Toggle Menu

‘Lost as Food and Won as a Coast’

Is a new, computer-generated poetry anthology as intriguing--and boring--as the lifework of any fairly prolific poet?

Barry Schwabsky

December 2, 2008

As a poet, I’m used to seeing my work appear in some pretty eccentric or recondite publications. But none of them, I think, have been quite as unusual as the one in which the following poem appeared on October 3:

Utmost as a stand

Calling Calling A stand Horned and hornless Disgust

Fine as a    station Like a hungry bee Of wilderness

A child of epochs

An utmost play To say an utmost hem

“Utmost as a stand” saw the light on page 2,039 of Issue 1, a 3,785-page anthology published online as a PDF file and containing poems signed by nearly that many contemporary poets (along with a few representatives of past generations and a surprising number of contributions by writers who have rarely, if ever, ventured into verse). Has there ever been such a massive gathering of poetry? I doubt it. Such a project could only have happened in the age of the Internet, when paper and bindings are no longer necessary for the creation of books, and when bookstores with shelves and staff have become inessential to their distribution.

But its bulk was not the only surprising thing about Issue 1. What first caught my eye was that I couldn’t recall ever having submitted my work to its editors. And when–thank goodness for that “find” function–I saw page 2,039, I knew why: “my” poem was one I’d never written. Neither had any of the other thousands of authors written theirs. The anthology’s “editors,” Stephen McLaughlin and Jim Carpenter, using a computer program of Carpenter’s devising, were responsible for its entire contents–and thereby for the most provocative hoax to hit the poetry world since the Araki Yasusada scandal in the early ’90s.

For those who let that one pass them by, Yasusada was, as Eliot Weinberger once put it, “both the greatest poet of Hiroshima and its most unreliable witness.” His wildly inventive and sometimes grotesque poems, reflecting his experiences as a survivor of Japan’s atomic holocaust, were published in esteemed venues like American Poetry Review, Grand Street and Conjunctions, until it became obvious that Yasusada had never existed. He is presumed to be the invention of Kent Johnson, an American poet and literature professor whose name is also included in Issue 1, under the following poem:

Glare

An image A sensation A mangrove A street

To obviate

Taking Meaning Putting Surrounding Questioning

Birthing above a place A spot At an immaterial material

A mile Of glare Shaking Seeming on a stream

The great difference between Yasusada and Issue 1, of course, is that Johnson, or whoever wrote the Yasusada poems, deceived quite a few people over a long period of time, whereas McLaughlin and Carpenter clearly meant to deceive no one. Their fiction was entirely transparent. That doesn’t mean they didn’t succeed in getting a lot of people hopping mad, at least for a few days. Ron Silliman, considered by some the dean of American avant-garde poets and a renowned blogger, muttered darkly of possible lawsuits. “Play with other people’s reps at your own risk,” he warned. Reading the poem Issue 1 attributed to Silliman, I can sort of see why he was unhappy:

Food

Lost as food and won as a coast Inefficient as a corner and efficient as a recess Lost as balance, won as a time Lost as a coast and found as a recess

It has been like becoming an .  idea, jewels, memories, . .   devils, the fearing highnesses

Haze has gone in your impotent trading-house You have been inefficient

Little and much Low and high Rotten and fresh

But as far as I know, our legal system has not yet been burdened with this matter.

The reason why deception was essential to the Yasusada hoax and not to Issue 1 may have something to do with the fact that the former was a product of the print era and the latter a creature of the web. It’s hard to remember now how long it used to take during the analog era for information to be shared. Even if Issue 1 could have been produced in those days, it might have taken months for the named poets to find out about it, and longer still for each of them to establish that he or she was not the only victim of the fake. Working through the issues the project raised might have taken years in print. Now, thanks to Google Alerts, one knew about Issue 1 instantly, and the chatter immediately multiplied on blogs and listservs throughout the English-speaking world. By the end of October, everyone had had their say and turned their attention elsewhere.

Arguably, there is yet a more important difference between Yasusada and Issue 1, which is that the faux-Japanese poems (written the old-fashioned way, by one or more people) were, in the eyes of many, very good poetry, whoever wrote them. What they lost in authenticity, once one realized they were most likely the work of a college professor in the American Midwest rather than a survivor of Hiroshima, they gained in novelistic virtuosity. The poetry in Issue 1, composed with the assistance of a computer, has not been received as well; Catherine Daly, for instance, writing to the University of Buffalo poetics listserv, complained that her poem was “really terrible…. Just awful,” describing the production as a whole as “a miserably aimed, sloppily controlled, misanthropic project.”

I wonder if that’s quite fair. I rather liked “my” poem, not as a poem in my own style, naturally, because it isn’t, but as an example of one particular present-day period style. But neither I nor anyone else has read the whole book, so none of us can come to grips with the totality of the oeuvre. Random dips into it also turn up some pretty boring pages–still in the same style–but then the same is sure to be true of any fairly prolific poet. You’d be shocked at what turns up in the collected Wordsworth or Whitman–or in Silliman, whose own voluminous production runs into the thousands of pages.

But you’d be shocked, too, at what turns up in well-respected magazines of contemporary poetry. The ones that take what Silliman would call a “post-avant” stance include a lot of poems that vaguely resemble what’s in Issue 1–highly paratactic, abstract and yet not without occasional hints at inchoate emotions of the sort that have long been the burden of lyric poetry. (Journals with a more traditional aesthetic have a different sort of period style, of course.) And a lot of that poetry is not as good as some of what I’ve read in Issue 1. That a computer can generate better poetry than some poets can write should not be shocking. Remember Kleist’s great essay on the marionette theater, where his interlocutor proves to him that a puppet has the potential to be made to dance more gracefully than any human, because “grace will be most purely present in the human frame that has either no consciousness or an infinite amount of it, which is to say either in a marionette or in a god.” There’s a glimmer in Issue 1 of what poetry written without consciousness might be–but just a glimmer, luckily, because were it entirely so, we flesh-and-blood poets might not stand a chance any more than the chess players do. For now, it’s a good reminder that we really ought to try and write better than a computer, while we still can.

Barry SchwabskyBarry Schwabsky is the art critic of The Nation.


Latest from the nation