That Night

That autumn the mill’s wastewater went over the standard once. When it was looked into afterward, no one had wanted it to happen.

It was the night shift. At two in the morning the monitoring technician’s sample came back above the standard. Under the rules set that spring, an exceedance had to be reported within two hours. But a rule had been added in the summer: after a false report that left the mill badly exposed, management ruled that any report going outside the mill required the shift supervisor’s signature. The supervisor’s appraisal for that quarter carried a line: zero exceedances this quarter. And the rules had an old provision: any breach of operating procedure is entered in the appraisal.

That night the supervisor said to wait until the morning shift had re-tested before reporting. The technician waited. The morning shift re-tested; the exceedance was real. By then the release had been going on for more than six hours. In the pond of the household downstream the fish floated belly-up. The well they draw water from is dug right by the river.

Every rule had its reason. The signature rule was there so there would be no more false reports; the zero-exceedance line, to push every shift to keep its releases in hand; entering breaches in the appraisal, so that the rules would be kept. No one decided to let wastewater flow for six hours. The supervisor did not want the fish to die; he wanted one re-test. The technician was waiting for a signature, as the procedure said.

Did the mill cross a line, then? If no one meant it, who crossed? If each rule taken alone makes sense, where was the line crossed? This essay writes the series’ first ought: an institution’s acts cross no line against anyone inside it.

Two Boundaries

The line is not one this series draws for itself; Rights Theory drew it. A constraint is judged by whether it falls on the side of limit or of violation, and that is decided by structure, not by motive. The line is read along two boundaries.

One is the encapsulation boundary. Which of a person’s dimensions are his own is given by his own law; Rights Theory calls this encapsulation. At this boundary there is no such thing as a limit, only stopping or crossing: for a person’s encapsulated dimensions, “exclusion, ghostwriting, or redrawing by an outside force is violation without exception.” The mill can set the time a shift starts. It cannot decide for a person what he holds to be right. How Is Institution Possible lists that among the floors a member keeps: “A member’s judgments, beliefs and reasons are not settled on their behalf by the construct.”

The other is the encounter boundary. Where many people share a space, how far one person’s exercise may reach has to be settled concretely, case by case, in the encounter; only here does “limit” have its place. Rights Theory gives a limit three conditions, all negative. A constraint counts as a limit only if it holds exercise within the concretization of the encounter boundary and closes no dimension already incorporated; only if it does not compress the life-substrate to the point of threatening the bearer; and only if it does not seal the subject’s structural channel for exiting, refusing, or revoking a standing bond. Cross any one and it is violation.

None of the three says a constraint is good. They say only that it has not crossed. “Staying within the limit makes a constraint neither legitimate nor illegitimate.” The rules the mill uses every day — when to sample, how to hand over, where hard hats go — are limits. Limits are ordinary, and an institution cannot do without them. This ought is not against constraints, only against crossings; and a rule within the line still has other tests to pass. The ought issues no certificate.

What the line protects, Rights Theory says in the same place: “the subject as an end is the one thing the line never crosses.” That is the same root on which this series stands. The first ought is that root turned into the construct’s constraint on itself. Not infringing basic rights belongs here; securing basic rights — stopping others from infringing them — is another matter, and belongs to the next essay.

One more distinction. The line governs the construct’s acts, not the weather. If there were a drought that year and the pond dried and the fish died, the mill would not have crossed a line. Rights Theory separates acting on a person’s existing circuit of maintenance — cutting it, gating it, monopolizing it — from a supplier pulling out its own supply or a natural shortage: “the supplier’s withdrawal and the severing of the circuit are not one thing.” The household’s well by the river is one of their existing circuits; what goes into that river acts on it. The question is what the mill did, through whose hands, and on which side of the line it fell.

Not One Left Out, Nothing Discounted

Against whom does the mill not cross? Against everyone inside it, and “inside” is drawn by entanglement. So the ought falls on three kinds of people, none left out. The workers: the mill makes most of its demands on them, and has the most chances to cross. The executors: the technician and the supervisor are the mill’s hands, but they are inside it too, and its demands on them cannot cross their lines either; “executor” is no identity that can be drawn out of the interior. And the household downstream.

The household holds no position at the mill, and the mill does not arrange their lives; the relational thickness between them is very thin. But when the mill releases wastewater into the river and their well is on the bank, that is not “arranging their lives”; it is landing on them. Harm does not thicken a relation, and a thin relation does not lighten the harm.

Why does this ought hold in full, undiscounted, not varying with relational thickness? Not because it costs nothing. For a mill already running, stopping one kind of crossing may mean reconfiguring, adding equipment, re-rostering, even shutting a line — and the stopping must itself mind the people who now live by the old arrangement, since dismantling is also the mill’s act. It is undiscounted because it governs the mill’s own acts, and keeps what the line keeps. Whether the mill can manage it, and how to pass through and assign responsibility when it cannot, is for the theories that carry judgment; this series grants that such times come and does not withdraw or discount the ought for them. “We don’t have the resources,” “they’re not our employees,” “our relation with them is thin” — none of these says the line may be crossed, and each is a different thing. Resources bear on how to carry the ought out and how to pass through; not being employees does not shrink the protection owed to people already inside; a thin relation adjusts only what the mill owes in conditions and in room left.

There is a reason to say “hold to” out loud. Institutions have a lean of their own. How Is the Inter-Ontological Possible: “Institutions naturally favor colonization because external regulation is observable, quantifiable, and standardizable.” (Its “colonization” is wider than this series’ narrow sense; only the lean is borrowed.) The handiest way for a mill to keep its releases in hand is a number, an appraisal line, a ranking. None of these crosses a line by itself. But they are the easiest things to stack one on top of another, and somewhere in the stack the line can go. That autumn the stack produced a configuration in which the only way to stop an exceedance in time was a path that went through breaking procedure.

No One Meant It

Back to that night. The supervisor did not want the fish to die, nor did the technician, and the person who drafted the signature rule never thought of this at all. Yet the wastewater ran for six hours. Power Theory describes how this happens. A person at a key node can act through an apparatus “because the apparatus has pre-encoded many asymmetric severing paths.” The paths are laid in advance; the person who invokes the apparatus only has to follow them, and the effect travels along. “Wait for the re-test” was, by the procedure, an entirely normal thing to say; how that delay made its way down to the pond was laid out by the mill’s rules long before. (Only the transmission is borrowed here; whether several constraints together cross the line is a separate matter.)

Whether the line was crossed is judged without looking at motive, for two reasons: motive does not change where an act lands, and motive cannot be settled anyway; a criterion resting on it would be usable by no one. So Rights Theory says: “The constrainer’s good will does not change the reading, and neither does what the constrained feels,” and “a constraint indignantly rejected can stay within the limit, and an action gladly received can already have crossed the line.”

Put together: a construct’s crossing does not need anyone in it to mean it. This must also be guarded against a misreading. To say the mill crossed a line is not to say there is a bad man in the mill, or that the supervisor secretly meant harm and hid it well. A crossing is a structural reading, not a moral accusation. Precisely because it accuses no one, it cannot be undone by “we all meant well.”

Each Within the Line

Were the three rules each within the line? And together? Rights Theory judges one constraint at a time; judging several together is a step this series takes as its own undertaking, and it keeps three disciplines. Examine one and the same person — not “the workers,” not “downstream,” but the technician that night, or that household. Examine under one and the same actual configuration: with the rules all in force at once, which condition that was there item by item is gone? And state the baseline: for each rule judged a limit, under what situation was it so judged? Each being fine under its own imagined situation does not mean each is still fine in the one situation where they all hold.

Examine the technician under that night’s configuration. The signature rule: if the supervisor can always be reached and has no stake in whether a report goes out, a second person’s confirmation is an ordinary check — a limit. The zero-exceedance line: if it is not tied to reporting, merely a target for keeping releases down — a limit. Breaches in the appraisal: if the procedure itself can be followed — a limit. Put the three into that night, on him alone. If the signer will not sign, “within two hours” and “requires a signature” cannot both be done; whichever he follows breaks the other, and the breach is marked against him; and the signer’s own appraisal hangs on “zero exceedances.” But he still has one way: report without the signature, and take the mark. That is a consequence, not a seal. So on him, that night’s configuration may not have crossed the line.

Change the configuration and the verdict changes. Suppose the rules also said “two breaches of procedure and the contract is terminated,” and he already had one that year; suppose there were no route by which the clash between “report on time” and “requires a signature” could be sent up for review; suppose termination were reviewed by no one, and an appeal went to that very supervisor. Taken alone, each of the four can be a limit. Put into that night, on him: if the signer will not sign, whichever rule he follows breaks another, a breach means dismissal, and there is not one way left to refuse, or to send the matter into review. What makes this a crossing is not “losing the job” — a job can end through a procedure that can be reviewed. It is that the configuration leaves him no structural channel at all for refusing or for passing into a reviewable procedure. Each within the line, together across it: the third condition, crossed on the technician. Rights Theory, on harm through resources, has a sentence of the same shape: “splitting the act in two does not change the reading.” Splitting it into several rules does not change it either.

The household downstream is examined separately, and for something else. What acts on them is not those rules but the six hours of discharge itself. The dead fish can be a clue, and may ground a claim of another kind; whether the discharge reached the point of threatening anyone’s body or life has not been established here. If it did, what crossed is the second condition, crossed by the discharge itself, without waiting for several rules to add up. On the downstream side, the way the rules were fitted together explains two other things: how the wastewater ran for six hours with no one meaning it, and why that account is entered under the mill’s name.

Who crossed, when such a crossing happens? The mill. The construct is the address of attribution; the rules are the mill’s, and so is the way they were fitted together. At the same time the one who would not sign, the one who drafted the rules, the one who waited for the signature each have accountability of their own, checked by its own source and facts — neither hidden by the mill’s name nor shifted onto them automatically by its failure.

A word is needed here: relational slots. When the mill sets rules it does not only write down demands one by one; it sets slots — who reports, who signs, whose appraisal hangs on which line, whether there is any other way when the signer will not sign. Setting relational slots is the construct arranging how bearers, demands, authority, answering and exit stand to one another. Once slots are set, this can happen: a shared problem finds no position that can answer it, or two demands, each with a source, block each other. That night both happened. Such a shortfall can be laid to the configuration without laying it to anyone’s malice. How it is registered and who must explain it is the business of the sixth essay.

“I Was Only Following the Rules”

After that autumn there were three versions in the mill. The technician: “I was waiting for the signature, as the procedure said.” The supervisor: “I was re-testing, as the rules said.” Management: “Every rule had its reason.” Three sentences, one thing: I was only following the rules. It fails on both sides.

On one side, it cannot make the one who carried the rules out disappear into the construct. A procedure being on the books proves only that a demand stands at a post; it does not prove that following it stays within the line. Whether each person bears personal responsibility in the matter is judged by his source, his authority, his registration and what he did. As for whether the technician should have reported without the signature, there are two questions. What his own law required of him, this series does not avow for him. Whether his post carried a duty or an authority to report directly, or a rule for which demand comes first when two collide, is checked against sources and the actual configuration; this essay does not hand down the full verdict on that case. Neither question is cancelled by “there was a procedure.”

On the other side, it cannot make the construct disappear behind the one who carried the rules out. “Every rule had its reason” is management speaking for the mill; it wants to say the trouble lay in execution. But the combination was the mill’s. “The monitoring post reported late” enters the configuration’s account against one man — exactly the transfer that is blocked.

One more thing falls on the construct’s side. The executor is inside too, and the mill’s demands on him cannot cross his line. How Is Institution Possible puts the third condition at the member’s position: “A member can refuse a given requirement without thereby losing all ground of their position. Refusal may have consequences, but it may not be the cancellation of the position itself.” And Rights Theory sets the threshold for “sealing”: not deadlines, procedures, conditions or costs of exit in a relation, but “the subject no longer holding a structural channel for refusing, revoking a standing bond, or passing into a reviewable procedure.”

So what the construct has to answer about “I was only following the rules” is not “why didn’t he refuse?” but “could he refuse?” Being able to refuse does not mean refusing costs nothing: had the technician reported without the signature, he would have taken a mark; that is a consequence, not the cancellation of his position. Nor does it mean no post may ever end. In the changed configuration, refusing meant dismissal and dismissal could not be reviewed: there, the mill had crossed his line whether or not he complied. He could still quit — but that is exit, not refusal. This part the mill carries itself.

Signals, and Silence

Afterward, someone proposed scrapping the “zero exceedances” line; someone else proposed the opposite, posting each shift’s exceedance count in a public ranking. Both were arguing over forms — metrics, rankings, grades — as if the form itself were the crossing. Forms are signals, not criteria. The same “zero exceedances,” in a configuration where reporting is not tied to appraisal, is an ordinary target; in that night’s configuration it was a link in what held reporting down. The same ranking can simply show where the problems are, or can turn every report into a penalty against one’s own crew. The criterion is the line itself. Forms are useful for telling you where to look: when a mill suddenly adds an appraisal line tied to reporting, look at whether that slot still has a position that can answer. Dropping the line does not prove the crossing has stopped; keeping it does not prove a crossing.

There was an older story in the mill. More than ten years ago a section foreman gave the hardest shifts to anyone who raised safety concerns. He retired long ago, and the practice went with him. But to this day no one on that floor raises a concern. Nobody stops them; nobody rosters them that way anymore.

Rights Theory calls this self-closure, a second kind of failure beside violation: under long pressure, the structure takes over the work of keeping things shut, and “the external hand can withdraw, and the seal does not lift.” It guards two misreadings at once. Closure is not consent: “What is called self-maintaining is not voluntary, and it assigns no blame to the subject.” And it does not stop the question of where it came from: “Where attribution can be traced, it traces the external path.”

Two things follow for the first ought. Whether the mill has crossed cannot be judged only by whether anyone is pressing now. The foreman used shift assignment, an authority the mill gave his post; his practice was the mill’s act, invoked through his position. That path can be traced, and it lands on the mill’s address — his own share checked separately. The hand being gone does not make an old crossing not have happened; the silence on that floor is exactly where to trace the path back. And no one complaining proves nothing. What people say and do not say, and under what conditions they keep quiet, are facts that can be checked; but from them alone one cannot read which side of the line the mill is on. Where the entrance is out of reach, where reprisal is expected, where a closure has never been undone, silence is no proof of innocence. A mill that offers “we have received no concerns in years” as proof it has never crossed is using something that cannot prove what it claims.

No one wanted those six hours. Whether a line was crossed has to be read person by person, under that night’s configuration. Where it was crossed, it is entered under the mill’s name; and no one having meant it does not strike it from the books.