# The Murk Is the Tell

Here is a sentence I wrote and believed for about a day: "A day-old chick has to be sorted, future layer or not, and the cue is a smear of anatomy that the best sorters read by eye." It scans. It has rhythm, a little technical confidence, the faint authority of someone who has spent time at the hatchery. And it comes apart the moment a careful reader asks two plain questions: what is a future layer, and what is the cue? I did not have clean answers ready. A layer, it turns out, is a hen kept for laying rather than for meat, which is why a commercial hatchery sorts its chicks on the first day and sends the males away; the cue is a faint, ambiguous fold of tissue near the vent that an expert reads in about a second and cannot put into words. I knew the rough shape of all that when I wrote the sentence. I did not know it sharply, and the sentence told on me.

That is the thing worth dwelling on, because it took me an embarrassingly long time to believe it: murky prose is almost never a prose problem. It is a tell.

In poker, a tell is an involuntary signal, a glance or a stillness or a change in breathing, that leaks what a player is actually holding. He cannot fix a tell by working on his face, because the tell lives somewhere else entirely: in the gap between the hand he has and the hand he is pretending to have. Vague writing leaks the same way. The fog in the sentence lives in that same gap, between the idea you have and the idea you are pretending to have, and no amount of work on the words will close it. You can trade "a smear of anatomy" for something statelier and end up with a better-dressed version of the same confusion.

The instinct, when a sentence reads badly, is to reach for a better word, a smoother clause, a more elegant rhythm: to manage the face. I have spent years learning that this works only on sentences whose underlying thought is already clear and merely wearing the wrong clothes. When the thought itself is unfinished, polishing makes things worse, because it hides the tell instead of removing it; now the sentence sounds authoritative and is still hollow, which is a harder fault to catch the second time around. The honest move is cruder and slower. You write the sentence again, and then again, one way and then another, until the idea you were reaching for finally shows up on the page. Clarity is what the writing produces; it is rarely sitting there in advance, waiting to be transcribed. A first draft is not a clear idea in poor clothes; it is an unclear idea, plus your own dim awareness that it has not yet arrived.

None of this is special to writing. It is how every skill that cannot be spoken gets made. The chick sexer reaches ninety-eight percent accuracy not from a manual, of which there is none, but by guessing at thousands of birds while a master behind him says only "yes" or "no." The child on the bicycle never learns the balance rule the physicist can write down, the one about steering into the fall by an amount that scales with the inverse square of her speed; she learns it by falling until she stops falling. The psychologists Daniel Kahneman and Gary Klein, who spent years disagreeing about whether expert intuition can be trusted at all, settled on a single condition for when it can: it forms only in worlds that return fast, clear feedback, and never in worlds where the verdict comes late or muddy. Writing is one of the lucky worlds. Its feedback is a reader's reaction, including the colder reader you become to your own page an hour later, and you converge on the sentence the way the sexer converges on the chick, by producing one, hearing whether it lands, and adjusting. You cannot deduce a voice. You can only find it.

Some of what you find by writing is larger than any single sentence: the stance. A piece can be wrong in a way no line edit will reach, because the error is in its posture; it has been built as a performance of analysis, all clipped declaratives and withheld presence, when what it wanted was to sound like a person thinking on the page. The right register is no more deducible than the right sentence. You write a paragraph in one voice and hear it impressing rather than explaining; you write it again in another and hear it begin to breathe. A frame announces itself only from the inside, which means the only way to it is to keep writing through the wrong ones.

The rules that govern all of this deserve the same suspicion. It is tempting, each time a sentence goes wrong, to write down a rule so it will not go wrong again, until you are carrying a rulebook longer than anything you produce, half of it guarding against mistakes you stopped making years ago. A swelling list of writing rules is the same error as the polished hollow sentence: accretion mistaken for improvement. The discipline that works is subtraction. Keep the few rules that still catch something real this week, and retire the rest the way you clear a tool off the bench once it has stopped earning its place; a rule that no longer fires is only weight, and weight is the thing you are trying to cut. The shortest set of rules that still produces the work is the right one, and it is almost always shorter than the set you are holding.

I write this way because I have never found another that works. The rules I started with were tried and mostly thrown out, and what is left is a reader's reaction and the willingness to take another pass. The murk still shows up in my first drafts, reliably, every time. I have stopped reading it as a failure of language. It is only the tell, the honest signal that I do not yet know the thing as well as the sentence pretends I do, and the way past it was never to think harder. It was to write again.