Publish the question

Every essay you’ve ever read was pretending a little.

It arrived after the thinking was finished. The doubt had been sanded off, the dead ends quietly deleted, the conclusion set down on the page like it had been obvious all along. The writer knew the answer before the first sentence, and the whole piece was a guided tour to a destination already reached.

This one isn’t going to do that. I have a claim I believe and can’t yet prove, and instead of waiting until I can, I’m going to publish it with the hole still in it. On purpose. Let me try to convince you the hole is the point.

The instinct to wait

When you land on an idea you believe in, the responsible-sounding move is to wait. Run the test. Get the number. Then write it up properly, with the answer already in hand, so nobody can poke you.

That instinct feels like rigour. Often it’s just delay wearing rigour’s clothes.

Because some questions take a long time to answer, and some you can’t answer alone, and a few you’ll never fully close. If the rule is “say nothing until it’s settled,” then the most interesting things, the ones still open, are exactly the things you’re forbidden to talk about. You end up only ever publishing the parts that were already safe.

The question, well-posed, is the gift

Here’s the reframe I keep coming back to.

The map-makers who matter aren’t the ones who colour in known land neatly. They’re the ones who reach the edge of what anyone has charted and, instead of turning back, draw the line and write: here is where the known stops, and here is how you would find out what comes next.

That note at the edge isn’t an admission of failure. It’s the most useful thing on the whole map. It tells the next person where to point their effort. A question that has been sharpened to the point where someone could actually go and answer it is a finished piece of work, even with no answer attached. The answer, when it comes, is almost a byproduct.

So posing the question well isn’t the warm-up before the contribution. Sometimes it’s the contribution.

The line you still can’t cross

This only works if you’re honest about which half you’re holding.

There’s a difference between what you understand and what you’ve merely assumed, and you’re allowed to publish the first, not the second dressed as the first. You can say: here is why I think this happens, here is the mechanism, here is the shape of it. You can’t say: here is how big it is, here is the proven result, when you haven’t measured it.

A companion essay here, The Verb Was Doing the Deciding, lives on exactly this line. It argues that the name you give an agent’s tools quietly sets how readily it reaches for them, and lays the mechanism out in full. Then it stops, because I haven’t measured how big the effect is, only described the test that would. The mechanism is on the page; the magnitude is openly missing. That gap is the essay being honest about which half it holds, not a flaw in it.

The trap is subtle, because it’s easy to wave a small “of course, I can’t be certain” flag at the bottom and then write the rest as if you were. That’s not honesty. That’s dishonesty borrowing honesty’s coat for the evening. The open question only earns its keep if the doubt sits exactly where you say it sits, and not one inch further in.

Why this is hard to pull off

Now the craft problem, which is real.

An essay full of open questions can just read as unfinished. As if you ran out of time, or nerve, and called the gap “an invitation to the reader.” Nobody’s fooled by that.

The fix is that the question has to be the thing you built, not the thing you skipped. A vague “who can really say?” is a shrug. A precise “here is exactly what I don’t know, and here is the test that would settle it” is the opposite of a shrug. Same missing answer, completely different read. The difference is entirely whether you did the work of making the not-knowing specific. Sharpen the question hard enough and the absent answer starts to look like discipline instead of laziness.

Yes, I notice

I should say the obvious thing before you do.

I’m telling you, with some confidence, that you can publish a well-posed claim before you’ve proven it. Do I have proof of that? A study, a clean result, a settled finding that says open questions are a real contribution? I do not. I’m publishing the claim that you can publish unproven claims, unproven.

If that whole thing folds in on itself, fine. At least it folds honestly, and you can watch it happen in real time, which is more than most essays let you do.

Write from the edge

The inside of the map is finished. Everything safely inside the lines has already been drawn, by someone, probably better than you’d draw it again.

The edge is the only part still worth a pen. So here’s a question I can’t yet answer, set down plainly, with the route to the answer attached. The not-knowing isn’t the apology at the bottom of the page.

It’s the reason the page exists.