Guarantee
保证

Introduction

In the winter of Year 7, the application counter of Apple City's Housing Department is open as usual.

An Apple City ordinance embeds a heating subsidy for low-income households: a household whose income falls below a certain line may receive a fixed amount each month in winter; it applies to the Housing Department, which decides within thirty days. This provision was written into the ordinance in Year 1. The application counter has stayed open, and applications are received as usual. But the post in the Housing Department responsible for assessment has stood vacant since the winter of Year 4, and no one has been appointed to it since. By this winter, twelve hundred applications have piled up, and no one is processing them.

One applicant handed in an application at the start of winter, and forty days later went to the counter to ask. The person at the counter said: "We take the applications. The assessment post is empty."

The ordinance is still there, the counter is still open, applications are still received. On paper, the subsidy lacks nothing. It simply does not operate.

Three things first. Whom it concerns: the households that have applied and meet the conditions. Which line: that a post stands empty and a pile of applications accumulates does not in itself cross anyone's line; an embedded recognition-based position has a gap in operation, and by the reading of §3 this belongs to the third reading of "no empty shell," and goes through the explanation and correction of the fifth ought. Whether a winter without heating presses on some household's life-substrate is checked separately (§2). Baseline: the decision within thirty days that the ordinance stipulates.

This subsidy has appeared before, in Paper 1. From Year 2 to Year 4 the Housing Department handed initial assessment to a contracted scoring company, and that passage asked where the bearing falls once judgment has been handed to a machine (Paper 1, §9). This paper does not rewrite that passage. There the question was who answers; here it is how far a government must guarantee something written into an ordinance, what the guarantee is made of, how its falling through is read, and how withdrawing or amending it is questioned.

Paper 2 wrote about compulsion. Beyond compulsion, what a government does most is another kind of thing: assessing an application, paying out a subsidy, seeing that a protection order has someone to enforce it, collecting tolls on behalf of a franchise or declining to. These do not look like compulsion, yet the weight they place on people is often not light. How the households behind those twelve hundred applications get through this winter rests on one empty post.

The paper proceeds in eight steps. §1 separates the two kinds of rights. §2 says what a government gives in the half that concerns basic rights. §3 says which three things make up the guarantee of recognition-based rights. §4 says that a guarantee is not an endorsement. §5 divides supply into slots: the embedded, the not embedded, bare entitlement. §6 and §7 treat administration, public finance and tax. §8 says what cannot be cut when conditions are limited. Conventions follow the Prequel, §8; this paper has no closing section of conversation. The paper has nine propositions (G17 to G24, with G20a), listed together at the end.

§1 Keeping the Two Kinds of Rights Apart

There are two ways of speaking about those twelve hundred applications.

One says: the subsidy is something the government gives; it gives when it can, and when it cannot, it cannot; there is no question of owing anyone.

The other says: heat in winter is everyone's basic right; if the government does not provide it, that is a violation.

Each is half right and half wrong: one treats the subsidy as a favor, the other treats it as a basic right.

Rights Theory divides rights into two kinds, running on two spines. One is basic rights: "Basic rights are paired with the Dao, come from one's own law, are ontologically prior, presuppose restraint-type recognition, and do not enter the T5 seam" (Rights Theory P1 F2); the recognition they presuppose is "the others' forbearance from violation" (ibid., §II), and they do not need to be constituted by anyone. The other is recognition-based rights: "Recognition-based rights sit with power on the two sides of the T5 seam" (ibid., F2); such a right "does not come by grace or permission; it is constituted by constitutive recognition" (Rights Theory P3 F10).

The half that concerns basic rights not being violated is not peculiar to government. A person's not being enslaved, his body's not being harmed, is not given by the government, nor by anyone at all. It comes from that person's own side and holds against everyone: against a neighbor, against a factory, against the government alike. Institution Theory says: "basic rights are not newly made by the institution" (Z17). Nor is not violating them a demand that falls on government alone: a neighbor may not violate them, nor may a factory, nor may the government. Institution Theory has already established the ought addressed to the construct: the construct's acts cross no line against any internal subject, and not violating basic rights belongs here (Z9); the construct guards the lines between its internal subjects, and securing basic rights belongs here (Z15). The Theory of Government takes this over and does not establish it, nor does it write it as an undertaking peculiar to government.

The half that concerns guaranteeing recognition-based rights is an undertaking peculiar to government. The subsidy is not a basic right. Rights Theory says it plainly: "If continuing supply is to become something claimable, it is constituted as an institutionally embedded position at the recognition layer; it is not constituted as the content of a basic right." (Rights Theory P5 §II) The households that have applied can claim the subsidy because Apple City wrote it into an ordinance.

So it is now clear where the two ways of speaking are right and where they are wrong. The first is right about the source: this subsidy is something Apple City wrote into an ordinance, not something that grew from each household's own side. It is wrong in its second half: once the subsidy is embedded, it is not a favor the government gives or withholds as it pleases; it is a position, on the strength of which applicants can apply, question, and demand an answer. The second is right about one thing: everyone has a floor that no one may cross, the life-substrate included. It is wrong in counting the subsidy itself as part of that floor: the subsidy's stopping is not on that account a violation of a basic right; whether there is a violation on other grounds is checked separately (§2).

Here this series establishes one ought: a government ought to guarantee the recognition-based rights it has embedded for its internal subjects. Without the guarantee, the government's power at that place is illegitimate.

This ought is not derived from Rights Theory. What Rights Theory gives is structure: recognition-based rights and power sit on the two sides of one seam—read from the side of exertion, it is power; read from the side of the subject's position, it is a recognition-based right (Rights Theory P1 §II; see also the Rights Theory Prequel, §III). A government holds the power side of the seam. If it holds only the power and does not let the rights side operate, then on Rights Theory's reading the position breaks down: either compulsion persists while the network no longer processes the recipient's requests, "the position degenerating into sheer force"; or backing is drawn away while the shell remains, "the position reverting to a declaration" (Rights Theory P3 F11, §II). The Housing Department's case is of the latter kind. The structure gives a reading, not an ought; Rights Theory itself says that its criterion for de-righting "judges the structure of network-processing, not whether the thing is normatively legitimate smuggled back in" (Rights Theory P3 §II). The ought is established by this series: guaranteeing the recognition-based rights embedded for internal subjects is a normative undertaking newly added on the government's side (declared in the Prequel, §5, and established here). It adds no sixth heading; taken apart, it is organized into Institution Theory's first, second and fifth oughts (§3). "Peculiar" means this: in this series the undertaking is established only for government, and it is the Theory of Government that establishes it; it does not assert that other constructs owe nothing at all for the positions they have embedded. Institution Theory does not record it; what other constructs owe belongs to their respective specific theories of institutions, and this paper does not establish it for them.

The guarantee of both kinds of rights can also be understood as part of anti-colonization: both lie on the side of holding to, not on the side of striving for (the Prequel, §8; Institution Theory Z15 means the same when it says that securing basic rights lies "on the side of anti-colonization"). "Colonization" is still taken in the narrow sense: a boundary-crossing disguised as cultivation (Institution Theory Paper 1, §8). Not every crossing of the line poses as cultivation, and an empty shell is not a crossing of the line either; these are read under the first, second and fifth oughts, each where it falls (§3).

"Illegitimate" is read as "must be questioned." The "illegitimate" of this ought has a definite landing place in this series: must be questioned, that is, it must be explained and must be corrected, through the channel of the fifth ought (Institution Theory Z34). Its range must also be stated precisely. Wherever the lapse is, the government's acts through the position at that place, within that scope, are illegitimate; the Housing Department's empty assessment post is such a place. Where lapses have become the normal state—one item hollowed out after another, or de-righting—the government's power must be questioned as a whole. Crossing the line does not land here; it is read separately under the first ought.

Two expressions must be kept apart here. "Open to questioning" means that the channel is always open: for every government, at every time, whether or not it has lapsed (Institution Theory Z34; Paper 6, G35). "Must be questioned" means that a verified lapse owes an explanation and a correction. The Housing Department's lapse has been verified: the post has stood empty for three years, twelve hundred applications have piled up, and the thirty-day limit has not been kept for a single one of them. It must be questioned. Other departments of Apple City may at this moment have no verified lapse anywhere, and they are open to questioning all the same; a government can never be perfect, and being questioned is the normal state (Paper 1, §8). The two are not mixed: being questioned is in itself no evidence of illegitimacy, and providing the channel of being questioned is no certificate of repair.

One more qualification must be stated here at the outset: with the guarantee, the power is not thereby legitimate. The guarantee is a necessary condition, not a sufficient one (§4).

With Honneth. The "recognition" in "recognition-based right" is a structural word of Rights Theory: a position is constituted at its source by constitutive recognition and operates through the interlock of three legs (Rights Theory P3 F10). It is not the recognition Axel Honneth speaks of. In Honneth, recognition is divided among three spheres, love, law and social evaluation, in which a person gains in turn self-confidence, self-respect and self-esteem (per the entry "Recognition" in the Stanford Encyclopedia of Philosophy (hereafter SEP), on which Institution Theory Paper 0, §6.1 relies; tier C). The dimension of law lies nearest to what is at issue here; even in that dimension, Honneth asks how legal recognition makes for a person's self-respect, while Rights Theory asks how a position is constituted at its source, and the two questions are not one question. Institution Theory Paper 0 has already conversed with Honneth on how institutions stand to recognition between persons, and at the point on "self-respect and standing" it too first separated the questions and pronounced no conflict (§6.1); here this paper only points back to it, noting only that the two places share a name but refer to different things.

G17 Guaranteeing the two kinds of rights: the two halves kept apart. That basic rights may not be violated is derived from the subject's side and holds against everyone; it is not peculiar to government; the ought addressed to the construct already stands in Institution Theory's first and second oughts (Z9, Z15, Z17), and the Theory of Government takes it over and does not establish it. Guaranteeing recognition-based rights is an undertaking peculiar to government, established by this series as a normative undertaking newly added on the government's side: a government ought to guarantee the recognition-based rights it has embedded for its internal subjects; without the guarantee, its power at that place is illegitimate, that is, it must be questioned; Rights Theory gives the object and the criteria of failure, and does not by itself issue this ought. "Peculiar" means that in this series this undertaking is established only for government; it does not assert that other constructs owe nothing at all. It adds no sixth heading, and is organized into the first, second and fifth oughts (G19). The guarantee of both kinds of rights can also be understood as part of anti-colonization, and "colonization" is still taken in the narrow sense. "Open to questioning" holds for every government and at every time; "must be questioned" means that a verified lapse owes an explanation and a correction; the two are not mixed.

§2 The Backing of Basic Rights

Paper 2 told of an incident at the Orchard Court residential complex: the resident upstairs blocked the resident downstairs in the stairwell and said, "Go to the management again and you won't get out of this door." The resident downstairs called the police; a patrol officer arrived, separated the two, and helped him apply to the City Court for a temporary protection order (Paper 2, §9).

That the resident downstairs is not threatened, not blocked at his own door, is not given by the protection order, nor by the City Police Department. What the protection order gives it is a written form that can be invoked; what the City Police Department gives it is someone who will come and enforce it.

Rights Theory has a two-by-two table that sets this out clearly. By source, basic rights are endogenous and recognition-based rights are registered; by direction, there is negation and emergence. The cell where registered meets negation is the institutional backstop for the negative floor: a constitution that writes the prohibition of slavery and of torture as untouchable clauses is "giving the basic-rights floor 'refuse to be reduced' a registered backing" (Rights Theory P1 §II). One and the same right has its core on the basic side and its backing on the recognition side.

Backing can be hollowed out. Rights Theory says that where a freedom is still written into law while enforcement bodies, procedure and the network's recognition are all drawn away, "What is hollowed out then is the institutional backing, what is suppressed is the exercise of this basic right, and what is untouched is its core." (Rights Theory P3 §II)

Suppose that the City Court issued the temporary protection order, but the City Police Department said it was short of staff and never sent anyone to enforce it. The protection order is written on paper and cannot be delivered. What is hollowed out is the backing; that the resident downstairs is not to be threatened stands undiminished. What is missing is its anchorage in Apple City.

So what a government gives in this half is backing and enforcement, not the core. This sentence has to head off two readings.

One reads the government as the source of the core: if the government does not give it, there is none. Not so. The core does not move; "basic rights are not newly made by the institution" (Institution Theory Z17).

The other turns this around: since the core does not move, the government owes nothing. Not so either. Backing and enforcement are on the recognition side, and fall exactly under the guarantee of the next section. Institution Theory says that securing is a set of conditions given public form, and that "Protection on paper is not securing; being able to enforce does not prove that what is enforced, or the authority to enforce it, is warranted" (Z17). A protection order signed with no one to enforce it goes, at this point, under the second ought: the construct guards the lines between its internal subjects, and the way it guards may not be reduced to paper alone. If a government does not guarantee this layer of backing, its power at that place is likewise illegitimate and must be questioned.

Back to those households. Does a winter without heating press on some household's life-substrate? This must be checked separately, and what is being checked must be stated precisely. Rights Theory draws a restriction: "A violation on the basic layer is confined to action upon the existing thick layer and contains no claim upon future supply"; "Bare natural scarcity is not a rights event." (Rights Theory P5 F20) When an embedded supply fails, it is read on the recognition layer (§3); it does not on that account directly become a violation on the basic layer. What must be checked separately are several other things. Whether anyone has laid hands on this household's existing maintenance circuit, by severing it, gatekeeping it or monopolizing it: if the hands are someone else's, it falls under the second ought; if they are Apple City's own, under the first (§3, the second reading). Whether Apple City set up a separate backing for this floor, and whether that backing has emptied too: this falls under the second ought. This paper does not pass a verdict on any household. Rights Theory says in the same place that a withdrawal with neither a hand laid nor an embedded position registers on no scale in that set of readings, "which is not to say it weighs nothing, but that its weight does not lie in this register of readings" (Rights Theory P5 §II). These households stand on an embedded position; their weight lies in the set of readings on the recognition side.

G18 The guarantee of basic rights itself lies on the recognition side. What a government gives is backing and enforcement, not the core: the inviolability of the core comes from the subject's side, and basic rights are not newly made by the institution. Backing can be hollowed out; the core does not move. Backing and enforcement lie on the recognition side and belong to the guarantee of G19: where the backing of a basic right is hollowed out, it goes under the second ought; where a government does not guarantee it, its power is likewise illegitimate and must be questioned.

§3 Recognition-Based Rights: Three Things

Back to the Housing Department.

Rights Theory says that a recognition-based right operates on three legs: "institutional embedding gives an invocable form, effective community recognition gives distributed predictability, real power backing gives the capacity to deliver" (Rights Theory P3 F10, §I). Applied to the subsidy: the ordinance wrote it in, which is the first leg; the applicants and the counter alike treat it as something that can be applied for and questioned, which is the second; someone sits in the assessment post, decisions are made within thirty days, and the money is paid monthly, which is the third. The third leg has fallen below threshold. Rights Theory calls this hollowing-out: "Hollowing-out is power backing falling below threshold while the institutional shell remains, the position reverting to a declaration, operation gone." (Rights Theory P3 F11, §II)

For embedded rights, a government keeps three things: no de-righting; no empty shell; withdrawal or amendment made in public and open to questioning. Taken apart, the three fall into the first, second and fifth oughts; they add no sixth heading.

One: no de-righting. De-righting is a different kind of failure: "De-righting is coercion persisting while the relevant network no longer treats the recipient as a holder who can, on the strength of this position, initiate requests, invoke procedures, and require response, the position degenerating into sheer force." (Rights Theory P3 F11, §II) The subsidy's case does not amount to de-righting: the antecedent of de-righting is that compulsion persists, and the Housing Department has applied no compulsion to the applicants; it simply does not assess.

The shape of de-righting needs another example. Suppose that the freeze on the account of the grocery store owner in Paper 2, the one in arrears on his taxes, goes on and on; the owner files an objection, and the Tax Office does not accept it; he goes to the administrative division of the City Court, and there too no one processes his representations; the case officer is replaced, and the new case officer likewise does not accept his objection; the freeze stays as it was. The freeze persists, the compulsion persists; but the position on which he, as a taxpayer, can make representations and demand an explanation is no longer processed by the network.

To judge whether this is de-righting, Rights Theory gives three tests: "whether the recipient can initiate a procedure on the strength of this position; whether the position can be replicated and processed identically across independent nodes; and whether, after the direct power-wielder is replaced, the position is still maintained by the network." (Rights Theory P3 §II) In this supposition, all three answers are no. The third must be read with care: what remains after the person is replaced is the freeze; the request he makes on the strength of his position as a taxpayer still finds no node to take it up after the replacement. Rights Theory says: "Only when it can be sustained solely by the same power-wielder's continued pressure, the network not processing the recipient's request-position, is it de-righting." (ibid.) What holds up the freeze here is only the Tax Office itself; no independent node processes him as someone who can make requests on the strength of a position. The government is then going on acting through a position that has already degenerated into sheer force. This case falls under the first ought: the construct's acts cross no line against any internal subject (Institution Theory Z9); once an act no longer passes through a rights position, its effect on the person must be checked against the line, and it cannot be waved through with "this was originally a lawful freeze."

The three tests serve only as criteria, not as norms. Rights Theory itself says that the criterion they make up "judges the structure of network-processing, not whether the thing is normatively legitimate smuggled back in" (Rights Theory P3 §II). They tell us where a position breaks; what a government ought to do there does not come from the three tests, but from the ought this series has established (§1).

Two: no empty shell. This is the main matter in the case of the subsidy. An empty shell has three readings, which must be kept apart.

The first reading: where the backing of a basic right is hollowed out, it goes under the second ought (§2; Institution Theory Z17).

The second reading: where the pre-existing self-maintained or substitute circuits are first actively sealed off, so that people can depend only on that supply, and the position is then hollowed out, it goes under the first ought. Institution Theory says that dismantling, too, is an act of the construct: "the dismantling itself may not cross the line, for instance by first sealing off people's pre-existing self-maintained or substitute circuits, so that they can depend only on that supply, and then withdrawing; this part falls under the first ought and is not discounted" (Z38). Suppose that the Housing Department had earlier, on the grounds of central heating, removed the heating stoves that households in public housing had installed themselves, so that the tenants could pay for central heating only with the subsidy; then the assessment post stood empty, and the subsidy stopped running. Three things first. Whom it concerns: the public-housing tenants whose stoves were removed. Which line: Apple City first closed off their existing way of heating, then let the one remaining way fall through; the step of closing it off acts on their existing maintenance circuit and must be checked against the line. Baseline: the means of heating each household had installed itself, before the stoves were removed. What is hollowed out then is not merely a position of supply: Apple City had first laid hands on their existing way and closed it off. This branch is judged under the first ought and is not discounted. Z38 speaks of withdrawal; here the later step is hollowing-out, which leaves the shell in place and only lets the backing stop. Rights Theory says: "Monopolization and induced dependency are themselves the severing; the withdrawal only completes it, and splitting the act in two does not change the reading" (Rights Theory P5 §II); On Rights and Entitlements (Rights Theory, Specialist Paper) likewise says that ceasing to give is then "only the completing of a prior cutting-off" (On Rights and Entitlements F30). Whether the later step is withdrawal or hollowing-out, the reading is the same, so it is read by Z38 (extension; own undertaking). This reading must hold its threshold: no factual dependence settles a case directly under the first ought. Even if those households could in fact get through the winter only with the subsidy, that alone is not enough; "Mere factual irreplaceability is not monopolization: monopolization requires the exclusive laying of hands" (Rights Theory P5 §II).

The third reading: any other gap in operation enters the explanation and correction of the fifth ought (Institution Theory Z34). The Housing Department's case falls under this reading (Introduction). The explanation must answer to these things: what has not been delivered, that is, how long each of the twelve hundred applications has been waiting, and for how many the thirty-day limit has already passed; whether the original ground still holds, that is, whether the matter of heating for low-income households in winter is still there; and the arrangements for restoring operation, for transition, or for a grounded withdrawal or amendment, that is, when the post will be filled, what becomes of the backlog until then, and, if the subsidy is no longer to be run, on what grounds and by which route.

Explaining is not in itself repairing. However clearly an explanation is written, while the post is still empty and the twelve hundred applications still go unassessed, the empty shell is still an empty shell. Permitting questioning is not having fulfilled the guarantee either. "No empty shell" does not shrink into "no hidden empty shell." Whether the first or second ought is also touched is checked separately: the second reading above is one such case, and the life-substrate spoken of in §2 is another.

Where it truly cannot be achieved, for instance because no one can be hired for the time being, Paper 5, G33 applies: that something cannot be achieved is no ground for cancelling the burden of explanation and correction, nor is something booked as achieved because it has been explained.

Three: withdrawal or amendment made in public and open to questioning. If Apple City really no longer wants to run this subsidy, the road it must take is withdrawal or amendment, not leaving the ordinance hanging and the post empty. Withdrawal or amendment must be checked for authority, grounds, scope and transition, and made in public and open to questioning; this falls under the fifth ought.

What a government withdraws is its own share. Rights Theory separates source and operation into two layers: at the source, a recognition-based right is constituted by constitutive recognition, which has two branches, vertical institutional conferral and horizontal mutual recognition; in operation, it runs on three legs, institutional embedding, effective community recognition and real power backing; "Source-side institutional conferral and operation-side institutional embedding are not one thing" (Rights Theory P3 §I). Apple City's share is its institutional conferral at the source, together with its embedding and its backing, all withdrawn together and in public, with no shell left behind. The subsidy as it now stands, with the shell there and the backing gone, is exactly the shape that withdrawal or amendment is meant to avoid.

A public announcement neither acquires legitimacy by itself nor proves by itself that the right has been outright withdrawn. Outright withdrawal is "the exit of all constitutive sources within the relevant scope, not merely the disappearance of the old bearer" (Rights Theory P3 F11, §II); and "After an institutional conferral is revoked, horizontal mutual recognition may still constitute the same position" (Rights Theory P3 §II). What Apple City can withdraw is its conferral, its embedding and its backing; if mutual recognition among people still constitutes this position, the withdrawal is not yet outright withdrawal. This is not for the government to declare.

Suppose that the City Council repeals the subsidy ordinance and sets up a new program under a different name, with applications starting over from the beginning. What this step is must first be read clearly. If what the new program constitutes is the same position, only with a change of carrier, Rights Theory calls this carrier migration: "Carrier migration is the bearer migrating while the right runs on in the new bearer." (Rights Theory P3 F11, §II) Then the position continues, the applications filed under the old ordinance move over with it, and "starting over from the beginning" itself has to be explained. If it is a different position, the old one is withdrawn by the third thing. Either way, three things do not vanish along with it. The floor of basic rights: whether some household's life-substrate is pressed on is checked separately as in §2, and the answer does not change because the program has changed its name. Accounts already incurred: the twelve hundred applications already filed and already past thirty days are owed under the old ordinance, and are not written off because the old ordinance is repealed. Other independently standing sources and commitments: if the same position is also constituted by some other source, such as mutual recognition among people, it is checked by its own source; commitments already made are checked by their own conditions (the "existing commitments" of Institution Theory Z38).

What is held to is the ways. These three things lie on the side of holding to: they are not discounted for lack of resources, and do not stop in an emergency ("holding to" in the wider sense of the Prequel, §8, undertaken by this series). What is held to is these three ways, not the continued existence of every embedded position. The subsidy can be amended and can be withdrawn; the ways of amending and withdrawing are not withdrawn. Institution Theory has stated the same shape: "what is held to is the ought, not a particular structure" (Z22). If any of the three is missing, the government's power at that place is illegitimate and must be questioned; where lapses have become the normal state, its power must be questioned as a whole (§1). Of these, the effect exercised on people through a degenerated position, and the part in which a way is first closed off and the position then hollowed out, where verified as crossing the line, are read in addition under the first ought, and are not merely matters that must be questioned (§1).

The scope of the guarantee is itself subject to review as well: that a position is constituted in structure and is operating does not of itself require the government to deliver its operation for it. That is the next section.

G19 The guarantee of recognition-based rights: three things. A recognition-based right operates on three legs: institutional embedding, effective community recognition and real power backing. For embedded rights, a government keeps three things. First, no de-righting, which falls under the first ought: in de-righting, compulsion persists, and the government goes on acting through a position that has degenerated into sheer force; the three tests serve only as criteria, not as norms. Second, no empty shell: where the backing of a basic right is hollowed out, it goes under the second ought; where the pre-existing self-maintained or substitute circuits are first actively sealed off, so that people can depend only on that supply, and the position is then hollowed out, it goes under the first ought (extension; own undertaking), and no factual dependence settles a case directly under the first ought; any other gap in operation enters the explanation and correction of the fifth ought, and the explanation must state what has not been delivered, whether the original ground still holds, and the arrangements for restoring operation, for transition, or for a grounded withdrawal or amendment. Explaining is not repairing, and permitting questioning is not having fulfilled the guarantee; "no empty shell" does not shrink into "no hidden empty shell"; whether the first or second ought is also touched is checked separately; where it truly cannot be achieved, Paper 5, G33 applies. Third, withdrawal or amendment made in public and open to questioning, which falls under the fifth ought: authority, grounds, scope and transition are checked; what a government withdraws is its own share, institutional conferral, embedding and backing withdrawn together and in public, with no shell left behind; a public announcement neither acquires legitimacy by itself nor proves by itself that the right has been outright withdrawn, and whether the horizontal constitutive sources remain is not for the government to declare; the floor of basic rights, accounts already incurred, and other independently standing sources and commitments do not vanish because a program is renamed or ended. What is held to is these three ways, not the continued existence of every embedded position ("holding to" is taken in the wider sense; see the Prequel, §8). If any of the three is missing, the government's power at that place is illegitimate and must be questioned; where lapses have become the normal state, its power must be questioned as a whole; whatever of this is verified as crossing the line is read, in addition, under the first ought.

§4 A Guarantee Is Not an Endorsement

In Apple City's old quarter stands the Stone Bridge. Long ago the City Council granted a private company a fifty-year toll franchise on the bridge. The franchise was embedded in an ordinance, toll collection has operated as usual, and the revenue long ago far exceeded the cost of building the bridge. In Year 7, some residents cross the bridge without paying, and the company asks the City Police Department to help collect the tolls.

Three things first. Whom it concerns: the company; the residents who cross the bridge. Which line: within the line. Crossing the bridge is an action that may be limited; neither collecting tolls nor crossing without paying crosses anyone's line; what is involved is an embedded recognition-based position, and the compulsion of helping to collect the tolls. Baseline: a public bridge with no franchise.

The company's reasoning can be written like this: the franchise is a recognition-based right embedded by ordinance; by §1, a government ought to guarantee the recognition-based rights it has embedded; so the City Police Department should come.

Of this inference, the first step is only half right: the franchise is a recognition-based position embedded by ordinance; the company is a construct, and according to On Accountability · Wei, "A company does not hold a recognition-based right in its own name"; on this position it serves as the address of attribution (On Accountability F42). The ought of §1, too, speaks of the recognition-based rights a government has embedded for its internal subjects. How holding is to be broken down further this paper does not unfold here; even if the holding behind the position is counted in full, the second step leaves something out: that a position is embedded and operating does not mean that it ought to be kept operating by the government; whether it ought to be must be checked separately. Nor does the third step follow: between "ought to guarantee" and "the City Police Department should come" there still lies the proof-task chain of compulsion.

When Rights Theory establishes how recognition-based rights operate, it adds one sentence at the end: "Neither constitution nor operation succeeding is the same as normative legitimacy." (Rights Theory P3 F10) That a franchise has been constituted and is operating says what it is in structure; it does not say whether it ought to exist, still less what the government ought to use to keep it going. §1 said: without the guarantee, the power is illegitimate; this section says the other side: with the guarantee, the power is not thereby legitimate. Read together, the guarantee is a necessary condition, not a sufficient one.

So the scope of the guarantee is itself subject to normative review. F10 says only that structure issues no legitimacy; the requirement of review is a step this series takes from there (own undertaking). What the review governs must be stated precisely. The three things of §3 are ways, and they are kept for every embedded position: no compulsion through a degenerated position, no empty shell in place of withdrawal, and withdrawal made in public and open to questioning. Scope review governs something else: whether a position ought to continue, whether the government ought to deliver its operation for it, and whether that delivery uses compulsion. If the review concludes in withdrawal, the withdrawal follows the third thing; if not, and the operation that ought to be delivered has fallen through, leaving only the ordinance hanging, only then is it an empty shell (§3). The police not coming to help with this one round of toll collection does not by itself show that the toll position has become an empty shell: toll collection is operating as usual, and the company's request is another matter, which must pass separately through the chain of Paper 2.

The review checks against that sentence of Institution Theory: "continuation and dismantling are both checked against the persons protected, the task, the authority, the cost and existing commitments" (Z38). Applied to the Stone Bridge: whom does it protect, the company recovering the cost of building the bridge, or the people who cross having a well-kept bridge; what is its task, and, with revenue long since far exceeding the cost, is that task still there; on whom does the cost fall, on everyone who crosses the bridge; what are the existing commitments: the fifty-year franchise is itself a commitment, and must be checked as such. Institution Theory also says that the construct's own continued existence "is not an independent sufficient reason to close off questioning, or to continue a particular arrangement automatically" (Z38). That an arrangement has existed for a long time is not in itself a reason for its continuation (this step is undertaken by this series; concurring: Z22 says that a temporary structure of guarding "is not renewed automatically on the strength of its origin").

This paper does not judge whether the franchise should be withdrawn or kept. It judges one thing: Apple City cannot answer the company with "it is embedded, it is operating." To withdraw it, the road is the third thing of §3, withdrawal or amendment made in public and open to questioning, with existing commitments among the matters to be checked; to keep it, keeping it must be explained too.

There is one more layer. When the City Police Department helps collect the tolls, it is enforcing this embedded position. Enforcement lands on the residents who cross the bridge without paying and converges their action; at this point the right this position carries shows its power-face. The Abstract of Rights Theory P4, stating F15, says that a running recognition-based right shows a power-face when it actually converges a recipient, "conditioned on convergence, not a standing state of any running right." Once the power-face shows, the City Police Department's move is compulsion, and Paper 2 applies: compulsion itself must be a limit (G13); enforcing an arrangement not made to block line-crossing must be retraceable to a negative root and must complete the proof-task chain (G11). That a franchise is embedded does not wave this compulsion through.

G22 A guarantee is not an endorsement. Neither constitution nor operation succeeding is the same as normative legitimacy; the scope of the guarantee is itself subject to normative review, which flows back into G19 (own undertaking); long existence is not in itself a reason for continuation (own undertaking). Without the guarantee, the power is illegitimate; with it, the power is not thereby legitimate: the guarantee is a necessary condition, not a sufficient one. Enforcing a recognition-based right shows a power-face when it actually converges a recipient, and Paper 2, G13 then applies.

§5 Welfare and Entitlements

The subsidy is a supply. The supplies a government gives out are not of this one kind only, and they do not all sit in the same slot. There are three slots.

What is embedded. The heating subsidy is in this slot. If continuing supply is to become something claimable, it is constituted as an embedded position at the recognition layer (§1). So it is guaranteed as a recognition-based right, under the three things of §3. It can be withdrawn only by the third thing of §3; and the withdrawal itself also passes through the first ought (Institution Theory Z38). When such a position fails, Rights Theory assigns it to the recognition layer: "The failure of an institutionally embedded position of supply belongs to the recognition layer's five types" (Rights Theory P5 F20). Whether it is outright withdrawal, hollowing-out or carrier migration depends on where it breaks (§3).

What is not embedded. Suppose that for the past five years the mayor has, every winter, issued "fuel vouchers" by a temporary decision, not written into any ordinance and not embedded in any other form either (a given of this branch). In the winter of Year 7, the mayor says that this year none will be issued.

Three things first. Whom it concerns: the households that received fuel vouchers in past years. Which line: stopping a supply does not in itself cross anyone's line; if someone had earlier closed off other ways, that is another matter (§3, the second reading); whether it presses on some household's life-substrate is checked separately. Baseline: the means of heating each household had before Apple City began issuing the vouchers.

This branch stipulates that they are not embedded, so the fuel vouchers are not in the embedded slot. They are an entitlement, or a condition in the sense of Institution Theory's third ought, that is, a condition Apple City provides, according to relational thickness, for the cultivation of those inside it (Z23). But not being embedded does not automatically mean that they can be stopped at will. Institution Theory says that continuation and dismantling are checked against the persons protected, the task, the authority, the cost and existing commitments, and goes on: "beyond that, existing rights, commitments and dependencies are checked as in the preceding sentence" (Z38). Conditions already undertaken, existing dependencies, and other sources must all be checked separately.

Bare entitlement. On Rights and Entitlements says that the granting party may cease to give, and that in the slot of bare entitlement ceasing to give is not violation, "but only so far as the granting end has not first exclusively monopolized, manufactured dependence, and sealed off alternative routes"; it goes on: "a giving already loaded through promise, institution, trust, or guardianship is no longer bare entitlement" (On Rights and Entitlements F30). The first sentence draws a limit around ceasing to give; the second moves a large tract of supply out of this slot.

The fuel vouchers must be gone through item by item. Did the mayor promise to issue them every year from then on? This branch stipulates no such promise; had there been one, they would no longer be bare entitlement. Did Apple City create dependence? Five years of issuing: if households had come to rely on them to get through the winter, that is factual dependence; factual dependence is not created dependence, and what has to be seen is whether Apple City first closed off people's existing means of heating. Whether there was monopolization, whether substitute circuits were sealed off, must likewise be checked down to whether there was an exclusive laying of hands. If all of these are passed, stopping the vouchers in the slot of bare entitlement is not read as violation. Not being violation, it must still be explained: Apple City's responsibility for conditions toward these households goes by how far it actually organizes their lives; as Institution Theory puts it, "For what the construct has organized, taken up or restricted, it must correspondingly explain how room is left and provided" (Z25). And in the same winter the subsidy is stuck at the Housing Department as well; where the two together press on some household is checked separately as in §2.

Carried over to government, this slot needs its range written out in full. In On Rights and Entitlements the granting end is originally a subject's own giving: "Entitlement is anchored in the giving of the granting end" (On Rights and Entitlements F27). Ceasing to give is not violation because a supplier's withdrawing its own supply "is disposing of one's own exercise under one's own law" (Rights Theory P5 §II). Apple City, as a construct, gives through positions such as the mayor and the Housing Department (Institution Theory Z7a); a construct has no law of its own (the Institution Theory Prequel, §2), and so it has no such exercise of its own to dispose of. So what is carried over is only the half that concerns the reading: in this slot, ceasing to give is not read as violation; whether it may cease, and how, goes back to the third ought and to the explanation of Z25 (extension; own undertaking).

Supply under guardianship. Paper 2, §6 told of a nine-year-old child taken from his home; the family division of the City Court appointed the Department of Children and Families as his temporary guardian. During the guardianship, his housing, meals and school fees are paid from the Department's funds. On Rights and Entitlements divides being reared into two faces: "Being reared divides, by whether the vector of action is converged, into two faces—the supply-face landing at entitlement, the converging-face at a delegated-holding recognition-based right" (On Rights and Entitlements F34). The converging face, the Department giving medical consent and handling school enrollment for the child, falls under substitute exercise in Paper 2 (G12a); housing, meals and school fees are the supply face, and belong in this section.

The supply face lands in the slot of entitlement, but it is not bare entitlement: it is embedded through guardianship. By the second sentence of F30, this giving is no longer bare entitlement, and the Department cannot stop the child's meals on the ground that "an entitlement can be stopped"; if it is to be withdrawn, it is placed by Rights Theory P3, P4 and P5 (On Rights and Entitlements F30); enforcement and substitution run on power backing (Rights Theory P5 §II). That sentence of Rights Theory originally speaks of the withdrawal of the holder of a position of guardianship or provision: "Where the position of care or provision is institutionally embedded — guardianship, maintenance — its holder's withdrawal is caught by the recognition layer: the position's failure reads on the five types, and enforcement and substitution run on power backing." (Rights Theory P5 §II) If here it is the Department itself that withdraws, the subject term has changed, and the recognition layer and the power backing that would catch it are probably within the government as well; this must be checked separately. F34 originally speaks of infants; the extension to a nine-year-old child under government guardianship has already been marked in Paper 2, G12a, and here only its division into two faces is borrowed (extension; own undertaking).

So whether a supply is in the slot of bare entitlement cannot be settled by its name; it must be tested item by item.

With Marshall. T. H. Marshall divides citizenship into three elements, the third being the social element: "By the social element I mean the whole range from the right to a modicum of economic welfare and security to the right to share to the full in the social heritage and to live the life of a civilised being according to the standards prevailing in the society" (Citizenship and Social Class, Part II, p. 8, per the excerpt of the 1992 Pluto edition; tier B). The view of citizenship that takes this essay as its principal reference treats citizenship first of all as a legal status, giving the members of a political community the same set of civil, political and social rights (per SEP, "Citizenship," §2.1; tier C). Both place welfare within the structure of rights, not as a favor; here only one point is contrasted: on this line of reading, social rights are drawn by membership and attach to citizenship; the two kinds of guarantee of this series are addressed to internal subjects, and the interior is drawn by entanglement, not by identity (the Prequel, §7). Who is a member belongs to the Theory of the State.

G21 Welfare and entitlements. An embedded position of supply is guaranteed as a recognition-based right; it can be withdrawn only as G19 provides, and the withdrawal itself passes through the first ought (Institution Theory Z38). What is not embedded is an entitlement, or a condition in the sense of the third ought; not being embedded does not automatically mean that giving may be ceased at will, and conditions already undertaken and other sources are checked separately. In the slot of bare entitlement, ceasing to give is not read as violation, so long as the granting end has not first exclusively monopolized, created dependence, or sealed off substitute circuits; not being violation does not exempt it from explanation. Once a giving is embedded through promise, institution, trust or guardianship, it is no longer bare entitlement. Whether a given supply is in the slot of bare entitlement is tested item by item. Reading the granting end onto a government's supply, and F34 onto a child under government guardianship, are both extensions (extension; own undertaking).

§6 Administration

The twelve hundred applications piled up there have their root in one post. The subsidy is delivered through the assessment post; the post standing empty is a matter of administration.

Administration is often seen as the tail end of carrying out rules: once the rules are set, one simply follows them. Rights Theory, speaking of the third leg, places it nearer. It says of real power backing: "It is not only the prevention and repair that follow a violation, but also everyday performance, the allocation of resources, access to procedures, and the honouring of qualifications." (Rights Theory P3 §I) Filling a post, ordering the assessments, making a decision within thirty days, paying the money out monthly: all of these are delivery. Whether a recognition-based right has gone empty often depends not on some single great violation, but on whether anyone is doing these everyday things.

This paper gives administration two roots.

One is making embedded rights really deliverable. This continues §3: "no empty shell," brought down to the everyday, is administration.

The other is providing conditions for cultivation according to relational thickness. Institution Theory's third ought says: "An institution provides conditions for the cultivation of its internal subjects according to relational thickness; it is concerned with conditions, not with outcomes." (Z23) Apple City opening a community arts center and paying for its funding is administration on this root (§8).

Neither root comes from Rights Theory. Rights Theory only tells us which everyday things delivery includes; why administration must do these things comes, for the first root, from the guarantee of §3, and for the second, from the third ought.

The two roots ground only administration that uses no compulsion. Assessing an application, filling a post, paying out money, opening an arts center: none of these uses compulsion. If administration is to use compulsion, for instance to verify a household's declaration by compulsory means, that step passes separately through Paper 2, G11: where a rule not made to block line-crossing is enforced, its compulsion must be retraceable to a negative root; compulsion that cannot be retraced does not come to stand by being renamed administration; that an administrative aim has grounds does not on its own give the means of compulsion grounds.

There is one more matter of accounts. When the person at the counter says "The assessment post is empty," that is the fact, not an evasion. The entry for the post standing empty is not at the counter; it is at the positions that set the staffing, provide the funding, and decide whether to fill the post (Paper 1, G4). What they must explain is exactly the list given in §3. Assessing an application is also a judgment, and what Paper 1 says about answering applies here as well; assessment can be done by a person, or handed to a tool (Paper 1, §9); the form is not fixed, but delivery must be there.

G20 The roots of administration. Real power backing includes everyday performance, the allocation of resources, access to procedures, and the delivery of qualifications. Administration has at least two roots: making embedded rights really deliverable (G19); providing conditions for cultivation according to relational thickness (Institution Theory Z23). These two roots ground only administration that uses no compulsion; where administration uses compulsion, it passes through Paper 2, G11, and that an administrative aim has grounds does not thereby give the means of compulsion grounds.

§7 Public Finance and Tax

The grocery store in Paper 2, in arrears on its taxes, had its account frozen with an amount left for basic living. Property tax is the main source of Apple City's general budget; the heating subsidy, too, is funded from this general budget. The store owner asks: "What did the taxes I paid get me?"

This question can be read in two ways.

One reads it as a price: so much paid, so much got back; if nothing comes back, one should pay less or not at all.

The other reads it as a demand for an explanation: where the money comes from, on whom it presses, where it is spent, and who bears the shortfall.

This paper takes the latter, and must also make clear why the latter is not simply the former.

The root of tax. On what grounds does Apple City collect taxes? This paper's answer: the root of tax and funding is the funding of the functions required to undertake the five oughts and the two kinds of guarantee. Someone has to sit in the assessment post, someone has to serve as review officer, someone has to answer the hotline, the courts have to open their doors, someone has to go and enforce protection orders; the five oughts and the two kinds of guarantee come down to functions, functions come down to positions, and positions need funding. How Is Institution Possible (Foundational Paper 6; cited as F6), in deriving the institutional increments at the nation-state tier, lists the same thing: "fiscal institutions (standing bodies require funding)" (F6 §3.1b). There it is an institutional increment of one tier, hung on a change in conditions; here the root of tax is hung on the functions that the oughts and the guarantees require, and this step is undertaken by this series.

Where the root is, the explanation follows (Paper 1, G2): on whom the burden of raising funds falls, that is, on which people property tax presses; where the funding is spent, that is, how much of the general budget reached the subsidy and how much the assessment post; who bears the shortfall, that is, where the shortfall fell during the three years the assessment post stood empty. If no position comes forward to explain the last question, the answer falls of itself: on the households that are waiting. A shortfall that no one explains must itself be questioned.

How the compulsion of tax is retraced. Freezing an account in order to collect tax must be retraced through tax's own root (Paper 2, §4). Law Series P1 requires that the legitimacy of affirmative provisions be retraceable to a negative root: "the prevention of suppression, the safeguarding of channels for questioning, or the maintenance of co-constructive space" (Law Series P1 BL3). The functions the funding supplies connect to these one by one: review officers, the hotline and the courts safeguard the channels for questioning; protection orders that someone enforces prevent suppression; embedded positions that can be delivered maintain the positions people already hold in common. Taxation, and freezing in order to collect tax, are retraced by way of these functions (as in Paper 2, G11; extension; own undertaking). The part levied for conditions on the side of striving for, such as the arts center's funding, can only be read as the maintenance of co-constructive space, and providing conditions goes one step further than "maintenance" (extension; own undertaking); whether this reading can be followed to the end, this paper does not claim to have answered. Until the retracing is complete, this series gives no warrant for the qualification of this part of compulsion; it is left as a remainder (Paper 6, §8, the first of the six slots). Compulsion of this kind, whose retracing reads only as far as "the maintenance of co-constructive space" and is not complete, occurs at three places in this series, and this is one of them; the other two are in Paper 2: the reading of the land requisition in §4, and the candidate for compensation in §5. The three are registered together in the first slot.

A sentence not adopted. F6's Appendix B, in a fictional constitution, has this sentence: "Taxation is the construct's legitimate chiseling of individual property; its legitimacy derives from the construct providing the individual with what the individual cannot obtain alone." (F6, Appendix B, Book Five) This sentence is not adopted, for two reasons. First, that paper itself says that the appendices are "not evidentiary for this version" (F6, the introduction to the Appendices). Second, the reason it gives is of the benefit kind: on it, the legitimacy of tax hangs on what the construct has given, and the store owner's question then easily reads as a price. This series does not borrow it as a sufficient argument for particular fund-raising and compulsory apportionment. The benefit rationale and a universal quid pro quo are not the same proposition either: what the proposition below rules out is an automatic law of exchange, not every particular charge arranged by benefit.

Paying taxes is not a universal quid pro quo. On Accountability has a proposition: "A recognition-based right is not got in exchange for recognizing a power." It goes on: "The generation, the continuance, and the termination of a chain's accountability are judged by that chain's own content and conditions; an accountability is not generated automatically because the other chain has been performed, nor released automatically because the other chain has not." (On Accountability F43)

Between Apple City and the store owner there are two things running in opposite directions. In one, Apple City collects taxes from him through the Tax Office, and in that slot he holds the accountability to pay tax. In the other, the embedded positions he holds in Apple City call on Apple City for delivery, and Apple City owes them the guarantee. The first sentence of F43 blocks joining the two into one bargain: the guarantee is not got in exchange for paying taxes, and paying taxes is not the price of buying the guarantee. Its second sentence blocks two automatic steps.

The store owner's having paid taxes does not automatically generate a claim of his to some particular benefit; Apple City's having let the subsidy go empty does not automatically release him from his accountability to pay tax. Conversely, his owing taxes does not automatically cancel the positions he already holds elsewhere; those positions are checked by their own sources and conditions. The amount for living left in the frozen account is not part of this account: what it governs is that the freeze not compress this household's life-substrate to the point of threat, which is a matter for the first ought (Paper 2, Introduction), not an item on another chain.

F43 originally speaks of two chains of accountability running in opposite directions between two nodes, and its example is a company and an employee. While the positions the store owner holds are operating, their power-face converges Apple City, and Apple City holds accountability in that slot (On Accountability F39, F42); F43 applies as it stands. Where a position has been hollowed out and that chain is no longer there, the owed guarantee remains (§1); only then is applying F43 to the side of the "ought" a step of this series (extension; own undertaking), and its ground is that what F43 blocks is the "automatic": whether what stands on the other side is a chain or an ought, paying taxes does not automatically get anything in exchange, nor automatically release anything.

The guarantee of the subsidy does not come from taxes either. Whether the households that applied paid taxes, and how much, is not a condition of this guarantee; the guarantee comes from the ordinance's embedding and from the ought this series has established (§1).

F43 keeps one more sentence: "Whether a particular arrangement makes some fact a condition of application is judged separately, by that arrangement." A fee charged per use, a service that states payment as its condition: these are judged by their own conditions. So paying taxes is not a universal quid pro quo; and particular claims that already stand are not thereby cancelled.

G20a Public finance. The root of tax and funding is the funding of the functions required to undertake the five oughts and the two kinds of guarantee (own undertaking). On whom the burden of raising funds falls, where the funding is spent, and who bears the shortfall must be explained, and are open to questioning. The compulsion of taxation passes through Paper 2, G11, and is retraced to a negative root by way of the functions it funds (extension; own undertaking); the part levied for conditions on the side of striving for can only be read as the maintenance of co-constructive space; until the retracing is complete, this series gives no warrant for the qualification of this part of compulsion; it is left as a remainder (Paper 6, §8, the first of the six slots), registered together with the two places in Paper 2, land requisition and compensation. Paying taxes is not a universal quid pro quo (G23). The benefit rationale (F6, Appendix B, Book Five) is not adopted as a sufficient argument for particular fund-raising and compulsory apportionment.

G23 Paying taxes is not a universal quid pro quo. A recognition-based right is not got in exchange for recognizing a power; the generation, the continuance and the termination of a chain's accountability are judged by that chain's own content and conditions; an accountability is not generated automatically because the other chain has been performed, nor released automatically because the other chain has not. Whether a particular arrangement makes some fact a condition of application is judged separately, by that arrangement. So paying taxes is not a universal quid pro quo; nor are particular claims that already stand thereby cancelled (extension; own undertaking).

§8 When Conditions Are Limited

In the winter of Year 7, Apple City's revenue falls, and the City Council discusses cutting next year's budget by thirty percent. There are three proposals on the table:

1. Abolish the Housing Department's complaints desk, and merge complaints about the subsidy into the Mayor's Office complaint hotline;
2. Suspend new applications for the heating subsidy for one year;
3. Cut the community arts center's funding by half.

Three things first. Whom it concerns: the people each item touches, that is, complainants, applicants for the subsidy, and users of the arts center. Which line: none of the three directly crosses anyone's line. The first touches the channel of being questioned under the fifth ought; the second touches an embedded recognition-based position; the third is a condition on the side of striving for. Baseline: last year's budget.

With less money, something always has to be cut. This section asks: what cannot be cut.

What cannot be cut. What cannot be cut is the oughts and the necessary channels, not any particular old post, program or budget figure. Institution Theory says: "structures can change, but the three conditions cannot be withdrawn" (Z37; it originally speaks of the three conditions of being questioned; the other items follow the grouping under "holding to" in the Prequel, §8). The items on the side of holding to are not discounted for lack of resources: the first and second oughts, basic rights not being violated included; the half of the fourth ought that does not seal the channels of refusing, revising, exiting and changing direction; the channel of being questioned under the fifth ought; the ways of guaranteeing recognition-based rights (§3); the limits of substitute exercise (Paper 2, G12a). This last item is listed separately because one of these limits, that of being oriented toward return, originally lies upstream in the assistance under the third ought (Institution Theory Z26); extended to substitute exercise, it is a limit of compulsion, which falls under the first ought by way of Paper 2, G13, and is not a condition on the side of striving for that can be prioritized; with less money, it is not pushed back either.

One qualification here must be stated precisely. Institution Theory Paper 1, §8 says: "This series grants that in practice an ought may not be achieved, and does not on that account withdraw or discount any of them." This sentence governs all five oughts, without dividing holding to from striving for. So the oughts on the side of striving for are not discounted either; with less money, what can be prioritized is the conditions on the side of striving for: whether they come earlier or later, whether more or less is given, not the ought itself. Dividing the five oughts into two sides, one not discounted for lack of resources and one whose conditions can be prioritized, is undertaken by this series (the Prequel, §8).

Item by item.

The first: merging the complaints desk into the hotline. The desk is a structure, and structures can change. The three conditions cannot be withdrawn: the channel must meet the conditions that "it is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation" (Institution Theory Z34). What must be checked is how things stand after the merger. The hotline is in the Mayor's Office, which is not the same party as the Housing Department; independence is functional, and "it is checked against the party actually questioned" (Z34). In this case the party questioned is the Housing Department, and also the positions that set the staffing and provide the funding (§6); if those positions are in the Mayor's Office, the hotline is not independent of them, and there must be another route within reach. Does the hotline accept complaints about the subsidy; once it accepts them, can it make the party questioned answer; if the answer misses the point, where can the complaint be sent next: if these check out, this item is a change of structure; if they do not, complaints about the subsidy have nowhere to be sent at Apple City's level (the Office of the State Ombudsman in the outer layer is still there and within reach, but cannot carry the day-to-day check; Paper 4, §3), and what is cut is not a desk but this level's channel under the fifth ought. Whether arrangements of the desk's kind may fall into the hands of the party questioned is taken up again in Paper 6, where it treats capture (G38).

The second: suspending new applications for the subsidy for one year. This touches an embedded recognition-based position, so it can be done only by the third thing of §3, withdrawal or amendment made in public and open to questioning. Authority: the ordinance was enacted by the City Council, and suspending part of it is within the City Council's authority. Grounds: falling revenue is a ground; whether the subsidy's original ground, heating for low-income households in winter, is still there must also be stated. Scope: only new applications are suspended; the twelve hundred already filed are not within the suspension; they are old accounts (§3). Transition: there must be an arrangement for households that newly meet the conditions during the year. The suspension itself must also pass through the first ought (Institution Theory Z38): has Apple City first closed off anyone's existing way? One more point: if the suspension becomes another way of leaving the empty shell hanging, with the post still empty and the backlog still untouched, then it has not answered the thing called "no empty shell."

The third: halving the arts center's funding. This is a condition on the side of striving for: the third ought, Apple City providing conditions for cultivation according to relational thickness (Institution Theory Z23). It can be prioritized, and it can be cut. If some item within it has already been embedded or promised, that item is checked separately as in §5.

Prioritizing is itself a judgment. With the three items side by side, which is cut first, and by how much, is prioritization. Law Series P1 says that law carries a structural discount at the enforcement layer, one source of which is resource priority: "not every instance of suppression can be processed"; it goes on to say that prioritizing itself brings the one who prioritizes into it (Law Series P1, Boundary Three; restated). That passage speaks of law's discount at the enforcement layer; extending it to a government's prioritization of allocations is a step undertaken by this series (extension; own undertaking). The ground for the extension is in Paper 1: when Apple City sets the order of its budget, that too is persons sitting in positions making a judgment, and a judgment must answer (Paper 1, G1, G2).

So prioritization must be explained: it must state the matters that call for judgment, the subjects concerned, and how the responsibility for making up the shortfall is questioned. Halving the arts center's funding must explain what is cut, who thereby has less of what, who takes up this shortfall and where, and where those who were cut can ask.

Relational thickness is of use here, but only at one point. It delimits the scope of the responsibility for conditions: as far as Apple City actually organizes a person's life, so far it owes him corresponding conditions and explanation (Institution Theory Z23, Z25, Z25a). It does not by itself set the order of all matters. Between the arts center and other matters on the side of striving for, which comes first, thickness does not do the judgment in its place.

Who makes up the shortfall below the threshold. Institution Theory Paper 0, in conversing with Sen and Nussbaum, left a question: "when a person falls below some threshold, which construct is to make it up, and by what division, is a question of distribution and design"; it handed this question to specific theories of institutions and to the Theory of Government (Institution Theory Paper 0, §5.1). This paper answers it as far as structure.

First, an embedded position is guaranteed by the level that embedded it. The subsidy was embedded by Apple City's ordinance; Apple City guarantees it, keeping the three things of §3.

Second, for conditions on the side of striving for, each construct owes corresponding conditions and explanation according to how far it actually organizes this person's life (Institution Theory Z23, Z25, Z25a). How each level's relational thickness is read is written in Paper 4 (G29); how the levels share the burden belongs to specific theories of institutions.

Third, the responsibility for making up the shortfall must be open to questioning. Who bears the shortfall must be written down by some position (§7); if no position writes it down, the shortfall falls on those who are waiting, and this itself must be questioned.

To whom in particular, how much, and how the levels share it, belongs to specific theories of institutions. This paper does not set up a separate scheme of distributive justice.

With Sen and Nussbaum, and with Rawls. Institution Theory Paper 0 has already conversed with the capability approach (§5.1): Nussbaum's thresholds are addressed to the state, and how the burden of realizing them is broken down was registered there, as it stands, as unanswered. This section takes up exactly that question, but only as far as structure: it gives no list and no threshold. With Rawls's basic structure, too, Institution Theory Paper 0 has already conversed (§3.2): there it says that this series does not distribute, and that "distributive justice is an adjacent jurisdiction." This section does not change that sentence; it takes up only the question of who makes up the shortfall below the threshold.

Where it truly cannot be achieved, for instance where a thirty percent shortfall cannot be filled however things are prioritized, Paper 5, G33 applies: admit that it cannot be achieved; do not withdraw or discount the ought; the transition must be explained; responsibility is attributed by Institution Theory Z6.

G24 When conditions are limited. The side of holding to is not discounted for lack of resources: the first and second oughts (basic rights not being violated included), the "not sealing the channels" half of the fourth ought, the channel of being questioned under the fifth ought, the ways of guaranteeing recognition-based rights, and the limits of substitute exercise (Paper 2, G12a, by way of G13); what cannot be cut is the oughts and the necessary channels, not any particular old post, program or budget figure. Institution Theory's sentence on not discounting governs all five oughts; placing the conditions on the side of striving for on the side that can be prioritized is undertaken by this series. On the side of striving for, relational thickness delimits the scope of the responsibility for conditions; relational thickness does not by itself set the order of all matters. Prioritizing is itself a judgment (extension; own undertaking), and must explain the matters that call for judgment, the subjects concerned, and how the responsibility for making up the shortfall is questioned. Who makes up the shortfall below the threshold is answered as far as structure.

Where This Paper Stops

This paper has proved only these things: the two kinds of rights must be kept apart; that basic rights may not be violated holds against everyone and is not peculiar to government; guaranteeing the recognition-based rights embedded for internal subjects is an undertaking peculiar to government; without the guarantee, the power must be questioned, and with it, the power is not thereby legitimate; what a government gives basic rights is backing and enforcement, not the core; for embedded rights a government keeps three things, no de-righting, no empty shell, and withdrawal or amendment made in public and open to questioning; what a government withdraws is its own share, and what is held to is the ways, not the continued existence of every position; the scope of the guarantee is itself subject to review, and the three things are kept for every embedded position; supply falls into three slots, and in the slot of bare entitlement ceasing to give is not read as violation, within the limits of F30, while what has been promised or embedded is not in that slot; administration has two roots, which ground only administration that uses no compulsion; the root of tax lies in the functions the oughts and the guarantees require, the compulsion of taxation is retraced by way of these functions, and paying taxes is not a universal quid pro quo; when conditions are limited, what cannot be cut is the oughts and the channels, prioritizing is a judgment and must be explained, and who makes up the shortfall below the threshold is answered as far as structure.

What it has not proved: distributive justice, which this series does not set up separately; what in particular is to be given, how, and how the levels share it, which belongs to specific theories of institutions; whether the retracing of taxation levied for conditions on the side of striving for can be followed to the end, which has not been fully answered; until the retracing is complete, this series gives no warrant for the qualification of this part of compulsion; it is left as a remainder (Paper 6, §8, the first of the six slots; registered together with the two places in Paper 2, land requisition and compensation); on what grounds compulsion such as taxation, freezing and help with toll collection stands, which is written in Paper 2; whether the complaints channel may fall into the hands of the party questioned, which is written in Paper 6 (G38); how to make the transition when it cannot be achieved, which is written in Paper 5 (G33); how each level's relational thickness is read, which is written in Paper 4 (G29); who is a member, which belongs to the Theory of the State.

Someone will still come, later, to ask about those twelve hundred applications.

The Nine Propositions of This Paper

G17 Guaranteeing the two kinds of rights: the two halves kept apart. That basic rights may not be violated is derived from the subject's side and holds against everyone; it is not peculiar to government; the ought addressed to the construct already stands in Institution Theory's first and second oughts (Z9, Z15, Z17), and the Theory of Government takes it over and does not establish it. Guaranteeing recognition-based rights is an undertaking peculiar to government, established by this series as a normative undertaking newly added on the government's side: a government ought to guarantee the recognition-based rights it has embedded for its internal subjects; without the guarantee, its power at that place is illegitimate, that is, it must be questioned; Rights Theory gives the object and the criteria of failure, and does not by itself issue this ought. "Peculiar" means that in this series this undertaking is established only for government; it does not assert that other constructs owe nothing at all. It adds no sixth heading, and is organized into the first, second and fifth oughts (G19). The guarantee of both kinds of rights can also be understood as part of anti-colonization, and "colonization" is still taken in the narrow sense. "Open to questioning" holds for every government and at every time; "must be questioned" means that a verified lapse owes an explanation and a correction; the two are not mixed.

G18 The guarantee of basic rights itself lies on the recognition side. What a government gives is backing and enforcement, not the core: the inviolability of the core comes from the subject's side, and basic rights are not newly made by the institution. Backing can be hollowed out; the core does not move. Backing and enforcement lie on the recognition side and belong to the guarantee of G19: where the backing of a basic right is hollowed out, it goes under the second ought; where a government does not guarantee it, its power is likewise illegitimate and must be questioned.

G19 The guarantee of recognition-based rights: three things. A recognition-based right operates on three legs: institutional embedding, effective community recognition and real power backing. For embedded rights, a government keeps three things. First, no de-righting, which falls under the first ought: in de-righting, compulsion persists, and the government goes on acting through a position that has degenerated into sheer force; the three tests serve only as criteria, not as norms. Second, no empty shell: where the backing of a basic right is hollowed out, it goes under the second ought; where the pre-existing self-maintained or substitute circuits are first actively sealed off, so that people can depend only on that supply, and the position is then hollowed out, it goes under the first ought (extension; own undertaking), and no factual dependence settles a case directly under the first ought; any other gap in operation enters the explanation and correction of the fifth ought, and the explanation must state what has not been delivered, whether the original ground still holds, and the arrangements for restoring operation, for transition, or for a grounded withdrawal or amendment. Explaining is not repairing, and permitting questioning is not having fulfilled the guarantee; "no empty shell" does not shrink into "no hidden empty shell"; whether the first or second ought is also touched is checked separately; where it truly cannot be achieved, Paper 5, G33 applies. Third, withdrawal or amendment made in public and open to questioning, which falls under the fifth ought: authority, grounds, scope and transition are checked; what a government withdraws is its own share, institutional conferral, embedding and backing withdrawn together and in public, with no shell left behind; a public announcement neither acquires legitimacy by itself nor proves by itself that the right has been outright withdrawn, and whether the horizontal constitutive sources remain is not for the government to declare; the floor of basic rights, accounts already incurred, and other independently standing sources and commitments do not vanish because a program is renamed or ended. What is held to is these three ways, not the continued existence of every embedded position ("holding to" is taken in the wider sense; see the Prequel, §8). If any of the three is missing, the government's power at that place is illegitimate and must be questioned; where lapses have become the normal state, its power must be questioned as a whole; whatever of this is verified as crossing the line is read, in addition, under the first ought.

G20 The roots of administration. Real power backing includes everyday performance, the allocation of resources, access to procedures, and the delivery of qualifications. Administration has at least two roots: making embedded rights really deliverable (G19); providing conditions for cultivation according to relational thickness (Institution Theory Z23). These two roots ground only administration that uses no compulsion; where administration uses compulsion, it passes through Paper 2, G11, and that an administrative aim has grounds does not thereby give the means of compulsion grounds.

G20a Public finance. The root of tax and funding is the funding of the functions required to undertake the five oughts and the two kinds of guarantee (own undertaking). On whom the burden of raising funds falls, where the funding is spent, and who bears the shortfall must be explained, and are open to questioning. The compulsion of taxation passes through Paper 2, G11, and is retraced to a negative root by way of the functions it funds (extension; own undertaking); the part levied for conditions on the side of striving for can only be read as the maintenance of co-constructive space; until the retracing is complete, this series gives no warrant for the qualification of this part of compulsion; it is left as a remainder (Paper 6, §8, the first of the six slots), registered together with the two places in Paper 2, land requisition and compensation. Paying taxes is not a universal quid pro quo (G23). The benefit rationale (F6, Appendix B, Book Five) is not adopted as a sufficient argument for particular fund-raising and compulsory apportionment.

G21 Welfare and entitlements. An embedded position of supply is guaranteed as a recognition-based right; it can be withdrawn only as G19 provides, and the withdrawal itself passes through the first ought (Institution Theory Z38). What is not embedded is an entitlement, or a condition in the sense of the third ought; not being embedded does not automatically mean that giving may be ceased at will, and conditions already undertaken and other sources are checked separately. In the slot of bare entitlement, ceasing to give is not read as violation, so long as the granting end has not first exclusively monopolized, created dependence, or sealed off substitute circuits; not being violation does not exempt it from explanation. Once a giving is embedded through promise, institution, trust or guardianship, it is no longer bare entitlement. Whether a given supply is in the slot of bare entitlement is tested item by item. Reading the granting end onto a government's supply, and F34 onto a child under government guardianship, are both extensions (extension; own undertaking).

G22 A guarantee is not an endorsement. Neither constitution nor operation succeeding is the same as normative legitimacy; the scope of the guarantee is itself subject to normative review, which flows back into G19 (own undertaking); long existence is not in itself a reason for continuation (own undertaking). Without the guarantee, the power is illegitimate; with it, the power is not thereby legitimate: the guarantee is a necessary condition, not a sufficient one. Enforcing a recognition-based right shows a power-face when it actually converges a recipient, and Paper 2, G13 then applies.

G23 Paying taxes is not a universal quid pro quo. A recognition-based right is not got in exchange for recognizing a power; the generation, the continuance and the termination of a chain's accountability are judged by that chain's own content and conditions; an accountability is not generated automatically because the other chain has been performed, nor released automatically because the other chain has not. Whether a particular arrangement makes some fact a condition of application is judged separately, by that arrangement. So paying taxes is not a universal quid pro quo; nor are particular claims that already stand thereby cancelled (extension; own undertaking).

G24 When conditions are limited. The side of holding to is not discounted for lack of resources: the first and second oughts (basic rights not being violated included), the "not sealing the channels" half of the fourth ought, the channel of being questioned under the fifth ought, the ways of guaranteeing recognition-based rights, and the limits of substitute exercise (Paper 2, G12a, by way of G13); what cannot be cut is the oughts and the necessary channels, not any particular old post, program or budget figure. Institution Theory's sentence on not discounting governs all five oughts; placing the conditions on the side of striving for on the side that can be prioritized is undertaken by this series. On the side of striving for, relational thickness delimits the scope of the responsibility for conditions; relational thickness does not by itself set the order of all matters. Prioritizing is itself a judgment (extension; own undertaking), and must explain the matters that call for judgment, the subjects concerned, and how the responsibility for making up the shortfall is questioned. Who makes up the shortfall below the threshold is answered as far as structure.

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.

Qin, H. (2026). What Rights Are Not (SAE Rights Theory, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20726894

Qin, H. (2026). On the Source and Spectrum of Rights (SAE Rights Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20743135

Qin, H. (2026). The Constitution and Withdrawal of Recognition-Based Rights (SAE Rights Theory, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20993658

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). On the Limitation and Violation of Basic Rights (SAE Rights Theory, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21212595

Qin, H. (2026). On Rights and Entitlements (SAE Rights Theory, Specialist Paper) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21314907

Qin, H. (2026). On Accountability · Wei (SAE Rights Theory, Specialist Paper) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22853810

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). What an Institution Is Not (SAE Institution Theory, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23269330

Qin, H. (2026). To Whom the Ought Is Addressed (SAE Institution Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23269635

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). Conditions (SAE Institution Theory, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270922

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

Qin, H. (2026). Institution Theory and the Tradition: The Address of the Ought (SAE Institution Theory, Paper 0) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23272468

Qin, H. (2026). One's Law Meets One's Law: The Structural Origin of Law from 14DD Showdown (SAE Law Series, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19548237

Marshall, T. H. (1992). Citizenship and Social Class. In T. H. Marshall & T. Bottomore, Citizenship and Social Class. London: Pluto Press. (First published 1950.) Excerpt of Part II: https://pages.nyu.edu/jackson/analysis.of.inequality/Readings/Marshall%20-%20Development%20of%20Citizenship.pdf

Leydet, D. Citizenship. In The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/citizenship/

Interfaces

• Main upstream: Rights Theory (P1 F2, §II; P3 F10, F11, §I, §II; P4, Abstract (F15); P5 F20, §II); On Rights and Entitlements F27, F30, F34; Institution Theory (Z9, Z15, Z17, Z22, Z23, Z25, Z25a, Z34, Z37, Z38; Paper 1, §8).
• Also used: the Rights Theory Prequel, §III; On Accountability F39, F42, F43; F6 §3.1b, and Appendix B, Book Five (a sentence not adopted, for contrast); Law Series P1 BL3 and Boundary Three; Institution Theory Z7a; Institution Theory Z26 (by way of Paper 2, G12a; §8); the Institution Theory Prequel, §2; Institution Theory Paper 0, §5.1.
• Interfacing: T. H. Marshall (a one-sentence contrast). Pointing back: Honneth (Institution Theory Paper 0, §6.1); Sen and Nussbaum (§5.1); Rawls (§3.2).
• This series: takes up the Prequel, §§5, 7 and 8; takes up Paper 1, G2, G4 and §9 (the algorithmic stage of the same subsidy); Paper 2, G11, G12a and G13 are pointed back to here, and the retracing of tax handed over from Paper 2, §4 is written in §7 (G20a); Paper 4, G29 takes up how each level's relational thickness is read; Paper 5, G33 takes up what cannot be achieved; Paper 6, G35 and G38 take up channels and capture, and the first of the six slots in Paper 6, §8 receives the remainder of taxation levied for conditions on the side of striving for, registered together with the two places in Paper 2, §§4 and 5.
• Handed to Paper 4: several levels of government have a hand in one and the same matter.

引言

第七年冬天，苹果城住房局的申请窗口照常开着。

苹果城的条例里载着一项低收入家庭供暖补助：家庭收入在一条线以下的，冬季每月可领一定数额，向住房局申请，三十日内决定。这一条是第一年写进去的。申请窗口一直开着，收件照常。可住房局负责审核的那个岗位，从第四年冬天出缺，此后一直没有补人。到这一年冬天，积压的申请有一千二百份，没人处理。

一位申请人入冬时递了申请，四十天后去窗口问。窗口的人说："收件归我们。审核的岗空着。"

条例还在，窗口还开着，申请照收。纸面上，这项补助什么都不缺。它只是不运行。

三样先写。对谁：已经申请、符合条件的家庭。在哪条线：空着一个岗位、积着一批申请，这本身不越谁的线；一项已经载入的承认性位置出现了运行缺口，照第三节的读法，这属于"不留空壳"的第三读，走第五条应当的说明与修正。冬季没有供暖是否压到了某一户的生命基底，另核（第二节）。基线：条例写明的三十日内决定。

这项补助，第一篇已经出现过。第二年到第四年，住房局把初审交给一家打分公司，那一段问的是判断交给机器以后，承担落在哪里（第一篇第九节）。这一篇不重写那一段。那里问谁作答；这里问，一项写进条例的东西，政府要保证它到什么程度，保证由什么组成，落空了怎样读，要撤要改又怎样受问。

第二篇写强制。强制之外，政府做得最多的是另一类事：审一份申请，发一笔补助，让一张保护令有人执行，替一项特许收费或者不替它收。这些事看上去不像强制，落在人身上的分量却常常不轻。一千二百份申请背后的那些家庭，这个冬天怎样过，压在一个空着的岗位上。

这一篇分八步。第一节把两类权利分开。第二节说基础权利那一半，政府给的是什么。第三节说承认性权利的保证由哪三件组成。第四节说保证不是背书。第五节给供给分格：已载入的，未载入的，裸权益。第六、七节说行政、财政与税。第八节说条件有限时，什么不能砍。体例照前置篇第八节；本篇不设篇末对话一节。全篇九条（G17 至 G24，含 G20a），末尾合列。

第一节　两类权利分开

关于那一千二百份申请，可以有两种说法。

一种说：补助是政府给的，给得了就给，给不了就是给不了，谈不上欠谁。

一种说：冬天有暖气，是每个人的基本权利；政府不给，就是侵犯。

两种说法各对一半，也各错一半：一种把补助当成恩惠，一种把它当成基础权利。

权利论把权利分成两类，走两根脊。一根是基础权利：「基础权利对偶道，源于自己的法，本体先在，预设克制式承认，不进 T5 接缝」（权利论第一篇 F2）；它预设的承认，「即他者克制、不侵犯，而不需要被谁构成」（同上 §II）。另一根是承认性权利：「承认性权利与权力同在 T5 接缝两侧」（同上 F2）；它「不是凭恩赐或许可而来，是由构成式承认成立的」（权利论第三篇 F10）。

基础权利不受侵犯那一半，不是政府特有的。 一个人不被奴役、身体不被伤害，不是政府给的，也不是任何人给的。它从这个人自己那边来，对谁都成立：对邻居，对一家厂，对政府，一样。制度论说：「基础权利不是制度新造的」（Z17）。不侵犯它，也不是政府独有的一条要求：一个邻居不得侵犯它，一家厂不得，政府也不得。对构的那一条应当，制度论已经立过：构的施为对每一个内部主体都不越线，不侵害基础权利属此（Z9）；构守住内部主体彼此之间的线，保障基础权利属此（Z15）。政府论承它，不立它，也不把它写成政府特有的承担。

保证承认性权利那一半，是政府特有的承担。 补助不是基础权利。权利论说得直：「持续供给若要成为可主张的东西，它成立为承认层的已载入位置，不成立为基础权利的内容。」（权利论第五篇 §II）递了申请的家庭可以主张补助，是因为苹果城把它写进了条例。

这样，两种说法对在哪里、错在哪里，就清楚了。第一种说对了来源：这笔补助是苹果城写进条例的，不是从各户自己那边长出来的。它错在后半句：补助一经载入，就不是政府随手给、随手不给的恩惠；它是一个位置，申请人可以凭它申请、追问、要求答复。第二种说对了一件事：每个人都有一条谁也不得越过的底线，生命基底在内。它错在把补助本身算进了那条底线：补助停了，不由此就是对基础权利的侵犯；另有没有侵犯，另核（第二节）。

这个系列在这里立一条：政府应当保证它为内部主体载入的承认性权利。不保证，政府的权力在那一处就不正当。

这一条不从权利论推出。权利论给的是结构：承认性权利与权力同在一条接缝的两侧，从施为一侧读出的是权力，从主体位置一侧读出的是承认性权利（权利论第一篇 §II；又见权利论前置篇第三节）。政府握着接缝的权力一侧。它若只握着权力，不让权利一侧运行，照权利论的读法，位置会坏掉：或者强制还在，网络却不再处理受位者的请求，「位置退化为强力」；或者支持撤走、外壳还在，「位置退为宣告」（权利论第三篇 F11，§II）。住房局这一件属后一种。结构给的是读法，不给应当；权利论自己说，它判脱权的那几问「判的是网络处理的结构，而不是把规范上的正当与否偷渡进来」（权利论第三篇 §II）。应当是这个系列立的：保证为内部主体载入的承认性权利，是政府侧新增的一项规范承担（前置篇第五节声明，这里立）。它不另增第六个总名，拆开看，组织进制度论的第一、二、五条应当（第三节）。"特有"说的是：这一承担在这个系列里只对政府立，由政府论来立；它不断言别的构对它们载入的位置一概不欠。制度论没有记它；别的构欠什么，归各自的具体制度论，这一篇不替它们立。

两类保证，也可以理解为反殖民的一部分：它们都在"坚持"一边，不在"争取"一边（前置篇第八节；制度论 Z15 说保障基础权利「在反殖民一边」，也是这个意思）。"殖民"仍取窄义，指伪装成涵育的越界（制度论第一篇第八节）。越线不都装成涵育，空壳也不是越线，它们分别照第一、二、五条读（第三节）。

"不正当"读作须被追问。 这一条说的"不正当"，在这个系列里有确定的落点：须被追问，即须说明，须修正，走第五条应当的通道（Z34）。范围也要说准。失守在哪一处，政府经那一处位置、在那一作用域的施为就不正当；住房局空着的审核岗，是这样一处。失守成了常态，一项又一项地掏空，或者脱权，政府的权力就整体须被追问。越线不走这一落点，另照第一条读。

这里要把两个词分开。"可被追问"说的是通道常开：对一切政府，一切时候，不论它有没有失守（Z34；第六篇 G35）。"须被追问"说的是一处已核的失守欠下了一份说明与修正。住房局这一处已经核得清：岗位空了三年，积压一千二百份申请，三十日的时限对这些申请一份也没有守住。它须被追问。苹果城的别的部门，此刻也许没有哪一处已核的失守，它们照样可被追问；政府永远做不到完美，受追问是常态（第一篇第八节）。两样不混：受问本身，不是不正当的证据；提供了受问的通道，也不是修复的凭证。

还有一处分寸先说在这里：保证了，权力也不因此就正当。保证是必要条件，不是充分条件（第四节）。

与霍耐特。 "承认性权利"里的"承认"，是权利论的结构词：一个位置经构成式承认在来源上成立，经三支咬合运行（权利论第三篇 F10）。它不是霍耐特（Axel Honneth）所说的承认。在霍耐特那里，承认分在爱、法律与社会评价三个领域，人在其中依次得到自信、自尊与自重（据制度论 Paper 0 第六节第一小节所据的斯坦福哲学百科《承认》条目，C 档）。法律那一维离这里最近；即便在那一维，霍耐特问的是法律上的承认怎样成就人的自尊，权利论问的是一个位置怎样在来源上成立，两题不是一题。制度与人与人之间的承认是什么关系，制度论 Paper 0 已经对过，"自尊与地位"一处在那里也是先分题、不判冲突（第六节第一小节）；这里指过去，只点明两处同名，所指不同。

G17　保证两类权利，两半分开。基础权利不受侵犯，从主体那边推出，对谁都成立，不是政府特有的；对构的应当已在制度论第一、二条（Z9、Z15、Z17），政府论承它，不立它。保证承认性权利是政府特有的承担，由这个系列立为政府侧新增的规范承担：政府应当保证它为内部主体载入的承认性权利，不保证，权力在那一处就不正当，即须被追问；权利论给对象与失效判据，不独自发出这一应当。"特有"指这一承担在这个系列里只对政府立，不断言别的构一概不欠。它不另增第六个总名，组织进第一、二、五条（G19）。两类保证也可理解为反殖民的一部分，"殖民"仍取窄义。"可被追问"对一切政府、一切时候；"须被追问"指一处已核的失守欠下一份说明与修正，两样不混。

第二节　基础权利的背靠

第二篇写过果园里小区的一件事：楼上的住户在楼道里堵住楼下的住户，说"再去找物业，就让你出不了这个门"。楼下的住户报了警，巡警到场把两人分开，并协助他向市法院申请临时保护令（第二篇第九节）。

楼下的住户不被威胁、不被堵在自家门口，这一条不是保护令给的，也不是市警局给的。保护令给它的，是一个写下来、可以援引的形态；市警局给它的，是一个会来执行的人。

权利论有一张二乘二的表，把这件事放得很清楚。按来源分，基础权利是内生的，承认性权利是注册的；按方向分，有否定与涌现。注册与否定相交的那一格，是对否定底线的制度兜底：宪法把禁奴役、禁酷刑写成不可触动的条款，是「给基础权利那条拒绝被化约的底线加一个注册式的背靠」（权利论第一篇 §II）。同一项权利，核在基础一侧，背靠在承认一侧。

背靠可以被掏空。权利论说，法上仍写着某项自由，而执行机构、程序与网络的承认都被抽走，「这时被掏空的是制度背靠，被压制的是这项基础权利的行使，未被动的是它的核」（权利论第三篇 §II）。

设若市法院签发了临时保护令，市警局却说人手不够，从不派人执行。保护令写在纸上，兑现不了。被掏空的是背靠；楼下的住户不被威胁这一条，一分也没有少。少的是它在苹果城的着落。

所以政府在这一半里给的，是背靠与执行，不是核。这一句要挡两种读法。

一种是把政府读成核的来源：政府不给，就没有。不对。核不动；「基础权利不是制度新造的」（Z17）。

另一种是倒过来：核既然不动，政府就没有什么可欠的。也不对。背靠与执行在承认侧，正归下一节的保证。制度论说保障是一组公共化的条件，「纸面上的保护不算保障；执行得了，也不证明执行的内容与权限正当」（Z17）。保护令签了没人执行，这一处走第二条应当：构守住内部主体彼此之间的线，守的办法不能只剩纸面。政府不保证这一层背靠，它的权力在那一处同样不正当，须被追问。

回到那些家庭。冬季没有供暖，是否压到了某一户的生命基底？这要另核，并且要说准核的是什么。权利论划过一道限缩：「基础层之侵限于对既有厚层的作用，不含对未来供给的请求」；「纯自然匮乏是非权利事件」（权利论第五篇 F20）。一项已载入的供给失效了，按承认层的读法读（第三节）；它不因此直接成为基础层的一次侵犯。要另核的是几样别的。有没有人对这一户既有的维持回路动了手，切断、设卡或者独占：动手的是别人，归第二条；是苹果城自己，归第一条（第三节第二读）。苹果城有没有为这一层底线另设过背靠，那背靠是不是也空了：这归第二条。这一篇不替哪一户下判。权利论在同一处说过，一次撤出若既无动手、又无已载入的位置，在那套读数里不落刻度，「这不是说它无足轻重，是说它的重量不在这套读数里」（权利论第五篇 §II）。这些家庭凭的是一个已载入的位置，它们的重量在承认侧那一套读数里。

G18　基础权利的保证本身在承认侧。政府给的是背靠与执行，不是核：核的不受侵犯从主体那边来，基础权利不是制度新造的。背靠可以被掏空，核不动。背靠与执行在承认侧，归 G19 的保证：基础权利的背靠被掏空，走第二条；政府不保证它，权力同样不正当，须被追问。

第三节　承认性权利：三件

回到住房局。

权利论说，一项承认性权利靠三支运行：「制度载入给可援引形态，共同体有效承认给分布式可预期性，真实权力支持给兑现能力」（权利论第三篇 F10，§I）。放到补助上：条例把它写了进去，是第一支；申请人、窗口都把它当作一项可以申请、可以追问的东西，是第二支；审核岗上有人，三十日内作出决定，钱按月发到，是第三支。第三支跌破了。权利论管这叫掏空：「掏空是权力支持跌破阈值而制度外壳仍在，位置退为宣告，运行消失」（权利论第三篇 F11，§II）。

对已载入的权利，政府守三件：不脱权；不留空壳；撤改公开受问。三件拆开看，落进第一、二、五条应当，不另增第六个总名。

一、不脱权。 脱权是另一种失效：「脱权是强制仍在，而相关网络不再把受位者作为可凭该位置发起请求、调用程序、要求响应的权利持有人，位置退化为强力」（权利论第三篇 F11，§II）。补助这一件谈不上脱权：脱权的前件是强制仍在，而住房局并没有对申请人施加什么强制，它只是不审。

脱权的样子要另举。设若第二篇那家欠税的杂货店，账户的冻结一直延续；店主提出异议，税务处不受理；他去市法院行政庭，也没有人处理他的申辩；换了经办人，新经办人照样不受理他的异议；冻结照旧。冻结还在，强制还在；可他作为纳税人能申辩、能要求说明的那个位置，网络已经不处理了。

判它是不是脱权，权利论给了三问：「一是受位者能不能以这个位置发起程序，二是这个位置能不能跨独立的节点被同一地复制和处理，三是换掉直接的施权者之后，这个位置是否仍由网络维持。」（权利论第三篇 §II）设若的这一支，三问的答案都是否定的。第三问要分清：换了人仍在的，是冻结；他凭纳税人的位置提出的请求，换了人仍没有一个节点接住。权利论说：「只有当它只能靠同一个施权者持续施压，网络并不处理受位者的请求位置，才是脱权」（同上）。这里撑着冻结的只剩税务处自己，没有一个独立的节点把他当作可凭位置发起请求的人来处理。这时政府是在经一个已经退化成强力的位置继续施为。这一件归第一条：构的施为对每一个内部主体都不越线（Z9）；一项施为已经不经由权利位置，它对人的作用就要照线去核，不能拿"这原是一项依法的冻结"替它过关。

三问只作判据，不作规范。权利论自己说，三问「判的是网络处理的结构，而不是把规范上的正当与否偷渡进来」（权利论第三篇 §II）。它们告诉我们一个位置断在哪里；政府在那里应当怎样，不从三问里来，从这个系列立的那一条来（第一节）。

二、不留空壳。 这是补助这一件的正题。空壳有三种读法，要分开。

第一读：基础权利的背靠被掏空，走第二条（第二节；Z17）。

第二读：先主动封掉原有的自维或替代通路，使人只能依赖该供给，再掏空的，走第一条。制度论说拆除也是构的施为：「拆除本身不得越线，比如先封掉其原有自维或替代通路，使其只能依赖该供给，再撤出，这一部分归第一条，不打折」（Z38）。设若住房局早先以统一供暖为由，拆掉了公租房里各户自装的取暖炉，住户只能靠补助去付统一供暖的费用；随后审核岗空着，补助停转。三样先写。对谁：被拆了取暖炉的公租房住户。在哪条线：苹果城先封了他们原有的取暖之路，再让剩下的唯一一条落空；封路这一步动的是他们既有的维持回路，要照线核。基线：拆炉以前，各户自装的取暖办法。这时被掏空的就不只是一个供给位置：苹果城先动手封了他们原有的路。这一支照第一条判，不打折。Z38 原说的是撤出；这里后一步是掏空，不撤外壳，只让支持停下。权利论说，「独占与致依赖本身已是切断，撤出只是它的完成，分两步走不改读数」（权利论第五篇 §II）；权益专论也说，这时停付「只是先前切断的完成」（权益专论 F30）。后一步是撤出还是掏空，读数一样，所以照 Z38 读（推及自担）。这一读要守住门槛：不由任何事实上的依赖直接判第一条。那些家庭即便事实上只能靠补助过冬，这本身也不够；「仅仅事实上无可替代不是独占，独占要有排他的动手」（权利论第五篇 §II）。

第三读：其余的运行缺口，进入第五条的说明与修正（Z34）。住房局这一件，落在这一读（引言）。说明要答到这几样：什么没有兑现，即一千二百份各压了多久，三十日的时限对多少份已经过去；原根据是否仍成立，即低收入家庭冬季的供暖，这件事还在不在；以及恢复运行、过渡，或者有根据地撤改的安排，即岗位什么时候补，补上之前积压的那些怎么办，若不想再办这项补助，凭什么、走哪一条路。

说明本身不等于修复。一份说明写得再清楚，岗位仍空着，一千二百份仍没人审，空壳仍是空壳。允许追问，也不等于已经履行了保证。"不留空壳"不缩成"不隐瞒空壳"。是否另触第一、二条，分别核：上面的第二读是一种另触，第二节说的生命基底是另一种。

真做不到时，比如一时招不到人，照第五篇 G33：不以做不到为由取消说明与修正之责，也不凭说明就记作已经做到。

三、撤改公开受问。 苹果城若真不想再办这项补助，它要走的是撤改这一条路，不是让条例挂着、让岗位空着。撤改要核权限、根据、作用域与过渡，并公开受问；这一件归第五条。

政府撤的，是它自己那一份。权利论把来源与运行分成两层：来源上，一项承认性权利由构成式承认成立，有纵向的制度授予与横向的相互承认两支；运行上，靠制度载入、共同体有效承认与真实权力支持三支；「来源侧的制度授予与运行侧的制度载入不是一回事」（权利论第三篇 §I）。苹果城那一份，是它在来源上的制度授予，连同它的载入与它的支持，一并公开撤，不留外壳。眼下这项补助的样子，外壳在，支持没了，正是撤改要避开的样子。

公开宣布，不独自取得正当性，也不独自证明这项权利已经明撤。明撤是「相关作用域里全部构成来源退出，不是仅旧载体消失」（权利论第三篇 F11，§II）；而「一项机构授予被撤回之后，横向的相互承认仍可能构成同一个位置」（权利论第三篇 §II）。苹果城能撤的，是它的授予、载入与支持；人们彼此之间的相互承认若仍构成这一个位置，这一撤就还不是明撤。这一点不由政府宣布。

设若市议会废止补助条例，另设一个名目不同的新项目，申请从头来过。这一步先要读清楚是什么。新项目构成的若是同一个位置，只是换了载体，权利论叫转载：「转载是载体迁移而权利在新载体上继续运行」（权利论第三篇 F11，§II）。那时位置续行，旧条例下递进来的申请随它一起移过去，"从头来过"本身就要说明。若是另一个位置，旧的那一个照第三件撤。不论哪一种，有三样东西不跟着消失。基础权利的底线：哪一户的生命基底是否受压，照第二节另核，不因项目换了名目就换了答案。已发生的旧账：已经递进来、已经超过三十日的那一千二百份，是旧条例下欠下的，不因旧条例废止就销账。独立成立的其他来源与承诺：同一个位置若另有来源构成，比如人们彼此之间的相互承认，照它自己的来源核；已经作出的承诺，照它自己的条件核（Z38 所说的「既有承诺」）。

坚持的是方式。 这三件在"坚持"一边：不随资源打折，不因紧急停下（"坚持"的宽义见前置篇第八节，由这个系列自担）。坚持的是这三样方式，不是每一个已载入位置的存续。补助可以改，可以撤；改与撤的方式不撤。制度论说过同一个形状：「坚持的是应当，不是构造」（Z22）。三件缺一，政府的权力在那一处就不正当，须被追问；失守成了常态，整体须被追问（第一节）。其中经退化的位置施于人的作用，以及先封路、再掏空的那一部分，核成越线的，另照第一条读，不只是须被追问（第一节）。

保证的范围本身也受审查：不因一项位置在结构上成立、在运行，就要求政府替它兑现运行。这是下一节。

G19　承认性权利的保证：三件。一项承认性权利靠制度载入、共同体有效承认与真实权力支持三支运行。对已载入的权利，政府守三件。其一，不脱权，归第一条：脱权时强制仍在，政府经已退化成强力的位置继续施为；三问只作判据，不作规范。其二，不留空壳：基础权利的背靠被掏空，走第二条；先主动封掉原有的自维或替代通路、使人只能依赖该供给、再掏空的，走第一条（推及自担），不由事实上的依赖直接判第一条；其余的运行缺口，进入第五条的说明与修正，要说明什么没有兑现、原根据是否仍成立，以及恢复运行、过渡或有根据地撤改的安排。说明不等于修复，允许追问不等于已经履行保证，"不留空壳"不缩成"不隐瞒空壳"；是否另触第一、二条，分别核；真做不到，照第五篇 G33。其三，撤改公开受问，归第五条：核权限、根据、作用域与过渡；政府撤的是自己那一份，制度授予、载入与支持一并公开撤，不留外壳；公开宣布不独自取得正当性，也不独自证明权利已明撤，横向的构成来源还在不在，不由政府宣布；基础权利的底线、已发生的旧账、独立成立的其他来源与承诺，不因项目改名或终止而消失。坚持的是这三样方式，不是每一个已载入位置的存续（"坚持"的宽义见前置篇第八节）。三件缺一，政府的权力在那一处不正当，须被追问；失守成了常态，整体须被追问；其中核成越线的，另照第一条读。

第四节　保证不是背书

苹果城老城区有一座石桥。很久以前，市议会授予一家私营公司这座桥五十年的过桥收费特许权。特许载入了条例，收费照常运行，收入早已远超建桥的成本。第七年，有居民过桥而不交费，公司要求市警局协助收费。

三样先写。对谁：公司；过桥的居民。在哪条线：线内。过桥是可限制的行动，收费与过桥而不交费都不越谁的线；牵涉的是一项载入的承认性位置，与协助收费的那一下强制。基线：没有特许的公共桥。

公司的理由可以写成这样：特许是条例载入的一项承认性权利；照第一节，政府应当保证它所载入的承认性权利；所以市警局该来。

这一推，第一步只对一半：特许是条例载入的一项承认性位置；公司是构，照职责专论，「一家公司不自持一项承认性权利」，它在这一位置上作的是归责地址（职责专论 F42）。第一节那一条，说的也是政府为内部主体载入的承认性权利。持有怎样细分，这一篇不在这里展开；就算把位置背后的持有算足，第二步也漏了一处：一项位置载入了、在运行，不等于它该由政府维持运行，该不该，要另核。第三步也不跟着来：从"应当保证"到"市警局该来"，中间还隔着一条强制的证明链。

权利论在立承认性权利怎样运行的时候，末尾加了一句：「构成成功与运行成功，都不等于规范正当。」（权利论第三篇 F10）一项特许成立了、运行着，说的是它在结构上是什么，不说它该不该在，更不说政府该用什么去维持它。第一节说，不保证，权力不正当；这一节说另一面：保证了，权力也不因此正当。两句并读，保证是必要条件，不是充分条件。

所以，保证的范围本身受规范审查。F10 只说结构不发正当性；审查的要求，是这个系列由此走出的一步（自担）。审查管的是什么，要说准。第三节的三件是方式，对每一个已载入的位置都守：不经退化的位置施强制，不以空壳代撤，撤要公开受问。范围审查管的是另一件：一项位置该不该续，政府该不该替它兑现运行，兑现时动不动强制。审查的结论是撤，就照第三件撤；不撤，而该兑现的那一项运行跌破了、只剩条例挂着，那才是空壳（第三节）。警局不来协助这一次收费，不独自说明收费的位置成了空壳：收费照常运行，公司那一请求是另一件，要另过第二篇的链。

审查照制度论那一句去核，「续存与拆除，都回到所保护的人、任务、权限、代价与既有承诺去核」（Z38）。放到石桥上：它保护谁，是公司收回建桥的成本，还是过桥的人有一座维护得好的桥；它的任务是什么，收入早已远超成本，那个任务还在不在；代价落在谁身上，落在每一个过桥的人身上；既有承诺是什么，五十年的特许本身就是一项承诺，要照它核。制度论也说，构自身的存续，「不构成关闭追问、或自动续存一项具体安排的独立充分理由」（Z38）。一项安排存续了很久，这本身也不是续存的理由（这一步由这个系列自担；同向：Z22 说临时的守卫构造「不凭来历自动续期」）。

这一篇不判这项特许该撤还是该留。它判一件事：苹果城不能拿"它是载入的，它在运行"回答公司。要撤，走第三节的第三件，撤改公开受问，既有承诺在要核的事项之内；要留，留也要说明。

还有一层。市警局协助收费，就是在执行这一项载入的位置。执行落到过桥而不交费的居民身上，收敛他们的行动，这时这一位置所载的权利显出了权力面。权利论第四篇摘要述 F15：一项运行的承认性权利在实际收敛一个受效者时显出权力面，「这是条件于收敛而不是常态」。一显出权力面，市警局的那一下就是强制，照第二篇：强制本身须是限（G13）；执行一项不为挡越线而设的安排，要能回溯到否定性的根，走完证明任务链（G11）。一项特许载入了，不替这一下强制过关。

G22　保证不是背书。构成成功与运行成功，都不等于规范正当；保证的范围本身受规范审查，回流 G19（自担）；久存本身不是续存的理由（自担）。不保证，权力不正当；保证了，也不因此正当：保证是必要条件，不是充分条件。执行一项承认性权利，在实际收敛受效者时显出权力面，这时照第二篇 G13。

第五节　福利与权益

补助是一项供给。政府发出去的供给不止这一种，它们不都在同一格里。分三格。

已载入的。 供暖补助在这一格。持续供给要成为可主张的东西，它成立为承认层的已载入位置（第一节）。所以它按承认性权利保证，守第三节那三件。能撤，只照第三节的第三件撤；撤本身也过第一条（Z38）。这样一项位置失效了，权利论把它归到承认层：「已载入供给位置的失效归承认层五型」（权利论第五篇 F20）。是明撤、掏空还是转载，看断在哪里（第三节）。

未载入的。 设若过去五年，市长每年冬天以一个临时决定发"燃料券"，没有写进任何条例，也没有以别的形式载入（这一分支的给定）。第七年冬天，市长说，今年不发了。

三样先写。对谁：往年领燃料券的家庭。在哪条线：停发一项供给，本身不越谁的线；先前若有人封了别的路，那是另一件（第三节第二读）；是否压到某一户的生命基底，另核。基线：苹果城发券以前，各户原有的取暖办法。

这一分支设定它未载入，所以燃料券不在已载入那一格。它是权益，或者是制度论第三条应当所说的条件，即苹果城按关系厚度为它内部的人的涵育提供的条件（Z23）。可未载入，不自动等于可以随手停。制度论说续存与拆除要回到所保护的人、任务、权限、代价与既有承诺去核，接着说：「此外既有的权利、承诺与依赖，照上一句去核」（Z38）。已经承担的条件、既有的依赖、别的来源，都要另核。

裸权益。 权益专论说，授予的一方可以停付，裸权益这一格里停付不是侵，「但以授予端未先排他性独占、未制造依赖、未封死替代回路为限」；它接着说，「付出若已承诺、制度、信托或监护载入，便已非裸权益」（权益专论 F30）。前一句给停付划了限；后一句把一大片供给移出了这一格。

燃料券要逐项过。市长有没有承诺过以后每年都发？这一分支设定没有；若有，它便已不是裸权益。苹果城有没有制造依赖？五年照发，若各户已靠它过冬，那是事实上的依赖；事实上的依赖不等于制造依赖，要看苹果城有没有先封掉人们原有的取暖办法。有没有独占、有没有封死替代回路，同样要核到有没有排他的动手。几样都过了，停发在裸权益格里不读作侵。不是侵，也要说明：苹果城对这些家庭的条件之责，按它实际组织他们生活的程度来，「构已组织、占用或限制的，须相应说明怎样留出与提供空间」（Z25）。何况同一个冬天，补助也积在住房局；两样合起来压到哪一户，照第二节另核。

这一格搬到政府身上，射程要写全。权益专论的授予端，原是一个主体自己的付出：「权益锚在授予端的付出」（权益专论 F27）。停付不算侵，理由是供给者撤出自己的供给，「是以自己的法处置自己的行使」（权利论第五篇 §II）。苹果城作为构，经由市长、住房局这些位置付出（Z7a）；构没有自己的法（制度论前置篇第二节），也就没有那样一份自己的行使可以处置。所以搬过来的只是读数那一半：停付在这一格里不读作侵；许不许停、怎样停，回到第三条应当与 Z25 的说明（推及自担）。

监护里的供给。 第二篇第六节写过一个被带离家的九岁孩子，家事庭指定儿童与家庭局为他的临时监护方。监护期间，他的住处、饭食、学杂费，由局里的经费支付。权益专论把受养育分成两面：「受养育按行动矢量收不收敛分两面，供给面落权益，收敛面落代持承认性权利」（权益专论 F34）。收敛的那一面，局里代孩子作医疗同意、入学登记，归第二篇的代行（G12a）；住处、饭食、学杂费是供给面，在这一节。

供给面落在权益一格，可它不是裸权益：它经监护载入。照 F30 的后一句，这一份付出已非裸权益，局里不能以"权益可以停付"为由停掉孩子的饭食；它要撤，照权利论第三、四、五篇归位（权益专论 F30）；强制与替补由权力支持处理（权利论第五篇 §II）。权利论那一句原说的是监护、供养位置上的持有者撤出：「已载入的监护、供养位置，其持有者的撤出由承认层接住，位置失效按五型读，强制与替补由权力支持处理」（权利论第五篇 §II）。这里若是局自己撤，主词换了，接住它的承认层与权力支持多半也在政府里，要另核。F34 原说婴儿；推到一个在政府监护下的九岁孩子，第二篇 G12a 已经标过推及，这里只借它的两面之分（推及自担）。

所以一项供给在不在裸权益格里，不能凭名目定，要逐项检验。

与马歇尔。 马歇尔（T. H. Marshall）把公民身份分成三个要素，第三个是社会要素：「我所说的社会要素，指的是从享有一份适度的经济福利与安全的权利，到充分分享社会遗产、按社会通行的标准过文明人生活的权利，这整个范围」（By the social element I mean the whole range from the right to a modicum of economic welfare and security to the right to share to the full in the social heritage and to live the life of a civilised being according to the standards prevailing in the society；《公民身份与社会阶级》第二部分，据 Pluto 1992 年版节选 p. 8，B 档）。以他这篇文章为主要参照的那一种公民身份观，把公民身份首先当作一种法律地位，给政治共同体的成员同样的一组公民、政治与社会权利（据斯坦福哲学百科《公民身份》§2.1，C 档）。两边都把福利放进权利的结构里，不当作恩惠；这里只对照一处：照这一路的读法，社会权利按成员资格划，挂在公民身份上；这个系列的两类保证说给内部主体，内部按纠缠划，不按身份划（前置篇第七节）。谁是成员，归国家论。

G21　福利与权益。已载入的供给位置，按承认性权利保证；能撤，只照 G19 撤，撤本身过第一条（Z38）。未载入的，是权益或第三条所说的条件；未载入不自动等于可任意停付，已经承担的条件与其他来源另核。裸权益一格里，停付不读作侵，以授予端未先排他性独占、未制造依赖、未封死替代回路为限；不是侵，不免说明。付出一经承诺、制度、信托或监护载入，便已非裸权益。某项供给在不在裸权益格里，逐项检验。授予端读到政府的供给、F34 读到政府监护下的孩子，均推及自担。

第六节　行政

一千二百份积在那里，根子在一个岗位上。补助靠审核岗兑现；岗位空着，是行政的事。

行政常被看成执行的末梢：规则定好了，照着办就是。权利论在说第三支时，把它放得更近。它说真实权力支持「不只是侵害发生后的制止与修复，也包括日常的履行、资源的配置、程序的接入与资格的兑现」（权利论第三篇 §I）。补一个岗，排一个审核的次序，在三十日内作一个决定，把钱按月发出去，这些都是兑现。一项承认性权利空没空掉，常常不取决于哪一次大的侵害，取决于这些日常的事有没有人做。

这一篇给行政两条根。

一条是让已载入的权利真能兑现。这接第三节：不留空壳落到日常，就是行政。

一条是按关系厚度为涵育提供条件。制度论第三条应当说：「制度按关系厚度，为它的内部主体的涵育提供条件；只管条件，不管结果。」（Z23）苹果城开一间社区艺术中心，付它的经费，是这一条根上的行政（第八节）。

两条根都不从权利论来。权利论只告诉我们兑现包括哪些日常的事；行政为什么必须做这些，前一条来自第三节的保证，后一条来自第三条应当。

两条根只给不动强制的行政。审一份申请，补一个岗，发一笔钱，开一间艺术中心，都不动强制。行政若要动强制，比如以强制手段核查一户的申报，那一步另过第二篇 G11：执行一项不为挡越线而设的规定，其强制须能回溯到否定性的根；回溯不到的，不因改称行政而成立；行政目标有根据，不独自使强制手段有根据。

还要说一处账。窗口的人说"审核的岗空着"，说的是实情，不是推脱。岗位空着这一笔，不在窗口，在定编制、给经费、决定补不补人的那些位置上（第一篇 G4）。它们要说明的，正是第三节列的那几样。审核一份申请，也是判断，第一篇说的作答在这里照样适用；审核可以由人作，也可以交给工具（第一篇第九节），形态不定，兑现要在。

G20　行政的根。真实权力支持包括日常的履行、资源的配置、程序的接入与资格的兑现。行政至少有两条根：让已载入的权利真能兑现（G19）；按关系厚度为涵育提供条件（制度论 Z23）。这两条根只给不动强制的行政；行政动用强制时过第二篇 G11，不因行政目标有根据，就使强制手段有根据。

第七节　财政与税

第二篇那家欠税的杂货店，冻结的账户里留了一个维持基本生活的额度。房产税是苹果城一般预算的主要来源；供暖补助的经费，也出自这一般预算。店主问："我交的税换来了什么？"

这一问可以有两种读法。

一种读成价钱：交了多少，换回多少；换不回来，就该少交或者不交。

一种读成要一份说明：钱从哪里来，压在谁身上，花在哪里，缺口由谁担。

这一篇取后一种，并且要说清后一种为什么不就是前一种。

税的根。 苹果城凭什么收税？这一篇的回答是：税与经费的根，是承担五条应当与两类保证所需的那些功能的经费。审核岗要有人坐，复核员要有人当，热线要有人接，法院要开门，保护令要有人去执行；五条应当与两类保证落到功能上，功能落到位置上，位置要经费。《制度如何可能》在推民族国家一层的制度增量时，列过同一样东西：「财政制度（常设机构需要经费）」（F6 §3.1 b）。那里是一层制度的增量，挂在条件的变化上；这里把税的根挂在应当与保证所需的功能上，这一步由这个系列自担。

根在那里，要说明的就跟着来（第一篇 G2）：筹资的负担落在谁身上，房产税压在哪些人身上；经费用在哪里，一般预算里有多少到了补助，有多少到了审核岗；缺口由谁承担，审核岗空着的这三年，缺口落在哪里。最后一问若没有一个位置出来说明，答案就自己落下来了：落在等着的那些家庭身上。没有人说明的缺口，本身须被追问。

税的强制怎样回溯。 为征税而冻结账户，回溯要经过税自己的根（第二篇第四节）。法学第一篇要肯定性条款的正当性回溯到否定性的根：「对压制的防止，对追问通道的保障，或对共构空间的维持」（法学第一篇 BL3）。经费所供的功能一样样接上去：复核员、热线、法院，保追问的通道；保护令有人执行，防压制；已载入的位置兑现得了，维持人们已经共有的那些位置。征税与为征税而冻结，经由这些功能回溯（同第二篇 G11，推及自担）。为争取一侧的条件征的那一部分，比如艺术中心的经费，只能读作对共构空间的维持，而提供条件比"维持"多走一步；这一读推及自担，走不走得到底，这一篇不宣称答完。回溯没有走完以前，这一部分强制的资格，这个系列不担保，留作余项（第六篇第八节，六格第一格）。这样回溯只读到"对共构空间的维持"、没有走完的强制，这个系列里有三处，这是其中一处；另两处在第二篇：第四节征地的读法，第五节赔偿的候选。三处在第一格合登。

不取的一句。 《制度如何可能》的附录二，在一部虚构的宪法里写过一句：「税是构对个体财产的合法凿，其正当性来源于构为个体提供了个体无法独自获得的东西」（F6 附录二卷五）。这一句不取。理由有二。其一，那一篇自己说，附录「不是本版的论证件」（F6 附录卷首说明）。其二，这条理由是受益式的：照它，税的正当性挂在构给出了什么上；店主那一问，就容易被读成一笔价钱。这个系列不借它为具体的筹资与强制分担作充分论据。受益的理由与普遍对价也不是同一个命题：下面那一条排除的，是自动的交换律，不是一切按受益安排的具体收费。

交税不构成普遍对价。 职责专论有一条：「承认性权利不由承认权力换来。」它接着说：「一链职责的发生、继续与终止，依该链自身的内容和条件判断，不因另一链已经履行就自动生成，也不因另一链未履行就自动免除。」（职责专论 F43）

苹果城与店主之间，有两样方向相反的东西。一样，苹果城经税务处向他收税，他在那一格持纳税之责。另一样，他在苹果城持有的那些已载入的位置，向苹果城要兑现，苹果城对它们负保证。F43 的第一句，挡住的是把两样接成一笔买卖：保证不是交税换来的，交税也不是买保证的价钱。它的第二句，挡住的是两个自动。

店主交了税，不自动生出他对某一项具体给付的请求；苹果城把补助办空了，也不自动免掉他的纳税之责。反过来，他欠了税，也不自动取消他在别处已经成立的位置；那些位置照它们自己的来源与条件核。冻结时留下的那个生活额度，不在这笔账里：它管的是冻结不把这一家的生命基底压到受威胁，那是第一条应当的事（第二篇引言），不是另一条链上的一项。

F43 原说的是两个节点之间方向相反的两条职责链，它举的是公司与受雇者。店主所持的位置运行着的时候，它的权力面收敛苹果城，苹果城在那一格持职责（职责专论 F39、F42），F43 原样适用。位置被掏空、那一条链已经不在时，欠下的保证仍在（第一节）；只有这时，把 F43 用到"应当"一侧，才是这个系列的一步（推及自担），根据是 F43 挡的是"自动"：另一边是链还是应当，交税都不自动换来什么，也不自动免掉什么。

补助这一项保证，也不从税来。递了申请的家庭交没交税、交了多少，不是这项保证的条件；保证来自条例的载入与这个系列立的那一条（第一节）。

F43 还留了一句：「具体安排是否把某项事实作为适用条件，另按该安排判断」。一项按次收的规费，一份写明以缴费为条件的服务，照它们自己的条件判。所以，交税不构成一条普遍对价；已经成立的具体请求，也不因此消除。

G20a　财政。税与经费的根，是承担五条应当与两类保证所需的那些功能的经费（自担）。筹资的负担落在谁身上，经费用在哪里，缺口由谁承担，要说明，可被追问。征税的强制过第二篇 G11，经由所供的功能回溯到否定性的根（推及自担）；为争取一侧的条件征的那一部分，只能读作对共构空间的维持；回溯没有走完以前，这一部分强制的资格，这个系列不担保，留作余项（第六篇第八节，六格第一格），与第二篇征地、赔偿两处合登。交税不构成普遍对价（G23）。受益式的理由（F6 附录二卷五）不取作具体筹资与强制分担的充分论据。

G23　交税不构成普遍对价。承认性权利不由承认权力换来；一链职责的发生、继续与终止，依该链自身的内容和条件判断，不因另一链已经履行而自动生成，也不因另一链未履行而自动免除。具体安排是否把某项事实作为适用条件，另按该安排判断。所以交税不构成一条普遍对价；已成立的具体请求，也不因此消除（推及自担）。

第八节　条件有限时

第七年冬天，苹果城的财政收入下降，市议会讨论下一年预算缩减三成。桌上有三项提案：

一、撤掉住房局的投诉窗口，补助的投诉并入市长办公室的投诉热线； 二、暂停供暖补助的新申请一年； 三、削减社区艺术中心的经费一半。

三样先写。对谁：各项所涉的人，即投诉者、补助的申请者、艺术中心的使用者。在哪条线：三项都不直接越谁的线。第一项碰第五条的受问通道；第二项碰一项已载入的承认性位置；第三项是争取一侧的条件。基线：上一年的预算。

钱少了，总要砍些什么。这一节问：什么不能砍。

不能砍的是什么。 不能砍的，是应当与必要的通道，不是哪一个旧岗位、项目或预算数额。制度论说：「构造可以改，三项条件不能撤」（Z37；原说受问的三项条件，其余几样照前置篇第八节"坚持"的合列）。坚持一边的几样，不随资源打折：第一、二条应当，基础权利不受侵犯在内；第四条里不封拒绝、修订、退出与改向的通道那一半；第五条的受问通道；承认性权利保证的方式（第三节）；代行之限（第二篇 G12a）。最后这一样单列，是因为其中朝归还而去一条，上游原在第三条应当的扶助里（Z26）；推到代行，它是强制之限，经第二篇 G13 归第一条，不是争取一边可以排序的条件，钱少了也不往后排。

这里有一处分寸要说准。制度论第一篇第八节说：「这个系列承认实际可能做不到，不因此撤回或打折哪一条应当」。这一句管五条，不分坚持与争取。所以争取一边的应当，也不打折；钱少了，可以排序的是争取一边的条件：排在前还是排在后，给得多还是给得少，不是应当本身。把五条分成两边，一边不随资源打折，一边的条件可以排序，这一分由这个系列自担（前置篇第八节）。

逐项读。

第一项，投诉窗口并入热线。窗口是构造，构造可以改。三项条件不能撤：通道要「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」（Z34）。要核的是并入以后的样子。热线设在市长办公室，与住房局不是同一方；独立是功能义，「它对着实际被问的一方核」（Z34）。在这一件里，被问的是住房局，也包括定编制、给经费的那些位置（第六节）；那些位置若在市长办公室，热线对它们就不独立，就要另有一条够得着的路。热线收不收补助的投诉，收了能不能要被问的一方答，答不到点上还能往哪里送：这几样核得过，这一项是构造的改动；核不过，补助的投诉在苹果城这一级就无处可送（外层的州监察专员仍在，够得着，但担不起日常的制约，第四篇第三节），砍掉的就不是一个窗口，是这一级的第五条通道。窗口这一类安排会不会落到被问的一方手里，第六篇讲捕获时再接（G38）。

第二项，暂停补助的新申请一年。这碰一项已载入的承认性位置，所以只能照第三节的第三件撤改，公开受问。权限：条例是市议会立的，暂停它的一部分，在市议会的权限之内。根据：财政收入下降，是一个根据；补助原来的根据，低收入家庭冬季的供暖，还在不在，也要说。作用域：只停新申请；已经递进来的一千二百份不在暂停之内，它们是旧账（第三节）。过渡：这一年里，新符合条件的家庭怎么办，要有安排。暂停本身还要过第一条（Z38）：苹果城有没有先封掉谁原有的路。还有一处：暂停若成了让空壳挂着的另一种说法，岗位照旧空着，积压照旧不动，那么它没有回答"不留空壳"那一件。

第三项，艺术中心的经费减半。这是争取一侧的条件：第三条应当，苹果城按关系厚度为涵育提供条件（Z23）。它可以排序，可以减。若其中某一项已经载入或者承诺，那一项照第五节另核。

排序本身是判断。 三项摆在一起，哪一项先砍、砍多少，是排序。法学第一篇说法在执行层有一种结构性的折扣，其中一重是资源的优先级：「不是每次压制都能被处理」；它接着说，排序本身就把排序的一方自己带了进来（法学第一篇边界三，转述）。那里说的是法在执行层的折扣；推到政府的分配排序，这一步由这个系列自担（推及自担）。推的根据在第一篇：苹果城排预算，也是坐在位置上的人在判断，判断就要作答（第一篇 G1、G2）。

所以排序要说明：写明须判断的事项，相关的主体，以及补缺的责任怎样受问。艺术中心减一半，要说明减的是什么，谁因此少了什么，这一缺口由谁、在哪里接，被减的人往哪里问。

关系厚度在这里有用，只用到一处。它界定条件之责的范围：苹果城实际组织一个人的生活到什么程度，就欠他相应的条件与说明（Z23、Z25、Z25a）。它不独自给所有事项定先后。艺术中心与别的争取一侧的事项谁先谁后，厚度不替判断作。

门槛之下由谁补。 制度论 Paper 0 与森、努斯鲍姆对话时，留下过一问：「一个人落在某条门槛之下，由哪一个构、按什么分担去补，是分配与设计的问题」，它把这一问交给了具体的制度论与政府论（制度论 Paper 0 第五节第一小节）。这一篇答到结构为止。

其一，已载入的位置，由载入它的那一级保证。补助是苹果城的条例载入的，苹果城保证它，守第三节那三件。

其二，争取一侧的条件，每一个构按它实际组织这个人生活的程度，欠相应的条件与说明（Z23、Z25、Z25a）。各层的关系厚度怎样读，第四篇写（G29）；各层怎样分担，归具体的制度论。

其三，补缺的责任要能被问。缺口由谁承担，要有一个位置写明（第七节）；没有一个位置写明，缺口就落在等着的人身上，这一处本身须被追问。

具体给谁、给多少、各层怎样分担，归具体的制度论。这一篇不另立一套分配的正义。

与森、努斯鲍姆，与罗尔斯。 能力进路，制度论 Paper 0 已经对过（第五节第一小节）：努斯鲍姆的门槛说给国家，实现的负担怎样细分，那里照实登记为未答。这一节接的正是那一问，只接到结构，不给清单，不给门槛。罗尔斯的基本结构，也已对过（制度论 Paper 0 第三节第二小节）：那里说，这个系列不分配，「分配的正义是相邻的辖域」。这一节不改这一句，只接"门槛之下由谁补"这一问。

真做不到时，比如三成的缺口怎样排都填不上，照第五篇 G33：承认做不到，不撤回、不打折应当；过渡要说明；归责照 Z6。

G24　条件有限时。坚持一边不随资源打折：第一、二条应当（基础权利不受侵犯在内）、第四条的不封通道、第五条的受问通道、承认性权利保证的方式、代行之限（第二篇 G12a，经 G13）；不能砍的是应当与必要的通道，不是哪一个旧岗位、项目或预算数额。制度论"不打折"一句管五条；把争取一边的条件放在可以排序的一边，由这个系列自担。争取一边按关系厚度界定条件之责的范围；关系厚度不独自给所有事项定先后。排序本身是判断（推及自担），要说明须判断的事项、相关主体，以及补缺的责任怎样受问。门槛之下由谁补，答到结构为止。

这一篇停在哪里

这一篇只证明了这些：两类权利要分开，基础权利不受侵犯对谁都成立，不是政府特有的，保证为内部主体载入的承认性权利是政府特有的承担，不保证即须被追问，保证了也不因此正当；政府给基础权利的是背靠与执行，不是核；对已载入的权利，政府守不脱权、不留空壳、撤改公开受问三件，政府撤的是自己那一份，坚持的是方式，不是每个位置的存续；保证的范围本身受审查，三件对每个已载入的位置都守；供给分三格，裸权益一格里停付不读作侵，以 F30 为限，承诺与载入的不在其中；行政有两条根，只给不动强制的行政；税的根在应当与保证所需的功能，征税的强制经由这些功能回溯，交税不构成普遍对价；条件有限时，不能砍的是应当与通道，排序是判断，要说明，门槛之下由谁补，答到结构。

它没有证明的：分配的正义，这个系列不另立；具体给什么、怎样给、各层怎样分担，归具体的制度论；为争取一侧的条件征税，回溯能不能走到底，没有答完；回溯没有走完以前，这一部分强制的资格，这个系列不担保，留作余项（第六篇第八节，六格第一格；与第二篇征地、赔偿两处合登）；征税、冻结、协助收费这些强制凭什么，第二篇写；投诉通道会不会落到被问的一方手里，第六篇写（G38）；做不到时怎样过渡，第五篇写（G33）；各层的关系厚度怎样读，第四篇写（G29）；谁是成员，归国家论。

那一千二百份，后面还会有人来问。

本篇命题（合列）

G17　保证两类权利，两半分开。基础权利不受侵犯，从主体那边推出，对谁都成立，不是政府特有的；对构的应当已在制度论第一、二条（Z9、Z15、Z17），政府论承它，不立它。保证承认性权利是政府特有的承担，由这个系列立为政府侧新增的规范承担：政府应当保证它为内部主体载入的承认性权利，不保证，权力在那一处就不正当，即须被追问；权利论给对象与失效判据，不独自发出这一应当。"特有"指这一承担在这个系列里只对政府立，不断言别的构一概不欠。它不另增第六个总名，组织进第一、二、五条（G19）。两类保证也可理解为反殖民的一部分，"殖民"仍取窄义。"可被追问"对一切政府、一切时候；"须被追问"指一处已核的失守欠下一份说明与修正，两样不混。

G18　基础权利的保证本身在承认侧。政府给的是背靠与执行，不是核：核的不受侵犯从主体那边来，基础权利不是制度新造的。背靠可以被掏空，核不动。背靠与执行在承认侧，归 G19 的保证：基础权利的背靠被掏空，走第二条；政府不保证它，权力同样不正当，须被追问。

G19　承认性权利的保证：三件。一项承认性权利靠制度载入、共同体有效承认与真实权力支持三支运行。对已载入的权利，政府守三件。其一，不脱权，归第一条：脱权时强制仍在，政府经已退化成强力的位置继续施为；三问只作判据，不作规范。其二，不留空壳：基础权利的背靠被掏空，走第二条；先主动封掉原有的自维或替代通路、使人只能依赖该供给、再掏空的，走第一条（推及自担），不由事实上的依赖直接判第一条；其余的运行缺口，进入第五条的说明与修正，要说明什么没有兑现、原根据是否仍成立，以及恢复运行、过渡或有根据地撤改的安排。说明不等于修复，允许追问不等于已经履行保证，"不留空壳"不缩成"不隐瞒空壳"；是否另触第一、二条，分别核；真做不到，照第五篇 G33。其三，撤改公开受问，归第五条：核权限、根据、作用域与过渡；政府撤的是自己那一份，制度授予、载入与支持一并公开撤，不留外壳；公开宣布不独自取得正当性，也不独自证明权利已明撤，横向的构成来源还在不在，不由政府宣布；基础权利的底线、已发生的旧账、独立成立的其他来源与承诺，不因项目改名或终止而消失。坚持的是这三样方式，不是每一个已载入位置的存续（"坚持"的宽义见前置篇第八节）。三件缺一，政府的权力在那一处不正当，须被追问；失守成了常态，整体须被追问；其中核成越线的，另照第一条读。

G20　行政的根。真实权力支持包括日常的履行、资源的配置、程序的接入与资格的兑现。行政至少有两条根：让已载入的权利真能兑现（G19）；按关系厚度为涵育提供条件（制度论 Z23）。这两条根只给不动强制的行政；行政动用强制时过第二篇 G11，不因行政目标有根据，就使强制手段有根据。

G20a　财政。税与经费的根，是承担五条应当与两类保证所需的那些功能的经费（自担）。筹资的负担落在谁身上，经费用在哪里，缺口由谁承担，要说明，可被追问。征税的强制过第二篇 G11，经由所供的功能回溯到否定性的根（推及自担）；为争取一侧的条件征的那一部分，只能读作对共构空间的维持；回溯没有走完以前，这一部分强制的资格，这个系列不担保，留作余项（第六篇第八节，六格第一格），与第二篇征地、赔偿两处合登。交税不构成普遍对价（G23）。受益式的理由（F6 附录二卷五）不取作具体筹资与强制分担的充分论据。

G21　福利与权益。已载入的供给位置，按承认性权利保证；能撤，只照 G19 撤，撤本身过第一条（Z38）。未载入的，是权益或第三条所说的条件；未载入不自动等于可任意停付，已经承担的条件与其他来源另核。裸权益一格里，停付不读作侵，以授予端未先排他性独占、未制造依赖、未封死替代回路为限；不是侵，不免说明。付出一经承诺、制度、信托或监护载入，便已非裸权益。某项供给在不在裸权益格里，逐项检验。授予端读到政府的供给、F34 读到政府监护下的孩子，均推及自担。

G22　保证不是背书。构成成功与运行成功，都不等于规范正当；保证的范围本身受规范审查，回流 G19（自担）；久存本身不是续存的理由（自担）。不保证，权力不正当；保证了，也不因此正当：保证是必要条件，不是充分条件。执行一项承认性权利，在实际收敛受效者时显出权力面，这时照第二篇 G13。

G23　交税不构成普遍对价。承认性权利不由承认权力换来；一链职责的发生、继续与终止，依该链自身的内容和条件判断，不因另一链已经履行而自动生成，也不因另一链未履行而自动免除。具体安排是否把某项事实作为适用条件，另按该安排判断。所以交税不构成一条普遍对价；已成立的具体请求，也不因此消除（推及自担）。

G24　条件有限时。坚持一边不随资源打折：第一、二条应当（基础权利不受侵犯在内）、第四条的不封通道、第五条的受问通道、承认性权利保证的方式、代行之限（第二篇 G12a，经 G13）；不能砍的是应当与必要的通道，不是哪一个旧岗位、项目或预算数额。制度论"不打折"一句管五条；把争取一边的条件放在可以排序的一边，由这个系列自担。争取一边按关系厚度界定条件之责的范围；关系厚度不独自给所有事项定先后。排序本身是判断（推及自担），要说明须判断的事项、相关主体，以及补缺的责任怎样受问。门槛之下由谁补，答到结构为止。

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。

秦汉. (2026). 权利不是什么（SAE 权利论系列 前置篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20726894

秦汉. (2026). 论权利的源与谱（SAE 权利论系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20743135

秦汉. (2026). 论承认性权利的构成与撤回（SAE 权利论系列 Paper 3） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20993658

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). 论基础权利的限与侵（SAE 权利论系列 Paper 5） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21212595

秦汉. (2026). SAE 权利论 · 专论：论权利与权益 [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21314907

秦汉. (2026). SAE 权利论 · 专论：论职责 [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22853810

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). 制度不是什么（SAE 制度论 前置篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23269330

秦汉. (2026). 应当说给谁（SAE 制度论 第一篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23269635

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). 条件（SAE 制度论 第四篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23270922

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). 制度论与传统：应当的地址（SAE 制度论 Paper 0） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.23272468

秦汉. (2026). 自己的法遇上自己的法：14DD 对决中法的结构起源（SAE 法学系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19548237

Marshall, T. H. (1992). Citizenship and Social Class. In T. H. Marshall & T. Bottomore, Citizenship and Social Class. London: Pluto Press. （原作 1950 年。）第二部分节选：https://pages.nyu.edu/jackson/analysis.of.inequality/Readings/Marshall%20-%20Development%20of%20Citizenship.pdf

Leydet, D. Citizenship. In The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/citizenship/

接口

• 主承： 权利论（第一篇 F2、§II；第三篇 F10、F11、§I、§II；第四篇摘要述 F15；第五篇 F20、§II）；权益专论 F27、F30、F34；制度论（Z9、Z15、Z17、Z22、Z23、Z25、Z25a、Z34、Z37、Z38；第一篇第八节）。
• 另用： 权利论前置篇第三节；职责专论 F39、F42、F43；F6 §3.1 b，附录二卷五（不取的对照）；法学第一篇 BL3、边界三；制度论 Z7a；制度论 Z26（经第二篇 G12a，第八节）；制度论前置篇第二节；制度论 Paper 0 第五节第一小节。
• 对接： T. H. 马歇尔（一句对照）。指过去：霍耐特（制度论 Paper 0 第六节第一小节）；森与努斯鲍姆（第五节第一小节）；罗尔斯（第三节第二小节）。
• 本系列： 承前置篇第五、七、八节；承第一篇 G2、G4 与第九节（同一项补助的算法阶段）；第二篇 G11、G12a、G13 在这里回指，第二篇第四节交来的税的回溯在第七节写（G20a）；第四篇 G29 接各层的关系厚度怎样读；第五篇 G33 接做不到；第六篇 G35、G38 接通道与捕获，第八节六格的第一格收为争取一侧的条件征税的余项，与第二篇第四、五节两处合登。
• 交给第四篇： 同一件事，几层政府都管。