Checks
制衡

Writing Declaration: This paper was co-drafted with Claude (Anthropic). All intellectual decisions, framework design, and final editorial judgments were made by the author.

Introduction

In Year 22, a statistician in Apple City took a sample of the search results on Crossways Market. Among goods of the same kind with similar prices and ratings, Crossways Home appeared on the first page at three times the rate of other sellers. Crossways Home is Crossways Group's own brand. Crossways answered: "Ranking looks at buyer satisfaction; Crossways Home has fewer returns."

Crossways Market grew step by step out of four stalls in the East Market of Apple City's old quarter (Paper 1, §1). By Year 20, about sixty percent of the Kingdom State's online retail passed through it, and about thirty percent of that of the neighboring Highland State; it had about forty thousand sellers, more than ten thousand of them small workshops of five people or fewer, and twenty-six of them porcelain workshops in Bluekiln. Buyers pay Crossways, and Crossways settles with the sellers after deducting its commission.

Of the terms between the sellers and Crossways, three need to be looked at. First, a seller's listed price on Crossways Market may not be higher than its listed price in any other channel. Second, only sellers who ship through Crossways logistics can enter the "Express" section on the first page. Third, in Year 22 the commission was raised from twelve percent to fifteen percent. Crossways said: "Rates are set by the market. A seller who is not satisfied can choose not to sell here."

Of those twenty-six workshops in Bluekiln, one left Crossways in Year 22 and sold only in town and in its small shop on the review platform Echo Wall. A year later its turnover was thirty percent of what it had been.

In Year 23, Crossways set up the Seller Appeals Committee, with seven members: five appointed by management, and two elected by and from among the "Gold Sellers." The "Gold Seller" badge is awarded by Crossways according to sales volume and the amount spent on paid promotion. The committee's decisions are issued without public reasons. In the same year, Crossways opened to each seller its own sales and visitor data. Also in Year 23, the Competition Review Division, newly set up in the Office of Trade of the United City-States of Ends (UCSE), began to investigate the ranking and the seller terms of Crossways Market; the UCSE's amended Fair Competition Act had taken effect in the winter of Year 22, and a breakup falls under the Office of Trade's procedure under that Act. The Fair Competition Act, and the draft Online Marketplaces Act, the Public Utilities Act and the Regulations on Departing Public Officials that appear later, are all examples set up by this series; the provisions this paper refers to serve only as facts, and this series does not claim that they are the best design.

In this matter, no one has been locked up, and no household has had its water cut off. The sellers came of their own accord and can leave; one did leave. But the one that left was left with thirty percent; and for the forty thousand that stayed, on which page each one's goods appear, whether it can enter "Express," and what price it lists elsewhere all lie within one set of Crossways arrangements. "Can choose not to sell here" is not false; it is only that what it answers is not necessarily the matter being asked about.

What this paper asks is this: when an economic position grows so large that many people cannot get around it, how strong must the checks be, and where do they fall? The previous paper wrote of chokepoints at the level of relations: the position others cannot get around has depth, and also breadth (Paper 2, E7). This paper turns to the level of scale and looks at breadth: how many must pass through it, and how wide is the face it converges.

The paper proceeds in eight steps. It first says that checks increase with the power-face, and what standing this sentence has (§1); then what, beyond checks, a construct that does the work of the Base Layer requires (§2); and whether the public and the private each have an exemption of their own (§3). It then writes within the line: when the line has not been crossed and the channels are there, whether those acted on have actual control over the position, and how the positions that check it are themselves questioned (§4); and that checks fall on use and do not aim at lowering capacity (§5). Next it writes that positions are not final: the purchase of a substitute road that has not yet grown (§6); measures, forms and thresholds (§7). The last section writes of the exit of the strong: a construct large enough to leave makes relocation a condition (§8).

The reading follows the Prequel: the construct takes the ought, the position takes accountability, the person takes his own law. Crossways Group, Willow Basket Market, the Competition Review Division and the State Legislature are constructs; directors, committee members, the head of the Division and legislators are in positions; the sellers, the care workers and the elderly residents of the homes each have their own law, and this paper writes no ought for them. The division into the normative layer, the rule layer and the judgment layer also follows the Prequel; this paper states which sentence falls at which layer only where it is used.

§1 Checks Increase with the Power-Face

With Crossways grown this far, how strong must the checks on it be?

Those acted on by this position are the forty thousand sellers who sell on Crossways, more than ten thousand of them with five people or fewer; the workshop that left Crossways; and the buyers as well, since ranking also decides what they see first. What is to be read is how ranking, terms and logistics converge the sellers' roads, and whether the road of walking away is actually open; it is not to decide in advance which of these has already crossed the line. The facts in hand are a few: the rate at which the in-house brand appears on the first page, the three seller terms, the raised commission, and the turnover of the one that left, a year later. A marketplace "with ranking and rates made public and appeals independent" can be used for comparison; but it is only a supposition, not this marketplace's earlier history, nor securing already achieved.

Why say that Crossways' position has a power-face? Rights Theory P4 has given the condition: once a recognition-based right has entered operation, "Only when the position-holder, by that position, initiates a request, calls a procedure, excludes, votes, exits, or sets a path, and the action-space of a recipient or of a field is thereby converged, does the right show a power-face." (Rights Theory P4 F15) Crossways Market is a marketplace that sellers and buyers alike have accepted; this is its recognition-based position. In ranking, it sets the paths by which buyers go to sellers; in using terms to require that sellers not list lower prices elsewhere, it narrows the sellers' room elsewhere; in letting only sellers who use its own logistics into "Express," it ties a prominent place to another service. Each of these is read as a power-face only where it actually converges the action-space of sellers and buyers. What shows a power-face is these exercises, not how large Crossways is as such.

How strong, then, must the checks be? This series' answer is: the larger an economic position's power-face, the stronger the checks on it.

It must first be made clear what this sentence is. It is not a theorem derived from structure, as if each further measure of power brought with it, of itself, a further measure of constraint. Appendix C of How Is Institution Possible (Foundational Paper 6; cited as F6) is a set of principles the author wrote for companies; it places the founder last in the priority sequence and first in the accountability sequence, and states at the same time: "This arrangement is not given by any structural theorem making accountability proportional to power"; placing the founder at the head of the accountability chain "is this framework's own design choice" (F6, Appendix C, Article 18; Appendix C is a working demonstration; used only as an example). This series' sentence is of the same kind: it is this series' normative undertaking (own undertaking). To call it an undertaking is not to make it lighter; precisely because it is not given automatically by structure, it must state its reasons at every point.

The reasons lie in two places. One lies with those acted on: the wider the power-face, the more people it converges, and the harder it is for them to bear, each on their own, the cost of walking away one by one. For more than ten thousand workshops of five people or fewer each to look for another marketplace on its own: that road exists, and its cost falls on every one of them; the one that left being left with thirty percent is one shape this cost takes. The other lies in the channels: the wider the power-face, the more a position's decisions resemble rules; and the more people a rule falls on, the more someone must be able to question it and change it.

It must also be made clear what is being compared. The power-face is not compared by total assets. What Crossways is worth, or how many years of wages the Qiao family's assets amount to: these numbers do not in themselves set how strong the checks are; the gap itself is not the matter to be asked about (the Prequel, §4). What is to be compared is the particular power-face: whom it converges, down to which dimension, how many must pass through it, how great the cost of the road of walking away. Sixty percent of the Kingdom State's online retail passing through it, and forty thousand sellers: this is breadth; the one that left being left with thirty percent: this is the cost of walking away. Of two companies with comparable assets, one holding a marketplace many cannot get around and the other merely one seller among many, the power-faces differ, and so do the checks.

F6's Proposition III gives the measure for this: "A construct's scale cannot by itself fix the necessary number and granularity of its institutional provisions." (F6 §3.2, Proposition III) Scale is one of the conditions, not the whole of them; at the same scale, where exit cost and relational density differ, the arrangements required differ too. From this follows one more sentence to be held to: scale and power-face affect the form and strength of checks; they do not decide from what moment a construct must guard the line and be questioned. Springhollow Water Company has only fourteen employees; its depth for the two villages of Upper Hollow and Lower Hollow is not shallower because it is small (Paper 2, E7); it does not have to grow as large as Crossways before it enters the address of the ought (the Prequel, §6). The level of scale speaks of increase, not of a starting point.

This series sets only a direction, not gradations. How large a power-face counts as large, how strong the checks are to be, and what form they take lie at the rule layer (§7).

Here a comparison can be drawn with one claim. In Limitarianism, Robeyns argues for an upper limit on individual wealth: she has proposed a political limit, about ten million euros, taking the Netherlands as the example, and an ethical limit, about one million euros per person; one of her reasons is that extreme wealth turns into political power and harms democracy (Robeyns 2024; paraphrased from an interview published by Utrecht University; original not consulted; tier C). Since the original has not been consulted, this paper lists only the points available for comparison. There is one point of contact: by the paraphrase, her reasons include the layer of accumulation turning into political power; this series too, in Paper 4, asks how money turns into control over questioning (Paper 4, E22, E23). Once the original has been consulted, two questions are to be decided: her limit falls on the quantity held, while this series' checks fall on use; are the two, for her, substitutes for each other, or can they run in parallel; and how does she treat use within the limit. These are to be written once the original has been consulted; no conclusion is drawn here as to where the two stand together or apart. There is also a layer that this series must face itself, and that is not put down to her: when accumulation grows large enough, the checks on use may themselves be bought. This series does not evade this side; it is precisely the reason Paper 4 sets the second gate and §4 of this paper has those who check themselves questioned; and where checks land on a particular use that seals off a road, that use may also be contracted (§6).

It may be said that the power-face is so broad, and no gradations are given, that "the larger, the stronger" has nowhere to land. The weight of this objection lies in this: a "stronger" said at the normative layer must still, in actual legislation and review, be measured item by item. But measuring belongs to the rule layer. What the normative layer must hold to is the direction and the objects of comparison: what is compared is whom it converges, how deeply and how widely, not assets; the direction is "the larger, the stronger," not "the larger, the more exempt."

That the workshop which left Crossways was left with thirty percent of its turnover does not automatically show that its exit was sealed off, nor that its exit was already really usable. Whether the exit is real is examined item by item under the criteria of Paper 2 (Paper 2, E13).

E14 Checks increase with the power-face (own undertaking: a normative undertaking, not a theorem). The larger an economic position's power-face, the stronger the checks on it. The power-face is compared by its particular effects: whom it converges, down to which dimension, how many must pass through it, how great the cost of walking away; no single scale of total assets is set. "A construct's scale cannot by itself fix the necessary number and granularity of its institutional provisions." (F6 §3.2, Proposition III) This is this series' normative undertaking, and is not written as a structural theorem making accountability proportional to power (F6, Appendix C, Article 18: "This arrangement is not given by any structural theorem making accountability proportional to power"; a working demonstration; used only as an example). Scale and power-face affect the form and strength of checks; they do not decide from what moment a construct must guard the line and be questioned. The gradations and the forms lie at the rule layer.

§2 By Layer: Checks and Securing

The previous section asked how strong the checks must be. This section asks something else: are there places where checks alone are not enough?

In the two villages of Springhollow, Upper Hollow with about four hundred households and Lower Hollow with about six hundred, tap water comes from Springhollow Water Company alone. In the spring of Year 23 it raised the price of water by sixty percent; in the autumn, about one hundred and twenty households in the two villages were three months or more in arrears; in the winter, it cut off water to twenty-three households, two of them with bedridden elderly people at home. Paper 2 has already read this matter: Springhollow Water is the sole position of supply at the life-substrate; it holds the accountability of supply, and the institution or government that manages it holds the accountability of creating conditions for this supply (Paper 2, E11a).

Southwindow Care Homes is another place. It has twelve homes and eighteen hundred beds in the Kingdom State, and had been run by one family for thirty years. In the autumn of Year 21, Elmseed Capital's buyout fund acquired it. In the two years after the acquisition, night-shift care workers per thirty beds were cut from three to two; the minimum set by the State Health Agency is two. Nine homes had their rooms renovated, two new dementia care units were set up, and monthly fees were raised by twenty-five percent. The State Health Agency's inspection in Year 23 found falls up thirty percent on the previous year, and two homes were warned because their response to night calls exceeded the time limit. Southwindow's return targets and the arrangements of the acquisition are written in Paper 5 (Paper 5, E27); only the care side of it is used here.

Those acted on in the two matters are the households in the two villages whose water was cut off, and the bedridden elderly; and also the elderly residents of Southwindow's homes, the night-shift care workers and the residents' families. What is to be read is how supply of this kind, care and water, relates to the existing minimum standards and to particular operating configurations; it is not to judge which item at Southwindow has already crossed the line. Southwindow's configuration before and after the acquisition is fact; the two per thirty beds set by the State Health Agency is a rule already made, not this series' normative line. A number of falls is not a rate of falls; what caused the thirty percent increase is not worked out in these matters.

Divide by layer. Where what a construct does falls in the Emergence Layer, as with how a marketplace ranks, how much commission it takes, which brand it promotes, checks are the main thing: they ask how it uses this position, keep the road of walking away open, and do not let it convert or seal off roads. Where it falls in the Base Layer, as with water, or with the bodies and safety of the elderly in care, beyond checks, the institution must also secure this supply.

Why is checking not enough? Checks govern use; they presuppose that the supply is still there. To ask Springhollow Water why it raised the price by sixty percent, and what it explained before cutting off water, is checking; but if it withdraws, or its purification station breaks down, however well the questions are asked, the water will not come of itself. Southwindow schedules its night shifts at the minimum standard; to ask why it cut staff is checking; but when a bedridden elderly person presses the bell at night, what is needed is that someone comes, not that afterwards there is a road for asking. Words such as "one can switch to another" still carry their weight in a marketplace; for a bedridden elderly person they carry almost none. When a supply in the Base Layer is cut off, what is harmed is the Base Layer itself, without waiting for checks to run their course.

So there must be securing. Institution Theory says: "Securing is a set of conditions given public form: making the line explicit and concrete, giving it institutional backing, and having real capacity of enforcement." It also says: "Protection on paper is not securing" (Institution Theory Z17). The Theory of Government says: "What a government gives is backing and enforcement, not the core" (Theory of Government G18). Both places originally speak of the securing of basic rights; here it is extended to the supply undertaken by a construct that does the work of the Base Layer (own undertaking). The word is the one Institution Theory Z17 uses. This series uses the word "securing" only for supply in the Base Layer; giving conditions for people's unfolding in the Emergence Layer is what Paper 6 calls "support," and the two are not used interchangeably (Paper 6, E35).

Securing is not limited to sole suppliers. Springhollow is the sole one, and this gives it a further layer of accountability in Paper 2; but the division by layer in this section looks at which layer what is done falls in, not at how many suppliers there are. Whether several suppliers together suffice to secure the Base Layer is to be examined separately, and cannot be inferred merely from "there is competition"; competition can also drive all of them down to the minimum. These matters do not record how the care homes compete with one another, and this section does not use Southwindow to prove that competition has already failed; what Southwindow shows is only the relation among the supply of care, the minimum standard and particular operating configurations. The state has already set a floor, two per thirty beds; the State Health Agency has inspected, and has warned two homes. This is a part of securing that is already there: an explicit line, and a capacity to enforce. At Springhollow too there is a part already there: the water-abstraction permit the State Water Resources Agency gave Springhollow Water states that this water is for the public water supply of the two villages, and there the institution has embedded this supply; on the side of the water price, it is known that the State Public Utilities Commission has no authority over it, and what other positions of questioning there are, the facts do not record (Paper 2, §6). What is to be asked is whether this floor is enough, and who is to ask. The care workers, the families, the State Health Agency and Southwindow's home directors each stand in their own positions; the answer to this question lies at the rule layer and the judgment layer, and this paper does not give it in their place.

Two more sentences must be held to. First, meeting the minimum standard does not automatically prove that something is within the line; conversely, meeting the minimum standard is not in itself judged to cross the line either. The minimum standard is a line at the rule layer; it tells people where things are certainly not acceptable, not where they are certainly acceptable. Second, the object of securing is the supply, not that construct. Securing a supply of care is not endorsing Southwindow's operation or Elmseed's acquisition; securing Springhollow's water is not guaranteeing that Springhollow Water Company, this particular company, will exist forever, and succession can be arranged separately by the manager (Paper 2, E11a). The Theory of Government says it plainly: "A guarantee is not an endorsement." With the guarantee, the power "is not thereby legitimate: the guarantee is a necessary condition, not a sufficient one" (Theory of Government G22). The original sentence says that a government's guarantee does not thereby make the government's own power legitimate; here it is extended to the institution's securing a supply without endorsing the construct that provides this supply (own undertaking).

There may be this objection: securing is the government's business; a company need only not cross the line, and to speak of layers is to treat a company as a government. What this objection guards against is public functions being pressed quietly onto a private construct; this must be guarded against. But the subject of the sentence about layers is the institution: the supply undertaken by a construct that does the work of the Base Layer is secured by the institution or government that manages it. That construct itself remains under its own oughts as before: it crosses no line, and must explain any cessation of supply; in the sole position of supply at the life-substrate, it holds in addition the accountability of supply (Paper 2, E11a). The division by layer does not turn a company into a government; what it says is that here the institution must do one more thing.

Which matters count as the Base Layer is given by examples at the rule layer, and particular cases lie at the judgment layer (Paper 2, E10); this section uses only two, water and care.

E14a By layer: checks and securing. Where what a construct does falls in the Emergence Layer, checks are the main thing; where it falls in the Base Layer, beyond checks, the institution must also secure this supply. Securing is "making the line explicit and concrete, giving it institutional backing, and having real capacity of enforcement" (Institution Theory Z17); "What a government gives is backing and enforcement" (Theory of Government G18); the original speaks of the securing of basic rights; here it is extended to the supply undertaken by a construct that does the work of the Base Layer (own undertaking). It is not limited to sole suppliers: even where there are several suppliers, whether they suffice to secure the Base Layer is examined separately. The object of securing is the supply, not that construct: a guarantee is not an endorsement (Theory of Government G22; the original says that a government's guarantee does not make its own power legitimate; here it is extended to not endorsing the construct that provides the supply; own undertaking). Meeting a minimum standard neither automatically proves that something is within the line nor of itself judges it to cross the line. "Securing" is used for supply in the Base Layer, and kept apart from "support," which gives conditions for people's emergence. Which matters count as the Base Layer lies at the rule layer.

§3 Public and Private Alike

How much a position can mobilize, and whether it is public or private: are these two separate matters?

Take a public example first. A state's ministry of finance holds large amounts of capital, and what it can mobilize exceeds what many companies can; yet its power is subject to checks at the level of the state: the powers are separated, the budget passes through other positions, and its decisions can be questioned. Private positions directing resources with a comparable power-face often have no comparable structure in reality. What this series sets out to supply is this slot; in supplying it, it gives neither side an exemption.

The shareholder register of Crossways Group is a private example, yet it contains a public share. Crossways is listed on the Apple City stock exchange, and its share capital is in two classes: Class A shares carry one vote each, Class B shares five votes per share. Class B shares were issued only to the founding family; they are held by the Qiao Family Office, and their votes are exercised by the chair of the Family Office. In the spring of Year 21, the Family Office held all the Class B shares, twenty percent of the share capital and about fifty-six percent of all voting power. The Kingdom State Public Employees' Pension Fund held nine percent in Class A shares; under its mandate agreement with its asset manager, Elmseed Capital, the votes on these shares are exercised by Elmseed on its behalf, and the fund's members, state civil servants and public school teachers, have no road for a say in how they are voted; Elmseed publishes its voting record once a year. Elmseed's own funds hold a further four percent in Class A shares, and the votes it exercises together come to about seven percent of all voting power.

Where the position that can mobilize lies, Paper 1 has already read: in the chair of the Family Office, and in Elmseed, which votes for the pension fund, not in the nominal holders (Paper 1, E2). What this section asks is: when checks land on these positions, is there a difference between public and private? The Family Office is private, Elmseed is private, the pension fund is public. This series' answer is: recognized by power-face, not by public or private. The chair of the Family Office uses fifty-six percent of the votes to set Crossways' direction; the power-face of this position is not exempted from checks because it is a private body. Nor is the seven percent Elmseed exercises exempted because it acts only as a trustee. The pension fund's side is public: the mandate by which it handed over its votes, and the trustee positions that made this mandate, are questioned according to their public source, and gain no separate exemption because what they vote in is a private company. These matters do not record whether the fund's members can question at other levels; this section does not stretch "no road for a say in how they are voted" into "cannot question at any level."

A position can also have both sources. The Stone Bridge Franchise Company is an old story: long ago, the City Council of Apple City granted it a fifty-year franchise to collect tolls on a bridge, and the franchise was embedded in an ordinance. Its annual report for Year 21 states that it invested part of its toll income over the years, and holds about one percent of Crossways Group's Class A shares and thirty percent of the shares of Springhollow Water Company.

Two things must be kept apart here. The bridge toll is a granted position, its power coming from the City Council's grant; it is subject to checks according to the source of the grant, as the Theory of Government has already written: "A public grant must retain the function of revocability" (Theory of Government G37). The shares bought with the tolls are an accumulated position; whom they actually converge at Crossways and at Springhollow is read under this series. Each is subject to its own checks. One source does not wash the account of the other: that the tolls have become shares does not wash out the account on the side of the original grant and of the investment decisions; nor do the shares take on, across the board, the revocability of the franchise because the money came from it. Stone Bridge holds thirty percent of Springhollow, and Springhollow is the sole piped-water supplier to the two villages; what part this thirty percent has played in Springhollow's decisions, these matters do not record, and this section does not supply it.

Settlement too comes in public and private forms. The Mirror Sea Clearing Network is run by the Federal Clearing Bureau of the Plateau Federation; set up by a state body, it is a settlement position on the side of grant. Crossways Pay is Crossways Group's settlement subsidiary; by Year 26, the receipts of about forty percent of the Kingdom State's small merchants passed through it, and it is a settlement position that has been accumulated. Both collect and release money for many people, and both can hold up others' money; the former is subject to checks according to the source of its grant, the latter according to its accumulated power-face, by one and the same rule: whomever it converges, there it is questioned. Crossways Pay later ran into trouble, and the state gave it a loan; that matter is written in Paper 4 (Paper 4, E25).

Private is private: this is a common objection; property rights are a boundary of freedom, and treating public and private alike would have private enterprises managed as governments. The ground of this objection is real: accumulation itself is not the problem, as this series set down in Paper 1 (Paper 1, E1); a person doing his own things with his own things need not answer to anyone. But treating public and private alike does not mean treating the private as public. The granted and the accumulated have different sources: a public grant must retain the function of revocability (Theory of Government G37); an accumulated position does not on that account have to be handed back at any moment (own undertaking). What is alike is the reading: where a position actually converges others, there it is subject to checks; the public is not exempted for being public, and the private is not exempted for being private. A private enterprise does not become a government; the place where it converges others, and the place where a ministry of finance converges others, are asked about by the same rule.

How the public side is subject to checks is the business of the Theory of Government and the Theory of the State; this series says only that it gains no separate exemption from this rule.

E15 Public and private alike: recognized by power-face, not by public or private (own undertaking). An accumulated position with a comparable power-face is not exempted from checks for being private; nor does the public side gain a separate exemption on that account. A granted position is subject to checks according to its source, and a public grant must retain the function of revocability (Theory of Government G37); an accumulated position is subject to checks under this series. A position can have both sources, and is subject to checks separately under each: granted income that becomes accumulated holdings does not wash out the account on the side of the grant and of the investment decisions; nor do accumulated holdings thereby take on, across the board, the revocability of a grant.

§4 Control within the Line; Those Who Check Are Also Questioned

Back to Crossways' Seller Appeals Committee. Seven members, five appointed by management, two elected by and from among the Gold Sellers; decisions issued without public reasons. In Year 25, the board under its new chair changed the two elected members, too, to appointment by management (this change of personnel is written in Paper 5; Paper 5, E29, E30).

This section first sets a premise: the line has not been crossed, and the channels of questioning are there. A seller in a dispute with Crossways can, under the contract, go to law in the state court (what the facts record are disputes over accounts); the Competition Review Division is investigating; the State Legislature is considering a bill. This premise is a condition this section uses in order to ask its question, not this section's finding about Crossways: whether the line has been crossed is examined separately, each under its own criteria. What is to be looked at is whether, under this premise, the sellers acted on by Crossways have actual control over its relevant decisions. This question falls on the sellers who bring objections to the Seller Appeals Committee. The facts available are the committee's composition, its manner of appointment, and its giving no public reasons; what it takes up, who brings matters to it, and what it can change, these matters do not record.

The Theory of Government, within a level of government, has given the functions of control, and of each it asks three things: "who can activate it; what actual efficacy it has on reasons, actions, rules, or the holding of office; how the party questioned cannot cancel it on its own." (Theory of Government Paper 6, §2) It also says that the four functions of control it lists are not claimed to be exhaustive; "Control does not mean that every person affected has a veto" (Theory of Government G36a). The original speaks of a government's decisions; here they are changed to economic positions (own undertaking).

Read by these three questions, the Seller Appeals Committee yields little. Who can activate it: what the committee takes up is not recorded. Efficacy: its decisions are issued without public reasons, so the sellers get nothing at the layer of reasons; whether it can change a ranking or a term is not recorded. The third question reads most clearly: the party questioned is Crossways' management, and five of the seven members are appointed by management; after Year 25, all seven are. As for the two elected members, the Gold Sellers who elect them are themselves rated by Crossways according to sales volume and paid promotion.

Here one proposition of Institution Theory is needed: an institution keeps open the channel through which it is questioned and corrected, and the channel meets three conditions: "it is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation. Independence is functional and does not require being outside the institution; what it guards against is the party under review sealing off review of itself on its own; it is checked against the party actually questioned" (Institution Theory Z34). These three conditions were first set by F6's Proposition IV; but the subject of Proposition IV is "An institution premised on non dubito"; and it says of itself: "This proposition establishes conditions only; it establishes no office and no ought." (F6 §3.2, Proposition IV) Whether Crossways Market and the appeals committee it set up are an institution premised on non dubito in that sense, this paper does not judge; the ought of this section does not rest on Proposition IV but on Institution Theory Z34, which states that the step "From conditions to an ought" is its own undertaking; Proposition IV is only the source of these three conditions.

Read under Z34, the appeals committee's being set up inside Crossways is not in itself the problem: independence does not require being outside the institution. What is to be examined is whether Crossways' management can on its own seal off review of itself. Appointing the majority does not mean that every review has already been sealed off; but in this arrangement the majority of the seats that answer the questions is held by the party questioned, and after Year 25 all of them are, so the kind of independence Z34 requires cannot be seen from the structure. This section writes only this far: what decisions the committee has actually made, these matters do not record, and this section does not supply them. Nor is this the sellers' only road: a dispute can be taken to the state court under the contract, and the Competition Review Division is investigating. These two lie outside Crossways; whether they actually reach the sellers, what they can change, and whether Crossways can on its own render them ineffective, the facts do not fully record, and this section does not certify them in their place. On what is recorded, the roads of questioning are not only the one Crossways set up itself; as a control, this one is thin.

There is another half, often left out: those who check are themselves questioned.

Crossways has those who check it. The Competition Review Division investigated, and in the spring of Year 24 issued a report recommending "separating the marketplace from its own retail": Crossways Market and Crossways Home would belong to two companies and be run separately. At the same time the State Legislature was considering the draft Online Marketplaces Act, which covers only marketplaces within the state: marketplaces would have to make their ranking rules public and set up a seller appeals body independent of the marketplace; a further clause would let the state government request the Office of Trade to start breakup proceedings on the basis of the Competition Review Division's report. The breakup itself falls under the Office of Trade's procedure under the Fair Competition Act. These places are, respectively, who investigates, who proposes the breakup, and who, under procedure, initiates or adjudicates. Those who bear them are Crossways, its approximately forty-one thousand employees in the Kingdom State, the roughly thirty percent of sellers in the state who sell only on Crossways, and the buyers as well.

The decisions of these positions are likewise acts of constructs. Institution Theory says that the ought of not crossing the line "governs the construct's own acts, guarding and dismantling included" (Institution Theory Z9); the Theory of Government says: "Positions cannot be final: no position may monopolize the power to shut the door" (Theory of Government G36). The arrangements that check are themselves under a channel they cannot close. So the three questions must be put to them as well: as to the Competition Review Division's report, can the party investigated and those who depend on it raise objections to it, what can the objections change, and can the Division close it on its own; by what road is the State Legislature questioned when it cuts or amends a draft; and when the Office of Trade starts or does not start breakup proceedings, who can require it to explain. In these matters, the routes for reviewing the report and the remedies against a breakup are not recorded. This section does not invent a review body, nor say that one already exists; it says only that these places, by the same rule, ought to have actual points of entry.

In the winter of Year 24, the head of the Competition Review Division who had led that investigation left office, and four months later became a vice president of Crossways Group, in charge of "Compliance and Government Relations." The UCSE's Regulations on Departing Public Officials require that for one year after leaving office one not deal with one's former body on matters one personally oversaw; they place no restriction on employment. In the autumn of the same year, the Economic Committee of the State Legislature struck out, by eight votes to seven, the clause on requesting breakup proceedings. Crossways has issued a statement that it can move away (§8), and also has a record of donations (Paper 4). These are all clues of risk, and can make rules need prevention (how far the departure regulations reach, for example, is a matter for the rule layer); but the facts now in hand do not serve as a final verdict that capture has already occurred. Whether those who check have been taken over by the party they are to check, Paper 4 examines under the second gate, of which political–economic separation is the extreme case at the level of public power (Paper 4, E22, E23).

Here this paper can interface with Anderson's Private Government. She calls the domination of employers over employees "private government": a government under which the governed are unfree, holding arbitrary power over them that is unaccountable to them; she does not accept that the freedom to quit is enough to protect employees from this domination, and proposes constraining it through new employee rights and ways of taking part in decisions, rather than full workplace democracy (Anderson 2017; paraphrased from Estlund's 2018 review; original not consulted; tier C). Since the original has not been consulted, this paper lists only the points available for comparison. There is one point of contact: by the paraphrase, she does not accept that the freedom to quit is enough to protect employees; this section asks whether "can choose not to sell here" answers the matter being asked about, which is an adjacent question. She speaks of employers and employees; this section uses a marketplace and its sellers; the employment side is in Paper 2 (Paper 2, E12). Once the original has been consulted, the questions to decide are these: she calls it "government," while this series keeps the granted and the accumulated apart and does not read an accumulated position as a government (§3); is this distinction, for her, a difference of name or of substance; she gives particular rights and ways of taking part, while this series sets only the functions of control and hands the forms to the rule layer; where do the two correspond. These are to be written once the original has been consulted.

The appeals committee was set up by Crossways voluntarily, and need not have been set up at all; if, once set up, it is still picked at, who will set one up in future? This question is not without reason: a voluntary arrangement gives the sellers more than having none, at least a place to go. But the three questions do not pick at its having been set up; they read what it is: where the three things can be stated, it is a control; where they cannot, that function is there in name only (Theory of Government G36a). Something there in name only does not become a control by being voluntary, nor is it exempt, by being voluntary, from being read in this way.

What the appeals committee takes up and what it has changed, and how the Competition Review Division's report is questioned, are not recorded in these matters; they are left blank, as they stand.

E16 Control within the line extended to economic positions; those who check are also questioned (own undertaking). Where the line has not been crossed and the channels are there, whether those acted on by an economic position have actual control over its relevant decisions is examined by three questions: "who can activate it; what actual efficacy it has on reasons, actions, rules, or the holding of office; how the party questioned cannot cancel it on its own." (Theory of Government Paper 6, §2) Control does not mean that every person affected has a veto (Theory of Government G36a). A complaints arrangement whose majority is appointed by the party questioned is examined for independence against the party actually questioned, under Institution Theory Z34: "independent of the party questioned," "Independence is functional and does not require being outside the institution"; the source of the three conditions is F6 Proposition IV, which stands within an institution premised on non dubito and establishes only conditions; the ought rests on Z34. The decisions of those who check (investigating, proposing a breakup, initiating or adjudicating under procedure) are also questioned by these three questions: checking too is an act of a construct (Institution Theory Z9; Theory of Government G36). Review and remedy that have not been given are not treated as already existing; risk can make rules need prevention, but does not serve as a final verdict that capture has already occurred. The original speaks of a government's decisions; here they are changed to economic positions and the positions that check them (own undertaking).

§5 Checks Fall on How It Is Used

The Competition Review Division recommended separating the marketplace from its own retail. Does this recommendation fall on Crossways' capacities, or on its use of them?

Crossways has many capacities. It has built its own logistics since Year 14, with warehouses on the outskirts of Apple City and in the Highland State; buyers' goods therefore arrive quickly. Its ranking finds things for buyers among forty thousand sellers. Its settlement collects and pays for both sides. These are production, service and the giving of conditions; it is by these that four stalls grew into a place all forty thousand sellers pass through. These same places can also hold people up: only sellers who use Crossways logistics can enter "Express"; in-house brands appear more often in the ranking; sellers may not list lower prices elsewhere.

This series' answer is: checks fall on how it is used, and do not aim at lowering capacity. Checks are set on uses that choke, convert, seal off exit, refuse questioning, or entrench positions; where there is production, building and the provision of conditions, the way is left open.

The difficulty is that the two often sit in one place. Logistics is production, and also the door to "Express"; ranking is a service, and also the place where in-house goods rank first. Where one place is at once production and choking, a check must state three things: which path-closing effect, or which effect insulated from questioning, it addresses; how it preserves as far as possible the capacity to produce, to serve and to give conditions; and, if changing the configuration also affects these capacities, the cost, the alternatives available and the reasons for its choice, stated as they stand. This series does not promise in advance that the two sides can be separated without loss at every point.

Read in this way, Crossways' several terms are several separate matters. In the "Express" term, the road-sealing effect lies in tying a prominent place to Crossways' own logistics, so that sellers using other logistics cannot get in; a check can land on this tie, and the logistics itself stays as it is. In the price term, the road-sealing effect lies in sellers' being unable to list lower prices in other channels, so that other channels also find it hard to grow by price; a check can land on this term. In the ranking, the threefold rate itself cannot be attributed uniquely to favoring in-house goods; fewer returns, as Crossways says, is also a possibility; here a check can first land on making the ranking open to questioning, and the draft's requirement that ranking rules be made public is one way of writing this. All these are examples at the rule layer, and this series does not claim that they are the best means.

Separating the marketplace from its own retail reaches deepest. The effect it is to handle is a marketplace selling within its own marketplace while ranking its own competitors; what it changes is the structure of ownership. Will it also affect capacities? It may: the synergies between the marketplace and its own brand, the scale of logistics, and the transition of staff and contracts may all carry costs. These matters have not reckoned these costs, and this section does not reckon them in their place. By this section's requirement, the position that chooses this means must state which effect it handles, what the cost is, whether there are lighter means, such as the draft's other two clauses, making ranking rules public and setting up an independent appeals body, and why it does not choose the lighter one. Whether to separate is for the rule layer and the judgment layer, and this paper does not judge it.

Two more sentences must be held to. One: calling an act "production" does not exempt it from being examined against the line. Crossways says "Ranking looks at buyer satisfaction," and this may be true; it does not exempt the ranking from being questioned. Conversely, the threefold rate does not thereby prove the opposite either. The other: too many veto points can also leave a position unable to do anything. Control does not mean that every person affected has a veto (§4); if each of forty thousand sellers could block an adjustment of the ranking once, this marketplace could no longer find anything for anyone.

Separating is dismantling, and dismantling is lowering capacity; what is said is "not aimed at lowering capacity," and what is done is the very same thing. The merit of this criticism is that it does not let a fine-sounding phrase cover a real cost; structural change does carry costs, and they fall on employees, sellers and buyers. But it is precisely here that the difference of aim shows itself: the aim is that a marketplace not rank its own competitors, not that Crossways become smaller. If this aim can be reached with less loss of capacity, choosing the heavier means must explain why. What "not aimed at lowering capacity" requires is precisely this explanation: effect, cost, alternative, reason. Where it cannot be given, that check must itself be questioned (§4).

The cost of separating the marketplace from its own retail is not measured in these matters; the judgment whether to separate lies with the position that makes it.

E17 Checks fall on how it is used, not aimed at lowering capacity (own undertaking). Checks are set on uses that choke, convert, seal off exit, refuse questioning, or entrench positions; where there is production, building and the provision of conditions, the way is left open. Where one place is at once production and choking, a check must state which path-closing effect, or which effect insulated from questioning, it addresses, and how it preserves as far as possible the capacity to produce, to serve and to give conditions; if changing the configuration also affects these capacities, it states as it stands the cost, the alternatives available and the reasons for its choice; it does not promise in advance that the two sides can be separated without loss at every point. Calling an act "production" does not exempt it from being examined against the line. Too many veto points can also leave a position unable to do anything.

§6 No Economic Position May Monopolize the Power to Shut the Door

In Year 20, three young people from Willow Ferry started Willow Basket Market, which sold only handicrafts from the towns and villages of the Kingdom State, at a commission of five percent. By Year 22 it had two thousand sellers; nine workshops in Bluekiln sold on both Willow Basket and Crossways.

In the summer of Year 22, Crossways offered to buy Willow Basket, at a price of about twenty times Willow Basket's revenue that year, and the three founders accepted. The acquisition took place before the amended Fair Competition Act took effect, and the filing threshold before the amendment applied; Willow Basket's revenue was below that threshold, and the acquisition went through no review. In the spring of Year 23, Crossways announced that Willow Basket would be "merged into Crossways Market," and Willow Basket stopped running; sellers could "move their shop to Crossways in one click," at Crossways' commission. One founder later told a reporter: "We were not forced to sell; it was just that we could not refuse that price." Another joined Crossways as head of the handicrafts channel.

Those acted on by this acquisition are Willow Basket's two thousand sellers, the nine Bluekiln workshops among them, and also the three founders. What is to be read here is what procedure this acquisition went through, by which a substitute channel actually in use was bought and shut down. The facts give only the two ends: before the acquisition, Willow Basket was running, with two thousand sellers; after it, it stopped. These matters do not record what other substitute marketplaces there were elsewhere in the Kingdom State, and this section does not say that every substitute has disappeared; nor does it use the founders' consent to answer for the situation of the other sellers.

What did Crossways buy? A company, and also a road. The company belonged to the three founders; in selling, they disposed of what was their own; there was no coercion, and the price was high: this was an exchange. A high price is not in itself force: a condition that asks for what is the holder's own is exchange (Paper 2, E11). But in the same transaction there was also something that did not belong to the founders: a road that two thousand sellers could have taken was closed. They could "move their shop to Crossways in one click"; once they moved, the road led only to the one that bought it, and the commission went from five percent to fifteen percent. The founders' consent answers the founders' own affairs; the road of those two thousand sellers is another account.

The Theory of Government has said, within a level of government: "Decisions can be final: a particular matter can be answered and closed. Positions cannot be final: no position may monopolize the power to shut the door" (Theory of Government G36). Here it is extended to economic positions (own undertaking). An acquisition can be completed and closed; but if a position can buy up and shut down substitutes one by one before they grow, what it holds is the power to shut the door, and the possibility of others taking another road is in its hands. Economics P1 says: "the ontological motivation of antitrust is: preventing one party from demoting others to means." (Economics P1 §5.1) Read by this sentence, what is to be looked at is which part of the sellers' room to find another way out this acquisition shut off; what is read is this operation, and no party is given a verdict on its character.

"May not monopolize the power to shut the door" does not require prohibiting acquisitions, nor forbidding Crossways to grow. What it requires is that such an act of shutting the door also lie under a channel Crossways cannot close. This channel has two common means. One is review before the fact: above the threshold an acquisition must be filed, and the Division can question it; the Willow Basket acquisition was not questioned because it was below the threshold, and how the threshold is set is a matter for the rule layer (§7). The other is breakup and reorganization after the fact, which keep a position that has already grown from being final. The Competition Review Division's recommendation to separate the marketplace from its own retail is one of the latter (§5). The means must state their effect, cost and alternative arrangements; where checks land on a particular use that seals off a road, that use may be contracted. What is contracted is that use; the aim is not to reduce Crossways.

This series does not judge that Crossways bought Willow Basket in order to shut it down. The boundary between limit and violation is decided by structure, not by motive (Institution Theory Z9); likewise, in reading an act of shutting the door, what is looked at is what it shut and who can question it, not what the buyer had in mind at the time.

It may be said that this was as voluntary a transaction as there could be: the founders received twenty times revenue, the sellers could move over, and no one was harmed. Half of this is right: the three founders were indeed not coerced, and the road of moving over was indeed open; to read a voluntary acquisition as a violation would be wrong, and this section does not read it so. But "no one was harmed" says too much. The sellers who moved went from a commission of five percent to fifteen percent; any who sold only on Willow Basket would have had either to move or to look for another place; those nine Bluekiln workshops lost a place where they could compare prices with Crossways. These do not make the acquisition a crossing of the line; what they show is where the power-face of a position that can buy up and shut down substitute roads lies, and why checks must land here.

Where Willow Basket's two thousand sellers went afterward, and what each one's turnover became, these matters do not record.

E18 No economic position may monopolize the power to shut the door (own undertaking). "Decisions can be final: a particular matter can be answered and closed. Positions cannot be final: no position may monopolize the power to shut the door" (Theory of Government G36), extended to economic positions (own undertaking). Buying a substitute that has not yet grown can close a road others could have taken; the seller's consent answers the seller's own affairs, and does not answer for those who depend on this road. "the ontological motivation of antitrust is: preventing one party from demoting others to means." (Economics P1 §5.1) Review before the fact, and breakup or reorganization after it, are means by which the rule layer keeps positions from being final; where checks land on a particular use that seals off a road, that use may be contracted, and the means must state their effect, cost and alternative arrangements.

§7 Measures, Forms and Thresholds

The Willow Basket acquisition went through no review because its revenue was below the pre-amendment filing threshold. At Springhollow too there is a threshold: the Kingdom State's Public Utilities Act covers only water suppliers serving five thousand households or more; the two villages of Springhollow have about one thousand households, and the State Public Utilities Commission has no authority over the company's water price. Both thresholds measure by scale: one measures revenue, the other households. In both places there is a position below the threshold: one can buy up and shut down a substitute road, the other is the sole piped-water supplier to two villages.

This section asks: how the power-face is measured, what form checks take, where a formal separation of powers begins, and who decides these.

This series' answer is: all of these lie at the rule layer, and this series gives only examples. Measures can look at the scope of entanglement, at irreplaceability, at irreversibility, at the capacity to convert one kind of accumulation into other effects, and at the length of time in position; forms can be the separation of powers, a duty of explanation that grows heavier with scale, breakup and reorganization, or countervailing among several sources. These are all examples, and this series does not claim that they are the best.

There are two ready examples, to be read together with their own conditions. The set of principles for companies in F6's Appendix C has at its end a retroductive table, whose "Power checks" row reads: "Companies are small in scale; priority and transparency substitute for formal separation of powers" (F6, Appendix C, closing retroductive table; a working demonstration; used only as an example). This sentence stands on the conditions of the "Company" row of the same set of tables: the source of cohesion is "Non dubito + symbiosis of interests," and the degree of exit freedom is "Highly free (voting with one's feet)." When the conditions change, this sentence cannot be transplanted as it stands. For the workers of Bluekiln, a single step away puts their livelihood at stake; the households of Springhollow have no second tap; exit is not high in either place; there, the ground for substituting priority and transparency for a formal separation of powers is gone. The other example is the priority sequence in those principles: employees, users, partners, shareholders, founder; the greater the power, the later the place. Economics P1 says that this sequence is derived from three independent logics, the power one among them being "the greater the power, the lower the priority, to prevent power from being used for self-interest" (Economics P1 §5.4); Appendix C's own wording is: "The more powerful the stakeholder, the later they are served." (F6, Appendix C, Article 2(b)) This is an example of one form: positions with greater power are placed later in benefit.

The Theory of Government, within a level of government, writes the threshold for a formal separation of powers as: "The threshold lies in conditions, not in head count: holding compulsion, where neither exit nor outer-layer routes can carry the day-to-day check" (Theory of Government G26). The original speaks of governments. The first half, "holding compulsion," is not carried over to economic positions; here only the second half is borrowed, as a measuring stick for comparison: the threshold looks at whether exit and outer-layer routes can carry the day-to-day check, not only at scale (extension; own undertaking). Measured with this stick, what do the two thresholds measured by scale lack? They measure revenue and households, not whether exit and outer-layer routes can carry the day-to-day check. How the threshold ought to be set is a matter for the rule layer, and this paper does not set it for the Kingdom State's legislators; it is only that both examples show that a threshold set by scale can leave a deep or a wide position outside the door.

Here the normative layer holds to only one sentence: below the threshold, guarding the line and being questioned are not weaker. Institution Theory says: "An institution keeps open the channel through which it is questioned and corrected, and the channel meets three conditions" (Institution Theory Z34); the original speaks of institutions; here it is extended to economic constructs, on whichever side of whichever threshold they stand (own undertaking). A threshold decides which kind of formal review applies and which body has charge of it; it does not decide from what moment a construct must guard the line and can be questioned (§1; the Prequel, §6). Springhollow Water Company is below the threshold of the Public Utilities Act; it remains under its oughts, and in the sole position of supply at the life-substrate holds an accountability in addition (Paper 2, E11a); the Willow Basket acquisition was below the filing threshold, and the road it shut can still be questioned.

Giving no gradations at all and saying only "at the rule layer," is the normative layer pushing the hard problem onto others? When legislators really have to set a number, the normative layer indeed cannot help with that number. But the normative layer gives two things that are not gradations yet govern the gradations: what is measured is the power-face, not only scale; below the threshold there is no exemption from guarding the line and being questioned. If a threshold is set so as to leave the price of the only piped-water supply of two villages outside all questioning, it must be questioned by these two.

Whether the two thresholds were later changed: of this, these matters record only the amendment of the Fair Competition Act and the time it took effect.

E19 Measures, forms and thresholds lie at the rule layer. How the power-face is measured, what form checks take, and where the threshold for a formal separation of powers lies all belong to the rule layer; this series gives only examples. The closing table of F6, Appendix C, reads "Companies are small in scale; priority and transparency substitute for formal separation of powers"; it stands on the conditions of that table's own "Company" row: source of cohesion "Non dubito + symbiosis of interests," degree of exit freedom "Highly free (voting with one's feet)" (a working demonstration; used only as an example); where exit is not high, it is not transplanted. "the greater the power, the lower the priority, to prevent power from being used for self-interest" (Economics P1 §5.4) and "The more powerful the stakeholder, the later they are served" (F6, Appendix C, Article 2(b)) each serve as examples. The Theory of Government says that the threshold lies "in conditions, not in head count" (Theory of Government G26); the original speaks of governments; its first half, "holding compulsion," is not carried over, and only its second half is borrowed here, as a measuring stick for comparison (extension; own undertaking). Below the threshold, guarding the line and being questioned are not weaker: on Institution Theory Z34, extended to economic constructs (own undertaking).

§8 The Exit of the Strong

In the spring of Year 24, the Competition Review Division's report came out, recommending separating the marketplace from its own retail. At the same time the State Legislature was considering the draft Online Marketplaces Act: marketplaces would have to make their ranking rules public and set up a seller appeals body independent of the marketplace; a further clause would let the state government request the Office of Trade to start breakup proceedings on the basis of the Division's report.

Crossways issued a statement: if the draft passed together with the breakup clause, Crossways would move its headquarters and its settlement center to a district in the south of the Nation Union of Telos (NUT); that district's government had given Crossways ten years free of tax and a tract of land for a business park. The statement said: "Crossways is not the property of the Kingdom State."

Crossways has about forty-one thousand employees in the Kingdom State, logistics included; of the sellers in the state, about thirty percent sell only on Crossways. At a hearing of the Economic Committee of the State Legislature, a small seller said: "If it leaves, we have to follow it, or we have nowhere to sell."

This statement of relocation acts on the State Legislature, which is in the middle of its deliberations, and also on the thirty percent of sellers who sell only on Crossways and on the forty-one thousand employees. What is in hand is the statement itself and a few clues of structural dependence: the number of employees, the thirty percent who sell only on Crossways, and that seller's words. "How the Legislature would have deliberated without this statement" can serve only for comparison, and cannot be used to fill in its possible outcome; moving the headquarters away is not the same as moving all the business and all the employees away.

Crossways says it is not the property of the Kingdom State, and this is true. It can leave; this is true too. What is to be asked is: what is its "can leave" here?

Rights Theory P4 has said how exit turns into power: "Only when structural dependence is present does Exit Rights evolve into power through F15." (Rights Theory P4 §V) If a relation has no structural dependence on a subject's continued participation, his exit is still only the exercise of a right; if it has, the exit shows a power-face. The premise of F18 is a registered or recognized position of exit: "Without such a registered or recognized relational position there may be flight, refusal, or slipping free, but no exit right in the strict sense" (Rights Theory P4 F18). The original speaks of a subject exiting a relation; a construct's tying its relocation to a piece of legislation is extended to here (own undertaking). F18 also states that the withdrawal shows a power-face "not because the subject intends collapse, but because the institution's structural dependence passively amplifies the withdrawal" (Rights Theory P4 F18); here, accordingly, it is the structure that is read, not Crossways' intentions.

Read in this way, the facts point to these places. The Kingdom State has a structural dependence on Crossways: forty-one thousand jobs, and the thirty percent of sellers who sell only on it. If Crossways leaves, these people's situation changes with it; that seller's "we have to follow it" says exactly this. So Crossways' "can leave" shows a power-face here: what it converges is the room the State Legislature has to choose in its deliberations, and also the future road of the thirty percent of sellers. What can leave is Crossways; what cannot are those who depend on it. This asymmetry is precisely the reverse of the side Paper 2 wrote about under "exit is real": of the exit of the weak, one asks whether it is real (Paper 2, E13); of the exit of the strong, one asks whom it converges.

Reading a power-face is not saying that this statement has crossed the line, nor saying that Crossways may not leave. Where a construct chooses to place its headquarters is its own decision; issuing the statement publicly is also a kind of explanation. Reading a power-face says this: where this use converges others here, it falls under the proposition of §1; the checks increase with it, and it must also be open to questioning. How the State Legislature deliberates under such a statement, and its striking out of the breakup clause in the autumn of Year 24, are questioned through its own channels; whether that cut was connected with the statement, or with the donations, Paper 4 examines under the second gate, of which political–economic separation is the extreme case at the level of public power (Paper 4, E23); this section does not judge it.

The conditions that district gave for relocating there, ten years free of tax and a tract for a business park, are that state's act, read under its own laws and through its own channels. Cooperation that arises among states in dealing with such a cross-border economic position follows the functional boundaries the Theory of Inter-State Cooperation has already set. Of "institutions between institutions" between companies, and between states and companies, the Theory of Inter-State Cooperation states explicitly that it does not take them up (Theory of Inter-State Cooperation Paper 8, XZ56); this series does not unfold them in this round either, and registers them as a remainder. This section reads only how the threat of relocation acts, and how jurisdiction and service of process change.

What follows is a supposed case. Suppose that in Year 25 Crossways really did move its headquarters, and its business in the Kingdom State came to be run by a subsidiary of that NUT company; a seller who sues must go through cross-border service of process. The road of questioning has grown longer: the suit must go through cross-border service; where and how a judgment would be enforced, the supposed case does not say, and it is left here only as a question to be asked. A change in procedure does not by itself alone prove that suing has become ineffective, nor that it still reaches as before. What is to be examined is still the same thing: whether the position questioned is still under a channel it cannot itself close. Cross-border service and enforcement rely on arrangements among states such as judicial assistance, which this paper does not set up.

The most direct objection to this section is this: a company's being able to vote with its feet is the best check on a state; to read its exit as a power-face is to lock it inside the Kingdom State. It hits on one layer: the mobility of constructs does indeed also constrain governments. If a state makes a bad law, businesses walking away is one of the costs it pays, and this layer should not be erased. But reading a power-face locks no one in. It does not deny Crossways' right to leave; it only reads on whom this use falls here. Public and private alike applies on both sides here: the State Legislature's acts are questioned, and so is Crossways' act of tying its relocation to a piece of legislation (§3). What is subject to checks is the use, not that one act of leaving.

Of the part that a single state cannot hold in check and that has crossed borders, this series reads only as far as the effect of the threat and the change in the road of questioning.

E20 The exit of the strong shows a power-face under structural dependence (own undertaking). "Only when structural dependence is present does Exit Rights evolve into power through F15." (Rights Theory P4 §V) The premise of F18 is a registered or recognized exit position; a construct's tying its relocation to a piece of legislation is extended to here, and marked own undertaking; under F18, what is read is structural dependence, not intent. What can leave is the construct; what cannot are those who depend on it. Reading a power-face is not judging that it crosses the line, nor does it forbid it to leave; this use falls under the proposition of §1, and is open to questioning. The conditions one state gives for relocation into it are that state's act; cooperation among states arising from this follows the functional boundaries the Theory of Inter-State Cooperation already has. The institutional relations between company and company, and between states and companies, this series does not unfold (Theory of Inter-State Cooperation Paper 8, XZ56, states explicitly that it does not take them up); they are registered as a remainder.

Where This Paper Stops

This paper has proved these things. The larger an economic position's power-face, the stronger the checks on it; this is this series' normative undertaking, not a structural theorem, and the power-face is compared by its particular effects, not by total assets (E14). Where what a construct does falls in the Emergence Layer, checks are the main thing; where it falls in the Base Layer, beyond checks, the institution must also secure that supply; this is not limited to sole suppliers, and securing is not an endorsement (E14a). An accumulated position with a comparable power-face is not exempted for being private, nor does the public side gain a separate exemption; a position with both sources is subject to checks under each (E15). Control within the line is examined by the three questions, and independence is examined under Z34 against the party actually questioned; the decisions of those who check are also questioned by the three questions (E16). Checks fall on how it is used, and do not aim at lowering capacity; where one place is at once production and choking, effect, cost, alternatives and reasons are stated as they stand (E17). No economic position may monopolize the power to shut the door; the seller's consent does not answer for those who depend on that road (E18). Measures, forms and thresholds lie at the rule layer; below the threshold, guarding the line and being questioned are not weaker (E19). The exit of the strong shows a power-face under structural dependence (E20).

What it has not proved: the gradations of scale and of the power-face; the particular forms of checks; the details of antitrust law and of review thresholds; the separation of powers within a company, which is handed to a theory of the firm; the structure of cross-border relations among companies.

Handed to the later papers: where gates are to be set when accumulation is converted into other effects is in Paper 4, and the public loan Crossways Pay obtained (Paper 4, E25) and the donations and the employment after leaving office (Paper 4, E23) are all there; the third generation and the change of appointment to the appeals committee, and the self-chiseling of control arrangements, are in Paper 5 (Paper 5, E29, E30); Southwindow's return targets are in Paper 5 (Paper 5, E27); the two layers, the Base Layer to be secured and the Emergence Layer to be supported, are in Paper 6 (Paper 6, E35).

The remainders left: for the workshop that left Crossways, whether its exit is real is examined item by item under Paper 2; what the appeals committee takes up and what it has changed, and how the Competition Review Division's report is questioned, are not recorded; the cost of separating the marketplace from its own retail has not been measured; where Willow Basket's sellers went afterward is not recorded; whether Southwindow's floor is enough, and who is to ask, is left to the rule layer and the judgment layer; and there is the part that has crossed borders, which a single state cannot hold in check. All these are recorded as they stand, and each has its place to go.

The Eight Propositions of This Paper

E14 Checks increase with the power-face (own undertaking: a normative undertaking, not a theorem). The larger an economic position's power-face, the stronger the checks on it. The power-face is compared by its particular effects: whom it converges, down to which dimension, how many must pass through it, how great the cost of walking away; no single scale of total assets is set. "A construct's scale cannot by itself fix the necessary number and granularity of its institutional provisions." (F6 §3.2, Proposition III) This is this series' normative undertaking, and is not written as a structural theorem making accountability proportional to power (F6, Appendix C, Article 18: "This arrangement is not given by any structural theorem making accountability proportional to power"; a working demonstration; used only as an example). Scale and power-face affect the form and strength of checks; they do not decide from what moment a construct must guard the line and be questioned. The gradations and the forms lie at the rule layer.

E14a By layer: checks and securing. Where what a construct does falls in the Emergence Layer, checks are the main thing; where it falls in the Base Layer, beyond checks, the institution must also secure this supply. Securing is "making the line explicit and concrete, giving it institutional backing, and having real capacity of enforcement" (Institution Theory Z17); "What a government gives is backing and enforcement" (Theory of Government G18); the original speaks of the securing of basic rights; here it is extended to the supply undertaken by a construct that does the work of the Base Layer (own undertaking). It is not limited to sole suppliers: even where there are several suppliers, whether they suffice to secure the Base Layer is examined separately. The object of securing is the supply, not that construct: a guarantee is not an endorsement (Theory of Government G22; the original says that a government's guarantee does not make its own power legitimate; here it is extended to not endorsing the construct that provides the supply; own undertaking). Meeting a minimum standard neither automatically proves that something is within the line nor of itself judges it to cross the line. "Securing" is used for supply in the Base Layer, and kept apart from "support," which gives conditions for people's emergence. Which matters count as the Base Layer lies at the rule layer.

E15 Public and private alike: recognized by power-face, not by public or private (own undertaking). An accumulated position with a comparable power-face is not exempted from checks for being private; nor does the public side gain a separate exemption on that account. A granted position is subject to checks according to its source, and a public grant must retain the function of revocability (Theory of Government G37); an accumulated position is subject to checks under this series. A position can have both sources, and is subject to checks separately under each: granted income that becomes accumulated holdings does not wash out the account on the side of the grant and of the investment decisions; nor do accumulated holdings thereby take on, across the board, the revocability of a grant.

E16 Control within the line extended to economic positions; those who check are also questioned (own undertaking). Where the line has not been crossed and the channels are there, whether those acted on by an economic position have actual control over its relevant decisions is examined by three questions: "who can activate it; what actual efficacy it has on reasons, actions, rules, or the holding of office; how the party questioned cannot cancel it on its own." (Theory of Government Paper 6, §2) Control does not mean that every person affected has a veto (Theory of Government G36a). A complaints arrangement whose majority is appointed by the party questioned is examined for independence against the party actually questioned, under Institution Theory Z34: "independent of the party questioned," "Independence is functional and does not require being outside the institution"; the source of the three conditions is F6 Proposition IV, which stands within an institution premised on non dubito and establishes only conditions; the ought rests on Z34. The decisions of those who check (investigating, proposing a breakup, initiating or adjudicating under procedure) are also questioned by these three questions: checking too is an act of a construct (Institution Theory Z9; Theory of Government G36). Review and remedy that have not been given are not treated as already existing; risk can make rules need prevention, but does not serve as a final verdict that capture has already occurred. The original speaks of a government's decisions; here they are changed to economic positions and the positions that check them (own undertaking).

E17 Checks fall on how it is used, not aimed at lowering capacity (own undertaking). Checks are set on uses that choke, convert, seal off exit, refuse questioning, or entrench positions; where there is production, building and the provision of conditions, the way is left open. Where one place is at once production and choking, a check must state which path-closing effect, or which effect insulated from questioning, it addresses, and how it preserves as far as possible the capacity to produce, to serve and to give conditions; if changing the configuration also affects these capacities, it states as it stands the cost, the alternatives available and the reasons for its choice; it does not promise in advance that the two sides can be separated without loss at every point. Calling an act "production" does not exempt it from being examined against the line. Too many veto points can also leave a position unable to do anything.

E18 No economic position may monopolize the power to shut the door (own undertaking). "Decisions can be final: a particular matter can be answered and closed. Positions cannot be final: no position may monopolize the power to shut the door" (Theory of Government G36), extended to economic positions (own undertaking). Buying a substitute that has not yet grown can close a road others could have taken; the seller's consent answers the seller's own affairs, and does not answer for those who depend on this road. "the ontological motivation of antitrust is: preventing one party from demoting others to means." (Economics P1 §5.1) Review before the fact, and breakup or reorganization after it, are means by which the rule layer keeps positions from being final; where checks land on a particular use that seals off a road, that use may be contracted, and the means must state their effect, cost and alternative arrangements.

E19 Measures, forms and thresholds lie at the rule layer. How the power-face is measured, what form checks take, and where the threshold for a formal separation of powers lies all belong to the rule layer; this series gives only examples. The closing table of F6, Appendix C, reads "Companies are small in scale; priority and transparency substitute for formal separation of powers"; it stands on the conditions of that table's own "Company" row: source of cohesion "Non dubito + symbiosis of interests," degree of exit freedom "Highly free (voting with one's feet)" (a working demonstration; used only as an example); where exit is not high, it is not transplanted. "the greater the power, the lower the priority, to prevent power from being used for self-interest" (Economics P1 §5.4) and "The more powerful the stakeholder, the later they are served" (F6, Appendix C, Article 2(b)) each serve as examples. The Theory of Government says that the threshold lies "in conditions, not in head count" (Theory of Government G26); the original speaks of governments; its first half, "holding compulsion," is not carried over, and only its second half is borrowed here, as a measuring stick for comparison (extension; own undertaking). Below the threshold, guarding the line and being questioned are not weaker: on Institution Theory Z34, extended to economic constructs (own undertaking).

E20 The exit of the strong shows a power-face under structural dependence (own undertaking). "Only when structural dependence is present does Exit Rights evolve into power through F15." (Rights Theory P4 §V) The premise of F18 is a registered or recognized exit position; a construct's tying its relocation to a piece of legislation is extended to here, and marked own undertaking; under F18, what is read is structural dependence, not intent. What can leave is the construct; what cannot are those who depend on it. Reading a power-face is not judging that it crosses the line, nor does it forbid it to leave; this use falls under the proposition of §1, and is open to questioning. The conditions one state gives for relocation into it are that state's act; cooperation among states arising from this follows the functional boundaries the Theory of Inter-State Cooperation already has. The institutional relations between company and company, and between states and companies, this series does not unfold (Theory of Inter-State Cooperation Paper 8, XZ56, states explicitly that it does not take them up); they are registered as a remainder.

References

All SAE works below are cited by concept DOI, which always resolves to the latest version. The version records used for sentence-by-sentence verification are given in the accompanying audit table.

I · This system

Qin, H. (2026). How Is Institution Possible: From Inter-Ontological Remainder to Co-Constructive Framework (SAE Foundational Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19328662

Qin, H. (2026). The Interface and Transformation between Recognition-Based Rights and Power (SAE Rights Theory, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21093939

Qin, H. (2026). Not Crossing the Line (SAE Institution Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270384

Qin, H. (2026). Guarding the Line (SAE Institution Theory, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270766

Qin, H. (2026). Open to Questioning (SAE Institution Theory, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23271324

Qin, H. (2026). Guarantee (SAE Theory of Government, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413112

Qin, H. (2026). Levels (SAE Theory of Government, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413192

Qin, H. (2026). Questioning and Bearing (SAE Theory of Government, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413511

Qin, H. (2026). The Close (SAE Theory of Inter-State Cooperation, Paper 8) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23815796

Qin, H. (2026). Four Layers of Economic Rationality: A Diagnostic Extension of the Rational Agent from Dimensional Sequence Theory (SAE Economics, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19358010

II · External works (paraphrased from secondary sources only, tier C)

Robeyns, I. (2024). Limitarianism: The Case Against Extreme Wealth. London: Allen Lane. (original not consulted; paraphrased from the interview "What is limitarianism? Professor Ingrid Robeyns explains," published by Utrecht University, https://www.uu.nl/en/in-the-media/what-is-limitarianism-professor-ingrid-robeyns-explains)

Anderson, E. (2017). Private Government: How Employers Rule Our Lives (and Why We Don't Talk about It). Princeton: Princeton University Press. (original not consulted; paraphrased from Estlund 2018)

Estlund, C. (2018). Rethinking Autocracy at Work. Harvard Law Review 131: 795. https://harvardlawreview.org/print/vol-131/rethinking-autocracy-at-work/

Interfaces

Main upstream: Rights Theory P4 F15, F18 and the threshold passage of §V (F18 extended to a construct's relocation; own undertaking); Theory of Government G36, G36a and the three questions of Paper 6, §2 (extended to economic positions; own undertaking); Institution Theory Z34 (extended to economic constructs; own undertaking); F6 §3.2, Proposition III.

Also used: Institution Theory Z9, Z17; Theory of Government G18, G22, G26 (for comparison), G37; F6 §3.2, Proposition IV (the source of the three conditions); F6, Appendix C, Article 2(b), Article 18 and the closing retroductive table (a working demonstration; used only as an example); Economics P1 §5.1, §5.4; Theory of Inter-State Cooperation Paper 8, XZ56.

Interfacing: Robeyns, Limitarianism (tier C, paraphrased from an interview); Anderson, Private Government (tier C, paraphrased from Estlund's review).

This series: the Prequel, §§3, 4 and 6; Paper 1, E1, E2; Paper 2, E7, E10, E11, E11a, E12, E13; Paper 4, E22, E23, E25; Paper 5, E27, E29, E30; Paper 6, E35.

Handed to Paper 4: where gates must be set when accumulation is converted into other effects.

写作声明：本文与 Claude（Anthropic）共同起草，所有思想决策、框架设计和最终编辑判断由作者做出。

引言

第 22 年，苹果城（Apple City）的一位统计学者抽样了通衢市集（Crossways Market）的搜索结果。同类商品价格与评分相近时，"通衢家用"（Crossways Home）出现在首页的比例，是其他卖家的三倍。通衢家用是通衢集团（the Crossways Group）自己的品牌。通衢答："排序看的是买家满意度，通衢家用退货少。"

通衢市集是从苹果城老城区东市的四个摊位一步一步长起来的（第一篇第一节）。到第 20 年，王国州（the Kingdom State）的线上零售约六成经它，邻近的高岭邦约三成；卖家约四万家，其中一万多家是五人以下的小作坊，青窑镇（Bluekiln）的瓷器作坊有二十六家。买家付款给通衢，通衢扣下佣金以后结给卖家。

卖家与通衢之间的条款，有三条要看。一，卖家在通衢市集上的标价，不得高于它在其他任何渠道的标价。二，只有用通衢物流发货的卖家，能进首页的"快送"栏。三，佣金第 22 年由一成二提到一成五。通衢说："费率是市场定的。卖家不满意，可以不在这里卖。"

青窑镇那二十六家作坊里，有一家第 22 年退出了通衢，只在镇上和点评平台回音壁（Echo Wall）上的小店里卖。一年以后，它的营业额剩原来的三成。

第 23 年，通衢设"卖家申诉委员会"，七名委员：五名由管理层任命，两名由"金牌卖家"互选。金牌卖家的标识，由通衢按销量与付费推广的额度评定。委员会的决定不公开理由。同年，通衢向卖家开放各自的销量与访客数据。也是第 23 年，目的联合城邦（the United City-States of Ends，UCSE）商贸署新设的竞争审查处，开始调查通衢市集的排序与卖家条款；UCSE 修订后的《公平竞争法》第 22 年冬生效，拆分属商贸署依这部法的程序。《公平竞争法》与后文的《线上市集法》草案、《公用事业法》、《公职人员离任规定》，都是这个系列设的例子；本篇引到的条文只作事实，本系列不主张它们是最好的设计。

这件事里，没有谁被关起来，也没有哪一户人家断了水。卖家是自己来的，也可以走，有一家就走了。可是走了的那一家剩下三成；留下的四万家，每一家的东西排在第几页，能不能进"快送"，在别处标什么价，都在通衢的一套安排里。"可以不在这里卖"这句话不假，只是它答的未必是被问的那件事。

这一篇问的是：一个经济位置大到许多人都绕不过去时，制衡要多强，落在哪里。上一篇在关系一级写过关口：别人绕不过去的那个位置，有深度，也有广度（第二篇 E7）。这一篇转到规模一级，看广度：多少人要经过它，它收敛的面有多宽。

这一篇分八步。先说制衡随权力面递增，以及这一句是什么身份（第一节）；再说做基础层之事的构，制衡之外还要什么（第二节）；公的与私的有没有各自的豁免（第三节）。然后写线内：线没有越、通道也在时，受作用的人对它有没有实际的控制，制衡它的位置自己又怎样受问（第四节）；制衡落在用法上，不以压低能力为目的（第五节）。接着写位置不终局：收购一条尚未长大的替代之路（第六节）；量法、形态与门槛（第七节）。末节写强者的退出：一个构大到走得了，以迁出为条件（第八节）。

读法照前置篇：构用应当，位置用职责，人用自己的法。通衢集团、柳筐市集、竞争审查处、州议会都是构；董事、委员、处长、议员在位置上；卖家、护理员、院里的老人各有自己的法，本篇不对他们写应当。规范层、规则层、判断层的分法也照前置篇，本篇只在用到的地方说明哪一句落在哪一层。

第一节　制衡随权力面递增

通衢大到这一步，对它的制衡要多强？

受这一位置作用的，是在通衢上卖东西的四万家卖家，其中一万多家五人以下；退出了通衢的那一家作坊；还有买家，排序也决定他们先看到什么。要读的是排序、条款与物流怎样收敛卖家的路，以及走开的那条路实际上通不通，不是先认定哪一条已经越了线。手里的事实有几样：自营品牌在首页出现的比例，三条卖家条款，提高了的佣金，退出者一年以后的营业额。一个"排序与费率公开、申诉独立"的市集，可以拿来对照；但它只是一个设想，不是这个市集的前史，也不是已经实现的保障。

为什么说通衢的位置有权力面？权利论第四篇给过条件：一项承认性权利进入运行以后，「唯当持位者凭该位置发起请求、调用程序、排除、投票、退出或设置路径，而受效者或场域的行动空间因此被收敛时，该权利才显出权力面。」（权利论第四篇 F15）通衢市集是卖家与买家都认下的场，这是它的承认性位置。它排序，是在设置买家走向卖家的路径；它用条款要卖家在别处不得标低价，是在收窄卖家在别处的余地；它只让用自家物流的卖家进"快送"，是把一处显眼的位置与另一项服务拴在一起。每一样，只在实际收敛了卖家与买家的行动空间时，才读作权力面。显出权力面的是这些行使，不是通衢有多大本身。

那么制衡要多强？本系列的回答是：经济位置的权力面越大，对它的制衡越强。

这一句要先说清它是什么。它不是一条从结构推出来的定理，好像权力多一分，约束就自然多一分。《制度如何可能》（下称 F6）的附录三是作者为公司写的一份原则，它把创始人放在受益次序的最后、职责次序的最前，同时写明：「这一安排不由一条权力与职责成正比的结构定理给出」，把创始人放在职责链之首，「是本框架自己的设计选择」（F6 附录三第 18 条；附录三是工作展示，只作样例）。本系列的这一句也是这样：它是本系列的规范承担（自担）。说它是承担，不是把它说轻了；正因为它不由结构自动给出，它要在每一处说出理由。

理由在两处。一处在受作用的人：权力面越宽，收敛的人越多，他们一一走开的代价越难由他们自己承担。一万多家五人以下的作坊，各自去另找一个场，这条路有，代价落在每一家身上；退出的那一家剩下三成，就是这一代价的一个样子。另一处在通道：权力面越宽，一个位置的决定越像规则；规则落在许多人身上，就越要有人能问它、改它。

还要说清比的是什么。权力面不按资产总额比。通衢值多少，乔家的资产相当于多少年的工资，这些数本身不定制衡的强弱；差距本身不是要问的那件事（前置篇第四节）。要比的是具体的权力面：它收敛的是谁，收敛到哪一个维度，多少人要经过它，走开的路代价多大。王国州六成的线上零售经它，四万家卖家，这是广度；一家退出以后剩下三成，这是走开的代价。两家资产相当的公司，一家握着许多人绕不过去的场，一家只是许多卖家中的一家，权力面不一样，制衡也不一样。

F6 的命题 III 给了这一层的分寸：「构的规模不能单独给出其制度条款的必要数量和粒度。」（F6 命题 III）规模是条件之一，不是全部；同一规模，退出成本与关系密度不同，要的安排也不同。由此还有一句要守：规模与权力面影响制衡的形态与强弱，不决定一个构从哪一刻起才须守线、受问。泉坳水业公司（Springhollow Water Company）只有十四个员工，它对上坳、下坳两个村的深度，不因它小而浅（第二篇 E7）；它不必长到通衢那么大，才进入应当的地址（前置篇第六节）。规模一级说的是递增，不是起点。

本系列只立方向，不立刻度。权力面多大算大，制衡强到什么程度，取什么形态，在规则层（第七节）。

这里可以与一种主张对照。罗宾斯在《限富主义》里主张给个人的财富设一个上限：她提出过一个政治上的上限，以荷兰为例约一千万欧元，和一个伦理上的上限，每人约一百万欧元；理由之一，是极端的财富会转成政治的权力，伤及民主（Robeyns 2024；据乌特勒支大学刊出的访谈转述，原书未核，C 档）。原书未核，本篇只列可对照之处。接触面有一处：据转述，她的理由里有积累转成政治权力这一层；本系列在第四篇也问钱怎样转成对追问的控制（第四篇 E22、E23）。原书核到以后要决定的问题有两个：她的上限落在持有的数量上，本系列的制衡落在用法上，这两样在她那里是互相替代，还是可以并行；上限以内的用法，她怎样处理。这些核到以后再写，这里不作同在、分在的结论。还有一层是本系列自己要面对的，不归到她名下：积累大到一定程度，用法上的制衡本身也可能被买走。本系列不回避这一面；它正是第四篇要设第二道闸、本篇第四节要让制衡者自己受问的理由；制衡落到某一项封路的用法上时，那一项也可以收缩（第六节）。

有人会说，权力面这样宽泛，又不给刻度，"越大越强"就落不了地。这一说的分量在于，规范层一句"越强"，到了具体的立法与审查里，还要一项一项地量。可是量归规则层。规范层要守的是方向与比较的对象：比的是收敛了谁、多深、多宽，不比资产；方向是越大越强，不是越大越免。

退出了通衢的那一家作坊，营业额剩三成，这不自动说明它的退出被封死，也不说明它的退出已经真实可用。退出真不真，照第二篇的判据逐项核（第二篇 E13）。

E14　制衡随权力面递增（自担：规范承担，不是定理）。经济位置的权力面越大，对它的制衡越强。权力面按具体的作用比较：它收敛谁，收敛到哪一个维度，多少人要经过它，走开的代价多大；不设资产总额的单一刻度。「构的规模不能单独给出其制度条款的必要数量和粒度。」（F6 命题 III）这一条是本系列的规范承担，不写成"权力与职责成正比"的结构定理（F6 附录三第 18 条：「这一安排不由一条权力与职责成正比的结构定理给出」；工作展示，只作样例）。规模与权力面影响制衡的形态与强弱，不决定一个构从哪一刻起才须守线、受问。刻度与形态在规则层。

第二节　按层：制衡与保障

上一节问制衡要多强。这一节问另一件：有没有一些地方，光有制衡不够。

泉坳的两个村，上坳约四百户，下坳约六百户，自来水只有泉坳水业公司一家。第 23 年春它把水价上调六成；秋天，两村约一百二十户欠费三个月以上；冬天，它对二十三户停了水，其中两户家里有卧床的老人。第二篇已经读过这件事：泉坳水业是生命基底处唯一的供给位置，它持供给的职责，管理它的制度或政府持为这一供给创造条件的职责（第二篇 E11a）。

南窗养老院（Southwindow Care Homes）是另一处。它在王国州有十二家院、一千八百张床，由一个家族经营了三十年。第 21 年秋，榆钱资本（Elmseed Capital）的并购基金收购了它。收购以后两年里，每三十张床的夜班护理员由三人减为二人；州卫生厅定的最低是二人。九家院翻新了房间，新设两处失智护理区，月费上调两成五。州卫生厅第 23 年检查：跌倒事件比上一年多三成，两家院因夜间呼叫响应超时被警告。南窗的回报目标与收购的安排，第五篇写（第五篇 E27）；这里只用其中照护的一面。

两件事里受作用的，是两村被停水的人家与卧床的老人，也是南窗院里的老人、夜班的护理员与家属。要读的，是照护与供水这一类供给，与已有的最低标准、与具体的经营配置之间是什么关系；不是判南窗哪一项已经越线。南窗收购前后的配置，是事实；州卫生厅定的每三十张床二人，是一条已立的规则，不是本系列的规范线。跌倒的件数不是跌倒率；增加了的三成由什么造成，这几件事里没有算出。

照层来分。构所做的事落在涌现层的，比如一个市集怎样排序、收多少佣金、推哪一个品牌，以制衡为主：问它怎样使用这个位置，让走开的路通，不让它兑换、封路。落在基础层的，比如水，比如照护里老人的身体与安全，制衡之外，制度还要保障这一供给。

为什么不够？制衡管的是用法，它预设供给还在。问泉坳水业为什么涨六成，问它停水前说明了什么，是制衡；可它若撤出，或者它的净水站坏了，问得再好，水也不会自己来。南窗的夜班照最低标准排，问它为什么减员，是制衡；可一个卧床的老人夜里按了铃，要的是有人来，不是事后有一条问的路。"可以换一家"这一类话，在市集上还有它的分量；对一个卧床的老人，几乎没有。基础层的一项供给断了，伤的就是基础层，不等制衡走完。

所以要保障。制度论说：「保障是一组公共化的条件：把线显明、具体化，给它制度的背靠，并有真实的执行能力。」又说：「纸面上的保护不算保障」（制度论 Z17）。政府论说：「政府给的是背靠与执行，不是核」（政府论 G18）。两处原说的都是基础权利的保障；这里推到做基础层之事的构所承担的供给（自担）。英文写作 secure、securing。本系列用"保障"这个词只说基础层的供给；为人在涌现层的展开给条件，第六篇叫"支持"，两样不混用（第六篇 E35）。

保障不限于唯一的供给者。泉坳是唯一的，这使它在第二篇另有一层职责；可这一节的按层，看的是所做的事落在哪一层，不看供给者有几家。几家同时供给，是不是就足以保障基础层，要另核，不能由"有竞争"三个字推出来；竞争也可以让几家都排到最低。这几件事里没有记载几家养老院彼此竞争的情形，本节也不拿南窗去证明竞争已经失灵；南窗展示的只是照护的供给、最低标准与具体经营配置之间的关系。州已经定了一个底，每三十张床二人；州卫生厅检查过，警告过两家院。这是保障已有的一部分：一条显明的线，一种执行的能力。泉坳那一处也有已经在的一部分：州水利厅给泉坳水业的取水许可，写明这份水用于两村的公共供水，制度在那里把这一份供给载了进去；水价那一路，已知州公用事业委员会对它没有权限，还有哪些受问的位置，事实里没有记载（第二篇第六节）。要问的是这个底够不够，由谁来问。护理员、家属、州卫生厅、南窗的院长，各在自己的位置上；这一问的答案在规则层与判断层，本篇不替它们下。

还有两句要守。其一，达到最低标准，不自动证明在线内；反过来，达到最低标准本身，也不判为越线。最低标准是规则层的一条线，它告诉人哪里一定不行，不告诉人哪里一定行。其二，保障的对象是供给，不是那个构。保障一项照护，不是替南窗的经营或榆钱的收购背书；保障泉坳的水，不是保证泉坳水业公司这家公司永远存续，接续可以由管理者另行安排（第二篇 E11a）。政府论说得直白：「保证不是背书。」保证了，权力「也不因此正当：保证是必要条件，不是充分条件」（政府论 G22）。原句说的是政府的保证不使政府自己的权力因此正当；这里推到制度保障一项供给，不替提供这项供给的构背书（自担）。

也许会有这样的反驳：保障是政府的事；一个公司只要不越线就够了，按层一说，把公司当成了政府。这一反驳要防的，是把公共的职能悄悄压到一个私的构身上；这一层要守。可是按层这一句的主词是制度：做基础层之事的构所承担的供给，由管理它的制度或政府保障。那个构自己，照旧受它的应当，不越线，停供要说明；在生命基底处唯一的供给位置上，另持供给的职责（第二篇 E11a）。按层不把公司变成政府，它说的是：制度在这里要多做一件事。

哪些事算基础层，在规则层举例，个案在判断层（第二篇 E10）；本节用的只是水与照护两样。

E14a　按层：制衡与保障。构所做的事落在涌现层的，以制衡为主；落在基础层的，制衡之外，制度还要保障这一供给。保障是「把线显明、具体化，给它制度的背靠，并有真实的执行能力」（制度论 Z17）；「政府给的是背靠与执行」（政府论 G18）；原说基础权利的保障，这里推到做基础层之事的构所承担的供给（自担）。不限唯一的供给者：即使有多家供给，是否足以保障基础层，仍须另核。保障的对象是供给，不是那个构：保证不是背书（政府论 G22；原说政府的保证不使它自己的权力正当，这里推到不替提供供给的构背书，自担）。达到最低标准，不自动证明在线内，也不自身判为越线。"保障"用于基础层的供给，与为人的涌现给条件的"支持"分用。哪些事算基础层，在规则层。

第三节　公私同一

一个位置能调动多少，与它是公的还是私的，是两件事吗？

先看一个公的例子。一国的财政部拥有大笔资本，它能调动的，比许多公司都多；可它的权力在国家一级受制衡：多权分立，预算经别的位置，它的决定可以被问。权力面相当的私人调度位置，现实里往往没有相当的结构。本系列要补的是这一格；补的时候，也不给任何一边豁免。

通衢集团的股东名册是一个私的例子，里面却有公的一份。通衢在苹果城的证券交易所上市，股本分两类：A 类每股一票，B 类每股五票。B 类只发给创办人一家，由乔氏家族办公室（the Qiao Family Office）持有，表决由家族办公室的主席行使。第 21 年春，家族办公室持有全部 B 类股，占股本二成，占全部投票权约五成六。王国州公共雇员养老基金（the Kingdom State Public Employees' Pension Fund）持有 A 类股九分；照它与资产管理人榆钱资本的委托协议，这部分股份的表决由榆钱资本代行，养老基金的成员，州公务员与公立学校的教师，对怎样投票没有发言的路；榆钱每年公布一次投票记录。榆钱旗下各基金另持 A 类股四分，它行使的票合计约占全部投票权的百分之七。

能调动的位置在哪里，第一篇已经读过：在家族办公室的主席，在替养老基金投票的榆钱，不在名义上的持有人（第一篇 E2）。这一节要问的是：制衡落到这几个位置上时，公私有没有分别。家族办公室是私的，榆钱是私的，养老基金是公的。本系列的回答是：按权力面认，不按公私认。家族办公室主席用五成六的票定通衢的方向，这一位置的权力面，不因它是一家私人机构而免于制衡。榆钱代行的那百分之七，也不因它只是受托而免。养老基金那一边是公的，它把投票交出去的那份委托，以及作出这份委托的受托位置，照它的公共来源受问，不因它投的是一家私人公司而另得豁免。这几件事里没有记载养老基金的成员在别的层面能不能追问；本节不把"对怎样投票没有发言的路"推成"在一切层面都不能问"。

一个位置还可以两种来源都有。石桥特许公司是旧事：很久以前，苹果城市议会授予它一座桥五十年的过桥收费特许，特许载入条例。第 21 年的年报写，它把历年过桥收费的一部分拿去投资，持有通衢集团 A 类股约百分之一，持有泉坳水业公司三成股份。

这里要分开两样。过桥收费是授予的位置，权力来自市议会的授予；它照授予的来源受制衡，政府论已经写过：「公共授予须保留可撤的功能」（政府论 G37）。拿收费买下的股份，是积累的位置；它在通衢、在泉坳实际收敛了谁，照本系列读。两样各受各的制衡。一种来源不替另一种洗账：收费变成了股份，不洗掉当初授予与投资决定那一边的账；股份也不因钱从特许里来，就一概带上特许的可撤。石桥持有泉坳三成，泉坳是两个村唯一的水；这三成在泉坳的决定里起了什么作用，这几件事里没有记载，本节不补。

结算也有公私两样。镜海结算网是高原联邦的联邦结算局所办，一国机构开办，是授予一侧的结算位置；通衢支付（Crossways Pay）是通衢集团的结算子公司，到第 26 年，王国州约四成小商户的收款经它，是积累来的结算位置。两样都替许多人收款、放款，都可能卡住别人的钱；前一样照授予的来源受制衡，后一样照积累的权力面受制衡，同一条：收敛了谁，就在谁那里受问。通衢支付后来出了事，州给了它一笔贷款，那件事第四篇写（第四篇 E25）。

私的就是私的，这是一个常见的反驳：财产权是自由的一道边界，公私同一，会把私人企业当成政府来管。这一反驳的根据是真的：积累本身不是问题，这是本系列第一篇就立下的（第一篇 E1）；一个人拿自己的东西做自己的事，不必向谁交代。可是公私同一说的不是把私的当成公的。授予与积累来源不同：公共授予须保留可撤的功能（政府论 G37）；积累来的位置不因这一条要随时交还（自担）。同一的是读法：一个位置实际收敛了别人，就在那里受制衡，公的不因公而免，私的不因私而免。私人企业不变成政府；它收敛别人的那一处，与财政部收敛别人的那一处，照同一条问。

公的那一侧怎样受制衡，是政府论与国家论的事；本系列只说，它不因这一条另得豁免。

E15　公私同一：按权力面认，不按公私认（自担）。权力面相当的积累位置，不因私有而豁免制衡；公的一侧也不因此另得豁免。授予的位置照它的来源受制衡，公共授予须保留可撤的功能（政府论 G37）；积累的位置照本系列受制衡。一个位置可以两种来源都有，照各自的来源分别受制衡：授予的收入变成积累的持有，不洗掉授予与投资决定那一边的账；积累的持有也不因此一概带上授予的可撤。

第四节　线内的控制；制衡者也受问

回到通衢的卖家申诉委员会。七名委员，五名由管理层任命，两名由金牌卖家互选，决定不公开理由。第 25 年，新任董事长主持的董事会把那两名互选的委员也改由管理层任命（这一换人第五篇写，第五篇 E29、E30）。

这一节先设一个前提：线没有越，问的通道也在。卖家与通衢有争议，照合同可以到州法院打官司（事实里记的是账上的争议）；竞争审查处在调查；州议会在审一部法。这一前提是本节用来提问的条件，不是本节对通衢的认定：线越没越，照各自的判据另核。要看的是，在这个前提之下，受通衢作用的卖家，对它的相关决定有没有实际的控制。这一问落在把异议交给申诉委员会的卖家身上。已有的事实，是委员会的组成、任命方式与不公开理由；它受理什么、由谁提出、能改什么，这几件事里没有记载。

政府论在一级政府里给过控制的功能，每一项都问三件：「谁能启动；对理由、行动、规则或任职位置有什么实际效力；被问的一方怎样不能独自取消它。」（政府论第六篇第二节）它也说，所列的四项控制功能不称穷尽；「控制不等于每个受影响者都有否决权」（政府论 G36a）。原说政府的决定，这里换作经济位置（自担）。

照这三问读申诉委员会，能读到的有限。谁能启动，委员会受理哪些事，没有记载。效力，决定不公开理由，卖家在理由这一层拿不到东西；它能不能改一项排序、一条条款，没有记载。第三问读得最清楚：被问的是通衢的管理层，而七名委员里五名由管理层任命；第 25 年以后，七名都由它任命。那两名互选的委员，选他们的金牌卖家，本身也由通衢按销量与付费推广评定。

这一处要用制度论的一条：制度保持被追问与修正的通道，且通道满足三项条件，「独立于被追问的一方，内部主体实际可以发动，发动不被报复。独立是功能义，不要求在制度之外，要防的是被审的一方独自封住对自己的审查；它对着实际被问的一方核」（制度论 Z34）。这三项条件，F6 的命题 IV 先立过；但命题 IV 的主语是「一个以不疑为前提的制度」，它自己说「本命题只立条件，不立职务与应当」（F6 命题 IV）。通衢市集与它设的申诉委员会是不是那样一种以不疑为前提的制度，本篇不判；本节的应当不靠命题 IV，据的是制度论 Z34，它写明了"从条件到应当"这一步是它自担的；命题 IV 只是这三项条件的出处。

照 Z34 读，申诉委员会设在通衢里，本身不是问题：独立不要求在制度之外。要核的是通衢管理层能不能独自封住对它自己的审查。任命多数，不等于每一次审查都已封死；可在这一安排里，答问的席位多数握在被问的一方手里，第 25 年以后全部握在它手里，Z34 所要的那种独立，从结构上看不出来。本节写到这里为止：委员会实际作了哪些决定，这几件事里没有，本节不补。卖家也不止这一条路：有争议可以照合同到州法院打官司，竞争审查处也在调查。这两处设在通衢之外；它们对卖家实际够不够得着、能改什么、通衢能不能独自让它们失效，事实里没有记全，本节不替它们认证。就已记的看，问的路不止通衢自己设的这一条；这一条作为控制，很薄。

还有一半，常被漏掉：制衡者自己也受问。

通衢有人制衡。竞争审查处调查，第 24 年春出了报告，建议"场与自营分开"：通衢市集与通衢家用分属两家公司，各自经营。州议会同时在审《线上市集法》草案，只管州境内的市集：市集要公开排序规则，设独立于市集的卖家申诉机构；另有一款，州政府可以请求商贸署依竞争审查处的报告启动拆分程序。拆分本身，属商贸署依《公平竞争法》的程序。这几处，分别是谁在调查，谁在提出拆分，谁依程序启动或裁决。承受它们的，是通衢，是它在王国州的约四万一千名雇员，是州内约三成只在通衢卖的卖家，也是买家。

这些位置的决定，同样是构的施为。制度论说不越线那一条「管构自己的施为，守卫与拆除在内」（制度论 Z9）；政府论说：「位置不能终局：没有一个位置可以独占关门之权」（政府论 G36）。制衡的安排，自己也在一条它关不上的通道之下。所以三问也要问到它们身上：竞争审查处的报告，被调查的一方与依赖它的人能不能就它提出异议，异议能改什么，审查处能不能独自把它关上；州议会删改一部草案，经哪条路受问；商贸署启动或不启动拆分，谁能要它说明。这几件事里，报告的复核路径、拆分的救济怎样，都没有记载。本节不虚构一个复核机构，也不说它已经存在；只说这几处，照同一条，应当有实际的入口。

第 24 年冬，主持那次调查的竞争审查处处长离任，四个月以后出任通衢集团副总裁，主管"合规与政府关系"。UCSE 的《公职人员离任规定》要求离任一年内不得就本人主管过的事务与原机关接触，没有限制受聘。同一年秋，州议会经济委员会以八票对七票删去了请求启动拆分的那一款。通衢发表过可以迁走的声明（第八节），也有捐款的记录（第四篇）。这些都是风险的线索，可以使规则需要预防，比如离任规定管到哪里，是规则层的事；但现有的这几样事实，不充作捕获已经成立的终判。制衡者有没有被它要制衡的一方拿走，第四篇照第二道闸核，政经分离是它在公共权力一级的极端情形（第四篇 E22、E23）。

这里可以与安德森的《私人政府》对接。她把雇主对雇员的支配叫作"私人政府"：一种使受治者不自由的治理，握着对受治者恣意的、不向他们负责的权力；她不同意辞职的自由足以保护雇员免于这种支配，主张以新的雇员权利和参与决定的方式去约束它，而不是全面的工作场所民主（Anderson 2017；据 Estlund 2018 的书评转述，原书未核，C 档）。原书未核，本篇只列可对照之处。接触面有一处：据转述，她不同意辞职的自由足以保护雇员；本节问"可以不在这里卖"答没答到被问的那件事，问的是相邻的一件。她讲的是雇主与雇员，本节用的是市集与卖家；雇佣的那一面在第二篇（第二篇 E12）。原书核到以后要决定的问题：她叫它"政府"，本系列把授予与积累分开，不把积累来的位置读作政府（第三节），这一分别对她是名称之别还是实质之别；她给出具体的权利与参与方式，本系列只立控制的功能、形态交规则层，两者在哪里对得上。这些核到以后再写。

申诉委员会是通衢自愿设的，本来可以不设；设了还要被挑剔，以后谁还设？这一问不是没有道理：一项自愿的安排给了卖家多于没有的东西，至少有一个地方可以去。可是三问不是在挑剔它设了，而是在读它是什么：说得出三件的，是一项控制；说不出的，那一项在那里只是名目（政府论 G36a）。一个名目，不因自愿而成为控制，也不因自愿而免于被这样读。

申诉委员会受理什么、改过什么，竞争审查处的报告怎样被问，这几件事里都没有记载，照实留空。

E16　线内的控制推到经济位置；制衡者也受问（自担）。线没越、通道也在时，受一个经济位置作用的人，对它的相关决定有没有实际的控制，照三问核：「谁能启动；对理由、行动、规则或任职位置有什么实际效力；被问的一方怎样不能独自取消它。」（政府论第六篇第二节）控制不等于每个受影响者都有否决权（政府论 G36a）。由被问的一方任命多数的申诉安排，独立照制度论 Z34 对着实际被问的一方核：「独立于被追问的一方」，「独立是功能义，不要求在制度之外」；三项条件的出处在 F6 命题 IV，它立在以不疑为前提的制度里，只立条件，应当据 Z34。制衡者自己的决定（调查、提出拆分、依程序启动或裁决）也照这三问受问：制衡也是构的施为（制度论 Z9；政府论 G36）。没有给出的复核与救济，不当作已经存在；风险可以使规则需要预防，不充作已经捕获的终判。原说政府的决定，这里换作经济位置与制衡它的位置（自担）。

第五节　制衡落在怎样使用上

竞争审查处建议场与自营分开。这一建议落在通衢的能力上，还是落在它的用法上？

通衢有许多能力。它的物流从第 14 年起自建，在苹果城郊与高岭邦各设仓库；买家的东西因此到得快。它的排序替买家在四万家卖家里找东西。它的结算替两边收款付款。这些是生产，是服务，是给条件；四个摊位长成四万家卖家都经过的地方，靠的就是这些。同样是这几处，也可以卡住人：只有用通衢物流的卖家能进"快送"；排序里自营的品牌出现得多；卖家在别处不得标低价。

本系列的回答是：制衡落在怎样使用上，不以压低能力为目的。制衡设在卡住、兑换、封死退出、拒绝追问、位置固化的用法上；生产、建设、提供条件的地方，放开。

难处在于，这两样常在同一处。物流既是生产，也是"快送"那一道门；排序既是服务，也是自营排前的那一处。同一处既是生产也是卡住时，制衡要说明三件：它处理的是哪一项封路或不可问的作用；它怎样尽量保全生产、服务与给条件的能力；若改变配置也影响这些能力，照实说明代价、可替代的办法与选择的理由。本系列不先承诺两面在每一处都可以无损地剥开。

照这样读通衢的几条，是分开的几件事。"快送"一条，封路的作用在于把一处显眼的位置与通衢自己的物流拴在一起，用别家物流的卖家进不来；制衡可以落在这一拴上，物流本身照旧。价格一条，封路的作用在于卖家不能在别的渠道标得更低，别的渠道也就难以靠价格长起来；制衡可以落在这一条上。排序一处，三倍的比例本身，不能唯一地归因于自营优待，通衢说的退货少也是一种可能；这里的制衡，可以先落在让排序可问上，草案要求公开排序规则，是一种写法。这几样都是规则层的例子，本系列不主张它们是最好的办法。

场与自营分开，碰得最深。它要处理的作用，是一个场在场里卖东西，同时替自己的竞争者排序；它改的是所有的结构。它会不会也影响能力？可能会：市集与自有品牌之间的协同，物流的规模，人员与合同的过渡，都可能有代价。这几件事里没有算过这些代价，本节也不替它算。照本节的要求，选这一手段的位置要说明它处理的是哪一项作用，代价是什么，有没有更轻的办法，比如草案里另外两款，公开排序规则，设独立的申诉机构，以及为什么不选更轻的那一样。分不分，是规则层与判断层的事，本篇不判。

还有两句要守。一句：不靠给一个动作命名为"生产"，就豁免核线。通衢说"排序看的是买家满意度"，这可以是真的；它不使排序免于被问。反过来，三倍的比例也不因此证明了相反的一面。另一句：否决点过多，也会使一个位置什么都做不了。控制不等于每个受影响者都有否决权（第四节）；四万家卖家若每一家都能挡住一次排序的调整，这个场就不再能替任何人找东西。

分开就是拆，拆就是压低能力；嘴上说"不以压低能力为目的"，手上做的是同一件事。这一批评的好处在于，它不让人用一句好听的话遮住实在的代价；结构的改变确有代价，落在雇员、卖家与买家身上。可是目的的分别正在这里显出来：目的是不让一个场替自己的竞争者排序，不是让通衢变小。若这个目的可以用更少的能力损失达到，选重的那一样就要说明为什么。"不以压低能力为目的"要求的，就是这一份说明：作用、代价、替代、理由。说不出的，那一项制衡自己就要被问（第四节）。

场与自营分开的代价，这几件事里没有量；分与不分的判断，在作出它的位置。

E17　制衡落在怎样使用上，不以压低能力为目的（自担）。制衡设在卡住、兑换、封死退出、拒绝追问、位置固化的用法上；生产、建设、提供条件的地方放开。同一处既是生产也是卡住时，制衡须说明它处理的是哪一项封路或不可问的作用，怎样尽量保全生产、服务与给条件的能力；若改变配置也影响这些能力，照实说明代价、可替代的办法与选择的理由；不先承诺两面在每一处都可以无损地剥离。不靠给一个动作命名为"生产"，就豁免核线。否决点过多，也会使一个位置什么都做不了。

第六节　没有一个经济位置可以独占关门之权

第 20 年，柳渡镇的三个年轻人办了柳筐市集（Willow Basket Market），专卖王国州乡镇的手工品，佣金五分。到第 22 年，它有卖家两千家；青窑镇有九家作坊，同时在柳筐和通衢卖。

第 22 年夏，通衢出价收购柳筐，价约是柳筐当年营收的二十倍，三位创办人接受了。收购发生在《公平竞争法》修订生效以前，适用修订前的申报门槛；柳筐的营收低于那一门槛，这笔收购没有经过审查。第 23 年春，通衢宣布柳筐"并入通衢市集"，柳筐停止运行；卖家可以把店铺"一键迁到通衢"，按通衢的佣金。一位创办人后来对记者说："我们没有被逼着卖；只是那个价钱，我们拒绝不了。"另一位进了通衢，任手工品频道的主管。

受这笔收购作用的，是柳筐的两千家卖家，青窑镇那九家作坊在其中，也是三位创办人。一条实际在用的替代渠道被买下、被关掉，这笔收购走过什么程序，是这里要读的。事实只有两头：收购以前，柳筐在运行，有两千家卖家；收购以后，它停了。这几件事里没有记载王国州别处还有哪些替代的市集，本节不说一切替代都已消失；也不拿创办人的同意，去替其他卖家的处境作答。

通衢买下的是什么？是一家公司，也是一条路。公司是三位创办人的，他们卖，是处置他们自己的东西；没有被逼，价钱高，这是一项交换。高价本身不是强力：要的是自己那份东西的条件，是交换（第二篇 E11）。可同一笔交易里，还有一件不属于创办人的事：两千家卖家本来可以走的一条路，关了。他们可以"一键迁到通衢"，迁过去，路就只通向买下它的那一家，佣金由五分变成一成五。创办人的同意答的是创办人的事；那两千家卖家的路，是另一笔账。

政府论在一级政府里说过：「决定可以终局：具体事项可以答完、结案。位置不能终局：没有一个位置可以独占关门之权」（政府论 G36）。这里推到经济位置（自担）。一笔收购可以作完、结案；可一个位置若能在替代者长大以前一一把它们买下、关掉，它握着的就是关门之权，别人另走一条路的可能，在它手里。经济学第一篇说：「反垄断的本体论动机是：防止一方把其他主体降格为means。」（经济学第一篇 §5.1）照这一句读，要看的是这一笔收购关掉了卖家另寻出路的哪一份余地；读的是这一项操作，不给哪一方定性。

"不能独占关门之权"要求的，不是禁止收购，也不是不许通衢长大。它要求的是：这样一项关门的施为，也在一条通衢关不上的通道之下。这一通道有两样常见的手段。一样是事先的审查，收购在门槛以上要申报，审查处可以问它；柳筐那一笔因为低于门槛而没有被问，门槛怎样定，是规则层的事（第七节）。另一样是事后的拆分与重组，它让一个已经长成的位置不终局。竞争审查处建议场与自营分开，是后一样的一种（第五节）。手段要说明它的作用、代价与替代安排；制衡落到某一项封路的用法上时，那一项可以收缩。收缩的是那一项用法，不是以减少通衢为目的。

本系列不判通衢买柳筐是为了关掉它。限与侵之界由结构定，不由动机定（制度论 Z9）；同理，读一项关门的施为，看的是它关了什么、谁能问它，不看买的人当时怎么想。

有人会说，这是一笔再自愿不过的交易：创办人拿到二十倍营收，卖家可以迁过去，没有一个人受损。这话有一半是对的：三位创办人确实没有被逼，迁移的路也确实开着；把一笔自愿的收购读成侵害，是不对的，本节也不这样读。可是"没有人受损"说多了。迁过去的卖家，佣金由五分到一成五；若有只在柳筐卖的，要么迁，要么另找；那九家青窑作坊，少了一个能与通衢比价的地方。这些不使收购成为越线；它们说明的是，一个能买下并关掉替代之路的位置，权力面在哪里，制衡为什么要落到这里。

柳筐的两千家卖家后来去了哪里，各自的营业额怎样，这几件事里没有记载。

E18　没有一个经济位置可以独占关门之权（自担）。「决定可以终局：具体事项可以答完、结案。位置不能终局：没有一个位置可以独占关门之权」（政府论 G36），推到经济位置（自担）。收购尚未长大的替代者，关掉的可以是一条别人本来可以走的路；卖方的同意答的是卖方自己的事，不替靠这条路的人作答。「反垄断的本体论动机是：防止一方把其他主体降格为means。」（经济学第一篇 §5.1）事先的审查与事后的拆分、重组，是规则层让位置不终局的手段；制衡落到某一项封路的用法上，那一项可以收缩，手段要说明作用、代价与替代安排。

第七节　量法、形态与门槛

柳筐那一笔收购，因为营收低于修订前的申报门槛，没有经过审查。泉坳那一处也有一道门槛：王国州《公用事业法》只管服务五千户以上的供水者，泉坳两个村约一千户，州公用事业委员会对它的水价没有权限。两道门槛都按规模量：一道量营收，一道量户数。两处都有一个位置，在门槛以下：一个能买下并关掉一条替代之路，一个是两个村唯一的水。

这一节问：权力面怎样量，制衡取什么形态，正式的分权从哪里开始，由谁定。

本系列的回答是：这些都在规则层，本系列只举例。量法可以看纠缠的范围，看不可替代性，看不可逆性，看能把一种积累兑换成别的作用的能力，看在位的时长；形态可以是分权，可以是说明的义务随规模加重，可以是拆分与重组，可以是多个来源的抗衡。这些都是例子，本系列不主张它们是最好的。

有两处现成的样例，要连同它们自己的条件一起读。F6 附录三为公司写的那份原则，篇末有一张反推的表，在"权力制衡"一行写：「公司规模小，用优先级和透明代替正式分权」（F6 附录三，篇末反推表；工作展示，只作样例）。这一句立在同一组表里"公司"一行自己的条件上：凝聚力的来源是「不疑+利益共生」，退出自由度是「高度自由（用脚投票）」。条件变了，这一句不能照搬。青窑镇的工人走一步就是生计，泉坳的人家没有第二个水龙头，退出都不高；在那里，用优先级和透明代替正式的分权，根据就不在了。另一处是那份原则里的优先次序：员工、用户、合作伙伴、股东、创始人，权力越大，排得越后。经济学第一篇说这条次序由三条独立逻辑推出，其中权力一条是「权力越大者排位越后，防止权力被用于自利」（经济学第一篇 §5.4）；附录三自己的写法是：「权力越大的主体，排位越后。」（F6 附录三第 2 条 (b)）这是一种形态的例子：权力大的位置，受益排在后面。

政府论在一级政府里，把正式分权的门槛写作：「门槛在条件，不在人数：持有强制，且退出与外层路径都承担不了日常的制约」（政府论 G26）。原说的是政府。前半"持有强制"不搬到经济位置上；这里只借后半作一把对照的尺子：门槛看退出与外层路径承担不承担得了日常的制约，不只看规模（推及，自担）。拿这把尺子看，那两道按规模量的门槛缺了什么：它们量的是营收、户数，不量退出与外层路径承担不承担得了日常的制约。门槛该怎样定，是规则层的事，本篇不替王国州的立法者定；只是两个例子都说明，按规模定的门槛，可以把一个深的或宽的位置留在门外。

规范层在这里只守一句：门槛以下，守线与受问不弱。制度论说：「制度保持被追问与修正的通道，且通道满足三项条件」（制度论 Z34）；原说的是制度，这里推到经济的构，不论它在哪一道门槛的哪一边（自担）。门槛决定的是哪一种正式的审查、哪一个机关管它，不决定它从哪一刻起才须守线、才可被问（第一节；前置篇第六节）。泉坳水业公司在《公用事业法》的门槛以下，它仍受它的应当，在生命基底处唯一的供给位置上另持职责（第二篇 E11a）；柳筐的收购在申报门槛以下，它关掉的那条路，仍可以被问。

规范层什么刻度都不给，只说"在规则层"，会不会是把难题推给别人？立法者真要定一个数时，规范层确实帮不上那一个数。可是规范层给了两样不是刻度、却管着刻度的东西：量的是权力面，不只是规模；门槛以下不豁免守线与受问。一道门槛若定得把两个村唯一的水的价钱留在一切受问之外，它就要按这两样被问。

两道门槛后来改没改，这几件事里只记了《公平竞争法》的修订与它的生效时间。

E19　量法、形态与门槛在规则层。权力面怎样量，制衡取什么形态，正式分权的门槛在哪里，都在规则层，本系列只举例。F6 附录三篇末反推表写「公司规模小，用优先级和透明代替正式分权」，它立在该表"公司"一行自己的条件上：凝聚力来源「不疑+利益共生」，退出自由度「高度自由（用脚投票）」（工作展示，只作样例）；退出不高处不照搬。「权力越大者排位越后，防止权力被用于自利」（经济学第一篇 §5.4）与「权力越大的主体，排位越后」（F6 附录三第 2 条 (b)）分别作样例。政府论说门槛「在条件，不在人数」（政府论 G26），原说政府；前半"持有强制"不搬，这里只借后半作对照的尺子（推及，自担）。门槛以下，守线与受问不弱：据制度论 Z34，推到经济的构（自担）。

第八节　强者的退出

第 24 年春，竞争审查处的报告出来了，建议场与自营分开。州议会同时在审《线上市集法》草案：市集要公开排序规则，设独立于市集的卖家申诉机构；另有一款，州政府可以请求商贸署依审查处的报告启动拆分程序。

通衢发表声明：草案若连同拆分一款通过，通衢将把总部与结算中心迁到目的民族联盟（the Nation Union of Telos，NUT）南面的一个区；那个区的政府给了通衢十年免税与一片园区。声明说："通衢不是王国州的财产。"

通衢在王国州雇员约四万一千人，含物流；州内的卖家里，约三成只在通衢卖。州议会经济委员会听证时，一位小卖家说："它走了，我们也得跟着走，不然就没处卖。"

这份迁移的声明，作用于正在审议的州议会，也作用于那三成只在通衢卖的卖家与四万一千名雇员。手里有的，是这份声明本身，以及几样结构依赖的线索：雇员数，只在通衢卖的三成，那位卖家的话。"没有这份声明时议会会怎样审"，只能作对照，不能补出它可能的结局；总部迁走，也不等于全部业务与雇员都搬走。

通衢说它不是王国州的财产，这句话是真的。它可以走，这也是真的。要问的是：它的"可以走"，在这里是什么。

权利论第四篇说过退出怎样变成权力：「只有当结构依赖在场时，退出权才经 F15 演化为权力。」（权利论第四篇 §V）若一个关系对一个主体的持续参与没有结构依赖，他退出，仍只是一项权利的行使；若有，退出就显出权力面。F18 的前提，是一项已登记或被承认的退出位置：「没有这样一个被登记或被承认的关系位置，可能有逃离、拒绝或脱身，但没有严格意义上的退出权」（权利论第四篇 F18）。原说的是主体退出一个关系；一个构把迁出系在一项立法上，推到这里（自担）。F18 还写明，退出显出的权力面，「不是主体怀着致塌的意图，而是制度的结构依赖把这次撤离被动放大」（权利论第四篇 F18）；这里照此读结构，不读通衢的用意。

照这样读，事实指向这几处。王国州对通衢有结构依赖：四万一千个岗位，三成只在它上面卖东西的卖家。通衢一走，这些人的处境跟着变；那位卖家说"我们也得跟着走"，说的就是这个。所以通衢的"可以走"，在这里显出权力面：它收敛的是州议会审议时能选的余地，也是那三成卖家往后的路。走得了的是它，走不了的是依赖它的人。这一不对称，正是第二篇写"退出真实"时那一面的反面：弱者的退出要问真不真（第二篇 E13），强者的退出要问它收敛了谁。

读出权力面，不等于说这份声明越了线，也不等于说通衢不许走。一个构选在哪里设总部，是它的决定；声明公开发表，也是一种说明。读出权力面，说的是：这一用法在这里收敛了别人，就落在第一节那一条之下，制衡随它递增，也要能被问。州议会怎样在这样的声明之下审议，它第 24 年秋删去拆分一款，经它自己的通道受问；那一次删改与声明、与捐款有没有关系，第四篇照第二道闸核，政经分离是它在公共权力一级的极端情形（第四篇 E23），本节不判。

那一区给迁入的条件，十年免税与一片园区，是那一国的施为，照它自己的法与它自己的通道读。各国为处理这样一个跨境的经济位置而发生的协作，照国家间协作论已经立下的功能边界。公司与公司之间、国家与公司之间的"制度间的制度"，国家间协作论明写不接（国家间协作论第八篇 XZ56），本系列这一轮也不展开，登作余项。本节只读迁移的威胁怎样作用，管辖与送达怎样变。

下面是一个设例。设若第 25 年通衢真迁了总部，王国州的业务改由那家 NUT 公司的一家子公司经营；一个卖家起诉，要经跨国送达。问的路变长了：卖家起诉，要经跨国送达；判决在哪里、怎样执行，设例里没有给，这里只作待问。程序的改变本身，不独自证明起诉已经无效，也不证明它照样够得着。要核的仍是同一件：被问的位置，是不是还在一条它自己关不上的通道之下。跨国送达与执行，靠的是各国之间的司法协助一类安排，本篇不设。

反对本节最直接的一说是：一家公司能用脚投票，是对一个州最好的制衡；把它的退出读成权力面，就是把它关在王国州。它说中了一层：构的流动确实也制约着政府。一个州若立了糟糕的法，企业走开是它付出的代价之一，这一层不该被抹掉。可是读出权力面，不关住任何人。它不否认通衢走的权利，只读这一用法在这里落在谁身上。公私同一在这里两边都用得上：州议会的施为受问，通衢把迁出系在一项立法上的施为也受问（第三节）。制衡的是用法，不是那一次走。

一州之内制衡不住、跨了国境的那一部分，本系列只读到威胁的作用与问路的变化为止。

E20　强者的退出在结构依赖下显出权力面（自担）。「只有当结构依赖在场时，退出权才经 F15 演化为权力。」（权利论第四篇 §V）F18 的前提是一项已登记或被承认的退出位置；一个构把迁出系在一项立法上，推到这里，标自担；照 F18，读的是结构依赖，不是用意。走得了的是它，走不了的是依赖它的人。读出权力面，不等于判它越线，也不禁止它走；这一用法落在第一节那一条之下，可被问。一国给迁入的条件，是那一国的施为；各国为此发生的协作，照国家间协作论已有的功能边界。公司与公司、国家与公司之间的制度关系，本系列不展开（国家间协作论第八篇 XZ56 明写不接），登作余项。

这一篇停在哪里

这一篇证了这些。经济位置的权力面越大，对它的制衡越强；这是本系列的规范承担，不是结构定理，权力面按具体的作用比较，不按资产总额（E14）。构所做的事落在涌现层的，以制衡为主；落在基础层的，制衡之外，制度还要保障那一项供给，不限唯一的供给者，保障不是背书（E14a）。权力面相当的积累位置，不因私有而豁免，公的一侧也不另得豁免；一个位置两种来源，各受各的制衡（E15）。线内的控制照三问核，独立照 Z34 对着实际被问的一方核；制衡者自己的决定也照三问受问（E16）。制衡落在怎样使用上，不以压低能力为目的；同一处既是生产也是卡住时，照实说明作用、代价、替代与理由（E17）。没有一个经济位置可以独占关门之权；卖方的同意，不替靠那条路的人作答（E18）。量法、形态与门槛在规则层；门槛以下，守线与受问不弱（E19）。强者的退出，在结构依赖下显出权力面（E20）。

它没有证的：规模与权力面的刻度；制衡的具体形态；反垄断法与审查门槛的细则；公司内部的分权，交公司论；跨境的公司关系的结构。

交给后面各篇的：积累换成别的作用时，哪里设闸，在第四篇，通衢支付取得的那笔公共贷款（第四篇 E25），捐款与离任受聘（第四篇 E23），都在那里；第三代与申诉委员会的改任命，控制安排的自修凿，在第五篇（第五篇 E29、E30）；南窗的回报目标，在第五篇（第五篇 E27）；基础层要保障、涌现层要支持的两层，在第六篇（第六篇 E35）。

留下的余项：退出了通衢的那一家作坊，退出真不真，照第二篇逐项核；申诉委员会受理什么、改过什么，竞争审查处的报告怎样被问，没有记载；场与自营分开的代价没有量；柳筐的卖家后来去了哪里没有记载；南窗的底够不够，由谁来问，留给规则层与判断层；跨了国境、一州之内制衡不住的那一部分。这些都照记，各有去处。

本篇命题（合列）

E14　制衡随权力面递增（自担：规范承担，不是定理）。经济位置的权力面越大，对它的制衡越强。权力面按具体的作用比较：它收敛谁，收敛到哪一个维度，多少人要经过它，走开的代价多大；不设资产总额的单一刻度。「构的规模不能单独给出其制度条款的必要数量和粒度。」（F6 命题 III）这一条是本系列的规范承担，不写成"权力与职责成正比"的结构定理（F6 附录三第 18 条：「这一安排不由一条权力与职责成正比的结构定理给出」；工作展示，只作样例）。规模与权力面影响制衡的形态与强弱，不决定一个构从哪一刻起才须守线、受问。刻度与形态在规则层。

E14a　按层：制衡与保障。构所做的事落在涌现层的，以制衡为主；落在基础层的，制衡之外，制度还要保障这一供给。保障是「把线显明、具体化，给它制度的背靠，并有真实的执行能力」（制度论 Z17）；「政府给的是背靠与执行」（政府论 G18）；原说基础权利的保障，这里推到做基础层之事的构所承担的供给（自担）。不限唯一的供给者：即使有多家供给，是否足以保障基础层，仍须另核。保障的对象是供给，不是那个构：保证不是背书（政府论 G22；原说政府的保证不使它自己的权力正当，这里推到不替提供供给的构背书，自担）。达到最低标准，不自动证明在线内，也不自身判为越线。"保障"用于基础层的供给，与为人的涌现给条件的"支持"分用。哪些事算基础层，在规则层。

E15　公私同一：按权力面认，不按公私认（自担）。权力面相当的积累位置，不因私有而豁免制衡；公的一侧也不因此另得豁免。授予的位置照它的来源受制衡，公共授予须保留可撤的功能（政府论 G37）；积累的位置照本系列受制衡。一个位置可以两种来源都有，照各自的来源分别受制衡：授予的收入变成积累的持有，不洗掉授予与投资决定那一边的账；积累的持有也不因此一概带上授予的可撤。

E16　线内的控制推到经济位置；制衡者也受问（自担）。线没越、通道也在时，受一个经济位置作用的人，对它的相关决定有没有实际的控制，照三问核：「谁能启动；对理由、行动、规则或任职位置有什么实际效力；被问的一方怎样不能独自取消它。」（政府论第六篇第二节）控制不等于每个受影响者都有否决权（政府论 G36a）。由被问的一方任命多数的申诉安排，独立照制度论 Z34 对着实际被问的一方核：「独立于被追问的一方」，「独立是功能义，不要求在制度之外」；三项条件的出处在 F6 命题 IV，它立在以不疑为前提的制度里，只立条件，应当据 Z34。制衡者自己的决定（调查、提出拆分、依程序启动或裁决）也照这三问受问：制衡也是构的施为（制度论 Z9；政府论 G36）。没有给出的复核与救济，不当作已经存在；风险可以使规则需要预防，不充作已经捕获的终判。原说政府的决定，这里换作经济位置与制衡它的位置（自担）。

E17　制衡落在怎样使用上，不以压低能力为目的（自担）。制衡设在卡住、兑换、封死退出、拒绝追问、位置固化的用法上；生产、建设、提供条件的地方放开。同一处既是生产也是卡住时，制衡须说明它处理的是哪一项封路或不可问的作用，怎样尽量保全生产、服务与给条件的能力；若改变配置也影响这些能力，照实说明代价、可替代的办法与选择的理由；不先承诺两面在每一处都可以无损地剥离。不靠给一个动作命名为"生产"，就豁免核线。否决点过多，也会使一个位置什么都做不了。

E18　没有一个经济位置可以独占关门之权（自担）。「决定可以终局：具体事项可以答完、结案。位置不能终局：没有一个位置可以独占关门之权」（政府论 G36），推到经济位置（自担）。收购尚未长大的替代者，关掉的可以是一条别人本来可以走的路；卖方的同意答的是卖方自己的事，不替靠这条路的人作答。「反垄断的本体论动机是：防止一方把其他主体降格为means。」（经济学第一篇 §5.1）事先的审查与事后的拆分、重组，是规则层让位置不终局的手段；制衡落到某一项封路的用法上，那一项可以收缩，手段要说明作用、代价与替代安排。

E19　量法、形态与门槛在规则层。权力面怎样量，制衡取什么形态，正式分权的门槛在哪里，都在规则层，本系列只举例。F6 附录三篇末反推表写「公司规模小，用优先级和透明代替正式分权」，它立在该表"公司"一行自己的条件上：凝聚力来源「不疑+利益共生」，退出自由度「高度自由（用脚投票）」（工作展示，只作样例）；退出不高处不照搬。「权力越大者排位越后，防止权力被用于自利」（经济学第一篇 §5.4）与「权力越大的主体，排位越后」（F6 附录三第 2 条 (b)）分别作样例。政府论说门槛「在条件，不在人数」（政府论 G26），原说政府；前半"持有强制"不搬，这里只借后半作对照的尺子（推及，自担）。门槛以下，守线与受问不弱：据制度论 Z34，推到经济的构（自担）。

E20　强者的退出在结构依赖下显出权力面（自担）。「只有当结构依赖在场时，退出权才经 F15 演化为权力。」（权利论第四篇 §V）F18 的前提是一项已登记或被承认的退出位置；一个构把迁出系在一项立法上，推到这里，标自担；照 F18，读的是结构依赖，不是用意。走得了的是它，走不了的是依赖它的人。读出权力面，不等于判它越线，也不禁止它走；这一用法落在第一节那一条之下，可被问。一国给迁入的条件，是那一国的施为；各国为此发生的协作，照国家间协作论已有的功能边界。公司与公司、国家与公司之间的制度关系，本系列不展开（国家间协作论第八篇 XZ56 明写不接），登作余项。

参考文献

以下 SAE 文献一律为概念 DOI，始终指向最新版本。逐句核验所据的版本记录，见随稿的转述审计表。

一 · 本系统

秦汉. (2026). 制度如何可能：从主体间余项到共构框架（SAE 基础论文 Paper 6） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19328662

秦汉. (2026). 论承认性权利与权力的界面与转化（SAE 权利论系列 Paper 4） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21093939

秦汉. (2026). 不越线（SAE 制度论 第二篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23270384

秦汉. (2026). 守线（SAE 制度论 第三篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23270766

秦汉. (2026). 可被追问（SAE 制度论 第六篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23271324

秦汉. (2026). 保证（SAE 政府论 第三篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23413112

秦汉. (2026). 层级（SAE 政府论 第四篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23413192

秦汉. (2026). 追问与承担（SAE 政府论 第六篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23413511

秦汉. (2026). 收束（SAE 国家间协作论 第八篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23815796

秦汉. (2026). 经济理性的四个层级：维度序列理论对理性人假设的诊断与扩展（SAE 经济学系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19358010

二 · 外部文献（只据二手转述，C 档）

Robeyns, I. (2024). Limitarianism: The Case Against Extreme Wealth. London: Allen Lane.（原书未核；据乌特勒支大学刊出的访谈《What is limitarianism? Professor Ingrid Robeyns explains》转述，https://www.uu.nl/en/in-the-media/what-is-limitarianism-professor-ingrid-robeyns-explains）

Anderson, E. (2017). Private Government: How Employers Rule Our Lives (and Why We Don't Talk about It). Princeton: Princeton University Press.（原书未核；据 Estlund 2018 转述）

Estlund, C. (2018). Rethinking Autocracy at Work. Harvard Law Review 131: 795. https://harvardlawreview.org/print/vol-131/rethinking-autocracy-at-work/

接口

主承： 权利论第四篇 F15、F18 与 §V 的阈值段（F18 推到构的迁出，自担）；政府论 G36、G36a 与第六篇第二节的三问（推到经济位置，自担）；制度论 Z34（推到经济的构，自担）；F6 命题 III。

另用： 制度论 Z9、Z17；政府论 G18、G22、G26（对照）、G37；F6 命题 IV（三项条件的出处）；F6 附录三第 2 条 (b)、第 18 条与篇末反推表（工作展示，只作样例）；经济学第一篇 §5.1、§5.4；国家间协作论第八篇 XZ56。

对接： 罗宾斯《限富主义》（C 档，据访谈转述）；安德森《私人政府》（C 档，据 Estlund 书评转述）。

本系列： 前置篇第三、四、六节；第一篇 E1、E2；第二篇 E7、E10、E11、E11a、E12、E13；第四篇 E22、E23、E25；第五篇 E27、E29、E30；第六篇 E35。

交给第四篇： 积累换成别的作用时，哪里要设闸。