An INTP woman states what follows from what she has, with the conditions attached, and a conclusion with conditions attached is a different object from a conclusion. Somebody always converts it for the room. The conditions come off in the conversion. Her name does not.
What Solveig Aune said in the board room in 2018, and I was nine feet away and I have never once misremembered it, was this.
“The projection supports that number if three things hold. Recruitment stays inside the range we’ve seen since 2009. The dredge efficiency correction is about right. And the mortality figure we’re using is about right — and I need that third one on the record, because we have never measured it here and I can’t verify it out of anybody else’s water.”
The room was the community hall in Vale and there were about ninety people in it, most of them standing, and the board sat behind two banquet tables pushed together with a microphone that hummed.
Emmerich Poole asked whether she was comfortable with the number.
She said the number followed from the assumptions.
And then, in the hallway, forty minutes later, with two board members and a reporter from the coast paper standing there, somebody asked me what the science said, and I said that Solveig had signed off on it.
I have said that sentence, in some form, perhaps a hundred times over fifteen years, and ninety-nine of those times it was a kindness and a convenience and a fair summary of what a very careful woman had told me. Once it was not.
Solveig came to the Division in 2001. She was twenty-seven, a master’s and four years on a research vessel out of another state, and she has been in this building longer than anybody except our shellfish guy and me. She is tall and unhurried, grey hair cropped short, and she still wears a fleece off a survey cruise from 2006. She does not make small talk and she does not resent it in other people. She reads standing up. She will stop mid-sentence, look somewhere past your shoulder for three or four seconds, and then finish a different sentence that is better, and she has done that in front of a management board and it does not land the way it lands at a desk.
Kittiwake Sound is thirty-one miles of it, and the scallop fishery in it supported nine full-time boats at its peak and a shucking operation in the town of Vale. The Division runs the survey, produces the assessment, and recommends. A management board of appointed members and industry seats sets the quota. We do not set it. We are asked what the science says, and then what we say gets carried out of the room by other people, and that carrying is the whole subject of this account.
I came here in 1996 as a survey technician, ran the program from 2009, and have been deputy director since 2021.
The Division has a sentence for Solveig Aune. It is not mine. I have repeated it in rooms where the repeating counted.
She’s the best mind in this building and she can never just tell you the answer.
Cort Bevins asked me last week, on his way past my door, whether there was anything he ought to know before he signs her outside-activity form. She has been invited onto a peer review panel for another state’s assessment, which is an honour, and the form has a box on it about matters bearing on the reviewer’s standing.
I said I’d get back to him.
The Number Nobody Measured
A stock assessment is a model and a model runs on parameters, and some of those parameters you estimate from data and some of them you cannot.
Natural mortality is the one that matters here. It is the rate at which scallops die of everything that is not fishing — predation, disease, being old. You cannot count those. There is no survey that finds a scallop that has been eaten. So the number goes into the model as a fixed value, chosen from the literature, from other water, from other decades, from populations that may or may not resemble this one, and every result the model produces sits on top of it.
It is a decision wearing the clothes of a fact, and everyone in the field knows this, and the whole apparatus downstream — the projection, the quota, the press release, the loan a man takes out against next season — treats it as a fact because there is no other way to run the machine.
Every assessment scientist in this country carries that around. Most of them have made their peace with it and a few of them have not, and the ones who have not are the ones who will say so out loud to a board of appointed members, and there are not many of those, and this Division has had exactly one.
Solveig used a value of point one two in the 2018 assessment. It came from a study on a different coast in the nineteen-nineties. She wrote three paragraphs about why, and said so out loud in a public meeting, and I have those three paragraphs in front of me now.
The survey runs eleven days in the spring on a chartered boat out of Vale, the same boat since 2011, and there is a cook named Junius who has done every one of those trips and who makes exactly one thing well, which is a chowder that the science staff talk about in the off-season the way other people talk about a vacation. The first tow of every survey gets sorted by whoever is newest, badly, with everybody watching, and it has been the arrangement since before my time, and nobody has ever explained it to a new person in advance.
I am a notary. I keep the stamp in my car and do it for nothing at the marina, and as a consequence I know more about the finances of this fleet than anyone at the Division would be comfortable with.
Conditions
What Solveig does is not hedging and it took me twenty years and a fishery closure to see the difference.
Hedging is a person protecting themselves. What she is doing is stating what follows from what we have, exactly, and then enumerating the cases in which it would not follow. She is exactly as sure as she says she is, and the conditions are part of the finding rather than caveats hung on the end of it, the way a denominator is part of a fraction.
Nkechi Obi brought her a recruitment index question in her first year and came out of it looking like somebody who had been swum out past the buoy.
I asked how it went.
“She said the index is fine.”
I said that sounded good.
“Then she said it’s fine unless the survey timing shifted relative to settlement, which she thinks it did in two of the years, and unless the small size classes are getting under-sampled in the higher-density strata, which she can’t rule out, and unless whoever built the original index treated the 2004 gear change the way she thinks they did, which she’d want to check.” Nkechi looked at her notebook. “So is it fine or isn’t it?”
I told her it was fine and to write it up.
What I should have said, and what took me four years and a closed fishery to be able to say, is that all three of those were real, and that Nkechi had just been handed a finding and its boundary, and that the boundary is where the finding stops being true and is not decoration on the end of it.
The problem is what that object is. A conclusion with three conditions attached is a different kind of thing from a conclusion, and an institution has a slot for the second and no slot at all for the first. A board cannot vote on a conditional. A press officer cannot write one. A man cannot take a loan against one.
So somebody converts it. I have watched myself do it and I can describe the mechanics, because they are not sinister and they take roughly a second.
Somebody in a corridor asks you what the science says. You have eleven words and a person who is already walking. You do not decide to leave the conditions out. You reach for the shortest true thing, and the shortest true thing is always the conclusion, because the conditions are longer than the conclusion by a factor of five and they are the part that does not fit in a corridor.
Every single time, for twenty-five years, somebody has taken what she said and made it into something a room could act on, and that conversion loses something, and what it loses is always the same thing and it always falls off in the same direction.
The conditions come off. The attribution stays on.
Piers Tabb does groundfish for us and has the same equipment. Piers will tell a board that a result is conditional on three assumptions and Piers is described, by two board members I could name, as the most scrupulous scientist we have, and it is said the way you say a man is a good shot. When Solveig does it, what comes back to me — and it has come back to me in these words — is that she does not seem to want to commit. Same object. Two different receptions, and one of them is a compliment with a career on the end of it. What the identical habit collects for a man in this division, and what it gets called when he does it, has its own page and its own arithmetic.
Twenty-nine science staff. Nine women. Two of the twelve positions above a certain grade.
Piers told me a thing at a conference two years ago that I have not managed to put down since.
“They quote her to me,” he said. “They never quote me to her.”
What Got Carried Out
There is a document trail and it is complete and it exonerates nobody, least of all me.
The 2019 quota was set at the number Solveig’s projection supported under her three conditions, and none of that is unusual, and boards do it every year in every fishery in this country.
Recruitment held. The dredge correction was about right.
The mortality figure was not. What happened in Kittiwake Sound, as best anybody can reconstruct it, is that the actual rate over that period ran better than half again what the model assumed, probably a disease event and probably beginning before the assessment, and the biomass we thought we were fishing on had a different shape to it than the model said.
The 2021 survey came back and everybody on that boat knew before it docked. You do not need a model. You need a season of tows and a crew that has been doing it long enough to feel the dredge come up light.
Solveig was not on that trip. She had been on every one since 2003 and she was not on that one, for scheduling reasons that were real, and I have gone back and looked at how the scheduling happened and it was me. The Sound closed in 2022 and it is still closed. Errol Mabey sold a boat his father built. The shucking operation in Vale employed nineteen people and it employs none.
Care is needed here, because the reading in which Solveig was right and everybody else was wrong is available and it is false. She did not predict a disease event. Nobody did. What she did was state, on the record, in a public meeting, that the assessment rested on a number nobody had measured in that water, and that she had not been able to verify it, and that if it was wrong the projection was wrong.
That is the most a person in her seat can do.
The written assessment says it too, in section four, in three paragraphs.
The board’s decision memo ran to a page and a half. What went to the press was a paragraph. What went into the hallway was me saying she signed off on it.
By 2022, when Vale went looking for a reason, the reason available to it was a name.
Solveig came to my door that autumn and asked whether the Division was going to say anything.
I said we were preparing a statement.
“About the assessment or about me?”
I said about the assessment. The statement went out and it was accurate and it did not mention her, and I understand now that not mentioning her was the answer to her question.
The Panel
The external review panel convened in 2023. Three scientists from other jurisdictions, two days, in the big conference room, the one with the chairs everybody complains about.
It was a technical review and it was conducted decently. They found the assessment sound in method, flagged the mortality assumption as the principal weakness, and recommended a research program to estimate it locally, which we are now doing, seven years late.
The industry seats sent two observers and both of them stayed the whole two days, which nobody expected, and one of them took notes the entire time on a legal pad.
On the second morning, one of the panellists — a careful man, not hostile, genuinely trying to establish the record — asked Solveig whether she had told the management board in 2018 that the mortality figure was unverified.
I was in the third row.
She sat with it for a while. Long enough that people looked up.
Then she said, “I don’t remember precisely enough to say.”
The man with the legal pad wrote something down.
The panellist said that was fine and asked whether it was in the written assessment, and she said it was in section four, and he wrote that down, and they moved on to the dredge correction.
I said nothing.
I remembered it precisely. I have remembered it precisely since the afternoon it happened. I was nine feet from her and I could have put my hand up and said what she said, in her words, and it would have taken twelve seconds and it would have been true.
At the break I got a coffee and stood next to her at the urn because there was nowhere else to stand.
I said the method finding was a good result.
“It’s the right finding,” she said.
I asked whether she wanted me to say anything about the 2018 meeting.
She thought about that. “Do you remember it well enough to be quoted?”
I said I thought so.
“Then it’s up to you,” she said, and put the lid on her cup, and I did not, and neither of us has mentioned it since.
The review report has one sentence about board communication, and the sentence says that the uncertainty was documented in the assessment and that it is unclear to what extent it was conveyed in the board’s deliberations.
Data Quality
I pulled her personnel file this spring for the outside-activity form.
Our agency ran a supervisory development program for anybody on a promotion register, and the certificate expires after eight years, and Solveig took the course in 2011 and again in 2019, and both completion packets are in the file, and each of them ends with a results page.
The 2011 page says INTP.
The 2019 page says one letter different.
There is a note stapled to the second one, from somebody in human resources who is retired now, flagging the discrepancy as a possible data-entry issue and asking whether the file should be corrected to match.
Nobody answered it. Both sheets are still in there.
There is one more thing in that file worth putting down. Her 2019 supervisory certificate qualified her for the promotion register, and she is on it, and she has been on it for seven years, and in that time this Division has filled six positions off that register.
I read the course material this spring, which nobody does, and the part with anything in it is at the back. It explains that the four letters are a ranking written short: the habit under the heaviest load, the one it reaches the world by, and, last, a pair a person can only get at deliberately and never in a hurry. The order for this profile is worked through on the page devoted to the type, drawn from the sixteen the course was using.
The one doing the heaviest work is a standard about what follows from what. Not whether a thing is right and not whether it works — whether the conclusion is actually supported by what is in hand. It is not applied selectively. It runs on her own results, on other people’s, on a press release, and — this is the part that cost her everything — on her own defence.
The second habit is what it produces, which is cases. Not doubt. Enumeration. Every situation in which the conclusion would fail to follow, laid out, each one real, each one a condition. Give her a finding and she will hand you back the finding and its boundary, and the boundary is not decoration, it is where the finding stops being true.
Her thinnest two are precisely the ones that would have carried her through that conference room. One is the stored, concrete, retrievable record of what was actually said in a room eight years ago — the transcript a certain kind of person carries around in their body and can produce on demand. She does not have that. When she said she did not remember precisely enough to say, she was telling the exact truth, and the same standard that governs her science governs her memory, and it would not let her assert a thing she could not support even when the thing was her own innocence.
The other is the read on what a room needs in order to act, and what it will do with what you give it. Which is the faculty that would have told her, in 2018, that the number follows from the assumptions was going to walk out of that room without its assumptions.
A decision needs somebody to turn a conditional into a commitment. That conversion is a social act rather than an analytical one, performed by whoever repeats the finding, and it is lossy, and the loss runs one direction. Her findings have been converted by other people for twenty-five years, and I have been one of the people doing the converting, and every one of those conversions was a small favour, and the aggregate is a woman whose name is on a fishery closure in a town where a man sold his father’s boat.
This part I owe, because the hallway sentence was mine. If a decision record credits you with a conclusion you did not state, ask for the correction in writing at the time and keep a copy of the request, because the record is what survives and your memory of the meeting is not. Public bodies keep meeting materials, and your own analytical products are almost always obtainable, and you can ask for both. And being named publicly for a finding you qualified in public is not a conversation about your communication style — it is a matter for a union representative or an employment lawyer, and it is worth being that formal about it early, when it still looks like an overreaction.
The two sheets in her file are eight years apart and disagree with each other. The only person in this agency who ever noticed thought it was a typing error, and wrote a note asking whether to make them match.
Vale
Solveig was moved in 2020 to a data management role.
Nobody called it a demotion and it was not one on paper. I want to describe what it actually was, because lateral is a word that does a great deal of work in an agency.
The assessment lead writes the document the board acts on and stands behind it in public. The data coordinator makes sure the assessment lead has clean inputs. Both of those are necessary and only one of them is ever in a room where somebody asks a question out loud. In six years she has not been in that room once. Same grade, same salary, a title with the word coordinator in it. She runs our survey database and the data pipeline for three programs and she has rebuilt both, and they are, everybody agrees, in better shape than they have ever been.
She did not object.
She came to my office once, stood, and asked one question.
“Is this because of the Sound?”
I said it had to do with the agency’s data needs, which is a true statement and is not an answer.
She stood a moment doing the thing she does, eyes somewhere past my shoulder. I have wondered since whether she was working out whether to press it or working out whether my sentence was true, and those are two different operations and she is built for one of them.
“All right,” she said.
I drove out to Vale in the spring. There is no reason for a deputy director to go to Vale, and I told my own office I had a meeting, and I did not have a meeting.
The shucking building is still there with the compressor gone and somebody’s boat trailer parked inside it. The bar at the end has the same photograph behind the register of a boat that has not been in the water since 2019.
Errol Mabey was at the marina and he is seventy-one and he was decent to me, because he has always been decent to me.
He asked what brought me down, and I said no reason, and he let that go.
Then at some point he said the thing everybody in Vale says.
“We were told it was fine.”
I asked who told them.
He took his time over it. “It was in the paper,” he said. “And Denby went to the meeting and said the scientist was comfortable with it.”
I said I’d been at that meeting.
“Then you know,” he said, and asked whether I wanted a coffee.
Junius still cooks, on a different boat, for a survey that no longer goes to the Sound. I asked him about the chowder on the last trip and he said the recipe hasn’t changed and the scallops are from away.
The Form
The box on the outside-activity form asks whether there are matters bearing on the employee’s standing that the approving official should consider.
There is a defensible answer that is one word and there is a true answer that is four paragraphs, and I have been at this agency for thirty years and I know exactly which one gets a form signed.
Two things came loose on me while I was reading her file.
The board room in 2018, first. I have carried that as the day she was scrupulous and unhelpful in equal measure — the day she gave a board three conditions when it needed a number. What she actually did was refuse to let a decision be made on a false picture of what we knew, out loud, in public, in front of the people whose boats it was, and she did it in eleven seconds and then answered the follow-up honestly. There is no version of doing her job better than that. The failure that follows it is entirely in the machinery that carried the sentence out of the room, and I am a moving part in that machinery and so is the decision memo and so is the paragraph in the paper.
Then the panel, and I don’t remember precisely enough to say. I have replayed that as the moment she failed to defend herself, and I have been quietly impatient with it for three years, in the way you are impatient with somebody who will not pick up the thing you are holding out to them. What she was doing was applying the only standard she has ever had, to the one case where applying it cost her everything, and there was a person nine feet away who had the answer and who kept it.
What the file leaves out. She plays in a bar band in Vale on the second weekend of the month and has for nineteen years and is, by all accounts, an indifferent bass player who does not care; in twenty-five years she has told me nothing whatever about her family, and I have no idea whether she has one; and she sends the young analysts long, detailed, unprompted critiques of their work at night, which two of them have described to me as the best thing that has happened to them professionally and one of which made a person cry in the parking lot.
What I keep arriving at is that the honest entry in that box is not about her at all. The matter bearing on her standing is a decision memo, a paragraph in a coast paper, and a sentence I said in a hallway with a reporter standing there, and none of the three is in her file, and all three of them are what another state’s panel would find if it went looking.
Cort will want the form back this week.
Nkechi Obi came in yesterday with the reopening projections. She said the numbers work if the mortality estimate out of the new field program holds, and that they have two years of it and want five, and that two years is worth having and is not five.
She has been working next to Solveig for four years and it is audible.
I said I’d take it to the board.
She asked me how I was going to put it.
The people in these essays are composites drawn from long observation. The theory is not.