HOW I WRITE A POEM
I always start a poem by trying not to think—which is one reason why I prefer to write first thing in the morning, before I have eaten, when I am still partly in the associative mode of dreams, and the cares of the day have not yet taken hold of my mind. Often I am deeply groggy when (coffee in hand) I sit down at my desk, and it sometimes astonishes me that I can write at all, given how utterly incapable I would be of talking to another person.
My goal is to sink into that part of my mind where inspiration seems to arise of its own, without the influence of my will. I don’t want to bully inspiration with expectations or ideas; I want only to be a conduit for whatever it has to offer. But, of course, I can never really blot out thought, so I try to begin writing as soon as my computer cursor starts to flash, before any of my expectations or anxieties can develop full force.
As I type, I do two things simultaneously—although in some ways they are the same thing: I try to write whatever comes into my head without thinking about it, and I try to put only those things that surprise me onto the page. The reason I see these two activities as the same thing is that what surprises me is what I have not anticipated or intended—that is, what I have not thought about in advance. And, as this thing is not already established in my mind, it is less likely to be a cliché, and more likely to be a revelation, both to me and to my reader.
Poetry (like all writing) is about revelation, about showing readers the world in a way they have never seen it, or about giving them a perceptual, emotional, intellectual or linguistic experience they have never had before, and then doing that over and over again, with every new word, new line, new sentence, and in the work as a whole.
Something else I am alert to as I write is the sound of the words: to the rhythms of rhyme, alliteration, consonance, assonance and stress, and to the textures of syllables. The way the words sound is part of the revelation they deliver—at least if it is not predictable, if it comes as a surprise (though, perhaps, only a very small surprise). But my attention to sound is entirely unconscious. I never think about sonic patterns and textures as I write. I only discover that I have created them after I look at what I have written, and often I am amazed by how complex they turn out to be.
I just keep on writing like this—as unconsciously as I can—until I either get tired or I feel that I have reached a conclusion or climax, or have created some sort of unity. Then I look back over what I have written, and realize, of course, that almost all of it is garbage. I make the finished poem by finding those bits amid the garbage that seem redeemable, that seem fresh and true, and that seem to have some sort of natural connection to other passages. I put the best bits together in whatever order makes sense, and then spend a long period—days, at the very least, but usually weeks and sometimes months—trying to amplify and clarify meanings, emotions and sonic effects until, all at once, the assembled bits turn out to be an actual, finished poem.
It might seem that the poems coming out of this process would tend more toward the surreal or Language Poetry end of the spectrum, and it is true that I like poems that remain mysterious and that are filled with unexpected, and sometimes not entirely comprehensible juxtapositions—just as I love music that sometimes approaches and even crosses over into noise. But that said, my poems often end up making fairly straightforward statements, and can be entirely realistic evocations of moments in my life or of nature. I try not to be doctrinaire about poetry. I don’t see the point of turning my back on any form of beauty or any way of making meaning. I think it is wonderful that we can do so many different things with words.
I read my poems aloud over and over and over as I polish them. I try to feel them on my tongue and in my ears. I try always for the most pleasing sounds and rhythms, ones that—ideally—amplify the feeling and/or significance of what I have written. Sound patterns matter immensely, but still, I avoid all regular rhymes and meters. I feel that subjecting my imagination to any pre-existing pattern diminishes the precision, subtlety, truth and imaginative elasticity of what I write. This is purely a personal preference, however, one that—when I am reading Keats and Yeats, or Frost and Auden—I am sometimes tempted to abandon.
I work very hard never to fake meaning, that is, never to pretend that something is meaningful when it isn’t, and never to say anything to my readers that I don’t think is true, or that doesn’t seem to express, in fact, something I feel passionately. Oddly, this is not difficult to accomplish, because I find that my unconscious (which is only interested in the things that matter most, in those things I don’t understand but need to) has very little time for bullshit. Often, all I am really doing as I write and edit is discovering what my unconscious is trying to tell me.
It is probably true that ninety-nine percent of the time I spend on a poem I am only tinkering: trading this word for that, monkeying with line brakes, and trying always to make the poem more concise, more true, more clear and more surprising. Surprise is of the essence, of course, but not merely for its own sake. My goal is always to surprise readers with some kind of truth—though it could be a very small truth, say, about the smell of dead leaves after a rain.
The clarity I am talking about is more clarity of image. I avoid vagueness as much as I can. Had I written above “a small truth, say, about the perception of nature,” that would have been vague. When I say “the smell of dead leaves after a rain,” I am giving my reader a perception of nature, one that I hope is clear, evocative and—in the context of the poem, at least—significant.
We don’t live in the general. We live only in the specific: in a specific instant, a specific place, with a specific history, and amid very specific sights, sounds, feelings, etc. Writing that imitates such specificity feels more like life and so is more vivid, and brings readers closer to the world.
Although I talk about clarity and truth, I never intend the meanings of my poems to be obvious or simple. On the contrary, I think that poems must always be mysterious to a considerable degree, that what meaning they have (and the term “meaning” itself if fundamentally mysterious), should only filter out through repeated readings and through reconsideration of the poem over time, and, even then, should never be wholly accessible.
When a poem is mysterious, readers have to open themselves up to it, which is to say they have to open themselves up to surprises, to receiving things they don’t already have. In a way, they have to get into the associative state I am in when I write, which means they are opening themselves up to their own minds, to language, and to the world. Also, when readers are required to actively engage with a poem (rather than passively receive it), the experience is more intense, and they are more likely to retain what the poem has to offer… or, at least, I hope they do.
So that’s it: I write and write, then I tinker, tinker, tinker, tinker—hoping that I am making the poem a better poem, and making myself—over time, at least—a better poet.
One final thought: When I was an undergraduate, I took a class with Kenneth Koch, who once suggested that the best way to test the quality of a new poem was to imagine reading it to your girlfriend. At the time, consummately insecure about both the quality of my writing and the impression I was making on women, I found this suggestion utterly terrifying. But it stuck with me, and, over the years, it has come to seem increasingly wise, if not entirely practical. One thing I like about Koch’s recommendation is that it encourages us to think of poetry as an exchange between one human being and another, or as a dialogue between two hearts and two minds. But, more importantly, it encourages us to think of a poem as a gift to our readers. I don’t believe we should ever write because we want something from our readers—their admiration, their pity, or even their money. We should write only to give our readers something they want or need—a fresh way of seeing the world, an unexpected twist of emotion, a new idea, or just a good time. And I believe that this gift is all the more valuable because it is one we give to strangers, as well as to those whom we love.
I always start a poem by trying not to think—which is one reason why I prefer to write first thing in the morning, before I have eaten, when I am still partly in the associative mode of dreams, and the cares of the day have not yet taken hold of my mind. Often I am deeply groggy when (coffee in hand) I sit down at my desk, and it sometimes astonishes me that I can write at all, given how utterly incapable I would be of talking to another person.
My goal is to sink into that part of my mind where inspiration seems to arise of its own, without the influence of my will. I don’t want to bully inspiration with expectations or ideas; I want only to be a conduit for whatever it has to offer. But, of course, I can never really blot out thought, so I try to begin writing as soon as my computer cursor starts to flash, before any of my expectations or anxieties can develop full force.
As I type, I do two things simultaneously—although in some ways they are the same thing: I try to write whatever comes into my head without thinking about it, and I try to put only those things that surprise me onto the page. The reason I see these two activities as the same thing is that what surprises me is what I have not anticipated or intended—that is, what I have not thought about in advance. And, as this thing is not already established in my mind, it is less likely to be a cliché, and more likely to be a revelation, both to me and to my reader.
Poetry (like all writing) is about revelation, about showing readers the world in a way they have never seen it, or about giving them a perceptual, emotional, intellectual or linguistic experience they have never had before, and then doing that over and over again, with every new word, new line, new sentence, and in the work as a whole.
Something else I am alert to as I write is the sound of the words: to the rhythms of rhyme, alliteration, consonance, assonance and stress, and to the textures of syllables. The way the words sound is part of the revelation they deliver—at least if it is not predictable, if it comes as a surprise (though, perhaps, only a very small surprise). But my attention to sound is entirely unconscious. I never think about sonic patterns and textures as I write. I only discover that I have created them after I look at what I have written, and often I am amazed by how complex they turn out to be.
I just keep on writing like this—as unconsciously as I can—until I either get tired or I feel that I have reached a conclusion or climax, or have created some sort of unity. Then I look back over what I have written, and realize, of course, that almost all of it is garbage. I make the finished poem by finding those bits amid the garbage that seem redeemable, that seem fresh and true, and that seem to have some sort of natural connection to other passages. I put the best bits together in whatever order makes sense, and then spend a long period—days, at the very least, but usually weeks and sometimes months—trying to amplify and clarify meanings, emotions and sonic effects until, all at once, the assembled bits turn out to be an actual, finished poem.
It might seem that the poems coming out of this process would tend more toward the surreal or Language Poetry end of the spectrum, and it is true that I like poems that remain mysterious and that are filled with unexpected, and sometimes not entirely comprehensible juxtapositions—just as I love music that sometimes approaches and even crosses over into noise. But that said, my poems often end up making fairly straightforward statements, and can be entirely realistic evocations of moments in my life or of nature. I try not to be doctrinaire about poetry. I don’t see the point of turning my back on any form of beauty or any way of making meaning. I think it is wonderful that we can do so many different things with words.
I read my poems aloud over and over and over as I polish them. I try to feel them on my tongue and in my ears. I try always for the most pleasing sounds and rhythms, ones that—ideally—amplify the feeling and/or significance of what I have written. Sound patterns matter immensely, but still, I avoid all regular rhymes and meters. I feel that subjecting my imagination to any pre-existing pattern diminishes the precision, subtlety, truth and imaginative elasticity of what I write. This is purely a personal preference, however, one that—when I am reading Keats and Yeats, or Frost and Auden—I am sometimes tempted to abandon.
I work very hard never to fake meaning, that is, never to pretend that something is meaningful when it isn’t, and never to say anything to my readers that I don’t think is true, or that doesn’t seem to express, in fact, something I feel passionately. Oddly, this is not difficult to accomplish, because I find that my unconscious (which is only interested in the things that matter most, in those things I don’t understand but need to) has very little time for bullshit. Often, all I am really doing as I write and edit is discovering what my unconscious is trying to tell me.
It is probably true that ninety-nine percent of the time I spend on a poem I am only tinkering: trading this word for that, monkeying with line brakes, and trying always to make the poem more concise, more true, more clear and more surprising. Surprise is of the essence, of course, but not merely for its own sake. My goal is always to surprise readers with some kind of truth—though it could be a very small truth, say, about the smell of dead leaves after a rain.
The clarity I am talking about is more clarity of image. I avoid vagueness as much as I can. Had I written above “a small truth, say, about the perception of nature,” that would have been vague. When I say “the smell of dead leaves after a rain,” I am giving my reader a perception of nature, one that I hope is clear, evocative and—in the context of the poem, at least—significant.
We don’t live in the general. We live only in the specific: in a specific instant, a specific place, with a specific history, and amid very specific sights, sounds, feelings, etc. Writing that imitates such specificity feels more like life and so is more vivid, and brings readers closer to the world.
Although I talk about clarity and truth, I never intend the meanings of my poems to be obvious or simple. On the contrary, I think that poems must always be mysterious to a considerable degree, that what meaning they have (and the term “meaning” itself if fundamentally mysterious), should only filter out through repeated readings and through reconsideration of the poem over time, and, even then, should never be wholly accessible.
When a poem is mysterious, readers have to open themselves up to it, which is to say they have to open themselves up to surprises, to receiving things they don’t already have. In a way, they have to get into the associative state I am in when I write, which means they are opening themselves up to their own minds, to language, and to the world. Also, when readers are required to actively engage with a poem (rather than passively receive it), the experience is more intense, and they are more likely to retain what the poem has to offer… or, at least, I hope they do.
So that’s it: I write and write, then I tinker, tinker, tinker, tinker—hoping that I am making the poem a better poem, and making myself—over time, at least—a better poet.
One final thought: When I was an undergraduate, I took a class with Kenneth Koch, who once suggested that the best way to test the quality of a new poem was to imagine reading it to your girlfriend. At the time, consummately insecure about both the quality of my writing and the impression I was making on women, I found this suggestion utterly terrifying. But it stuck with me, and, over the years, it has come to seem increasingly wise, if not entirely practical. One thing I like about Koch’s recommendation is that it encourages us to think of poetry as an exchange between one human being and another, or as a dialogue between two hearts and two minds. But, more importantly, it encourages us to think of a poem as a gift to our readers. I don’t believe we should ever write because we want something from our readers—their admiration, their pity, or even their money. We should write only to give our readers something they want or need—a fresh way of seeing the world, an unexpected twist of emotion, a new idea, or just a good time. And I believe that this gift is all the more valuable because it is one we give to strangers, as well as to those whom we love.