to arrange my own pieces into a pretty diamond shape on the board, or in weighing whether, instead of trying to get the ball through the hoop, I ought to see how many cars in a row I could roll the ball under in the parking lot. In chess or in basketball, doing better what we were doing really would be the only meaningful possibility.
But writing is not chess or basketball, and in this respect at least, writing is not like chess or basketball. In writing, it is legitimate to seek to fulfill received standards, but also to ask after the available or possible standards, with the prospect left open that, upon deliberation, I might elect and enact standards that differ from those that previously I took as given. To quote Sen once more, this time from The Idea of Justice: “We can not only assess our decisions, given our objectives and values; we can also scrutinize the critical sustainability of these objectives and values themselves.”
One might distinguish, then, between two approaches to reflection on poetry. An ethics-related approach would ask about poetry’s motivations (posing such questions as “What ought a poem achieve?”) and about poetry’s effects (posing such questions as “What is a poem about?,” meaning both “What is a poem up to?” and “What is a poem speaking of?”). An engineering-related approach would ask after the techniques and processes that transform “normal” language into “poetic” language, prose into poetry. Those approaches host different questions about craft. The ethics-related approach suggests questioning along the lines of “What can I do now, that I could not do before?” The engineering-related approach suggests such questioning as “How can I do better what I am doing?” Sen believes that economics has been impoverished by keeping the ethics-related and engineering-related traditions separate and attending almost exclusively to the engineering tradition; analogously, poetry has been impoverished by keeping the two approaches separate and devoting much more attention to the engineering approach. Emphasis on the engineering approach shows itself in the unanimity, the givenness, of the sense that the point of an MFA program would be to help us as poets get better at what we do. But both approaches have validity. Both are necessary.
Another way to get at this point would be to assert that in the teaching of craft one ought to push students toward not writing better poems. The formulation is willfully perverse, but I mean by it that any concept of “better” presupposes an ideal. There may be enterprises for which the ideal is settled, such as chess and basketball, but poetry is not one of those enterprises. It’s why poetry matters more than basketball or chess: in poetry, the ideal is not given, but ever at stake. In our critical reception of works of art, we often acknowledge the variability of ideals. If, for example, I were to ask a cinephile which is the better movie, Taxi Driver or Standard Operating Procedure, she surely would respond, rightly, that the two films are trying to do very different things. You can’t say which is better until you specify what you mean by “better.” But our willingness to describe what we are doing in an MFA as “learning to write better poems” is analogous to asking which movie is better; it is an engineering approach that needs in complement an ethics approach.
If by “craft” we typically denote something analogous to Sen’s engineering-related approach (an attempt to internalize, and to replicate in our work, given poetic ideals and techniques), then it seems to me valuable to supplement “craft” with “metacraft” (a self-conscious questioning of ideals and techniques that keeps open, rather than foreclosing, the issue of what poetry is and what it might be and do). Let “metacraft” designate such a poetic practice, one in which ideals and techniques are not given once and for all, but remain ever at stake in the poem and for the poet.
In an essay, Robert Creeley reports having been told once by John Frederick Nims “a lovely story” about another poet’s having been asked, after a reading, “that next to last poem you read — was that a real poem or did you just make it up yourself?” The anecdote is funny because it reveals the limitations of (by reducing to absurdity) a certain understanding of what a “real poem” is. The questioner probably conceived of a “real poem” in terms given by what someone with formal education in literature might name “canonicity”: a poem, on such an account, is a literary artifact that has been preserved, and has had conferred on it cultural status of a sort that exacts reverence, because it was written by an historical figure long dead, and since deemed by relevant authorities (textbooks, teachers) deserving of the honorific “Poet.” Creeley and Nims could share a laugh over the story because their ideas of a “real poem” resembled one another more than either one resembled the idea of a “real poem” held by the questioner in the story. And Creeley can count on our laughing with him, because he can reasonably assume that any reader of his essays will think of a “real poem” in terms more like his own and Nims’s than like the questioner’s.
But.
If I grant the questioner his conception of a “real poem,” then the question stops being a false dilemma: something that “you just made up yourself” in fact couldn’t be a real poem. Consequently, far from being funny or absurd, it would be perfectly reasonable and appropriate to ask a reader which kind of thing he had just read. Similarly, if I don’t grant Creeley his conception of a “real poem,” it becomes clear that Creeley does have a conception, one that, no less than the questioner’s, has limitations. Creeley’s conception, too, picks out certain things as poems, and not others; it affords poetry certain powers but denies it others. Then the rub: once I recognize that Creeley’s conception of a “real poem” is a conception, not the conception, I see that my conception, too, is a conception.
This realization suggests that working at my craft might take the form of refining my craft (doing even better what I am doing), but it also might take the form of renewing my craft (doing differently what I have been doing). That is, the recognition that my conception of a “real poem” is a conception, not the conception, invites me to ask (even obliges me to ask) what possibilities are opened (and what ones closed) if I adopt another conception. In Sen’s terms, it invites me to add an ethics approach to my study and my practice of poetry, not confine myself exclusively to an engineering approach. Recognition that my conception of poetry is a conception urges me to complement my attention to and pursuit of craft with attention to and pursuit of metacraft.
George Lakoff’s Women, Fire, and Dangerous Things offers a construct that helps toward this aim. Lakoff speaks of “idealized cognitive models” (ICMs), structures by means of which we organize our knowledge. ICMs function in a given human context, but do not correspond to preexisting realities. The concept of a “weekend,” for example, “requires a notion of a work week of five days followed by a break of two days, superimposed on the seven-day calendar,” but this reveals that it is idealized, not “real,” since “seven-day weeks do not exist objectively in nature.” Lakoff further distinguishes “cluster models,” in which “a number of cognitive models combine to form a complex cluster that is psychologically more basic than the models taken individually.” An example is “mother.” One would think that for so important a concept we would be able to “give clear necessary and sufficient conditions” that would “fit all the cases and apply equally to all of them.” But in fact no possible definition can “cover the full range of cases,” because “mother” employs various ICMs, including such divergent models as: the birth model (the mother is “the person who gives birth”); the genetic model (the mother is “the female who contributes the genetic material”); the nurturance model (the mother is “the female adult who nurtures and raises a child”); the marital model (the mother is “the wife of the father”); and the genealogical model (the mother is “the closest female ancestor”). So if, say, I was adopted by one woman, who died soon after, and raised by the woman who raised her, then the birth model identifies one person as my mother, the marital model picks out another, and the nurturance model yet another.
Lakoff goes on to point out that “when the cluster
