OK, back to this. In Part 1, I traced out how in conceptual art, the concept lies outside whatever artwork is produced—how, strictly speaking, the concept itself is the artwork, and whatever thingamabob the artist then uses the concept to go on to make (if anything) counts more as a record or a product of the originating concept. (This is according to the teachings of Sol LeWitt, as practiced by Kenneth Goldsmith.) Thus, we arrived at the following formulation:

• Artist > Concept > Artwork (Record)

Now, I’m not going to argue that every conceptual artist on Planet Earth works according to this model. But LeWitt’s prescription has proven influential, and continues to be revolutionary—because choosing to work with either a concept or a constraint will lead an artist down one of two very different paths. To see how this is the case, let’s try defining what a constraint is, aided by the Puzzle Master himself, Georges Perec . . .

It’s 1967, and Perec has just come up with the notion to write La disparition, a novel that doesn’t use any words that contain the letter E. And at first glance, it might be tempting to call this book a conceptual work of literature, as some have in fact done. But does Perec actually have a concept? In other words, does he have an idea that becomes a machine that makes the artwork for him?

No, he does not. Perec’s decision, rather than eliminating all other decisions for him—and thereby making the production of the artwork a perfunctory affair—instead necessitates even more choices. Broadly speaking, Perec’s problem next becomes: which words not containing the letter E should he include? And in what order should he put them? Those are questions he needs to answer on a word-by-word, and sentence-by-sentence, basis.

And even then, he’s still not done. Should Perec call attention to his own constraint? Or should he try writing a book that, despite the constraint, reads as naturally as possible?

It’s easy to imagine the former:

“I’m stolidly commissioning you, buddy,” said Mark. This was at a bar, or a pub, in Scranton. “Mantissas reoccupy slugs, so watch out! I’m Christian.”

A fractious bushman in a miniskirt* was warbling as drunk nuns sang hymns and a manta ray shat on my hat. Truth told, I didn’t know what to do.

So I said: “Man, your normalisation distils aircraft. Hullabaloo!”

“I don’t furnish a toss**,” was his back talk, taking a topmost cultivation (and fab psychosis) and nibbling a tapir’s spawn on a luminal flotilla.

I wrote that example quickly, using this random word generator. I simply accepted whatever E-less words I was given, then stitched them together with little concern for meaning or coherence. And it would be easy to continue in this fashion, accumulating pages of loosely coherent sentences. Some of them might even be amusing! (“a manta ray shat on my hat”). But this project would also get pretty random pretty quickly—and it’s not at all what Perec did.

Nor did Perec operate the way Goldsmith did when making No. 111. There, you’ll recall, Goldsmith collected “any word or phrase ending in the sound of ‘r’ or the ‘schwa.'” Perec could have done something similar—he could have wandered the boulevards of Paris, collecting any and all words lacking the letter E. Just like Goldsmith, he may have come to realize (after a while) that none of those words were any better or any worse than any other—that all of the words were equally good. And so he could have produced a list like the following one (albeit in French):

amount spots dull invasion attracting fatcat magnanimity cobra owl fobs mitochondrial mumblings salvaging disorganising alkalis constraint staving buckshot classicists twofold contradictions tumultuous vulgarly quarterstaffs ***

(I once again leaned on the random text generator. And I swear I didn’t add “constraint”! It really was one of the results—no editing here.)

Once again, this is a far cry from what Perec did.

Another option. Perec could have turned his idea into a procedure, and applied it to an existing text. Here, for instance, are the opening three paragraphs of Harry Mathews’s Cigarettes, stripped of any word containing an E (I preserved the capitalization and punctuation):

“What’s, ‘I you want an?’ anything.

us a buzzard in mid flight. still arriving. Through lilac of smoothly, and swinging of light that us along a bank of dogwood, a man in a stood Allan’s with a.

around. it was my turn I, in of, “…I was in—you taking away . . . blinding light . . . I couldn’t a.” I too was. by, could this Allan?

But this is also nothing like what Perec did (and not just because he didn’t have access to a copy of Cigarettes in the 1960s).

Instead, Perec sat down and wrote this:

Anton Voyl n’arrivait pas à dormir. Il alluma. Son Jaz marquait minuit vingt. Il poussa un profond soupir, s’assit dans son lit, s’appuyant sur son polochon. Il prit un roman, il l’ouvrit, il lut; mais il n’y saisissait qu’un imbroglio confus, il butait à tout instant sur un mot dont il ignorait la signification.

Il abandonna son roman sur son lit. Il alla à son lavabo; il mouilla un gant qu’il passa sur son front, sur son cou.

Son pouls battait trop fort. Il avait chaud. Il ouvrit son vasistas, scruta la nuit. Il faisait doux. Un bruit indistinct montait du faubourg. Un carillon, plus lourd qu’un glas, plus sourd qu’un tocsin, plus profond qu’un bourdon, non loin, sonna trois coups. Du canal Saint-Martin, un clapotis plaintif signalait un chaland qui passait.

Sur l’abattant du vasistas, un animal au thorax indigo, à l’aiguillon safran, ni un cafard, ni un charançon, mais plutôt un artison, s’avançait, traînant un brin d’alfa. Il s’approcha, voulant l’aplatir d’un coup vif, mais l’animal prit son vol, disparaissant dans la nuit avant qu’il ait pu l’assaillir.

Here’s how Gilbert Adair (loosely) translated it (much later on in 1995, as A Void):

Incurably insomniac, Anton Vowl turns on a light. According to his watch it’s only 12.20. With a loud and langorous sigh Vowl sits up, stuffs a pillow at his back, draws his quilt up to his chin, picks up his whodunit and idly scans a paragraph or two; but, judging its plot impossibly difficult to follow in his condition, its vocabulary too whimsically multisyllabic for comfort, hurls it from him in disgust.

Padding into his bathroom, Vowl dabs at his brow and throat with a damp cloth.

It’s a soft, warm night and his blood is racing through his body. An indistinct murmur wafts up to his third-floor flat. Far away, a church clock starts chiming — a chiming as mournful as a last post, as an air raid alarm, as an SOS signal from a sinking ship. And, in his own vicinity, a faint lapping sound informs him that a small craft is at that instant navigating a narrow canal.

Crawling across his windowsill is a tiny animal, indigo and saffron in colour, not a cockroach, not a blowfly, but a kind of wasp, laboriously dragging a sugar crumb along with it. Hoping to crush it with a casual blow, Vowl lifts up his right hand; but it abruptly flaps its wings, flying off without giving it assailant an opportunity to do it any harm.

. . . As we can see, what Perec did—and what Adair also did, even as he preserved the constraint—was write what Goldsmith might call a “readable” novel: one concerned with “more traditional linguistic concerns of readerliness, rhythm, phrasing, song, etc.” And we should recall that what Goldsmith found especially valuable about devising and obeying a concept—of not editing one’s results, but just “[letting] the language fall where it may within that specified context”—is that those more traditional linguistic concerns (readerliness, rhythm, phrasing, song, etc.) were “no longer of importance.”

Not so for Perec, for whom those factors were of utmost importance. In other words, as Perec constructed (crafted) La disparition—decision by decision, sentence by sentence, word by word—he was constantly judging certain E-less words to be better or worse than other ones—all for the purpose of determining which one should go next. Indeed, Perec proved so good at this that (as the story goes—perhaps apocryphally), some readers entirely missed the fact that the novel lacked E’s.

Perec’s efforts lead us to a situation opposite Sol LeWitt’s. The production of La disparition was not a perfunctory affair that removed the exercise of conscious decision-making or ego. Instead, in writing it, Perec was required to make innumerable decisions. Constant decision-making. Constant ego.

Another central difference between Perec and LeWitt’s works is that the constraint, unlike the concept, lies not outside but rather inside the artwork:

• Artist > Artwork > Constraint

As such, the constraint does not eliminate decisions for the artist. Rather, it constrains the decisions that the artist goes on to make.

The constraint is a component of the artwork’s form. It is, to be sure, one component among many. However, it is privileged in that it is essential to the artwork’s formal integrity. In this regard, it is much like the Russian Formalist notion of the dominant, which “may be defined as the focusing component of a work of art: it rules, determines, and transforms the remaining components. It is the dominant which guarantees the integrity of the structure.” (I won’t go on about that here, having written a great deal about the dominant elsewhere, though I’ll be happy to return to this point if there is interest.)

To sum up, then. We’ve arrived at two different approaches to making art—two different notions as to how an artist can proceed:

• concept: eliminates all other decisions for the artist, and makes the execution of the artwork a perfunctory affair; the work’s value lies in its demonstration of the concept
• Artist > Concept > Artwork (Product)
• constraint: constrains the decisions the artist makes, but does not eliminate them; the work’s value hinges on how it is executed (i.e., how successfully the artist negotiates the constraint plus the traditional impediments of constructing an artwork)
• Artist > Artwork > Constraint

Now let’s return to Goldsmith’s criticisms of the Oulipo, which I think will make much more sense. Specifically, Goldsmith’s said:

One of the greatest problems I have with OULIPO is the lack of interesting production that resulted from it. While I like the idea of “potential literature,” it strikes me that their output should have remained conceptual — a mapping, so to speak; judging by the works that have been realized, they might be better left as ideas. On the whole, they embraced a blandly conservative narrative fiction which seems to bury the very interesting procedures that went into creating the works.

Goldsmith isn’t criticizing the ideas for texts that the Oulipians have come up with. Presumably, he would have liked Perec’s work better if Perec, rather than actually writing La disparition, had instead proposed something like, “Write a novel without using the letter E”—if he had left it as an idea (though we should now be able to see how that wouldn’t have worked, since Perec’s idea wasn’t really a concept). In other words, what Goldsmith dislikes is the actual books that the Oulipians went on to write, which he calls “blandly conservative narrative fiction.” In doing so, what Goldsmith is criticizing is Perec’s decision-making, and Perec’s concern with “readerliness.”

I will have more to say about this in Part 3. In the meantime, I hope these two posts have clarified how Goldsmith and Perec are committed, ultimately, to very different ideas of what should count as art. Because we’ve arrived at a rather substantial debate in contemporary art—the distinction between formalism and anti-formalism.

More about which soon. Until then, may you manage to find your way about, regardless of what constrains you.

*I originally had “total spectrum” here, until the eagle-eyed Devin King spotted the E—thanks Devin! I went to the Random Text Generator for a replacement, set it to 5 and received: “bushman, evaporating, fractious, miniskirt, sported.”

**I originally had “I don’t give a toss.” Obviously I suck at telling whether or not words contain E’s!

*** I originally had “father” between “buckshot” and “classicists.” This is why I don’t write lipograms in E—I obviously have no idea what an E is . . .