Optimal Contraction Target

You have always loved the numbers.

Not the way a mathematician loves them, clean and self-sufficient, but the way a farmer loves weather — as a language that tells you what’s coming, whether or not you want to hear it. You learned the structural-demographic equations at the Halavet Institute when you were nineteen, sitting in a long hall of polished coralstone where northern light made the chalk on your instructor’s hands glow phosphorescent. The equations were old. Older than the Institute, older than the city of Duvenne that surrounded it, older than every institution you could name except the one that mattered: the Office of Cyclical Administration, which had been generating columns of numbers for eleven thousand years, and which would, your instructor told you with a calmness you mistook for wisdom, continue generating them for eleven thousand more.

You believed her. Why wouldn’t you? The evidence was in the ground beneath your feet. Duvenne sat on seventeen layers of previous occupation, each one documented. Beneath the coralstone of the Institute lay fired brick from the Forty-Second Cycle, reed-and-daub from the Thirty-Ninth, and beneath that — you had seen the core samples — a dark ribbon of ash from the single unmanaged crisis in the Continuance’s history, the Nineteenth Cycle, when the Office failed and the collapse ran wild. Two hundred years of silence in the archaeological record. No pottery. No seed caches. No written language for six generations. The population of the Duvenne basin dropped from four hundred thousand to fewer than twelve thousand. The civilization survived, barely, because institutional knowledge had been cached in the highland monasteries. But the cost — you could see it in the core sample, that dark band. You could hold it in your hands.

This is what you know. This is what you have always known: that the managed crisis is not cruelty. It is the alternative to cruelty. The equations describe a process as natural as a heartbeat — the elite fraction grows, opportunity compresses, intra-elite competition escalates until the state cannot contain it, and then either the state manages the correction or the correction manages itself. Every civilization in the archaeological record that failed to manage its cycles is gone. Not diminished. Gone. Their cities are under sand, under jungle, under water. Their languages are reconstructed fragments. Their children’s children’s children don’t know their names.

The Continuance knows its name. It has known it for fifty-five cycles. And you — you, Kiveste Sahl, Senior Historian of the Office of Cyclical Administration, you have spent thirty-one years making sure it continues to know it. You generate the column.

Here is what the column looks like, since the people in the streets of Duvenne have been demanding transparency for nine months and you have been arguing against it in committee meetings with a fluency that surprises even you:

A spreadsheet. An actual grid of numbers, produced by the actuarial bureau on the third floor of the Office, updated quarterly, tracking the following variables for each of the Continuance’s eight hundred and eleven administrative districts: population, elite fraction, wealth concentration index, institutional trust metric, agricultural surplus-to-demand ratio, birth rate, education-to-employment gap, and a composite score called the Pressure Index which synthesizes all of the above into a single number between zero and one hundred.

When the Pressure Index exceeds seventy in more than forty percent of districts for three consecutive quarters, the managed crisis is triggered. No committee votes on whether to trigger it. The threshold was set during the Twenty-Third Cycle and adjusted only twice since. The trigger is mechanical, as impersonal as a river cresting its banks.

What is not mechanical is what comes after. The column — your column — specifies which districts undergo primary contraction, which absorb displaced populations, which receive stabilization support. You generate it based on a formula that weighs each district’s absorptive capacity, infrastructure resilience, and demographic flexibility. The formula is published. Anyone can read it. Anyone can check your work.

What the formula does not specify is what primary contraction means for the people inside it. The closing of secondary schools. The suspension of medical subsidies. The withdrawal of agricultural extension services. The reallocation of water rights. For one generation — twenty-five to thirty years — a community becomes poorer, less healthy, less educated, and smaller. Families leave. Birth rates drop, partly through emigration and partly through the quiet calculus of people who look at their prospects and decide not to bring children into them. The community does not die. That is not the same as saying it lives.

You have generated the column for the last four cycles. Cycles Fifty-Two through Fifty-Five. You took over from your predecessor, Doret Valenn, who generated it for three cycles before you and who retired to a farmstead in the western uplands and, according to the last letter she sent you, raises goats now and does not read the news. You understood why when you generated your first column and saw the district names resolve into places you had visited, streets you had walked, markets where you had bought bread and spoken to vendors whose children’s schools you were about to close. Doret had warned you about that. She said: the formula is correct. The names are a distraction. Focus on the formula.

You focused on the formula. For four cycles, you focused.

The committee meets on the first Tensday of every month, in a windowless room on the sixth floor of the Office. There are seven of you. You are the only historian. The others are an economist, two demographers, a systems analyst, a public health officer, and the Director — Ouen Rast, who has held the position for eleven years, long enough to have overseen the transition from Cycle Fifty-Four to Fifty-Five.

You are fourteen years into the interstitial period, the rest phase between managed crises. The Pressure Index is at nineteen. By every measure the Office tracks, the Continuance is healthy, growing, and stable.

The committee is meeting because of the refusers.

Ouen calls them the Uncooperative Districts. There are thirty-one, spread across four provinces, formally declaring their intention to refuse participation in the next managed crisis. Not this one — the next one. The one that is one hundred and fifty-six years away.

They are refusing something that won’t happen in their lifetimes. And this is what you understand perfectly and cannot accept: they are refusing the premise. Not the timing, not the magnitude, not the formula by which districts are selected. The premise. The idea that managed crises are necessary. The idea that the column should exist at all.

“They’re asking us to guarantee that no future crisis will be managed,” Ouen says, reading from a petition that arrived last Tensday and that you have already read four times, each time finding it more lucid and less dismissible than the time before. “They want the Office dissolved. The actuarial bureau reassigned to census work. The Pressure Index discontinued.”

“They want to pretend the cycles don’t exist,” says Tesk Arunn, the economist, who is sixty-three and who has, in your experience, never been wrong about a number or right about a person.

“They don’t think the cycles don’t exist,” you say. You hear your own voice as if from a slight distance, the way you always hear it in this room, as if the windowless walls create a small acoustic lag. “They think the cycles exist and should be allowed to run unmanaged.”

The room goes quiet. Not the quiet of agreement. The quiet of six people looking at you and trying to determine whether you have lost your mind.

“Allowed to run unmanaged,” Ouen repeats. “Kiveste. You’ve seen the Nineteenth Cycle ash layer.”

“I’ve held it in my hands.”

“Then you know what unmanaged looks like.”

“I know what unmanaged looked like eleven thousand years ago, in a civilization that had no warning, no institutional knowledge, and no preparation for what was coming. The refusers are not proposing ignorance. They’re proposing that we track the cycles, publish the data, educate the population, and then let communities make their own decisions about how to respond.”

“That’s not a plan,” Tesk says. “That’s an abdication.”

“They would call it autonomy.”

“I don’t care what they would call it.”

You don’t respond to Tesk. You have learned, over four cycles, that Tesk’s certainty is architectural — it holds the room up, it keeps the agenda moving, but it does not bend, and you cannot argue with a load-bearing wall. Instead you look at the petition on the table and think about the woman who wrote it, whose name is Lesse Dain, who is a district administrator in Rovenne, and who — this is the part that keeps you up at night, the part you have not told the committee — holds an advanced degree from the Halavet Institute. She studied the same equations you did. She sat in the same hall. She held the same core samples.

She generated her own column once, as a graduate exercise. You have seen it in the Institute’s archives. It was competent. Methodologically sound. Her instructor’s notes say: Clear thinker. Understands the necessity.

And now she is leading thirty-one districts in refusing the necessity she once understood, and you cannot dismiss her as ignorant, because she is not ignorant. She is something worse. She is informed and unconvinced.

You go home. Your apartment is in the Lenten Quarter, in a building old enough that the coralstone walls hold the summer heat at bay until evening. You have lived here for twenty years. The building is in a district that has never been selected for primary contraction — Duvenne’s absorptive capacity is too high, its institutional density too great. The formula protects cities like Duvenne. The refusers have pointed this out, in language that is measured and devastating: the column is generated by people who will never appear in it.

You sit at your desk and open the actuarial projection for the next two centuries. You stare at the numbers the way you have stared at numbers your whole life, waiting for them to tell you something.

In one hundred and fifty-six years, the Pressure Index will cross the threshold in approximately three hundred and forty districts. The managed crisis will require primary contraction in eighty to one hundred and twenty of them. The affected populations will lose fifteen to twenty-five percent of their people through emigration and reduced birth rates. The elite fraction will decrease. The Pressure Index will drop below thirty. The recovery will begin.

This has happened fifty-five times. The numbers are very good at this.

What the numbers do not say is whether the people in the contracted districts experience their contraction as a managed crisis or as the destruction of their lives. The numbers track populations. They do not track the woman in Rovenne who ran a school that closed, or the family in the Geshen corridor that relocated to a district they had never seen and will never love. The numbers call this acceptable. The numbers are, in the strict mathematical sense, correct.

You close the projection. You open it again. You will be dead in one hundred and fifty-six years, which is a fact that has never before struck you as relevant and now strikes you as the most relevant fact in the world. You will generate the column and you will die and the column will outlive you by a margin so vast that your existence is, in the actuarial sense, a rounding error.

The refusers know this too. That is what Lesse Dain’s petition says, buried in the seventh paragraph, in language so calm it took you three readings to hear the fury underneath: We are asking you to stop calculating our lives as though they are instruments of a continuity we did not choose and from which many of us do not benefit. We understand the mathematics. We reject the conclusion that our reduction is a service we owe to the future.

You read that paragraph again. You try to find the flaw in it. You have been trying for nine months.

The committee meets again with observers — two representatives from the refuser districts, seated along the wall, not at the table, a distinction no one acknowledges and everyone understands.

Lesse Dain is one of them. Younger than you expected. Dark hair, short, practical clothes cut for weather rather than appearance. She does not look angry. She looks like she is listening with a precision that makes you want to be more careful with your words.

Ouen presents the updated projection. Tesk walks through the economic modeling. You present the historical analysis: all fifty-five previous managed crises, with particular attention to the four that were delayed by more than a decade and the consequences of delay. Deeper contractions. Longer recovery periods. Higher population loss. Your analysis is thorough, well-sourced, impossible to argue with on its own terms.

Lesse Dain does not argue with it on its own terms.

“Your analysis assumes the managed crisis is the only available response to the Pressure Index crossing the threshold,” she says, when Ouen opens the floor for comment. Her voice is level. “It assumes that because managed crises have been the response for fifty-five cycles, they must remain the response. This is an argument from tradition dressed up in actuarial language.”

“It’s an argument from evidence,” you say, and hear the defensiveness in your own voice like a crack in a wall you thought was solid.

“It’s an argument from the evidence you chose to collect. You track population. You track elite fraction. You track wealth concentration. You do not track what the contracted districts lose that your metrics cannot measure — the social cohesion, the cultural practices, the relationships between families who lived beside each other for generations and were scattered by an administrative decision made by people who have never lived in an outer district. You have eleven thousand years of data and none of it measures what actually matters to the people in the column.”

You want to say: we measure what can be measured. You want to say: if you have a better metric, propose it.

What you say is nothing. You sit in the windowless room and think about the seventeen layers of occupation beneath it, and whether those layers represent survival or something else — seventeen versions of Duvenne that were made and unmade and made again, each one convinced it was the same city, none of them asking who decided.

Here is the question you cannot stop asking yourself, in the weeks that follow:

If the Continuance had collapsed during the Nineteenth Cycle — truly collapsed, not survived in the highland monasteries, but gone completely, the way every other civilization in the archaeological record has gone — would the people who lived in the next civilization to arise in the Duvenne basin have missed it? Would they have known what they were missing? Would the accumulated knowledge and art and agriculture of the first eighteen cycles have mattered to them, or would they have built their own knowledge and art and agriculture and lived lives that were, in every way that mattered to the people living them, complete?

This is not a historical question. You cannot answer it with evidence. It sits in the space between your discipline and something older than your discipline, something that the equations were built to make unnecessary.

You go to the deep archive in the basement, where core samples are kept in climate-controlled cases and the records from the earliest cycles are stored on tablets of fired clay. The air smells of mineral dust and the faint chemical sharpness of preservation agents. You pull the Nineteenth Cycle core sample and hold it in your hands. The ash layer is thin. Maybe eight centimeters. The worst catastrophe in the Continuance’s history, compressed into a width you can span with your thumb and forefinger.

You think: two hundred years of silence, in eight centimeters.

You think: every life lost in that silence is in this dust.

You think: and if the managed crises prevented fifty-four repetitions of this, then the column — your column — has saved more lives than any other document in history.

You think: and if the managed crises created fifty-four smaller, slower, more dispersed versions of this — versions that don’t leave ash layers because they happen over decades instead of years, because the people don’t die but merely leave, because the schools close but no one burns them — then the column has done something that doesn’t have a name in your discipline’s vocabulary. Something that is not saving and not killing but a third thing, a thing that your numbers do not measure because you never thought to build a metric for it.

You put the core sample back. Your hands are dusty. You wash them in the basement sink and watch the water run gray and then clear.

Ouen calls a special session. The refuser districts have grown to forty-seven. Three provincial governments have issued statements of support. Ouen distributes a document titled CONTINGENCY PLANNING: SCENARIOS FOR NON-PARTICIPATION and you understand, with a coldness that settles in your chest, that Ouen is not asking whether to proceed with the next managed crisis. Ouen is asking how to proceed if a significant portion of the population refuses to cooperate.

“We have one hundred and forty years,” Ouen says. “Political movements rarely sustain themselves across generations.”

“This one has sustained itself for eleven thousand years,” you say, and then stop, because that is not what you meant. The Office has sustained itself for eleven thousand years. The refusers are fourteen months old. But the sentence came out with a conviction that does not match its content, as if your mouth knows something your mind has not yet articulated.

Ouen looks at you. Tesk looks at you. The demographers look at each other, which is worse.

“I need to visit Rovenne,” you say.

“You’re a historian, Kiveste, not a field anthropologist.”

“Then let me be a field anthropologist for a month. We have one hundred and forty years. We can spare thirty days.”

Ouen considers. Tesk objects. The systems analyst and public health officer vote in your favor. Ouen breaks the tie. You pack a bag. You do not bring the actuarial projection.

Rovenne is forty thousand people in the eastern uplands, seven hours by rail from Duvenne. Dry grassland cut by seasonal rivers reduced, in late summer, to braided trickles in wide gravel beds. The town center has a market, a district hall, two schools, a medical clinic, and a library larger than you would have predicted for a community of this size. Lesse Dain built it — organized the funding, designed the shelving, spent two years of weekends laying the foundation stones.

You spend a month in Rovenne. You attend three district council meetings where people argue about water allocation with the same procedural intensity as your committee, except they are arguing about their own water. You eat at a table with a family of seven who have lived in the same house for nine generations and who show you a wall where the heights of children have been marked since before the Fifty-First Cycle. You visit the school Lesse Dain built the library next to and watch a class of twelve-year-olds learn the structural-demographic equations from a teacher who knows exactly what those equations could do to her town.

You do not write the report.

You sit in Lesse Dain’s library on your last evening in Rovenne and she brings you tea and sits across from you and says, “You see it now.”

And you say, “I saw it before. I always saw it. The formula accounts for it. Social cohesion index. It’s a variable.”

“No,” she says. “It’s a number. A number that stands in place of something that isn’t a number. And you’ve been treating the number as if it were the thing.”

You drink the tea. It is a local blend, sharp and faintly sweet, made from a plant that grows only in the eastern uplands, in soil that would revert to scrubgrass within a decade if the community that tends it dispersed. This is the kind of fact that does not appear in the actuarial projection. This is the kind of fact that you have been trained, for thirty-one years, to set aside.

You return to Duvenne. You go back to the windowless room. Ouen asks for your report. You say it is not finished. Ouen gives you another month. Tesk, in the hallway afterward, says: “Don’t go soft on us, Kiveste. The numbers are the numbers.”

The numbers are the numbers. You have loved them your whole life. You love them still. The Pressure Index will cross the threshold in one hundred and fifty-five years. The managed crisis will require contraction in eighty to one hundred and twenty districts. The populations will decrease. The elite fraction will normalize. The civilization will survive.

You sit at your desk and open the column. The empty column, the one that will be filled by you or your successor. Eight hundred and eleven rows. Each one a district. Each district a place where people grow a particular plant for tea, or build libraries on weekends, or tell their children a specific story about why the river floods in spring. Your formula will evaluate each row and assign a number, and the formula will be correct — mathematically, historically, defensibly correct — and you will generate it the way you have always generated it, with care, with diligence, with the quiet conviction that the work you do is necessary.

You look at the column. The column looks back.

You do not close the file. You do not write the report. You sit in your apartment in the Lenten Quarter, in a district that will never be contracted, at a desk that has been yours for twenty years, and outside your window the city of Duvenne goes on doing what cities do in the evening — cooling, quieting, turning on its lights one by one, while somewhere in Rovenne, Lesse Dain is shelving books in a library that your formula says should not exist in forty years, and the numbers wait for you to tell them what they mean, the way they have always waited, patient as ash.