I spent four months assembling a place to write. The desk was the last piece and the most expensive, and it arrived in November, and between November and February I wrote eleven hundred words at it.
In the same period I wrote about nine thousand words in a café on Rua da Boavista, at a table that wobbles, on a chair that is too low, in noise.
What I thought I was buying
I thought I was buying the removal of obstacles. The theory was that writing is hard, that every friction between me and the work costs something, and that a good chair, a good lamp and a surface at the right height would leave nothing between me and the sentence but the sentence.
This was correct about the frictions and completely wrong about what remained. What remained, with all the frictions removed, was the writing — undiluted, with nothing to blame. I had built a room in which failure could only be my fault, and I could not work in it. The café is full of reasons the work might go badly today, and every one of those reasons is a small mercy.
The permission of a bad table
I’ve come to think that a slightly bad working environment does something specific and useful: it lowers the stakes of the session. At a wobbling table you are not Beginning The Work. You are just getting something down before the coffee’s finished. Nobody sits at a wobbling table and expects the definitive draft.
The good desk makes an implicit promise, and the promise is a demand. Everything is ready. Where is it?
- At the desk, a bad morning is a verdict.
- At the café, a bad morning is the table’s fault, or the noise, or the fact that I only had forty minutes.
These are both fictions. The second one is a useful fiction, and I no longer feel bad about needing it.
Anything that makes a session feel provisional makes the session possible. This is why so much good writing happens on trains, and why almost none happens on sabbaticals.
What the desk is for
I have not sold the desk. It turns out to be excellent for the second half of the work — the reading back, the cutting, the checking of a quotation, the part where you need to see the whole piece at once and be a bit ruthless. That work wants exactly the conditions that kill the drafting: quiet, space, good light, no reason to stop.
So the arrangement now is that the first draft happens somewhere slightly wrong and everything after it happens at the desk. This sounds like a system. It is really just a description of what was happening anyway, which I resisted noticing for four months because I had spent the money.
That’s the actual essay, I think. Not that good desks are bad. That I bought a tool for a problem I had not correctly identified, and then spent a season trying to have the problem the tool solved.