Dalkey Archive has always been an interesting project. While its track record publishing poetry has been more erratic than not – there is a book Cecil Giscombe, another by Gerald Burns, but mostly it has printed books of poetry written by novelists, even if one is by Harry Mathews & another by his Oulipo colleague Jacques Roubaud – and its record on critical writing even spottier – Viktor Shklovsky’s Theory of Prose is one of the great critical texts – Dalkey’s track record on publishing innovative fiction is unassailable. It is flat out the best publisher of innovative fiction the United States has ever had.

David Markson’s Wittgenstein’s Mistress is a novel, a booklength monolog by a woman who might (or might not) be the only surviving person on the planet. It’s an enjoyable book that I would happily encourage you to read. But that’s not really my intent here at the moment. Rather, I want to look at what a book like this tells me about writing & thus about poetry.

Fiction’s primary trick is to convince a reader than the syntax of its sentences integrates up not just grammatically or logically into an argument, but ultimately into an extra-linguistic phenomenon: a character or narrated world. This leap, from syntax to “voice” & through it to “character,” a nebulous concept at best, is the displacement that accounts for much of fiction’s reality effect. The power of grammar is thus transferred and felt by a reader as the power of the world “coming through” the emptied vessel of language.

Wittgenstein’s Mistress plays with these possibilities. For one thing, the narrator suggests that the book is being composed, rather like a letter. She announces the passage of days between paragraphs, many of which are only a single sentence long. Also, right at the end, in ways that I won’t go into here (so as to limit the number of “spoilers” I might inadvertently insert), Markson (or the narrator) also plays around with the possibility of who she in fact is.

But what is most powerful and telling about Markson’s book is that which goes palpably unsaid. In this way, Markson is miles ahead of most other novelists who explicate way too many of the details. Here Markson does just the opposite, raising other characters and contexts that scream out for explication without ever offering any. It’s not sloppy in the slightest – if anything, his consistency is an index of his meticulousness. The result gives Markson’s text the feel of a real person to a degree almost unimaginable in fiction. It’s that old Zen garden trick of making a circle of stones by pulling one out of position so that the displacement forces the viewing mind to cognitively “make the circle.” I’ve never seen it done better in a work of fiction than how Markson does it here.

I bought Wittgenstein’s Mistress after reading some extravagant review of one of Markson’s other novels, then let it sit for awhile, put off I suspect by the idea of that title, given just how the entire world knows Wittgenstein to have been gay. But the title works on multiple levels – it is in fact explained in the narrative, but even more so alludes to the narrator’s painful attempts to be exact with her language, not unlike Wittgenstein’s own books of philosophy. But what for him is an investigative method in a work of fiction becomes a series of extraordinary quirks. It’s remarkable just how well this works. I have a friend back in San Francisco – not a part of the arts scene as such – who is a great deal like this narrator & I ended up hearing her voice throughout my reading of the book.

Reading Markson’s novel made me think of my late dear friend Kathy Acker & how she used to worry during the composition of her early books about such issues as the construction of character. She was always clear in her own mind that character was just that – a construction. Each chapter in her book, The Adult Life of Toulouse Lautrec, presents a different conceptualization of truth & “the real” might be in fiction, which also includes the concept of character. It’s what Ulysses would have been if Joyce had actually felt some commitment to any of the styles he employed in each chapter. Well, maybe if Joyce had been committed to the idea that plagiarism plus porn equals autobiography.

There is an entire stratum of novels of course written with the idea that readers will identify with a character. A second stratum of novels is written with the idea that the readers will identify not with a character, but as readers & will remain aware of their own presence in front of the text, as though in a conversation with the author.* Finally, there is a tiny stratum of fiction written with the idea that readers will in fact identify with the author, not as a character, but as author. I tend to think that many of the books that I would characterize as fiction for poets – which would include works by Acker, Jack Kerouac, Bill Burroughs, Samuel R. Delany, Gilbert Sorrentino, Kevin Killian, Fee Dawson, Sarah Schulman & Harry Mathews – fall into this final category, or at least waver between it & the second. That border is also where someone like Markson, or W.G. Sebald, seems to fit.

When I read a novel I’m always think about how (or why) the author did this or that. Can Proust get the Madeline into the cup of tea? Can Kerouac really imitate the tape in Visions of Cody? I found myself thinking this way a lot with Markson, whom I’m not sure really expects that in the way these other novelists-for-poets seem to, but for whom it’s a perfectly reasonable & rewarding approach.

All of which made me think of my conversation here on the blog with Daisy Fried. A lot of what I don’t care for in the school of quietude is that presumption of readers falling principally, or only, into the first stratum when it comes to poetry. Most post-avant writing falls into the second category – that’s certainly where I would put Kelly & O’Hara & even Levertov. But much of the writing that compels me most is that which falls into or nearest that third stratum. And that’s where I would put Coolidge & Watten & Hejinian & Armantrout.

* I suspect that Jonathan Franzen’s to do with Oprah had very much to do with a concern on Franzen’s part that his book not be confused with that first stratum of writing.

At his reading Sunday with Chris McCreary and Rosmarie Waldrop at the Painted Bride, Lewis Warsh referred to the stories in his Singing Horse Press book Touch of the Whip as poems, then stopped & corrected himself. Perhaps he shouldn’t have.

Poets’ prose is a glorious & little understood jumble. The genre(s) can be traced back through Burroughs, Stein & Joyce at the very least to Baudelaire & Aloysius Bertrand*, to the origin of the prose poem. I would invoke Melville’s Moby Dick not only as a further instance, but as a superb example of the ways in which poets’ fiction almost invariably move beyond the tidy constraints of what is normatively fictive (which I might then trace back, at least in the U.S., to Twain). Let me map out what I see as six distinct tributaries of this phenomenon.

First is the prose poem itself. It by itself has multiple manifestations. One is the closed, one page or less prose piece that can be traced back to Max Jacob, but which in the United States comes heavily through the pernicious influence of Robert Bly’s journal(s), The Fifties and The Sixties, abetted by George Hitchcock’s Kayak and the numerous books of Russell Edson.

The second, far more interesting mode is the lengthier poet’s prose that remains clearly poetry, which begins in American English with Stein & then Williams’ Kora in Hell, but which really takes off after John Ashbery’s Three Poems, Clark Coolidge’s “Weathers” & Robert Creeley’s Mabel. This tendency has important French cousins in the work of St.-John Perse and Francis Ponge. This is where I would put Lyn Hejinian’s My Life, or Beverly Dahlen’s A Reading or even Jack Spicer’s Heads of the Town Up to the Aether. Questions of the serial poem and the epic will eventually expand this category even further.

After the prose poem comes a mode of poetic fiction that would include Warsh’s marvelous Touch of the Whip, much of the writing by Carla Harryman, Creeley’s stories, the short fiction of Gil Ott, the narratives of Bobbie Louise Hawkins. And Samuel Beckett most of all. These are all writers clearly interested in the traditions and devices of fiction itself, but written with a poet’s sense of literary value. There are few (if any) moments where, say, character or plot, which may in fact be both present & pertinent, are more important than the pleasures & problematics of the words immediately on the page in front of the reader. I think that these may be the most difficult works of all for people to gauge, because they truly transcend either of their source genres. Where I think you can test my own work as poetry, and, say, Paul Auster’s as fiction, these writers clearly are on their own. This thus may be the bravest prose of all.

A close cousin to this intergenre prose is more truly what I would call poet’s fiction, works by poets that genuinely aim for the goals of fiction, but often employing many of the devices (& pleasures) of their home form: Gilbert Sorrentino & Toby Olson would be good examples. So would almost all the writing of the so-called new narrative: Dodie Bellamy, Kevin Killian, Robert Gluck, Bruce Boone, Michael Amnasan. I would place Harry Mathews here, although I’d put the bulk of Oulipo fiction into the next category.

These would be those fiction writers who clearly identify as such, but who write as though their readers were going to be, if not poets per se, at least the readers of poetry. This is where Burroughs & Kerouac fit in (& Melville at his best also). Kathy Acker, Walter Abish, Lydia Davis, Sarah Schulman, Samuel R. Delany, Julio Cortázar, Italo Calvino, Joyce of course; one could make a case for W.G. Sebald, as for Carole Maso.

Finally there are poets who work hard to make a transition all the way to the values of fiction – the problematics of plot-centric narrative, for example – but whose prose still retains some surface features of their past as poets. Auster fits here, as I think does the later work Michael Ondaatje (tho his first works fit closer to the poet’s fiction category).

There are of course many other kinds of creative prose & fiction. These are merely the types that touch on poetry as a genre & tradition. None of this has to do with quality per se, but I do think that it has to do with certain questions of literary judgment. It’s a mistake, for example, to compare the prose of Lewis Warsh with the novels, say, of Paul Auster, or with the prose poetry of Clark Coolidge. Rather I suspect that over time, as we have more readers & writers and more works in each of these tributaries of excellence, we will eventually have a cleaving between the various categories far more decisively than we have today. In 2002, it is still possible to call both Russell Edson & Lyn Hejinian prose poets, Carla Harryman & Michael Ondaatje fiction writers. Fifty years from now, such clusterings will simply seem like nonsense.

* When is somebody going to publish Merrill Gilfillan’s superb collection of translations from Bertrand’s Gaspard  de la Nuit?