publisher colophon

Is a poem equivalent to the particular words of which it seems to be made? I argue that it is not. If we take a poem as constituted by its particular words in their particular order, we produce a surprising (often ignored) logical pressure on how those words can function and mean. If we follow out this particularity, we discover that poems are odd logical constructs. The more explicit we attempt to make the logical particularity of a poem, the more logically strange it becomes. In fact, if poems are logically constituted by what they say, and not by what they mean, they become parodies of language: faux language.


A poem is not particular in the way a painting is particular. A copy of a poem is still the poem, while a copy of a painting is not the painting. But a poem is still particular, since it seems to be constituted by a specific set of words in a specific order such that to alter that order or any of those words is to make a new poem. Marianne Moore begins her poem “An Egyptian Pulled Glass Bottle in the Shape of a Fish” in the following way:

Here we have thirst and patience, from the first, and art, as in a wave held up for us to see in its essential perpendicularity1 [End Page 416]

She might have revised this (she often radically revised her poems) and replaced “perpendicularity” with “verticality.”

Here we have thirst and patience, from the first, and art, as in a wave held up for us to see in its essential verticality

This says the same thing, but it is not the same poem. The particularity of a poem, relative to the words that make it up, constitutes a minimal description of that poem, a description independent of almost all aesthetic ideologies and philosophical commitments. The point is not that poems, like everything, are self-identical, but that for poems made of words, what matters is not what is meant but what is said.

The particularity of a poem motivates various aesthetic ideas. For example, Malcolm Budd begins his account of poetry by recognizing the particularity of a poem relative to its words. He writes, “The value of a poem as poem does not consist in the significance of the thoughts it expresses, for if it did, the poem could be put aside once the thoughts it expresses are grasped.” This fact prompts him to suggest that what matters to us is “the imaginative experience [we] undergo in reading the poem, not merely the thoughts expressed by the words of the poem.” This is not a necessary conclusion, but it a very common way of understanding the value of reading a poem, which would consist, then, “in an awareness of the words as arranged in the poem.”2

Many literary critics and a number of poets imagine that the particularity of a poem allows for the expression of new kinds of thoughts (either ineffable or grasped in some putatively non-normative way). Poems are called particular and not general, and thus poems become situated in arguments about knowledge (which is general) and freedom (often understood as particular). I want to turn aside from these aesthetic ideologies and theories in order to investigate the logic of a poem’s particularity. If we take a poem as constituted by its particular words in their particular order, we produce a surprising (often ignored) logical pressure on how those words can function and what they mean. If we follow out this particularity, we discover that poems are odd logical constructs. The more explicit we make the logical particularity of a poem, the more logically strange it becomes. In fact, if poems are logically constituted by what they say (their particular words) and not by what they mean, they become parodies of language: faux language. [End Page 417]


I understand what each sentence in Middlemarch (or any novel) means, but by virtue of that do I know what Middlemarch as a novel means? Clearly not. Because we can know and understand the specific meaning of each and every sentence of a novel, we do not by virtue of that know and understand the meaning of the novel. The meaning of a novel is not a concatenation of the meanings of its sentences (or phrases), partly because a sentence has a different kind of meaning than does a novel. We can replace “novel” with “story” and make the same argument.

I can make my argument more general. If I were asked to describe my day in ten sentences, I could. And those ten sentences could be perfectly intelligible. If I were then asked what my day means, or rather, what my account of my day means, I might legitimately not know. I might not even know what would count as a meaning. This is seemingly the kind of meaning we are after when we talk about the meaning of a novel or a story. We could reduce these ten sentences to one sentence, but still the meaning of that sentence need not be the meaning of my day or of my account of it in that one sentence. The sense of “meaning” is different in each case. And thus, the sentence, what it says and means, is not equivalent to the story; it is something different than the story it tells.

One can make the same argument about poems: that knowing the meaning of the phrases that make up a poem does not mean that one understands what the poem means, suggesting again that this second use of “meaning” signifies something different than the first. But a problem arises in extending the argument to poetry. A poem need not be a story. And if a poem does tell a story, that does not mean the poem just is that story, which could, of course, be told nonpoetically. Before we can extend the argument from story to poem, we therefore have to show that they are similar kinds of things (which seems unlikely), or we would have to show that a poem is not equivalent to the phrases and words of which it seems to consist.3 With this last possibility, I have returned to the question of how a poem is defined by the words that seem to make it up. Maybe a poem just means what its sentences mean, especially if it tells no story, but expresses some emotion or makes a statement of some kind. I can reformulate this challenge into a simple question: is a poem (however defined) equivalent to its sentences? Or, to make it clearer and more difficult: if a poem consists of one sentence, more or less intelligible, is that poem equivalent as a thing and in meaning to that sentence? [End Page 418]

Before I answer that question, I will examine the easier case of a poem made of multiple sentences (or phrases). If the relationship among the sentences and phrases is unclear, ambiguous, and multiple, then those disjunctions and their effects prevent any equivalence between those phrases and the poem of which they are a part. The poetry of John Ashbery offers many examples, one of the more extreme being “A Last World”:

Yet having once played with tawny truth Having once looked at a cold mullet . . . He wished to go far away from himself. There were no baskets in those jovial pine-tree forests, . . . In that foam where he wished to be.4

Is playing with the tawny truth the same as looking at a cold mullet? Who is the “he” that wished to go far away from himself? Are we relieved that there are no baskets in those jovial pine forests? Maybe, whoever the “he” is, being far away from himself is being in some kind of foam; that sounds like a joke. The poem, in effect, includes these questions and the problems of sense they reveal. The poem cannot be the sentences because we do not know how these sentences fit together as thoughts, except as possible thoughts (or effects) in a strange thing called a poem. The disjunctive character of these sentences allows us to take the poem as separate from the sentences that make it up.

But poems are not all of one kind. A more challenging case would not be a poem of fragments, but a poem of one sentence, the sentence perfectly intelligible, thus supporting the idea that the meaning of the sentence would be the meaning of the poem. Ben Jonson writes a poem in praise of Benjamin Rudyerd, consisting of a single intelligible, if complex, sentence:

If I would wish, for truth, and not for show, The agèd Saturn’s age, and rites to know; If I would strive to bring back times, and try The world’s pure gold, and wise simplicity; If I would virtue set, as she was young, And hear her speak with one, and her first tongue; If holiest friendship, naked to the touch, I would restore, and keep it every such; I need no other arts, but study thee: Who prov’st, all these were, and again may be.5 [End Page 419]

Thus, the poem forms a clear argument: If I would wish for truth, if I would recover the good of the past, if I would try the world’s purity and wise simplicity, if I would return to the uncorrupt virtue of our nativity, if holiest friendship were restored without disguise and dissemblance, then you (Rudyerd) would be the person I could study to show what these were and could be. The repeated phrase “if I would . . .” is part of a rhetorical fiction that situates the speaker within the degradation he would wish to overcome.

Why not say the poem means what the sentence means, and in the same way? If I utter the sentence in conversation, even with its cadence and ambiguities (“try” in line three), the poem will not sound like a poem; I will be accused of being pretentious, making speeches. But I could say it in praise of my admired friend, Benjamin Rudyerd. I could say it as a general comment on our degraded society. I could say it. Or write it in a letter. I could send it to anyone as a comment on how things are and how I wish they would be.

If a poem is simply highly rhetorical language, organized according to particular notions of genre, diction, and modes of address, as it seems in this case, then poems would not have any special meaning beyond their constituent sentences. They might have effects (like any rhetorical statement), producing feelings, engendering associations, suggesting allusions, and so on. But this just means that some other sentence saying the same thing would mean the same thing, although it might have a different set of rhetorical effects (or similar effects)—in which case, the initial poem would not mean what its sentence means. It in fact would not mean at all—only its sentences would mean. There would be no poem. Until we say how a poem is different than a sentence, it is a sentence and not a poem.

If we ask how Jonson’s poem-sentence is more than a sentence, we might point to its rhetorical effects and putative allusions, its implications and associations. None of these, however, need to be understood as poetic to say they make a poem. They can just be clever ways of speaking. Nothing in the sentence makes it a poem; only our taking it as such makes it that—and when we do that, we take it as not simply a complex statement that could have been written in a letter to Rudyerd but as a complicated and subtle poetic performance that is part of a particular social and moral context, and given its particularities of form, is separable from that context, recoverable for us as an attitude, as a set of thoughts about both morality and poetry. The sentence is not just a sentence when we take it as a poem, which we have to justify in some way. This may not seem earth-shattering, but in effect it makes a poem [End Page 420] not only contingent and undefinable, relative to the words of which a poem is made; it means that there is no integral relationship between poems separate from our construing them relative to each other, bound to our ideas of poetry. And if we take a sentence as the poem, then we take it as meaning more than what it means as a sentence, but that is to step away from sentence meaning and take it as meaningful in some other way. What is this other way of being meaningful?

Jonson’s words are equivalent, as they stand, to a sentence that could be spoken nonpoetically. For example, one could say the same words in a speech, not as a quoted poem but as prose, in which case we might still infer that the point was to praise Rudyerd as a way of criticizing more common behavior, the form itself giving grandeur to the wit of criticizing by praising. Or we might embed the sentence in an oration with other goals and motives. In both cases, we would evaluate the words relative to their effect and our putative goals. We might conclude that the words were perfectly chosen. The same poetic effects would be present in the speech, but none of these would turn the sentence so presented into a poem, even if we found the sentence had a special kind of interlocking integrity of effects.

The words of the speech, of course, constitute the oration as what it is. A speech is formalized, after all—and is also defined by the words of which it is constituted. This last is to say that everything is identical to itself, which is not interesting in this case. The particularity of the words, in a speech, is justified relative to the point of the speech and to its effect.

This is a general point. The particularity of words (or of forms) has to be justified relative to some criteria. In many cases, the criteria constitute the particular language practice as generally understood, such that there can exist broad agreement about what those criteria are. This is the case with orations and, in many ways, with stories. It is not the case with poems.

One can react to a poem without understanding the criteria by which to judge its words as necessary, but one cannot justify those words without having a set of criteria by which to do that. When this is not the case, when the criteria are contested and thus subject to some prior commitment or investment, then either the aesthetic manifested by the poem (and our commitment to it) or the words in themselves count as the criterion of justification. In the first case, a poem is just what we take as a poem; there is no such thing as poetry separate from our ideas about it. In the second case, when we take the words themselves [End Page 421] as constituting the criteria by which the poem is defined as what it is, the poem is self-defining and is understood separate from effects and from any justification we might offer for those words.

The poem is equivalent to the words that make it up or it is constituted by the idea of poetry that turns its words and phrases into more than words and phrases. In the first case we have a sentence, not a poem, and in the second case we have an idea and not a poem. We seem to have a problem: we can’t find the poem.

To ask what Jonson’s poem “To Benjamin Rudyerd” is about—and to answer “Rudyerd” or “the moral degradation of society”—will not be enough to capture the content of the poem as a poem. If Jonson’s praise of Rudyerd is more than a rhetorical exercise, more than a way of making it more vivid and memorable, then the poem-sentence has some significance beyond what it says. Why write a poem of praise? Why not send Rudyerd a letter, or comment about him in conversation? And, of course, how you write the praise matters. Figuring out why is to go beyond what the sentences mean—but to do so is also to leave behind the idea that the poem means (at least in the way sentences do).

A poem qua poem is not logically distinct from its sentences in the way fictions and stories are. For a poem to mean more and other than a sentence might, it must be other than a sentence, even if it is a poem of one sentence. Poets can create logical complexities of ambiguity, disjunction, distortion, and nonsense. None of these, however, necessarily make a poem but rather prompt a challenge to make sense of the phrases so de-formed in themselves and relative to each other. To take the complexities of form as part of a poem, a poetic whole, is a further idea, and one we bring to the poem. Our attitudes and beliefs (ideas) give a sentence a promise greater than what the sentence offers as a sentence. Unlike fictions and stories, poems are conceptually other than their sentences in contingent ways, dependent on our attitudes and beliefs.6


In ordinary communications, I can fail to say what I mean (and fail to mean what I say). I can also mean (assert or state) the same thing using different, logically synonymous sentences. In all of these cases, there is no necessary relationship between what is said and what is meant; we can say and mean whatever in multiple ways. The particularity of a poem, on the other hand, seems determined by the particularity of its words. [End Page 422] We can revise a poem, alter it significantly, but when we do so, we have just made another poem with a variable relation to a previous version. There is no ontological mystery in this. A poem is a string of words in a particular order in which that order defines that poem as that poem.

Can this particularity of words of which the poem seems to consist function as the primary criterion for its existence and recognition? (These are two different things, but they are strongly related in the case of a poem.) We might balk at using the particularity of words that make up a poem in this way, since we might describe any sentence in a similar way. If we understood a sentence simply as the words that make it up, then of course it would be self-identical. But a sentence is not simply the words that make it up (just as a word is not simply constituted by its orthographic marks or its phonemes). A sentence also involves in some essential way what it expresses, its meaning, reference, significance, and so on. (We can be agnostic about this for my argument.) A poem is somehow different from how we ordinarily use sentences. If we take its particular words in their particular order as constitutive, we do so by refusing to understand another sentence as equivalent (in the relevant ways) to the original poetic sentence. And thus we would be taking a poetic sentence as not equivalent to a proposition that might be stated in some other way.

If such is the case, then the determinate form of the poem underdetermines the putative meaning of the poem. What are the consequences and significance of the asymmetry between the determinacy of form and the consequent underdetermination of meaning? In other words, if we take seriously the possibility that the words of a poem constitute it in this radical way (so that the meaning of a sentence would be radically underdetermined relative to those words), then what does that show a poem to be? To answer this question, we can set aside the complexities of poems, and simply examine the logical consequences of treating a simple sentence as if it were a poem, constituted by its words such that there would exist no equivalent sentence in the same language. I will use the sentence “The bookcase is filled with objects.”

“The bookcase is filled with objects” means that there exists a bookcase such that it is filled with objects. We can find many examples of such bookcases, although there might be some questions about what “filled with” means, i.e., how many objects on a bookshelf count as “filled with objects.” This does not mean that “filled with objects” has an indeterminate meaning but rather that its vagueness underdetermines the meaning in such a way that it can become problematic. If we have [End Page 423] no way to adjudicate what is meant (because, say, the person who said this is dead), then we can guess or have some general sense of what “filled with objects” means; but in so doing we further underdetermine what the sentence means (again, the meaning is not indeterminate but is generalized among various possible judgments).

We could take (in the relevant circumstances) the new sentence “The bookcase is filled with things” as logically equivalent to our previous sentence: “The bookcase is filled with objects.” Both sentences could be used to refer to the same bookcase, and the truth of the sentences could be determined in the same ways (beset by the same ambiguities). If, however, we insist that in changing “objects” to “things” we have changed the meaning, then we are claiming that the sentence is not about any particular bookcase filled with objects, since we would either be suggesting that there would be no other way of referring to such bookcases except with this sentence (which is nonsensical) or we would be using the sentence in some new and mysterious way (which may be the hope of the speaker).

If the object-sentence and the thing-sentence cannot be used as synonyms in the way we would ordinarily use them, that is, if they cannot express the same proposition, either there is some special distinction being made about things and objects, or else these sentences cannot pick out what they ordinarily do anymore. If we mean what we ordinarily do by “objects” and “things,” and if these cannot be used as synonyms, then sentences have lost the world.

There is an alternative: if the object-sentence and the thing-sentence are seen as not synonymous and having no referential overlap, we might understand them as names, in which case each sentence would attach to some particular bookcase. If we take the sentence as a name, it has a determinate referent, but no meaning. If we do not take the sentence as a name, and we hold to the necessary particularity of the sentence, then it is no longer a sentence about bookcases, but would express some special and specific thought that we have yet to determine.

What kind of thought? If we insist that the sentence be treated like a poem, such that the thought about bookcases filled with things is different than the thought of one filled with objects, then we might imagine that by so insisting on the difference, we make the thoughts expressed more particular in each case: object-thoughts as distinct from thing-thoughts. The issue for me is not whether these are different thoughts but whether by making a specific set of words essential to the thought, such that just those words are the thought, we make that thought (the meaning, in this case) more or less determinate. [End Page 424]

We cannot grasp the difference between object-thoughts and thing-thoughts by example, since we lost our examples once we made the sentence determinate in form (again, unless we treat it as a name). Even if we used examples to differentiate between these two thoughts, we would require further explanations, and not just a general explanation. We might argue that the word “objects” marks a special category of stuff (decorative stuff, for example), but that the word “things” describes anything—paper, old tickets, coins. Such an explanation would allow for cases of overlap (objects are always things, but not all things need be recognized as objects), and, since these are thoughts (and not statements about the world per se), the only way to express our thoughts would be by further talk, open to further ambiguities. In effect, even if the thought putatively expressed or produced by the sentence had a particular, repeatable, and thus determinate form for me, I could not know it, nor could you, without explanation—all of which means that the thought would be logically dependent on the explanation. It would be logically dependent because the truth or falsity of the sentence, let alone its meaning, could not be determined separate from these explanations, all of which would be subject to various ambiguities. The thought and the sentence would not be and could not be determinate.

Therefore, a sentence, if radically dependent on such explanations (since no example will count as decisive), would logically underdetermine our explanations, and thus the thought it would express. Consequently, the sentence could not differentiate between various thoughts I might have, all of which could be prompted by the sentence, nor could it differentiate, in itself, among various thoughts it might be taken to be about. My concern is not with how the sentence means, nor does my argument require a decision about the logical priority of sentences or mind-states relative to our general linguistic understanding (an important question in the philosophy of language). The initial premise of my thought experiment is that the sentence, in its particularity, has logical priority relative to any thought or content we might ascribe to it. This priority is what makes it like a poem.

The more specific and particular and necessary the words are, the more difficult it is to support the idea that a specific meaning to those words could be correct or incorrect. Instead, the specific sentence has possibilities of meaning. In effect, a sentence taken as necessary in its particular form cannot be necessary relative to its meaning, since it has only possible meanings. We might, therefore, construe the sentence as not like any ordinary sentence we might use, but as like a word (the [End Page 425] sentence as a single word). Under such a construal, the sentence would, like a word, have possible meanings, not a determinate meaning. Words gain specific sense within the context of a sentence. A sentence construed as a poem has a sense within the context of my interpretations of it, which is to say my interpretations have meaning, the sentence (so construed) does not; it is merely meaningful. It might be meaningful in the way an action or event can be meaningful. This argument is not an argument for any general indeterminacy of meaning, either Quinian or deconstructive, but for the peculiarities that follow from making the form of a sentence primary and not what we might mean by means of it.

Although mathematical statements are not equivalent to sentences, I can illustrate the effect of an assertion of the radical uniqueness of a sentence using mathematical formulae. The situation is like this. I believe 3+3 = 6, and I take that statement as radically unique. Consequently, I deny that it is equivalent to 3×2 = 6 and I deny that it is equivalent to 4+2 = 6. If so, what can I mean by “6,” let alone by “3+3 = 6”? By so asserting, I have removed 3+3 = 6 from mathematics, I have denied addition, equality, and the relationships of numbers. I must either imagine some special, unspecified quality or meaning to 3+3 = 6, or what I mean by that statement is nonsense. Similarly, a unique sentence, if it could be uttered or read, would be displaced out of language, becoming a faux sentence, like a mask is a faux face.

Thus, any ordinary sentence—“this bookcase is filled with objects”—if it has no relevant equivalent (synonymous) statement that, relative to its truth value (or whatever rubric of content one wants to appeal to), could mean the same thing, then such a sentence would no longer express a thought; it would become a target for interpretative speculation. If an ordinary sentence is understood as articulating a unique thought (according to whatever theory), and if no other sentence, no other logical structure can be the means of articulating that same thought, even if the sentence or thought seems ordinary to us, then we can only understand that thought by grasping not only what we take it to say but by excluding any other single equivalent thought (no sentence could say the same thing, under some reasonable interpretation of “same”).

It might seem to express a thought, but since it has to be a unique thought, it would also have to exclude, and be in its content different from, any other seemingly equivalent thought-statement. (This follows from taking the words as decisive.) Such a sentence would mean something we could never grasp, and it would cease to function as a sentence within our language. Such a unique thought-sentence might be [End Page 426] meaningful in some aesthetic way, relative to some theory or interpretation, but it would no longer be a sentence. It would be a faux sentence, whose sense we do not know (and about whose justification we could only speculate). To read “this bookcase is filled with objects” under these special conditions would be equivalent to understanding the words but not what was meant (or could be meant) by the sentence. The sentence “this bookcase is filled with objects” would become nonsense, although it would still look similar to an ordinary sentence. To read the unique kind of sentence would require that we find and then reject all other possible synonymous-seeming sentences, which would then reveal the sentence as nonsensical.


I can now collect my arguments about the attempt to allow the particularity of words to constitute a sentence, as those words might constitute a poem. My arguments lead to three conclusions, three different ways we can respond or construe such a unique sentence: (1) we can exclude it from the language; (2) we can treat it as a parody of a name; or (3) we can understand it as analogous to words. If the sentence is excluded from the language, it is simply to say it is nonsense, or to imagine some occult sense of meaning. If the unique sentence is construed as a name, it then has a referent and not a meaning. Reading such sentences, and such poems, would be like reading the phone book; it would be a mere name of whatever we might take it to stipulate. If we take the unique sentence (the entire sentence) as analogous to a word (a single word), we would understand it as having possible meanings, not yet specific meanings. Such a sentence would not be meaningless; rather, it would be meaningful in ways that have to be determined. Such a sentence might lack any semantic sense, but could be meaningful in ways analogous to how actions or events can be meaningful (which is not to say that it need be understood as an action or event). In other words, how such a sentence-poem would be meaningful (not what it semantically means) is what we have to determine through our reading of it.

A poem, if it is constituted by its words, becomes an analogous structure relative to ordinary language, and not only if we understand it as analogous with a word, as in option three above. If we understand the poem as excluded from language, for example, then it looks like an example of language, but isn’t; it is nonsense. To see the poem as a parody of a name, since it would never be clear what it named, is again [End Page 427] to see it as analogous with a name. And, of course, if we take a poem as an analogue for a word, it is analogous with one aspect of language. In all of these possible cases, the poem has become a parody or caricature of the sentence of which it is made, and thus of language itself. A poem can look linguistic, but it isn’t quite. If it is defined by the particularity of its words, then it has become a parody of a sentence and of language. So what a poem is is already an analogy, and this before we make any sense of it, and discover (by analogic means, of course) other analogies it might offer.

When we read a poem as if it were constituted by its words, it is as if we are mimicking ordinary reading, imagining that the words we read are ordinary words. But if we also insist that those particular words constitute the poem, then the words in the poem only approximate ordinary words. We read them that way because we have no other way to read, but we must then remember: “These are not really words but only like words, I must decide what else to do with them.” With this reminder in mind, we can then offer an interpretation of the specialness of the words, again modeled often on what we know of language, but also relative to our ideas of poetry. The oddness of what a poem is (and of where it is) requires that we figure out what to do with it. The poem becomes an ethical challenge—first given in the simple phrase “What should I do with this thing?” but then modulating into “What can this thing have to do with me?” I take this last question to be one of the gifts of poetry.

If we define a poem by its particularity (insisting that it is equivalent to the specific words of which it seems to be made), we find that this definition fails, but it reveals how poems, if understood as radically particular, pressure poetic words into analogic words. In so doing, our relationship to language shifts (or aspects of it become more visible). If we try to reduce poems to their words, we fail and find poems elsewhere; if we give up that attempt, then we begin with an idea that poems are elsewhere. No matter what poems are, they are elsewhere than their words.

The attempt to define the poem by its particularity displaces it elsewhere. A poem as a name refers to what it names, and thus the poem is elsewhere; if we take the name as without a referent, as a mere name, then it collects whatever we associate with it; and again the poem is elsewhere. If we imagine there is no adequate synonymous proposition for any particular poetic phrase, then we exclude that sentence (that poem) from the normative modes of language. In that exclusion we would have [End Page 428] to reject the poetic phrase as pointless, alter our own involvement with language, or find ourselves held by the phrase as a kind of nonsensical event. The poem would be elsewhere than language and we, too, might find ourselves elsewhere, in something like Wonderland. If a poem is like a word, then we must fit it into a sentence, what we call an interpretation. So again, the poem needs an elsewhere to be somewhere.

What we see in all three of these possibilities is that poems are dispersed, never more so than when we try to make them as particular as we can. They escape us. And yet what motivates the attempt to get the words to be the poem is the fact that those words, that particularity, claim us, constrain us. With poems we can find ourselves intimate with words whose sense we may not be able to fathom, and yet we are held by those words, by their form and promise. This is what poems do—they trade off determinate meaning for the intimacy of being held by the particularity of form.

Poems are in-between: a spoonful of intersubjective soup, and with each bite we eat the spoon. Poems are not in one place, and yet they are someplace: this is the riddle of poetry. But the words? Do specific words in a particular order make a poem? Nothing in the words makes a poem. If a poem is a set of phrases, the poem is not simply these phrases or even their meaning. If a poem were these, then a poem would simply be those phrases, each phrase like any other phrase. A poem—if it is a sentence—is always more than that sentence; otherwise a poem would not be a poem but a sentence.

If a poem is not an act of communication, which it need not be, and if it is not a mere illustration of a poetic idea, which it should not be, then it is natural to imagine that it is constituted by the words of which it seems to be made. If it is, however, then we find the poem displaced out of language, reduced to a name, or transformed into a word. In all three cases, when a poem is understood as constituted by its words (logically and thus essentially), it becomes faux language, analogous to the words and phrases that seem to constitute it.

A poem is more than its words and phrases because it is also analogous with those words and phrases. It is elsewhere than simply in its sentences or phrases or words. If the words are strange or the words make more non-sense than sense, then we can either normalize those words and phrases into sentences that make sense (and thus they are simply code for some other sentence) or we give up taking those words as giving us the poem in themselves. Their de-formation, like a parody, gives us analogies or demands theories of justification or relies on our reactions, [End Page 429] and thus the poem shifts elsewhere from the words that prompted it. A poem is potentially something somewhere, potentially itself, just not yet.

As are we.

Brett Bourbon
University of Dallas


1. Marianne Moore, The Complete Poems of Marianne Moore (New York: Penguin Books, 1981).

2. Malcolm Budd, “The Uniqueness of Poetic Value,” Values of Art: Pictures, Poetry and Music (New York: Penguin, 1995), p. 83. For Budd, the particularity of the poem matters because that particularity produces (or facilitates) the poetic experience. The thoughts of the poem could be said in another way, but those thoughts do not constitute the poem—the words do.

3. The meaning of the sentences that make up a story should be distinguished from the meaning of a story, and that would be true as well for whatever story might be told or shown or expressed through a lyric poem (let alone a narrative poem). The meaning of a story, however, is different in kind from the meaning of a poem qua poem.

4. John Ashbery, Selected Poems (New York: Penguin, 1986).

5. Ben Jonson, The Complete Poems, ed. George Parfitt (New York: Penguin Books, 1996).

6. These attitudes are not aesthetic attitudes of the kind criticized by George Dickie; they are simply a disposition to take something as a poem. See George Dickie, “All Aesthetic Attitude Theories Fail: The Myth of the Aesthetic Attitude,” in Aesthetics: A Critical Anthology, ed. George Dickie and R. J. Sclafani (New York: St. Martin’s Press, 1977), pp. 342–55. [End Page 430]

Additional Information

Print ISSN
Launched on MUSE
Open Access
Back To Top

This website uses cookies to ensure you get the best experience on our website. Without cookies your experience may not be seamless.