Friday, October 16, 2009

Message and technique

Explaining why Madame Bovary was a scandal to French readers in 1857 but not to moderns, Kenneth Burke says that “we demand technique where they inclined to content themselves with ‘message.’ ”[1] In recent days, I have unexpectedly found myself in sympathy with Flaubert’s contemporary French readers. My strongest literary convictions have been unsettled by the evidence of Gordon Lish’s mucking with Raymond Carver’s fiction, even though Mark McGurl says in The Program Era that the “controversy surrounding Lish’s editorship” has been “considerably overblown.”[2]

I am not sure whether McGurl means that the extent of Lish’s mucking has been overblown or that the consequences for criticism are overblown, no matter how extensive Lish’s mucking. In either case, McGurl is wrong, I think. Lish’s revisions were deep and pervasive; they were sufficient, as I have argued, to damage or destroy several of Carver’s stories.

But the consequences for criticism are what really exercise me. What Lish’s mucking calls into question is the basic presupposition of literary analysis since the rise and flowering of the New Criticism—that is, the assumption that the literary text is an integer. Even deconstructionist criticism, which seeks to demonstrate the incoherence of a text, despite the author’s best efforts to submit it to his discipline, assumes that the published text, quoted against itself by them, has an integrity of intention: the incoherences they trumpet are not accidents of the publication process, but have been inserted into the text, on purpose, by an author who wants to reduce his work to a unified whole. An incoherence that results from the cross-purposes of author and editor does nothing whatever to establish or confirm the larger deconstructive claim that literary texts are unsuccessful attempts by a powerful ideology to subdue the inconvenient facts of the world.

For three generations, academic literary critics have been trained to handle the text as a woven tissue, which is all of a piece. Sometimes the assumption makes it possible for scholars to correct errors, as when J. V. Cunningham shows that Marvell’s To His Coy Mistress is not a “succession of images,” as T. S. Eliot would have it, but a logical syllogism; or when, later in the same essay, he restores Nashe’s original wording of a line in Summer’s Last Will and Testament, which Stephen Daedalus had misread in A Portrait of the Artist as a Young Man with “trembling joy.”[3]

More often, though, the errors are taken as established fact, because literary critics are not taught to raise questions about the provenance and authority of the texts in their hands. Everything about their professional training and status encourages them to rush to interpretation. It makes no difference to them whether Raymond Carver’s stories, in the published form with which critics are familiar, are the record of Carver’s thinking or a joint product of Carver’s drafts and Lish’s revisions. The condition of the text does not alter their responsibility to interpret it, nor does it leave critics with no option but to adopt new methods. By definition, any doubts about the condition of the text are considerably overblown.

Even when they believe themselves to be discussing a text’s meaning, then, literary critics are really examining technique, the evidence of its weaving (or its unraveling, if their allegiances are deconstructionist). Although McGurl argues that The Program Era succeeded and replaced “the Pound era” in American literature, and though he points to creative writing with its “prideful attention to ‘craft’ ” as the cause, the truth is that the demand for technique—that is, craft—is what links the program era to the Pound era. Hugh Kenner, whose 1971 magnum opus gave McGurl his name for the earlier era, wrote that modernist literary texts are “self-similar” or “scaling objects,” terms derived from Benoit Mandelbrot, the inventor of fractal geometry. That is, great texts are fractals—“works with something of interest to offer at varying scales of attention.” Their larger design is reproduced in their smallest details. The whole is implied in the parts.[4]

Here is the apotheosis of technique, the prideful attention to craft. But what if literary texts (or at least a significant number of them) are not self-similar tissues of self-consistent details, but something looser, more informal, perhaps even more extemporaneous? An endeavor not to design something, but to say something? What if a novel or even a poem is not a verbal icon, but a voluble discourse? What if the most important thing about a literary work, in short, is not its technique, but its message?

At a stroke, textual errors become of deeper concern, because the problem is to make certain that you are getting the author’s message straight. At the same time, however, it becomes less pressing to settle upon an authoritative text, reflecting the author’s ultimate intention, because it is no longer the perfection of technique, the scaling of design, that is at issue—but what the author is trying to say.

The controversy over Carver’s stories has begun to convince me that literary criticism took a wrong turn some time ago, and needs to start paying less attention to texts and more attention to authors, which would mean (among other things) listening to what they have to say.

[1] Kenneth Burke, Counter-Statement [1931] (Berkeley: University of California Press, 1968), pp. 59–60.

[2] Mark McGurl, The Program Era: Postwar Fiction and the Rise of Creative Writing (Cambridge: Harvard University Press, 2009), p. 446n.

[3] J. V. Cunningham, “Logic and Lyric,” in The Collected Essays (Chicago: Swallow, 1976), pp. 162–79.

[4] Hugh Kenner, “Self-Similarity, Fractals, Cantos,” in Historical Fictions (San Francisco: North Point, 1990), pp. 317–27.


Miriam said...

Is this really true of early modern studies? Not my field, obviously, but Book Two required me to spend a fair amount of time reading in that area, and the past couple of decades have seen a lot of work on the non-unity of various works, or works that can't be said to have a final form (for reasons having to do with editing, silent authorship, and production). Shakespeare's plays are the obvious example, but so too is John Foxe's Book of Martyrs.

By the same token, Romanticists have become very self-conscious about this topic, especially thanks to the Cornell Wordsworth. And Blake poses a major problem to anyone wanting to talk about works as an "integer," since a) the poems and their illustrations are interdependent, and b) no two editions are the same.

D. G. Myers said...


You are surely right that the more historical is the study then the less likely that it will fall prey to the assumption of the integral text. Before the rise of literary modernism, writers were little concerned to transform their texts into well wrought urns. Looser conceptions of literary form, except for meter and certain poetic set-pieces, were the norm.

Yet there is no denying that the assumption continues to influence even historically minded scholars. A glance at the tables of contents of scholarly journals reveals as much. Scholars still carefully lay out patterns of symbolism, imagery, metaphor, allusion, and meaning in early modern texts. And this pattern-tracing, a ready form of discovery in scholarly interpretation, is the legitimate offspring of the integral text.

Dan Green said...

Perhaps before deciding that Carver's case shows that we should be looking at what the author has to say, we should look more closely at the Carver case. What, exactly, is it that he's trying to say in a particular story that he is or isn't saying in the alternative version? In what two versions of the "same" story is he saying different things? Or in what two versions is Lish saying one thing and Carver another?

D. G. Myers said...

Alas, I have been unclear—if you can ask, Dan, “What, exactly, is it that he’s trying to say in a particular story. . . ?”

The whole problem, in my view, is conceiving the basic unit of literary criticism as “the particular story”—that is, the stand-alone text. You want to compare Carver’s original version of a story to the version that is left on the table after Lish had finished mucking with it. You assume, that is, that each version has its own integrity.

I wish to restore integrity to the author. It is Raymond Carver that I want to know about, not Gordon Lish! And the biographical fact that I happened to know Carver personally once upon a time is merely a proxy for my larger point. If I am going to compare versions, I will want to compare two different stories, with the editor’s revisions discarded.

For the simple truth is that stories do not “say”; persons do. It is time to start asking what a particular author is saying.

Dan Green said...

I really don't want to know about either Carver or Lish. I *am* interested in Carver's particular stories.

"For the simple truth is that stories do not “say”; persons do."

Exactly. Which is why I don't finally understand the effort to make stories say something.

D. G. Myers said...

We are talking at cross-purposes. You are a latter-day New Critic, uninterested in Carver but interested in his stories.

If you are interested in the stories, though, but not interested in their provenance and authority—if you give no mind to what condition they arrive in—what exactly are you interested in? The problem is this, Dan: you can’t say. They are for you something like an oracle or the Wizard of Oz.

No more than you am I trying to “make stories say something.” As I just observed, they don’t say anything at all, although a good writer can occasionally say something by means of a story.

Dan Green said...

"If you are interested in the stories, though, but not interested in their provenance and authority"

I'm interested in the stories primarily, their provenance only secondarily. I'm not completely uninterested in their origins or the conditions of their creation, but only after I've come to terms with the stories as works of literary art. To the extent that attention to "provenance" can occasionally help me come to terms with them on later readings, I would consider it.

D. G. Myers said...

I'm not completely uninterested in their origins or the conditions of their creation, but only after I've come to terms with the stories as works of literary art.

This is broken-backed. How can you even know that a story is a “work of literary art” without coming to grips with its provenance? What warrants that assumption?—other than that is how the story comes to you prepacked, that is, by the publisher.

And how can it be a work of art if it is a botched assembly of various incompatible parts by diverse hands? And how can you know that without inquiring into the condition of the text?

You take too much on faith, Dan—depend too naively upon editors and publishers.

Dan Green said...

I come to terms with it by determining, through the experience of the text, whether it is worth my time as literary art. Editors and publishers have nothing to do with it, except insofar as, collectively, they make some texts available to me--although they no longer have as much control over that, and will continue to have less.