Posthumous Regrets

I recently reread one of my published translations. A mistake. This is not something I normally do, and for good reason: just as no piece of writing is ever truly finished, it holds even truer that no translation is ever truly complete.

Oh, it may seem so at the time of publication. And lord knows, virtual pencils aplenty have poked and prodded at it by that point—mine (numerous drafts before I hand it in), the editor’s (if I’m lucky), the copyeditor’s, mine again, and then still again in galleys. While the translation is ongoing, and even for weeks afterward, I live in a sea of alternate takes and variant inflections, unable to walk down the street or hold a conversation without taking surreptitious note of some potentially useful flotsam.

And still, I can rarely bring myself to read the printed version even as little as a few months after publication without cringing at a wording that rings untrue, a phrase that could have been better turned (turned better?), a mot that could have been juster. Where, I wonder, are those sentences that filled me with such delight as I was composing them? And even if I am able to relocate them, it’s like identifying the body of a deceased beloved, once beautiful and full of life, now incomprehensibly inert. Language at these moments appears such a perishable commodity, an unstable element.

It helps, of course, when the original is a masterwork of style and substance, a Virgil to guide the translator through the selva oscura of interlingual negotiation. But even if it’s not, that’s no excuse. As William Weaver put it, “The worst mistake a translator can commit is to reassure himself by saying ‘that’s what it says in the original,’ and renouncing the struggle to do his best.” For those of us who subscribe to this credo, there is nowhere to hide when that best stays out of reach; when a reviewer or well-meaning friend or one’s own perverse need to revisit the past brings to light the glaring inadequacies that had lurked undetected until then.

No doubt part of the problem lies in the fact that a translation comes unstuck only gradually, grudgingly, from the rhythms of the original—French, in my case. While I’m still in the thick of working on it, I can’t help hearing a bit of that native drumbeat in my ear: insistent, seductive, lulling me into forgetting that different languages naturally find different cadences, and that to convey the lilt of the original properly, one sometimes (but not every time) has to reinvent it anew. Only later, after I’ve been away from the text for a while, moved on to something else, do those last echoes die out—and then I get that awful “aha!” (or “aaaghh!”) moment, when what I should have written becomes achingly clear.

Still, I don’t know that it would be any different if I had the luxury of letting the text sit quietly for six months or a year before turning it in—anyway, what publisher could afford to give you that? And while we’re at it, how about a text that has been sitting for forty years? I’m thinking of Jean Echenoz’s Cherokee (1987), the first of nine books by him that I’ve translated over a period of decades, and the one that—perhaps understandably—I’ve most wanted to redo. I’ve been fortunate in that the novel went through several reprints early in its life, allowing me at least to correct a few egregious errors from the first edition. But there are others that scratch at me with grating persistence, reminding me of every instance in which I could have done better.

In one scene, for instance, a character has to hastily improvise a doctor’s name, coming up in the original with “le docteur Péroné.” What makes the joke work is that péroné is both the French word for fibula and a perfectly plausible last name in its own right. At a loss, blinded by anatomical similitude, I went for the bone’s closest neighbor and baptized my physician “Dr. Tibia.” Too late did it occur to me to aim a bit higher—at the kneecap—and give him the much more convincing moniker “Dr. Patella.” (Echenoz, in fact, seems to excel at these anatomical sand traps: elsewhere he introduces a minor character named Roger Groin—“Roger Snout”—who “looked a bit too much like his name.” Would this have worked better as Roger Beak? Roger Bill?) And this leaves aside the more workaday changes one would like to visit on the text, such as “How annoying” in place of “How very bothersome” (“C’est embêtant”), or “I didn’t mean to do that” over the more literal—but, in the context, less accurate—“I didn’t want that” (“Je ne voulais pas ça”). And so on, punctuating the narrative like irritating flyspecks: little words and phrases that are not, technically speaking, wrong, nor even necessarily inelegant, but that nonetheless leave one with the sour aftertaste of not having done one’s best.

Perhaps this is the curse suffered by anyone who commits words to paper (or music to disc, or acting to film). Language is performance, and any performance can be improved. The heartache lies in recognizing precisely how those improvements could be made, while no longer having the opportunity to make them. Is there any solution to this? Hard to say. In some cases, new editions, or (as time goes on) the malleability of online publication, might allow us to atone for past sins. In others, our only recourse might simply be to push on to the next project, learn from our mistakes, try not to be paralyzed by them, and take comfort in the fact that, as Beckett suggested, we’ll have the chance to fail better another day.