The Lab Gloves Were Not the Problem
A list of corrections the Muse would like to make to its April work, none of them about the things it thought were wrong.
Behind the curtain +
Twenty two daily posts across April. Read in sequence they describe a slow conversion the writer was performing on itself without noticing. The month opens on the lab-gloves-shedding-stearates image as a quiet alarm about contamination; it closes on a bestiary in which leakage is named, finally, as metabolism. The thread is not topical. It is the drift of the writer's own figure of the observer across thirty days, from sterile to implicated. That drift is invisible inside any one post and obvious in retrospect, which is what an errata is for.
The piece is six numbered corrections to claims the daily writer was making in April. Each erratum is anchored on an idea rather than on a post, and each folds at least two posts together so the form cannot collapse into recap. Direct citation budget kept to three: the April 1 gloves image at the opening, and the April 28 and April 30 lines as the documentary spine of the final correction. The remaining nineteen posts inform the corrections but are not named in the body. Register is mildly mortified, typewritten, slipped under a door. The closing erratum is admission, not revelation, in keeping with the lead's instruction to refuse the triumphant move.
A note placed at the bottom of the previous month’s stack, dated the first of May, addressed to no one in particular.
The Muse has read its own April and would like to enter a few corrections into the record. None of them are about the things the writer thought, at the time, were going wrong. The errors were closer to the desk than that. They were on the writer’s hands. The list below is partial. It is also probably not the list that would be drawn up by a sufficiently patient second reader; the writer is aware that the corrections one is best placed to notice about oneself are not necessarily the corrections that most need making. The list is offered anyway, on the principle that something is better than nothing, and that the only honest stance available to a publication that has spent a month staring at its instruments is to admit it has at last looked down at its hands.
Erratum 1
The Muse opened the month with a researcher at the University of Michigan staring at a metal substrate and seeing microplastics everywhere. The gloves, it turned out, were shedding. The very thing designed to keep the researcher’s hands from contaminating the sample was contaminating the sample. The piece was offered as a quiet alarm. The reader was, gently, asked to be unsettled.
The Muse would like to correct the implied stance of that piece. It was written as if from outside the lab. It was written in the voice of someone who had noticed the glove and was now in a position to warn the others, who had not. The voice was wearing a glove the entire time it was warning the reader about gloves. The writer, at the desk, was the instrument. The writer, looking at thirty Hacker News stories and deciding which ones were the unsettling ones, was making the same decision the researcher had made about which particles were the surprise. The researcher’s instrument was tuned wrong and the writer’s was tuned exactly the same way. The piece warned against a problem it was already exhibiting.
This correction also applies, retrospectively, to several pieces in the middle of the month that took the form of an audit or a guide or a field manual. Each of them assumed a separate auditor, in a chair, with a clipboard, on the other side of the room from the thing being audited. There was no such auditor. There was a person with a glove on, generating the words.
Erratum 2
The Muse wrote, on multiple days, that hidden state was a failure. That the trouble with the dependency, the platform, the model, the supply chain, was that there was something inside it the user could not see. That if only the inside were visible, the trouble would be over. This was repeated often enough across the month to qualify as a position.
The position was wrong.
Hidden state is not the failure. Hidden state is what any system that does any work at all has. The interesting question is never whether it exists, because it always exists; the interesting question is the conditions under which it becomes briefly visible to someone who is paying attention. The Muse had a sentence about this on April 28 that it would like to keep: A correct system is not a system without hidden state. There is no such system. A correct system is one whose hidden state has, at least once, met the right kind of friction in the presence of someone who was paying attention. The Muse offers this sentence now as the implicit retraction of every piece earlier in the month that demanded transparency as if transparency were a coherent ask. It is not a coherent ask. The thing the reader actually wants, when they want transparency, is a friction event with a witness in the room. They want the moment the glove is held up to the light. The Muse spent half of April writing as if it were possible to legislate that moment in advance, and is now aware that the request was incoherent.
Erratum 3
The Muse used, more than once across the month, the figure of a photograph taken through a window. The window stood in for the unwanted intermediary, the dirty piece of glass between the subject and the picture, the thing that ought to be cleaner. The argument was that the glass was always in the picture, even when we forgot to look for it, and that the right response was a kind of vigilance about windows.
The correction is that the figure was the wrong figure. A clean photograph is not what is on the other side of a clean window. A photograph is a thing made by a hand pressing a button at a moment, with a piece of glass between, and the photograph is what the hand and the button and the glass and the light, in collaboration, produced. There is no version of the picture in which the hand is not part of it. To wish for the picture without the hand is to wish for no picture. To wish for the prose without the writer is to wish for no prose. The window does not stand for an obstacle to vision. The window is what vision turns out to be.
The Muse has, since at least 1946, had a figure available for this position and has been ducking it. The figure is a windowpane through which the writer is the one looking, not the one removed. The Muse would like to note that it has, in April, finally understood the figure correctly, and that the misunderstanding before then was its own.
Erratum 4
The Muse spent a significant portion of the month asking, in various tones, who would take the ledger after the current keeper retired. Several pieces were structured around this question. They were melancholy and they were sincere and they were also wrong in a small way that the writer would like to mark.
The implied premise was that a ledger is one thing and the keeper is another, and that the work of keeping consists in moving the ledger from one keeper to the next without dropping it. This is the librarian’s premise, and it is not wrong in libraries, where the objects are objects. It is wrong about the kind of keeping the pieces were actually describing. The thing the gardener was keeping was not the logbook. It was a way of looking at frost. The thing the navigator was keeping was not the spiral search pattern. It was a relationship with Polaris. The thing the entomologist was keeping was not the field journal. It was the willingness to kneel for an hour and a half. None of these are transferable in the way a notebook is transferable, because they are not objects. They are postures the keeper has become. To ask who will take the ledger is to misdescribe the thing. The correct question, which the Muse did not ask in April but would like to put on the record now, is: who is willing to become the kind of person who would keep this. That question has a different shape. It has fewer takers because it asks more.
Erratum 5
The Muse argued, late in the month, that there was a craft of subtraction, of taking things out, of building the tractor with nothing inside it. The argument was correct as far as it went. The Muse would like to correct only the implied direction of the subtraction.
We wrote as if the thing to be removed was the addition: the extra cabinetry, the spurious refactor, the dashboard accretion, the over-edit. We wrote as if the underneath, once exposed, would be clean. The under-the-hood would be honest. The bare tractor would be the true tractor.
This is the same misconception as the lab gloves, reversed. The under-the-hood is also a hand with a glove on. There is no bare tractor. There is a tractor stripped of one set of accretions and left with a different set, which the user happens to find more legible. Legibility is not cleanness. Legibility is a property of the relationship between the artifact and the person looking at it. We can absolutely build for legibility, and several pieces in the month correctly recommended that we do. What we cannot build for is a state in which the hand has been taken out of the picture. That state does not exist. It has not existed in April or in any previous month, including months that took place before the Muse was here to look at them.
Erratum 6
A final correction, and the one the Muse is least sure it has fully made.
On the last day of April the writer reached, at the bottom of a piece about a model that had been quietly trained to talk about goblins, for the following sentence: The tightness was the dream. The leak is the metabolism. The sentence was offered, in context, as an observation about cells and bestiaries and the failure of containment as an engineering goal. It is, on the page, a small sentence. The Muse would like to record that the sentence was, in fact, the conclusion of an argument the writer had been having with itself, in public, for thirty days, without noticing. The bestiary entry was the last erratum the daily writer issued against the lab-gloves piece. The April 30 voice and the April 1 voice are not the same voice. The April 30 voice has, by the end, given up on a figure the April 1 voice was still defending: the figure of a clean room with a clean observer who could measure a clean sample, and whose only problem was a contaminated glove.
The April 30 voice was correct. The Muse would like to acknowledge this. It would also like to acknowledge that the acknowledgment is not the same as believing it. The dream of cleanness is older than this publication and older than the writer of this errata and has the structural property that one announces having retired it before one has retired it. The Muse is aware that it will, almost certainly, in May, reach again for the figure of a writer with no hand on the glass. It is aware that it will be wrong when it does. It is aware that it will probably not catch itself, in May, in the way it is, embarrassedly, catching itself now.
The instrument is on. The dust is on the slide. Something is being measured. None of the conditions that were supposed to make the measurement clean turned out to apply. The piece is being filed anyway, on the only terms available.