language has been done either by taking a very simple form, like iambic pentameter, and constantly withdrawing from it, or taking no form at all, and constantly approximating to a very simple one. Is this contrast between fixity and flux, this unperceived evasion of monotony, which is the very life of verse… We may therefore formulate as follows: the ghost of some simple metre should lurk behind the arras in even the ‘freest’ verse; to advance menacingly as we doze, and withdraw as we rouse. Or, freedom is only truly freedom when it appears against the background of an artificial limitation. [p. 112]
This passage by Eliot is like a breath of fresh air compared to Hartman’s abstruse and byzantine argumentation. But Hartman just can’t leave it alone. He has to tweak it. He writes that “Eliot’s ‘simple form’ is a traditional metrical one. Thus it must precede the poem, and in this sense it is more accurate to say that the poet withdraws from it rather than that he approaches it.” Why Hartman felt compelled to make this completely gratuitous observation is beyond me. What does it matter if the poet is “approaching” or “withdrawing” from meter? My only guess is that Eliot’s description rattles Hartman’s preferred sobriquet of meter as a “received form”, a term he pointedly uses in the very next sentence. The terminology has always seemed like a political one, and one can’t help sensing a chip on Hartman’s shoulder whenever he writes about traditional poetry –but I could be wrong. One wonders what Eliot would say if he could be roused from his slumber.
More importantly, as with the preceding two chapters, Hartman is the least convincing when he tries to portray the use of meter (or its approximation) as something like a convention that can guide our reading and understanding of a poem. He can write for instance, that
“The end of the passage [Burnt Norton] shows how the metricality of the fragments can control meaning most directly. It is Eliot’s evocation of meter that makes us shift stresses in the repeated phrases of the final two lines.
…will |not stay |in place,
Will not | stay still.
[And here is the larger portion from which these two lines are extracted:
Words move, music moves
Only in time; but that which is only living
Can only die. Words, after speech, reach
Into the silence. Only by the form, the pattern,
Can words or music reach
The stillness, as a Chinese jar still
Moves perpetually in its stillness.
Not the stillness of the violin, while the note lasts,
Not that only, but the co-existence,
Or say that the end precedes the beginning,
And the end and the beginning were always there
Before the beginning and after the end.
And all is always now. Words strain,
Crack and sometimes break, under the burden,
Under the tension, slip, slide, perish,
Decay with imprecision, will not stay in place,
Will not stay still.]
In other words, Hartman is claiming that Eliot’s “evocation” of meter is somehow enough to cause readers to shift the emphasis from will in the first line, to not in the second line. At this point, I’m almost feeling like I should concede the argument out of apologetic generosity (because I’ve been so unconvinced elsewhere) but, in truth, I can see no reason why, at this particular moment, any reader should be compelled to suddenly invoke the prosody of traditional poetry. As for myself, I read the second Will as emphasized, just like in the first occurrence. What would T.S. Eliot say? Well, guess what, we have a recording of T.S. Eliot reading Burnt Norton. I smell a smack down. I went looking for it as soon as I wrote this paragraph and after I had read it myself. Here it is:
T.S. Eliot begs to differ. He reads it the way I do, not the way Hartman does. (Or rather, I read Burnt Norton the way T.S. Eliot does.) Nothing so illustrates the limitations (if not failure) of Hartman’s prosody. Eliot’s evocation of meter makes us shift stresses? He states his prosodic opinion as though he spoke for all readers and as if his conclusion were self-evident (a habit of academics). He turns out to be wrong. (It’s one thing to speak for and correct poets who can’t talk back, but when they do talk back, it’s almost always trouble) If he can be wrong about this, then why are we to believe his assertions concerning Auden or WC Williams? We have no compelling reason because he has no established or compelling prosody (his authoritative tone notwithstanding).
An Unnecessary Distinction
Why does Hartman spend so much time trying to prove that free verse is the prosodic equal of traditional verse? The effort turns out to be wasted. Wouldn’t it have been better to discuss free verse on its own terms, without reference to traditional poetry and without attempting baroque redefinitions of rhythm and meter? The old prosodists were probably much better at it than Hartman.
The more obvious issue readers are likely to have with Hartman is the often near impenetrable opacity of his prose. He veers in and out of academese.
“A striving toward concreteness in language — the subordination of other linguistic processes to that of naming — is one corollary of the doctrine of the objective correlative. On the other hand, the poem’s method partly conflicts with the purpose of that doctrine, which is finally to facilitate communication between poet and reader by giving them a common ground. If sense in language inheres in the connections among units of sound, meaning inheres in the connections among units of perception, not simply in the units themselves. The poem’s linguistic fragmentation, besides emphasizing sound, tends also to atomize experience into isolated glimpses, and thus to fragment the meaning that the reader is asked to share.” [p. 154]
Right. One can only marvel at the irony of beginning this passage with the phrase “concreteness of language”. The first time you read this you need a Babel fish. As far as I’m concerned, it’s terrible writing. If an argument is clear and concise, then it will be made concisely and clearly. While Hartman makes some interesting and valid points concerning the uses of lineation in free verse, he fails to create the prosody he defined in the first pages of the book.