Earlier today John pointed toward Nigel Beale’s cleverly-titled criticism of my post “Tiny Shocks: Uncovering the Reductive Plot of James Wood’s How Fiction Works.” I’m looking forward to Nigel’s longer criticism; in the meantime I thought I’d reply regarding the mistakes Wood makes in his readings of Viktor Shklovsky and William H. Gass, since Nigel asked specifically about them:
Does Wood ‘misunderstand’ Gass? Is his reading of Shklovsky ‘demonstrably wrong’? Are these ‘intellectual errors’ or are they mischievous ploys to argue (successfully I’d say) points which you just don’t agree with? Who’s being Ad hominem here?
(Nigel, I hope you don’t mind my calling you by your first name; since we’re Facebook friends now, and I hope you’ll call me Adam. It keeps things friendlier!)
And let me say that it’s certainly fair for Nigel to take issue with my calling Wood’s account “smug and small” etc. Those are critical words, granted. I stand by them, however, as fitting descriptions of Wood’s argument and the manner in which he makes it: Wood’s reading of fiction in How Fiction Works is reductive, and I believe that a critic of his stature is capable of far better. (He studied with Frank Kermode!)
OK, on to the formalists.
So, Wood’s reading of Shklovsky. I called it demonstrably incorrect. I think my post clearly demonstrated why, but I’ll reiterate, and try to say something new about this in the process. The relevant section in Wood is here:
[T]he novel soon showed itself willing to surrender the essential juvenility of plot, in favor of what Viktor Shklovsky calls ‘unconsummated’ stories with ‘false endings’ (he was referring to Flaubert and Chekhov, respectively). (149)
I’ll define some of these terms briefly, for clarity’s sake (although interested parties should refer to Shklovsky’s Theory of Prose, as it’s fascinating reading).
A consummated or open closed ending is one where the plot experiences denouement. (Again, denouement means an untying, and the implication behind that word is that, following it, there is nothing left to do.) Most stories end this way: e.g., a comedy ends with a marriage, a tragedy ends with a death, or deaths. The audience is aware that “more could be said” (more can always be said), but the main questions raised by the narrative have been answered, and there is a sense of catharsis, or at least completeness. Hamlet is a classic example of a narrative with a closed ending. Shakespeare could have written more—we don’t stick around to see whether Fortinbras becomes the next king—but we have a pretty strong sense that the story is finished. We leave the theater satisfied.
An unconsummated or open ending is one where the narrative ends before denouement. In my original post, I cite François Truffaut’s well-known film The 400 Blows (1959). I apologize for not having cited a literary example, but I thought the film one easier to see: I can embed the actual ending of the film in the post:
Here’s another example: the ending of the great 1971 film Two-Lane Blacktop (spoiler alert!):
Literary examples are, in any case, abundant. For instance, see the ending of the novel Nog (1968) by Rudolph Wurlitzer (which is on my mind as I just read it—an amazing book). (Wurlitzer also wrote the screenplay for Two-Lane Blacktop.)
A false ending is something else entirely: when a story ends not with plot consummation but a bit of writing that gives the impression of a closed ending. For example, the author might append a recognizably clichéd bit of ending to the narrative, creating the impression of closure but without really resolving the narrative. (Imagine an open-ended film that then features a camera craning skyward toward the sunset, and you’ll get the idea.) Shklovsky actually discusses this type of ending before he discusses unconsummated endings (which he calls negative endings).
Here are the corresponding pages in Theory of Prose, for anyone who wants to read more about all of this. I encourage you to do so because it’s just such great stuff!
OK. As I wrote at the time:
Wood is mistaken in his reading: Shklovsky does not identify “false endings” with Chekhov, but rather with the Satyricon and Gogol’s “Tale of How Ivan Ivanovich Argued with Ivan Nikiforovich” (56–7). It’s only later that Shklovsky mentions Chekhov’s short stories at all—and then he does so in order to celebrate that author’s innovative uses of closed plots!
This has nothing to do with opinion; it’s pure reading comprehension; Shklovsky doesn’t begin his discussion of Chekhov until after he’s finished defining these different types of endings; see the second half of page 57. (But before we move on, note that Shklovsky is not writing about “‘unconsummated’ stories with ‘false endings’”—those are two different types of endings. Indeed, Wood’s use of “respectively” doesn’t even make sense unless those are two different types of endings. This is an example of the kind of vague and lazily-written sentence one finds throughout How Fiction Works.) (There are, admittedly, many very good sentences, but they don’t excuse the blunders. To paraphrase Howard Hawks on what makes a good film: three great scenes—but no bad ones.)
I think this is a pretty straightforward matter, and it’s a mistake that Wood should account for. It’s also something that shouldn’t have gone to print; an editor or fact-checker should have caught it. (This flub and the fact that the index is often incorrect makes me doubt that the book was very carefully fact-checked or edited.)
(Adding to all of this that Chekhov is, in fact, a master of closed-plot narratives, as Shklovsky’s reading demonstrates—Shklovsky attributes Chekhov’s success as a storyteller to “distinct plotlines with unexpected resolutions” (58, my emphasis).
Chekhov routinely wrote plots that developed and closed off—remember his oft-repeated bit of advice that the gun from Act I should go off in Act IV? Firs dies as the cherry orchard is chopped down at the end of The Cherry Orchard. Konstantin shoots himself at the end of The Seagull. Serebryakov and Vanya make up at the end of Uncle Vanya, and then Serebryakov and Yelena take their leave, and Sonya delivers her heart-breaking closing monologue:
Now that is a closed ending! I cry every time I see it, due to its astonishing finality.
This is what Wood calls “the essential juvenility of plot.” This is what he would prefer that artists “surrender” in favor of unconsummated endings. (“[T]he novel soon showed itself willing to surrender the essential juvenility of plot, in favor of what Viktor Shklovsky calls ‘unconsummated’ stories with ‘false endings.’”)
I myself am happy that both open and closed endings (as well as other types of endings) exist as options available to fiction writers. If you’ll permit me a personal example: In my most recently completed story collection, “Distress” (as yet unpublished), some of the stories feature open endings, and some feature closed endings. But I don’t think one type of ending is inherently superior to the other; they’re different options that writers have. (François Truffaut gave The 400 Blows an open ending, then the very next year gave Breathless (1960)—he wrote the screenplay—a closed one.)
When writing one of the stories in “Distress,” “5000 Units of Product,” I had a great deal of trouble figuring out how to end it. I tried for many years to give it an open ending. But last year I realized that it would work better with a closed ending; within a week, I had finished the piece. [Update: it's since been published, in case you're curious.]
Wood doesn’t explain why open-ended plots are superior to closed plots, or why plot is “essentially juvenile”; as is typical in How Fiction Works, he simply proclaims this to be the case. (It’s not unlike earlier in the book, where he decrees third person limited superior to all other points of view. That claim he at least provides more reasoning for, although I find that reasoning specious.)
Wood is certainly free to have his preferences, but to argue that any perspective or type of ending is objectively superior to another is a silly business. It’s not unlike arguing that stories with dogs in them are better than the canine-free variety. This is precisely the kind of nonsense argument that I consider beneath a critic of Wood’s stature—”smug and small.”
Furthermore, it’s particularly egregious to campaign for one particular point of view or type of ending in a book that purports—at least initially—to be a primer on how to write fiction. What good does it do to tell younger writers to not use a particular perspective? Or that their stories should end only in certain ways?
The novel is not evolving toward a higher state of being, some ideal future where every writer worth his salt produces only third-person limited narratives with open endings! And the novels of today are not necessarily better than the novels of yesteryear. However, this is the cartoonish history of the novel that Wood paints in How Fiction Works.
Attributing such telos to the novel compounds Wood’s misunderstanding of the Russian Formalist concept of ostranenie (enstrangement or estrangement or defamiliarization). Again, see my original post for more on this; see also here, here, and here. Wood reduces this very powerful concept to personification, which I find extremely wrong-headed. It is also, once again, demonstrably wrong—ostranenie is not, as Wood claims, a kind of metaphor. As I noted, for Shklovsky ostranenie was a vastly larger concept, and the very thing that makes art art. Shklovsky notes repeatedly throughout Theory of Prose that exposure to art desensitizes us to it (just as exposure to life inures us to life).
And so it is valuable for artists have a wide range of devices and techniques to choose from, because that allows them to continuously refresh art, and therefore life. Shklovsky’s oft-cited words are well worth repeating:
And so, held accountable for nothing, life fades into nothingness. Automatization eats away at things, at clothes, at furniture, at our wives, at our fear of war. […] And so, in order to return sensation to our limbs, in order to make us feel objects, to make a stone feel stony, man has been given the tool of art. The purpose of art, then, is to lead us to a knowledge of a thing through the organ of sight instead of recognition. By ‘enstranging’ objects and complicating form, the device of art makes perception long and ‘laborious.’ (5–6)
If everyone around you is writing stories with closed endings, and you write a story with an open ending, that helps make your story feel fresh, and reminds everyone of a possibility in writing (and maybe also in life). But the converse is also true: the closed plot will feel revolutionary when perceived against a field stuffed with more existential, open-ended works. This is the basis by which old ideas can be brought back and made new again.
Here’s an example of an unconsummated ending that surprised a lot of people—because in this type of movie, we expect the good guy to win:
By way of comparison, the ending of Last Tango in Paris (1972) owes at least some of its power to the fact that we don’t expect European art films to end so definitively. Roger Ebert is so fond of that scene’s definitive power that he ends both his 1995 and 2004 essays on the film by mentioning it. (Spoiler warnings for all of these films, by the way! But they’re all great films that you should check out if you haven’t seen… Last Tango in Paris is especially brilliant.)
And some authors, perversely, like to have it both ways. Here’s the famous definitive and yet ambiguous ending of Antonioni’s Zabriskie Point (1970)—one of my favorite endings in any film ever:
I stand by my claim that Wood deeply misunderstands Shklovsky’s point, due both to his trivialization of ostranenie, and due to his view of some literary devices as inherently superior to others (which he is compelled to hold because he considers some genres of fiction inherently superior to others; I disagree here as well). The result is that Wood’s view of fiction is weaker than Shklovsky’s, and less useful to anyone who wants to understand how fiction actually works.
As for Gass, and Wood’s misreading of him, that is less straightforward. The relevant section from Wood is this:
“I find this deeply, incorrigibly wrong. Of course characters are assemblages of words, because literature is such an assemblage of words: this tells us absolutely nothing, and is like elaborately informing us that a novel cannot really create an imagined world, because it is just a bound codex of paper pages.” (103–4)
To understand what Wood finds so “deeply, incorrigibly” wrong in Gass, please see my original post; see also Gass’s original essay, “The Concept of Character in Fiction.” I won’t reiterate my full criticism here; suffice it to say:
1. Gass’s formalist analysis of character is in fact extremely useful, and tells us many useful things. I’ve used it myself on many occasions. For instance, I just completed an interview where I explain in more detail how Gass’s analysis helped me overcome some technical and conceptual problems when writing my first story collection. When the interview goes up, I’ll add a link.) Of course some people don’t find formalist criticism useful—some people don’t find mathematics useful—but to claim that it therefore isn’t useful is ignorance plain and simple.
2. Wood’s summary of Gass is misleadingly reductive. Gass does define characters in terms of the words that comprise them (and in truth it’s more complicated and subtle than that; characters are recurring nouns that mnemonically gather other words under them), but Gass doesn’t then proceed to say something tautological, like, “And that’s all they are! And that’s the end of it! Huff!”
3. Gass never, ever, anywhere (and I have read a lot by William H. Gass) denies that the words that comprise characters (and stories) can’t be illusionistic, or can’t be applied toward mimetic ends (“cannot really create an imagined world,” as Wood puts it). Wood is wrong to imply that Gass believes and argues this.
On that last point, Wood has fallen victim to a trap realists often fall into when debating formaliststs: he has repeated a straw-man argument of formalism. But just because you define the devices of fiction in terms of language doesn’t mean you are denying the communicative power of that language. How anyone can read Gass and come away with the conclusion that he doesn’t think characters can be illusionistic, or words communicative—well, it baffles me. As I worded it in my original post:
I would argue that Gardner’s concept (of realistic, plausible character) is not precluded by Gass’s; Gass’s concept readily admits Gardner’s. One way to construct a “linguistic source” is by conventionally depicting psychologically plausible characters with whom an intended audience can relate, and upon whom they can project their experiences and biases. And Gass does not deny this.
Wood’s argument against Gass is, in my view, discredited a priori; see the debates between Gass and John Gardner for an extensive—and I do mean extensive!—argument on this very subject, where Gass responds to this very criticism many times. (I linked to some information about these debates in my original post; a full transcript of one debate is transcribed in the wonderful interview collection Anything Can Happen.)
Indeed, Wood’s repetition of this hackneyed criticism calls to mind John Barth’s complaint about his contemporaries who wrote as if “the last sixty years or so hadn’t existed”; Wood is making an argument that ignores more than thirty years of substantial debate. As James Tadd Adcox wrote in my original post’s comments section:
This sentence–”I would argue that Gardner’s concept (of realistic, plausible character) is not precluded by Gass’s; Gass’s concept readily admits Gardner’s”–does an excellent job of expressing where I’d argue fiction actually *is* right now. In the seventies, the question of who was right between Gass and Gardner made sense, but I think that writers coming up now no longer see those two positions as opposing, in any real sense…
Wood’s mischaracterization of Gass’s argument is an embarrassing moment in his book, and another example of the whole project’s small-mindedness.
…Well, I fear that I’ve mostly repeated myself here, but I’ll hope that I’ve done so with additional clarity (and at least we got in some movie clips). Again, I look forward to Nigel’s longer and more substantive complaint with my post, as well as to the ensuing discussion.
Until then, best regards to all, Adam
P.S. Dear James Wood! Please drop me a line! I want to hear what it was like to study with Frank Kermode!