Outside the Remit

In the autumn of Year 5, Levee Street High School was still requisitioned. That spring the river had risen and the mayor had declared an emergency and taken the school; the water went down in two weeks and the curfew was lifted in the fifth; by summer the school was still requisitioned, the order renewed month by month, the students taught in the temporary building across town. What time leaves behind needs someone to ask about it. The parent of one of the students raised an objection: the water had long since gone down, so why had the school not been returned, and why was the order renewed again and again? She first called the Mayor’s Office complaint hotline. The answer was: “The emergency order is renewed under the ordinance; it is outside the hotline’s remit.”

She then went to the Office of the State Ombudsman of the Kingdom State. The Office takes complaints from residents against the city and needs no consent from the party complained against; she went, and it took up her complaint. The ombudsman called for the file, required the mayor to explain in writing the present ground for the renewals, and in the spring of Year 6 published a report. A month later the mayor ended the emergency order, and the school was returned. The ombudsman closed the case. Whom it concerns: the parent and the students. Which line: requisitioning the school acts on actions that may be limited, within the line; what is involved is the channel of the fifth ought. Baseline: before the flood.

It ended well enough, but it shows up several questions. The hotline said it was not its business: does that road count as a channel? The ombudsman is within reach but cannot itself annul the city’s decisions: how far does its efficacy go? The order ended after the report: was that the efficacy of questioning, or the mayor’s own decision? Once the case is closed, can the matter never be asked again? The mayor’s emergency power was given by the Council: who can take it back? Action Theory handed one of these questions to this series by name: “who calls power to account, and whether calling to account has a final bearer, is the business of the Theory of Government”; and with it the concern about exit offered in form and sealed in fact by ruinous cost. When the first essay stopped on Levee Street, it too left a question: the question can be asked and has been answered; can those affected actually move anything? This is the last essay of the series, and it takes these up.

At Every Level

Institution Theory’s fifth ought sets three conditions for a channel: “it is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation.” By these three, look at the two places the parent went. The hotline sits in the Mayor’s Office, and what she asked about was the mayor’s own renewals; even had it taken the complaint, its independence would have to be checked against the mayor, since independence is such that “what it guards against is the party under review sealing off review of itself on its own.” Here it did not even take the complaint, nor pass it on, nor say where to go. A hotline need not handle everything; the question is not whether it handles this, but whether, on the question of whether the renewals had a present ground, the parent had a road within reach and not in the mayor’s hands.

The state ombudsman is another matter. It is not a city body, and takes complaints without the mayor’s consent, so independence passes at the layer of intake; whether the mayor can sway how far it looks and when it closes is checked separately. She went and it took the complaint, so reach passes too. As for retaliation, nothing in this case shows her being made to suffer for it; but the absence of such facts is not the same as having checked. Nor is independence certified once and for all: were the party questioned a part of the state itself, say the state law that took Riverbank Park, the same Office’s independence would have to be checked against the state: “when the party questioned changes level, independence must be checked again.”

This holds at every level. The form of the separation of powers varies with conditions; being questioned does not, and the fifth ought “does not vary with relational thickness”: however thin a relation, the channel must be there. So every government, of whatever size, must be open to questioning. Check Orchard Court by the three: a household in arrears makes its representations to the committee, which is the party that took the measure, so they are not independent; the City Court is independent of the committee, but a first-instance case takes a year, too late for a fine that grows monthly. Even if the covenant were changed so that fines could be enforced only through the City Court, and the separation of powers could weaken there, a household objecting to the committee’s measures would still need somewhere to ask that the committee does not keep alone, and asking must not cost it its parking space.

Two more words must be kept apart. The parent asking, the ombudsman looking into it, the mayor explaining: that is the channel in use, being open to questioning, there for every government at all times; it is no evidence of illegitimacy, and being questioned does not itself show that the mayor was wrong. Must be questioned says something else: a verified lapse owes an explanation and a correction; if the renewals after the water went down are found to have had no present ground, that place is not merely open to questioning but must be questioned. Whether to ask is the parent’s own affair: “Whether to ask is the internal subjects’ own affair.” The series does not say residents ought to ask; it says that when she wants to ask, the government must let her reach somewhere, and must answer. But the ombudsman, independent and within reach, cannot annul the city’s decisions; the question can be asked and the answer has come, and what efficacy this channel has on the order itself has not yet been said.

Can Anything Move?

The first essay stopped on Levee Street: the objection reached someone, the answer addressed the point, and the decision still could not be moved; it went as far as what has to be asked. This essay answers. Control, divided by what it acts on, has four candidates: on decisions and rules, reversal, rectification and reopening the agenda; on positions, both the grant and the holding of office, revocation of authorization; on the relation itself, exit; on the forward conduction of acts, interruption. Of each the same three things are asked: who can activate it; what actual efficacy it has on reasons, actions, rules or the holding of office; and how the party questioned is unable to cancel it on its own. Where these cannot be stated, the item there is only a name.

Back to the stall on Levee Street. The party questioned is the Permits Office: the decision was made in its name, and the stall rules were set by it under the Council’s ordinance; the permit examiner and the review officer are both its positions. Activation: the applicant files a written objection within thirty days; it does not pass through the examiner or the section chief but is taken up by the review officer in the Regulations Unit; activation is in her hands and needs no consent from the Permits Office. Efficacy: it stops at reasons. The review officer replied in writing within the time limit, answered the very matter she asked, and verified the note the examiner had missed; but the review officer can only recommend reconsideration, and whether to reconsider is decided by the Permits Office, which decided not to. Not being cancelled must be looked at layer by layer. At the layer of reasons, the Office cannot cancel it: the reply is signed by the review officer and sent to the applicant directly, and the Office cannot alter it; the review officer sits inside the Office, but independence “does not require being outside the institution,” and here the Office cannot seal it off. At the layer of actions, the Office can cancel it alone, because whether reconsideration starts is in its hands. In refusing, it gave its reason in writing: “The note says only that it may be recognized at discretion, not that it ought to be; the original decision stands.” Whether this answers the point she raised, the missed note, can be asked: the note leaves the matter to the discretion of whoever decides, and since the examiner never saw it, no discretion was exercised at all. But in this configuration there is no position outside the Office to ask about this reason; at the layer of actions, the Office both decides whether efficacy comes into play and checks the reason it has itself given. That is what “could not be moved” means.

Suppose reconsideration could instead be activated by a position outside the Permits Office, say the City Court’s administrative division, or the state ombudsman by referral, and the Office could not refuse. Activation stays with the applicant; whether reconsideration starts falls to a position outside the Office. Efficacy: once activated, the decision must be determined afresh, an efficacy on action. Not being cancelled: it is not enough that the Office cannot refuse to start; if the reasons it gives on redoing the decision fail to answer the missed note, that position must be able to send it back, or the Office could still empty this layer of efficacy with a boilerplate line upholding the decision. Whether that position starts reconsideration must itself be explained, and lie under a channel it cannot shut. Changed so, the objection on Levee Street would go from answered to movable. Movable does not guarantee she wins: after reconsideration she may still be treated as a new applicant, so long as the reasons answer that point and survive that position’s check. The layer of rules this supposition does not touch; the rules are written by the Permits Office itself, so on a road that sends the gap in the rules back to the Office, the party questioned at the layer of rules is the Office again. That is what reopening the agenda must answer.

Control Is Not a Veto

Two things control is not. It is not a veto: if the residents opposite demand that every stall on the block be removed and the Permits Office does not comply, that does not show they have lost control. Nor is it change every time: if a procedure has the efficacy of reconsideration but in five years no new material happens to warrant reversal, no reversal does not by itself diagnose failed control. Conversely, a change does not by itself prove control. The order ended a month after the report; whether the ending came from the ombudsman’s report or from the mayor’s own reading of the water, the case does not say, and the paper infers no cause from sequence. Control is seen in the structure: if what was asked would warrant reversal, is there a road not in the hands of the party questioned that makes the reconsideration actually happen?

How does control connect with the three conditions of the channel? Institution Theory left this question in its comparison with Pettit: “how the control that non-domination requires connects with the three conditions of the channel.” Levee Street gives the shape of the answer. On the side of being questioned, the Permits Office’s objection procedure passes all three: she can reach it, the Office cannot seal off the review officer’s intake and reply, and no one made things hard for her. But the three conditions check precisely the side of being questioned; the side of efficacy, whether reconsideration happens, was never touched, and so fell back into the Office’s hands. Institution Theory says questioning must be independent because “the party under review may not monopolize the decision on whether a challenge exists and whether it is over”; what the Office monopolizes is the same kind of thing, moved one step later: whether the challenge takes effect.

Institution Theory’s fifth ought already speaks of “the channel through which it is questioned and corrected,” and says an institution may not “close off the conditions of questioning and correction”; it is only that it gives the three conditions their concrete form mainly on the side of being questioned. The series reads the three conditions of the correcting half concretely and carries them over to efficacy: independence, whether efficacy comes into play is not monopolized by the party questioned; initiation, the step that activates efficacy is within reach of those affected, or of a position that can activate it without the questioned party’s consent; no retaliation, using the function brings no penalty. The grounds are two: Institution Theory already writes correction into the channel; and its summary says that “the party questioned may not monopolize the power to shut the door”; the door can be shut at the layer of being questioned or at the layer of efficacy, and watching only the first leaves the second free to shut. The fifth essay took the same step for ending an emergency; here is its general form. The step has three limits: it says only that the side of efficacy must at least watch these three, not that with them control is enough; it is a reading within the fifth ought, not a new ought; and which decisions must carry which control, beyond public grants being revocable and an emergency’s confirmation and ending not being monopolized by the one who took the measures, the paper does not prove.

The matter in the opening can be read this way too: the ombudsman, independent of the city and within reach, has efficacy on reasons, able to require an explanation and to publish; on action, it can refer a case to the state courts, another position not in the mayor’s hands. Here efficacy is formed by two positions joined: one within reach but reaching only reasons, one reaching action but distant, expensive and slow for everyday matters. The functions of control may be carried by several positions together; joined, the three things must still each be statable. As the first essay said, going from contestability to efficacy is not a step beyond Pettit; the series’ own step is to join control to the efficacy side of the three conditions. They part at one more point: by the abstract of chapter 5 of On the People’s Terms, he rests the unconditioned part of the people’s influence on the citizens’ disposition to resist; the series’ third question looks only at the construct’s configuration, at whether the party questioned can cancel the function alone; whether residents ask, and whether they resist, is their own affair and does not enter the criterion. Whether the two answer the same question, the paper registers without judging.

After the Case Is Closed

After the report the order ended, the school was returned, and the ombudsman closed the case. Once closed, can the matter be asked again? It can be closed, and should be: Institution Theory says “a particular matter can be answered and closed”; what persists are the conditions of the channel and of the duty to explain, “not the endless re-asking of the same matter.” But closing a case closes this matter, not the position; after it, “under new evidence or new problems the matter can be questioned again.” One case in Apple City falls right here. In Year 2 the riverside vegetable plot was taken for the levee, with compensation at market price, and the man who farmed it objected that market price could not buy land as close to the river and as easy to water; the second essay read it: market price answered how much the loss was worth in money, not where the alternative path lay. After the flood he asked about compensation again. Is asking again re-asking? What was answered then was the money; what he asks is where the alternative path lies; that went unanswered then, so asking now is not re-asking but the original item still standing: “When something raised goes unanswered, the original item remains.” What matters is that whether it can be asked again is not settled by the place that made the decision alone. Decisions can be final; positions cannot.

Who questions the questioner? The ombudsman is a position too, and its closing a case may be wrong; when it is questioned, independence is checked against it. But the question need not be pushed ever upward to a last position no longer questioned: questioning does not need an unquestionable end in order to stand, a claim that is the series’ own. In answering Schmitt, the fifth essay left for here why no final judge is needed, and this is the answer: the construct’s configuration should put every position under some channel it cannot shut itself. Several channels together make a net, not a ladder with one man sitting at the top. Hobbes closes the chain of questioning with an unquestionable sovereign; the Law Series Supplement has said that in the ashes of civil war “an unquestionable sovereign is better than an unbounded showdown,” a judgment rational in his situation. The series does not close the chain; it leaves it open to questioning everywhere within a net, and bearing still lands somewhere.

Where does the bearing fall? There are several accounts in this, each from its own source. The mayor renewed every thirty days, and renewal needs a present ground: that is his position’s entry. The Council could end the order by two-thirds and never did; whether it has an entry, and which, is checked by what it knew then and could do. The hotline refused the complaint; whether it should have passed it on and did not is the entry of the place where the hotline sits. How Apple City fitted these places together is the entry of Apple City as a construct. Each is checked by source, authority, registration and the facts of conduct, and none settles another. So the question Action Theory handed over, whether calling to account has a final bearer, is answered in two readings: if it means a place where all accounts are gathered and the door shut, the series has none, and wants none; if it means that every account lands at a position that can be named, yes, and the construct’s configuration ought to see that it does. Keeping the two readings apart is the series’ own step. Bearing can be located; exemption from questioning cannot be made final. Ending the order and returning the school is itself a handling; if it divides by rules, categories and belonging, say who may stay in the shelter and who returns to class first, it is a new division, and leaves its own unaccommodated residue: if anyone in the shelter still had nowhere to go, where would they go? Questioning does not finish off dividing; this matter closed, the next will come. That is not questioning’s failure; it is why it cannot stop.

Can It Be Taken Back?

The mayor’s emergency power was granted by the Council in the Emergency Ordinance, which kept two hands free: ending an order by two-thirds, and amending or repealing the grant through the procedure for making ordinances. Rights Theory, discussing whether encapsulation can be surrendered, says: “Revocable authorization is exercise: the subject commits certain decisions to another position while retaining revocation and exit — this is one way of disposing of its own exercise under its own law, and authorization lies within the space of limits. The surrender of position that seals revocation and exit is another matter”. The subject of that sentence is a subject disposing of its own exercise; the Council granting the mayor emergency power is not such a thing. From a subject disposing of its own exercise to grants between public positions, subject and object have both changed, and a bridge is owed, which the series supplies on its own account: every public grant must be able to state who grants, what is granted, and who ends or alters it by what procedure.

In Apple City revocation can fall in three places: on a single exercise, the Council ending one order by two-thirds; on the grant itself, the Council amending or repealing it through the procedure for making ordinances; on the holding of office, the mayor being elected and residents able to start a recall under the ordinance. Their efficacies differ: the first on action, the second on rules, the third on the holding of office. Why must a public grant remain revocable? Rights Theory’s sentence gives a structure, but its judgment of crossing the line cannot be carried over directly, since no one’s encapsulation is rewritten here. The reason lies on the side of Institution Theory’s fifth ought: an irrevocable public grant would put the grant itself beyond correction for good; the grantee would hold a power no one could take back, and the door would be shut at the layer of the grant; and an institution may not, on the strength of claiming cultivation, reform or a right aim, “close off the conditions of questioning and correction”; from a construct’s conditions of questioning to whether a grant within it is revocable, subject and object change again, and this too is the series’ own step, resting on the principle that the party questioned may not monopolize the power to shut the door. To grant without keeping revocation is to hand the grantee the key to the door as well. What this reason supports is that the grant itself can be changed, its scope, conditions and existence amended or taken back; not that the public function itself must be abolishable as a whole. Being able to revoke only a single exercise, or only to replace the officeholder, does not by itself satisfy the layer of the grant.

The requirement has its scope. Revocable does not mean every affected person is the direct grantor of every public power: the parent’s child attended that school and the requisition fell on her, but the emergency power was not hers to grant, and she has no power alone to take it back; what she has is questioning. Nor does revocable mean an individual can void a decision at any time; revocation runs through procedure, and the procedure itself must be open to questioning. Suppose the Council’s majority belongs to the mayor’s party, two-thirds can never be found, and the majority, checked, acts only on the mayor’s will: ending needs two-thirds, amending the grant needs that majority, and the function is written on paper while its efficacy lies with a position standing with the one questioned. Then the slot must be filled by others together: within the city, a position that confirms or ends emergencies independently of the one taking the measures, so that when it finds the emergency gone the order stops, a step passing neither through the mayor nor through that majority; in the outer layer, the ombudsman within reach but reaching only reasons, the state courts reaching action but distant; on the voters’ side, recall and the next election, which revoke office rather than end an order. Whether this is enough is checked case by case. The form revocation takes, an order left to lapse, an ordinance amended, a recall, annulment by a court, belongs to specific theories of institutions.

With Rousseau

On how public power once granted can be taken back, Rousseau has a passage close to this. He says the act by which a people puts itself under a prince is not a contract: “It is simply and solely a commission, an employment ... This power it can limit, modify or recover at pleasure,” the “it” being the Sovereign. He says too that “the depositaries of the executive power are not the people’s masters, but its officers; that it can set them up and pull them down when it likes.” To guard against usurpation he wants periodical assemblies, preferably needing no summons, opening with “two propositions that may not be suppressed”: whether the Sovereign wishes to keep the present form of government, and whether the people wishes to leave the administration with those now charged with it. They agree in two places: government is entrusted, and the grant can be taken back; and the occasion of revocation is not in the government’s hands, the assemblies needing no summons and the two questions not to be suppressed, which is exactly the third question: revocation as a function cannot be cancelled by the party questioned alone.

They part in three places. First, who revokes. For Rousseau it is the Sovereign, the people assembled, the general will; the series sets up no “we” to revoke: “Consensus does not automatically generate a third first-person legislator,” and who gives voice to what the people want must itself be questioned; revocation falls on positions and procedures that can be named, the City Council, the recall vote. Second, whether revocation is itself questioned. For Rousseau the Sovereign may as of right take back at will; he grants that such changes are always dangerous, but “the circumspection this involves is a maxim of policy and not a rule of right,” and what he requires is the observance of “all the formalities necessary to distinguish a regular and legitimate act from a seditious tumult.” The series wants a further layer: whatever position revocation falls on, that position and that procedure are themselves open to questioning; the parting lies in the authority of revocation and the manner of review, not in one side having reasons and procedures and the other none. Third, form: Rousseau gave periodical assemblies and two questions; the series fixes no form. The paper drafts his strongest objection: if revocation is not in the people’s hands but falls on a position like two-thirds of the Council, revocation itself can be held by those in power, as in the supposition where two-thirds can never be found. The reply: that the position of revocation can be held, the series admits; the remedy is not another unquestionable revoker, but that revocation as a function itself meet the three conditions and be open to questioning, and that when it is held, positions independent of the party questioned and the outer layer fill it together. That is a difference of route, not a refutation; how the Sovereign’s authority is kept from usurpation by the government is just what Rousseau himself set out to deal with.

Capture

The Economics series has a sentence: “Institutions can be captured.” Its examples: “Judicial independence can be eroded; FOI requests can be stalled by bureaucratic procedure until they are useless; whistleblower hotlines can exist on paper only; protection orders can go unenforced.” The series focuses on two readings of capture: a channel of questioning controlled by the party questioned, which can then seal off review of itself on its own; and the party questioned controlling delivery, so that a rights position is hollowed out, the embedding still there and the delivery gone. Capture can cause hollowing-out; hollowing-out does not certify capture: when staff and procedure fall away, delivery can drop below threshold with no one controlling anything. The two readings are focal, not exhaustive.

Back to Apple City. Complaints about the heating subsidy are taken by a complaints desk inside the Housing Department; the desk is in the Department, and the party complained against is the Department. Is that capture? Not on this ground alone: independence is functional and does not require being outside the institution. By the matter in the opening, residents can go straight to the state ombudsman, about the subsidy as about anything else; the desk is not the only door, and the Department can close its own desk but not the ombudsman’s door; that is not capture. Suppose instead the ordinance said that complaints about the subsidy may be made only through the Department’s desk, that the desk may close cases itself, and that referral to the ombudsman needs the Department’s consent: the only door, self-closing, referral at its nod; the party questioned can seal off review of itself on its own, and that is the first kind of capture. In both cases the desk bears the same sign and sits in the same place; the cases differ in whether the door is the only one, who can close cases, and whether matters can be sent out. Sitting inside the department complained against does not prove capture; being called oversight or review does not prove independence. The assessment post left empty for three years is entered first as hollowing-out; why it was never filled there is no material to say, so it is not read as capture. But suppose complaints could then go only through that controlled door: the hollowed place would be shut behind the door, invisible from outside, and the two together are harder to question than either alone. The judge nominated by the mayor is checked by the second essay’s sentence: nomination does not in itself prove capture; what must be examined is whether the judge still acts solely on that side’s will, and holds independent procedure and applicable authority.

Twelve Months’ Rent

The lease for Apple City’s public housing says: leaving early costs twelve months’ rent. A tenant found work in Willow Ferry and wants to move; she cannot pay. Whom it concerns: this tenant. Which line: the line of not sealing off the channel of exit; in form she can leave, in fact she cannot afford to. Baseline: leaving after a reasonable notice period. The criterion is already there, in Rights Theory: a constraint counts as a limit only if it “does not seal the subject’s structural channel for exiting, refusing, or revoking a standing bond”; “Whatever crosses these is violation.” But sealing must be read precisely: “it is not the presence in a relation of terms, procedures, conditions, or costs of exit; it is the subject no longer holding a structural channel for refusing, revoking a standing bond, or passing into a reviewable procedure.” A penalty for leaving early is not in itself sealing; a cost is not the same as being unable to leave.

The Economics series puts the same question bluntly: “‘possible in principle’ does not mean ‘affordable.’” Exit costs “can be high enough to be practically equivalent to lock-in”; and they break into three components forming one closed loop: “retaliation makes you afraid to leave, evidence difficulty makes you unable to leave effectively, economic lock-in makes you unable to afford leaving.” Put the two together and the question is clear: not whether there is a cost, but whether the cost is so high that she no longer holds a structural channel, to leave, or, if she cannot, a road that brings her case into a reviewable procedure. Twelve months’ rent is, for her, economic lock-in. Exit by the three questions: activation is in the hands of the one leaving; the efficacy acts on the relation itself, the lease ending and she free to take the job in Willow Ferry; as for not being cancelled, what must be checked is whether the party being left can cancel it alone by pricing or approval. Who sets the penalty, whether it can be waived or paid in installments, whether there is a review that can take her case, a job found in Willow Ferry and a penalty she cannot pay, to a place the Housing Department cannot shut alone: the paper does not stipulate these; what it makes clear is what is to be checked. Not why she did not save, but can she leave, and if not, is there a road. Orchard Court’s consent to sell is the same question: consent being required is not in itself sealing, and what matters is what road the owner has left once refused; approval of exit lies with the party being left, which can refuse on a reason with no stated bounds; it resembles capture, except that the door shut is not the door for asking but the door for leaving. How protective arrangements are designed belongs to specific theories of institutions.

On the side of exit, Hirschman said: “The voice option is the only way in which dissatisfied customers or members can react whenever the exit option is unavailable”; “This is very nearly the situation in such basic social organizations as the family, the state, or the church.” Where they part must be written against his question: he asks how, when an organization declines, exit and voice bring it back; this asks how exit is kept from being offered in form and sealed in fact. The first looks at effects, the second at the line: sealing off exit is a violation whether or not sealing it helps the organization recover. The paper drafts his strongest objection: he describes how, when public schools deteriorate, the parents who care most about education move their children first, so that the public schools lose exactly those most willing to fight for them; by this reasoning, lower the cost of leaving public housing and the tenants with the most means leave first, and those who remain have still fewer to speak for them. The reply: the series does not require exit to cost nothing; the threshold of its criterion is sealing, and where the penalty is set is design, where effects may enter the judgment; voice is not kept up by locking people in, for the channel of questioning is there for every government; and sealing off one person’s exit to keep up an organization’s voice trades her line for the organization’s recovery, and the first ought is discounted for no such reason.

The Voting-Position

Apple City’s mayor is elected for four years, with an election when the term ends; residents may start a recall under the ordinance, and with signatures from a set share of voters a recall vote is held. Power Theory, at its close, stops at a position it calls the voting-position: “the institutionalized position at which a subject retains the interrupt-right standing within a group pressure field.” The one position has two readings: relative to the common outcome it is power, “my one vote takes part in changing the common outcome”; relative to group pressure it is a right, “the majority of others cannot fold my position into an uninterrupted forward conduction.” The series takes only its structure. Power Theory itself says: “Whether and how an actual institution approaches this structural position is not within this paper’s evaluation; the anchor is a structural type, not an institutional instance.” And: “The anchor does not serve as an argument that democracy is the natural terminus of power.” Apple City’s elections and recall are a fictional arrangement used to show a structure; they prove no electoral system good, and do not presuppose that actual democracy has been achieved.

Forward conduction, in Power Theory, means a power passing on down the line through others’ silence and operational recognition, meeting no resistance. The series applies the voting-position to the construct’s acts, as a step of its own: an act is made through the Mayor’s Office and passed on by others’ tacit acceptance; at the end of a term or in a recall, voters re-register their positions at this point, and the bearer of the acts can be replaced. Interruption here means only that tacit forward acknowledgment, and the incumbent’s continued occupation of the position, are broken off, not that an act under way stops: an emergency order does not stop because election day has come. By the three questions: activation, periodic elections need no one to activate them, and a recall is activated by voters gathering signatures; not being cancelled, once the signatures suffice the recall vote is held, without the mayor’s consent; efficacy, on the holding of office it is replacement or removal, which is the efficacy of revoking authorization, while re-registering one’s position is something on the voters’ side; on particular decisions it has no direct efficacy, and no vote reverses the requisition of Levee Street High School. So in the slot of interruption, activation and non-cancellation can be stated, but every efficacy that can be stated falls in other slots; its own efficacy would have to land on reasons or rules, and this paper did not find it. By the standard set above, it has a structural position here and remains, as to efficacy, a name, listed with rectification and reopening the agenda.

This reading has two limits. Voting does not issue legitimacy for every decision of the one elected: that the mayor was elected does not mean every order he renewed had the voters’ consent; being elected does not exempt from questioning. And interruption is periodic, questioning always open: elections come every four years, recalls need signatures, and meanwhile the order is renewed again and again; the parent cannot wait for the next vote; what she needs is the channel of the beginning. Having a voting-position does not replace being open to questioning; being open to questioning does not replace the voting-position.

Where the Series Stops

This is the last essay of the series. In the prequel paper the series registered two slots it claimed, neither handed to it by name upstream. One is control within the line, from a research candidate Institution Theory left: “The line is not crossed and the channel is there: do those affected have actual control over the relevant power? Wide discretion within authority is one instance.” The first essay posed the question and this one answers, this far: the functions of control have four candidates, each needing three things stated; reversal was laid out on the Levee Street case and its supposition, with whether the procedure can be withdrawn or brings retaliation marked as not stipulated; revocation of authorization and exit were each walked through on a case; interruption was given a structural position; rectification, reopening the agenda and interruption remain names by the paper’s own standard; the connection between control and the three conditions is written as an extension within the fifth ought, and whether it is enough is not judged; nor is any actual government certified as having achieved it. The other is the slot in the guardian network where government stands as a construct, limited to the branch of the subject-to-be, answered in the second essay under substitute exercise; the branch of those whose gate already self-exercises but who for a time cannot is the series’ own extension, not part of the claim; it claims no ought for the individual guardian, and settles nothing on the critique series’ behalf.

The remaining work is registered in six slots: “What the six slots distinguish is the nature of what was not continued; they are not six blanks that every series is to fill.” First, the remainder this construction itself left: control has no scale, the three questions saying whether a function is there but not whether it is enough; whether another ought is needed beyond the fifth, the series cannot itself judge; for three instances of compulsion, land-taking, compensation and taxing for conditions on the side of striving for, the retracing reads only as far as the maintenance of co-constructive space and is not finished, and their standing is not vouched for; “from the city level up” and “a residential complex that enforces its own covenant reaches the threshold” are normative undertakings, with no general method for checking how outer layers actually stand, which does not change the undertaking’s conclusion; and grouping things under holding to is itself a division, whose remainder has not been counted item by item. Second, objects adjacent and not handled here: distributive justice; whether machines can be participants in an institution. Third, conclusions at their end: rules cannot judge in anyone’s place, so the series will never be a handbook of judgment; a guarantee is not a sufficient condition; exemption from questioning cannot be made final, not because the last man has not yet been found, but because the series sets up no such position and wants none. Fourth, research candidates raised and not delivered: whether there are functions of control beyond the four, and which each kind of decision at each level must at least carry, beyond the two already settled: that public grants be revocable, and that an emergency’s confirmation and termination not be monopolized by the one who took the measures. Fifth, handovers with clear addresses: the Theory of the State, specific theories of institutions and jurisprudence, the Critiques, the critique of judgment; a clear address is not the same as settlement. Sixth, what came to light only looking back: Pettit rests the unconditioned on the citizens’ disposition, while the series puts its third question on the construct’s configuration; whether the two answer the same question, resting on an abstract only, is not judged. Of the six slots only the fifth points forward; what is registered in the others is no promise of future writing.

Place by Place

Back to the main line. The whole series is one sentence: Toward its internal subjects, a government holds to anti-colonization and strives for cultivation; its judgments cross no line, must be explained, and remain open to questioning. What it holds to are the things not discounted for lack of resources and not stopped in an emergency, a grouping wider than Institution Theory’s word and the series’ own undertaking: its acts cross no line, it guards the lines between its internal subjects, and compulsion itself is a limit; the recognition-based rights it has embedded are not de-righted, not left as empty shells, and are withdrawn or amended in public and open to questioning; the channel of questioning is not withdrawn; the channels of exit and refusal are not sealed; in an emergency all this holds. What it strives for is providing conditions for cultivation according to relational thickness, concerned with conditions, not outcomes. Judgments are made by people sitting in positions; answering is stated in three layers, to the construct, to the position, and to the person, to whom no ought is written; every level is open to questioning, and when the level changes, independence is checked again; and once the question has been answered, one must still see whether anything can move.

The direction is the Kingdom of Ends. The series takes it at Institution Theory’s layer: the Kingdom of Ends is a direction, not an ought, nor a state to be signed off; “it is made by persons; an institution cannot make it on their behalf, nor does it count as made because an institution declares it.” So with a government: it cannot make it on people’s behalf, nor declare it made. Here the series undertakes once more, as in the first essay: a government can never be perfect; any claim to perfection is itself a place to be questioned. So the close lies not in completion but in being open to questioning. The state Apple City lies in is called the Kingdom State; that a state is called a kingdom does not mean the kingdom has arrived; a name does not testify for a direction.

The river still runs past Apple City, and downstream still lies Willow Ferry. Levee Street High School was returned long ago; the hotline is still in the Mayor’s Office, the Office of the State Ombudsman still in the Kingdom State. Seven essays written, and Apple City has not become a good person for it; it has no first person, and cannot. What has been written down is only what the constructs along this river must keep: judgments have someone to make them, and once made, someone to answer for them; compulsion has a root, and a limit; embedded rights leave no empty shell; levels do not swallow one another; emergency measures are re-examined when conditions change; and when a question can be asked and has been answered, one must still ask whether there is a road not in the hands of the party questioned. Whether all this has been done must be asked, place by place.