On How Rights Presuppose No Duty
论权利不预设义务

0. Introduction

SAE Rights Theory posits no category of duty. A subject has no duty toward others; it has only one inward duty, its answerability to its own law, which is the autonomy that belongs to the domain of the moral law (the termination lemma; SAE's sole axiom is Negativa, the negative method, from which autonomy and the theorems of the moral law all follow as lemmas and theorems [8]). What is called a duty toward others is the pushing-outward of a subject's inner law into a demand upon other people; that is ethics, which SAE judges a structurally illicit overreach and deletes. So the "presuppose no duty" of this paper's title does not say that a right needs no particular duty; it says that the word "duty," in SAE Rights Theory, has no object it could presuppose — the thing the correlativity axiom asks for has, here, no referent.

Let the reader carry a difference of coordinates. Contemporary discussion of rights mostly proceeds in a coordinate one could call market-utilitarian, where the weighing of input against return, of cost against benefit, has become the default way of looking at nearly everything. This paper does not stand in that coordinate. Its source is Kant — the subject as an end, the Kingdom of Ends. Some of this paper's conclusions will read as counterintuitive within the market-utilitarian coordinate, precisely because it stands in another coordinate, from another source. To be clear: market utilitarianism is the legitimate and powerful logic of the market layer; this paper does not criticize it, but only points out that it is one particular coordinate and not the only one. A subject as an end may perfectly well invest, exchange, and pursue gain. The paper says only that the source of a right does not lie in that coordinate, but in standing.

This is the meta-paper of SAE Rights Theory. The main line, P1 through P6, has already built the internal structure of the rights spectrum; this paper walks out of the main line and faces Western rights theory directly, taking up the one axiom the main series kept avoiding, the correlativity axiom. As a meta-paper it is no longer bound by the main series' restraints of not naming theorists and not pointing to particular institutions and historical events. It will name Hohfeld, Hart, Raz, MacCormick, Kant, Nozick, Dworkin, Sen, Nussbaum; it will cite the UDHR, the two Covenants, Vasak's generations; it will draw on postwar examples. But this exemption releases only the naming and the exemplifying, not the cold structure itself: an example is used only as a demonstration of structure, showing where a claim falls in the SAE architecture — whether it is a basic or a recognition component, a limit or a violation, whether it corresponds to a power or has no correlate at all. The paper adjudicates no institution's or policy's rightness, takes no side, and steps around the most contested value-questions of the day. The SAE Jurisprudence series contains a posterior dialogue with Western legal philosophy; this paper is its sibling on the rights-theory side — that one converses with legal philosophy, this one with rights theory, and the theorists they share are each engaged in their several different works.

The paper advances five propositions, in two views. M1 through M4 look at the constitution of rights from the structural view, asking on a single time-slice what a right is in structure, where it comes from, and how it divides: M1 takes up the correlativity axiom, that rights presuppose no duty; M2 takes up will theory, which grounds rights in choice; M3 takes up interest theory, which grounds rights in a protected interest; M4 takes up the negative-positive distinction, which cuts by duty-form. M5 looks at the diachronic unfolding of the rights spectrum from the dynamic view: that unfolding is the necessary dynamics of any subject-society. The dynamic view watches how the same structure moves in time; it does not set up a second structure apart from it — what M5 sees unfold in time is the recognition-based right as a structure, and what opens a new dimension is the subject as an end as a structure. The five propositions each engage one traditional current, and across source, division, and time they move the relation of right to duty out of the hand of the correlativity axiom and onto SAE structure.

I. On How Rights Presuppose No Duty

> M1. Rights presuppose no duty. Rights presuppose no duty, because SAE has no concept of a duty toward others. SAE replaces ethics with one's own law [8]: a horizontal duty is the pushing-outward of an inner law into a demand upon others, structurally an overreach, and is deleted. SAE's one duty faces inward — the termination lemma, that one cannot-not legislate only for oneself, the subject's answerability to its own law (SAE's sole axiom is Negativa; self-legislation and the four theorems all follow from it as lemmas and theorems, not as axioms). The paper argues at the level of the subject as an end, at the 15DD transparency-state's checking-network, the structural depiction of Kant's Kingdom of Ends: each subject at 15DD is an end to the others, and its own law already looks out over the whole field and includes the other (P1 §V; the first theorem of the moral law) [2][3][9]. So that it cannot-not oppose the suppression of any subject's remainder is a matter internal to its own law: the constraint faces inward, and inward the law already includes the other. A horizontal duty is superfluous in the Kingdom of Ends; the horizontal duty the correlativity axiom (correlativity; claim-duty) asks for has no object. Of Hohfeld's four pairs, SAE claims power-liability: a recognition-based right corresponds to power [5], the one bound holds a liability, and liability is not a duty. Claim-duty does not apply, for SAE has no horizontal duty. A basic right is pre-relational — it stands with a single subject, awaiting no second party, anchored in ρ ≠ ∅ — and so falls outside Hohfeld's whole net; it is at most cognate to immunity, not homologous, for immunity presupposes another party while a basic right awaits none. The constraint on others is carried not by duty but by the violation-reading and its consequences: to cross is to violate (P5 F21) [6], and the aftermath is a restorative act, the re-registration of the recognition layer, the withdrawal of power's support — not the discharge of a duty.

Western rights theory has a proposition close to an axiom: correlativity. Every right corresponds to a duty; the two entail each other. That I have a right means someone owes me a duty; that a right is violated means a duty is breached. Hohfeld made this the micro-grammar of legal relations: a claim-right and a duty are a correlative pair, and to have the one is to have the other [13]. From P1 through P6 the main series never took up this axiom directly; this paper takes it up.

To take it up one must first be clear about the level at which the paper argues. SAE inherits Kant, and also develops Kant — unfolding Kant, in the twenty-first century, with a finer structural analysis. Kant said that the human being is an end, and set up a regulative ideal, the Kingdom of Ends (Reich der Zwecke) [18]: all rational beings as ends to one another, each legislating and each subject to a common legislation. SAE gives that ideal a structural depiction: the checking-network of the 15DD transparency-state. When a subject's rights unfold to the crown of the spectrum, it enters a state of being unavoidably open to the whole field (P1 §V) [2]; in a centerless, distributed checking, each node checks whether the parties still acknowledge one another as ends at the 15DD position (P2) [3]. That each subject is an end to the others at 15DD — this is the Kingdom of Ends, only turned from a regulative ideal into a structural description. This paper speaks of rights, of the rights-structure of the subject as an end, and argues at just this level — as Kant, speaking of the human being as an end, spoke of the normative structure of the human being as a rational being, adding for it no empirical presupposition. (The moral law series stratifies in the same way: its earlier papers argue within the pure 15DD ideal type, while the mixed reality of 14DD growing toward 15DD is handled separately [11].)

At this level there is exactly one duty, and it faces inward. SAE replaces ethics and morality with one's own law [8]. Morality is the law a subject generates from within, at its own DD level, valid only for itself; ethics pushes that inner law outward into a demand upon others, a structurally illicit overreach. So SAE has no horizontal duty, no duty a subject owes to others. What SAE has is one vertical duty, the termination lemma of the axiom: that one cannot-not legislate only for oneself, the subject's answerability to its own law. This duty faces inward; it is autonomy.

Inward — and yet the other is not shut out. For in the Kingdom of Ends a subject as an end has an own law that already looks out over the whole field, that already includes the other. In the checking-network of the 15DD transparency-state each node acknowledges the others as ends at the 15DD position, and this acknowledgment is written into its own law, not added from without. The first theorem of the moral law has already fixed this: a 15DD subject cannot-not acknowledge a specific other as an end, and read from the symmetric side, it halts before the other's basic rights (P2) [9][3]. So that a subject cannot-not oppose the suppression of any subject's remainder — this law is given to itself, a matter internal to its self-legislation as an end, because to suppress the other's remainder violates precisely that acknowledgment, in its own law, which includes the other. The constraint faces inward, and inward the law already includes the other. This is no contradiction; it is the structure, in the Kingdom of Ends, of a member who legislates and is subject at once, put more finely. And so there is one thing to be distinguished: that a subject, at some moment, does not acknowledge a specific other as an end is not a 15DD subject in breach of a duty; it is a subject not operating at 15DD in that moment [9]. Restraint is not a breachable obligation; it is the difference between operating within the Kingdom of Ends and not.

And so the very thing the correlativity axiom asks for is, here, superfluous. It asks for a horizontal duty, my duty toward you, as the correlate of my right. But in the Kingdom of Ends the constraint is already internal to each member's inward law; each member is subject to the law it, as an end, has itself legislated, a law that includes the other; there is no need to erect a further layer of duty toward others. A horizontal duty is not something SAE deletes; it is a layer the Kingdom of Ends never needed — Kant's Kingdom of Ends already runs not on external duties among its members but on the fact that each member's self-legislation is already universal, and SAE makes this plain with the structure of 15DD. So the horizontal duty the correlativity axiom asks for has no object: SAE's duty faces entirely inward, and matches no other's claim.

This is not to set SAE wholly against Hohfeld. Of Hohfeld's four pairs, SAE claims one — power and liability. A recognition-based right corresponds to power (P4) [5]; one who holds a recognition-based right faces, across from it, a liability held by the one whom the power converges upon, who is subject to that power's demand and, failing to exit, bears it (P4 F18) [5] — and a liability is precisely not a duty, a distinction Hohfeld himself drew. What SAE does not claim is the pair claim and duty, for on this side SAE has no horizontal duty. As for basic rights, they are not in Hohfeld's net at all. Hohfeld's four pairs are all positions between two parties, while a basic right is pre-relational: a subject alone — the sole subject in the universe — still has its basic rights, because the core's inviolability comes from the operational fact ρ ≠ ∅, not from the other party's lacking power (P1) [2]. A basic right is at most cognate to Hohfeld's immunity, not homologous: immunity is relational — I have immunity if and only if you have a corresponding disability — and presupposes another party, while a basic right awaits none. So a basic right is not even an immunity; it falls outside Hohfeld's whole net of relations. The Jurisprudence series' dialogue confirms this from the other side, mapping immunity to the region outside law's range (Jurisprudence supplement §6) [12].

The three positions can be seen more clearly in concrete cases. For the pre-existence of basic rights, the Nuremberg trials are a cold example. Nuremberg refused defenses of the "it was lawful at the time under domestic law" and "I was only following orders" kind; it asserted that certain violations, even where lawful under the positive law of the time, were still wrong. On SAE's reading this is not an appeal to a higher natural law, but the claim that the basic rights violated are ontologically prior [1], anchored in ρ and in encapsulation, not manufactured by positive law and awaiting no in-presence other's duty — so that positive law's counting them lawful does not make them other than a violation. This walks a structural third road, neither reducing rights to the positing of positive law nor grounding them in a natural-law endowment (following Jurisprudence supplement §7). For the side where a recognition-based right corresponds to power, odious debt is a cold example. After a dictatorial regime falls, the new regime and people may declare the odious debts contracted by the former regime void; what is at work here is not a duty laid on creditors but a power to alter a systemic legal state, and the recognition-based right here just is that power, with the creditor across from it holding a liability, not a duty owed to anyone. The paper takes from these two examples only their structural position, adjudicating neither every point of the Nuremberg trials nor the specific attribution of any particular debt — that lies on another axis.

There is a further position SAE Rights Theory acknowledges, a concept related to but distinct from a right: entitlement. Rights come in two layers, basic and recognition-based (P1) [2]; entitlement is related to rights but is neither of those two layers — it is another position. Entitlement is not a basic right: a basic right is pre-relational and stands with a single subject, while entitlement requires another's giving — without the other there is no such entitlement — and so is relational. Entitlement is not a recognition-based right either: a recognition-based right corresponds to power, its holder acknowledging the other's power to converge upon it, while entitlement acknowledges none of the other's power — the moment the receiving end acknowledges the granting end's power, entitlement is elevated into a recognition-based right, which is another matter. What entitlement corresponds to is another's giving: a subject gives, and the other end holds the position that the giving points toward. That a person has an entitlement to obtain another's knowledge is one instance — without the other there is no other's knowledge, so it is not basic; and to obtain knowledge one need not acknowledge the other's power, so it is not recognition-based; what it corresponds to is the other's giving of knowledge. The granting end of an entitlement must be a subject; the receiving end need not — it may be a subject, or a quasi-subject, an animal, or even a non-subject, non-living whole such as a stretch of sea, so long as some subject is giving. And here one thing should be marked: the receiving end is not a subject that owns an entitlement; it is the end that the subject's giving points toward. The structure of entitlement is anchored in the giving of the granting end, not in the receiving end; so to say a stretch of sea holds an entitlement-end is not to say the sea has acquired subjecthood — only that some subject gives to it, and it is what that giving points toward. And this position, too, corresponds to another's giving, not to another's duty: giving is the other's act, not a debt the other owes anyone; the other may give, and may also not give. So the positions the SAE Rights Theory architecture can name are complete: a basic right with no correlate, a recognition-based right corresponding to power, entitlement corresponding to another's giving — and a duty sits in none of these three, so the horizontal duty the correlativity axiom asks for has no place in the whole picture. These three are the positions the architecture can name at present; the paper does not claim they exhaust every possibility, and the future may yet find new positions. But one thing can be settled: whether or not positions are added later, a duty will not be any of them, because a duty is an ethical concept, belonging to the layer replaced by one's own law, and outside the range of the 15DD one's own law. As for interest — that is a concept of another range, belonging to 14DD law, not to the 15DD one's own law, and not in this picture; more on this when interest theory is engaged.

The phase transition by which an entitlement is elevated into a recognition-based right has, in trust law, a cold example: the unborn beneficiary. A trust may reserve a property position for a future beneficiary not yet born, not even yet conceived; before the beneficiary is born it has no present will, and no existing body and therefore no existing interest, so will theory and interest theory are both awkward here, and yet it already holds a position. This position is, for now, an entitlement; it corresponds to the giving of the settlor and of the trust system that follows the will, and the beneficiary's own end has made no investment. After the beneficiary is born, it has the position of a negation-gate, and a trust often carries conditions — the trust system follows the testament, and a testament may attach conditions — and this attached condition is a convergence of power: the beneficiary enjoys only on satisfying the condition. Here there are two roads. If the beneficiary does not turn back this convergence with its negation-gate, and accepts the benefit, that entitlement is elevated into a recognition-based right, corresponding precisely to the trust system's conditioned, converging power. Or the beneficiary may refuse: since the attached condition is a convergence of power, the beneficiary who refuses that power, who does not accept the condition, exits the relation and no longer enjoys the right — and this is just the exit right (P4 F18) [5], the one bound holding a liability, subject to the conditioned power's demand, bearing it if it does not exit, and no longer holding the right if it does. Nowhere in the whole chain is there a horizontal duty: the beneficiary owes no duty to the trust, and the trust system owes no duty to the beneficiary; there are only entitlement, the convergence of power, and the exit right. This single case strings the three positions of the rights architecture together dynamically: how a position transitions from entitlement, which corresponds to another's giving, into a recognition-based right, which corresponds to power, and how it exits by the exit right. The paper takes from this only the structure; the position of the unborn beneficiary is no claim whatever about the moral status of the unborn.

To have no horizontal duty is not to have no constraint upon others. The constraint is carried not by duty but by structure. A subject that crosses another's encapsulation boundary has not breached a duty toward the other; it has triggered a reading of violation (P5 F21) [6] — and, on the side of its own law, this crossing violates its inward law that includes the other. Violation has its structural consequences: a restorative act that restores the over-drawn encapsulation dimension to a registrable state, the re-registration of the recognition layer, the withdrawal of power's support — and these are not the discharge of a duty; they are the aftermath of the violation-reading. Everything the tradition says with "duty" finds its correlate in structure. Where the tradition says A has a duty of non-interference toward B, SAE says B's encapsulation boundary is violated if A crosses it. Where the tradition says A has a duty of provision toward B, SAE says B holds a recognition-based right corresponding to an institutional power, and A is the bearer of that power; failing to provide, the power's demand goes unmet (read on the bound one's side as liability, not on A's side as duty). Where the tradition says A has breached a duty toward B, SAE says a reading of violation is triggered, or the recognition layer fails. Where the tradition says A must compensate after a violation, SAE says: a restorative act. What "duty" can say, the language of structure can say too, and more precisely — this is SAE's constant practice: in dismantling a traditional concept it never merely subtracts, but always supplies a replacement in structure.

The Jurisprudence supplement, in its dialogue with Western legal philosophy, once wrote, in discussing Hohfeld and in Hohfeld's own language, the line "I have a right not to be suppressed; you have a duty not to suppress." This does not conflict with the present paper. That paper speaks of law, at the level of the two-party law, giving the correspondence of legal relations in Hohfeld's language, and it says of itself that this is a correspondence, not a rewriting in SAE's terms (Jurisprudence supplement §6) [12]. In SAE's own terms, "you have a duty not to suppress" just is "if you cross, you violate (P5 F21), and the actor has violated its own inward duty that cannot-not oppose the suppression of any remainder." And the levels differ: the Jurisprudence supplement, at the 14DD-collision level of the two-party law, uses Hohfeld's language for a jurisprudential-side correspondence, while this paper, at the rights-theory level, makes plain that SAE's one duty is inward and vertical, with no horizontal one. So the paper's citing of the supplement is a continuation, not a quarrel — it shows the correspondence (Hohfeld's language, the jurisprudential layer), the paper shows the ontology (SAE's one vertical duty). One might ask: since the Jurisprudence supplement used the word "duty," does it not contradict this paper's no-duty? It does not. That is a correspondence-description in Hohfeld's language at the jurisprudential layer; the use of Hohfeld's word does not mean SAE's ontology holds duty as a primitive. The two papers are two languages for one structure, not two competing claims. Also: the Jurisprudence supplement §6 maps immunity to the region outside law's range, which resonates with this paper's pre-relational basic right that falls outside the net.

One rejoinder is worth taking in hand first: since SAE claims that a recognition-based right corresponds to power, is not power's convergence a duty? It is not. Power is a conditional convergence — only upon actual convergence, only upon the specific one it converges (P4 F15) [5]; a duty is a universal mapping — one claim answered by everyone's one duty. The one is conditional, lodged in a specific relation; the other is unconditional, spread toward all — and this difference is structural, not verbal. And there is a temporal layer: a right is ontologically prior (P1) [2], prior to any remedy. The tradition readily mistakes the symmetry of post-violation remedy for the ontological symmetry at the right's origination — as if right and duty were born together, each the other's ground. But in SAE the right is prior, the violation is a later matter, the remedy a later matter still; no horizontal duty is born together with the right.

The correlativity axiom cannot find the horizontal duty it asks for, because SAE's duty faces entirely inward. With this settled, the two finer traditions on the subject-side of correlativity should be handled. They do not assert correlativity directly, yet each hangs the holding of a right on something SAE would relocate: will theory on the exercise of choice, interest theory on a protected interest. First will theory.

II. Will Theory Grounds Rights in Choice

> M2. Will theory grounds rights in choice. Will theory (Hart, and Steiner and Sumner) [14] grounds rights in the protected exercise of choice: the right-holder may assert or waive the correlative duty and is a small-scale sovereign — with the consequence it faces that children, the severely incapacitated, and the comatose fall outside the holding of rights. What this grasps is the assertable-and-waivable capacity of a recognition-based right (the exercise of a corresponding power), which it mistakes for the holding-condition of all rights. SAE grounds rights in standing (13DD+), not by the capacity for choice: the exercise of choice belongs to the bandwidth of exercise, and standing is not bandwidth (P2 F6) [3]. Sleep is the internal reductio of this mistake: an adult asleep cannot exercise choice, and by will theory's letter holds no right, yet will theorists will not accept that a sleeper may be disposed of at will — sleep is a temporary interruption of the bandwidth of exercise, not an exit; standing is maintained, and the criterion is not whether one is at this moment awake, speaking, or reflecting, but whether the load-bearing structure still maintains this subject's 13DD+ holding-position (P5 F19) [6]. A child is the same: standing equal, bandwidth yet to unfold, a future subject (a quasi-subject, subject-to-be; its full theory left to a dedicated paper). That a sleeper, doing nothing and asserting nothing toward the whole world, has its basic rights intact confirms in turn that basic rights are pre-relational, hanging on no present act, assertion, or choice, anchored in standing and ρ, ontologically prior.

Will theory hangs rights on the exercise of choice, where interest theory takes another road. To have a right, on will theory, is to have a protected choice; the right-holder may assert or waive the duty of which the right is the correlate, and is a small-scale sovereign (Hart). This picture grasps something real — but what it grasps is the recognition-based side.

What it grasps as real is that a recognition-based right does carry a certain assertable, waivable capacity. One who holds a contractual right may assert it, and may waive it, may set a questioning in motion, may withdraw an assertion — this is the exercise of a corresponding power, the true look of a recognition-based right at the operational level. Will theory sees this capacity keenly.

The challenge will theory faces turns on where that capacity is placed. What it grasps — the assertable and waivable — is the capacity of a recognition-based right at the operational level; the moment that capacity is placed in the position of the holding-condition of all rights, the challenge arrives: whoever cannot exercise choice falls outside rights. This position sets children, the severely incapacitated, and the comatose all outside the holding of rights, because they cannot exercise that choice. Some will theorists accept this conclusion, preferring to let the theory exclude children rather than change the criterion for grounding rights — the well-known price will theory pays for its criterion of choice.

But there is one case will theorists themselves will not accept. An ordinary adult falls asleep; at this moment it cannot exercise any choice; by will theory's letter it holds no right while asleep. No will theorist will accept the conclusion that follows — that the sleeper may be disposed of at will. This is no challenge from outside; it is a contradiction internal to will theory: its letter demands the exclusion of the sleeper, its adherents never exclude. Sleep splits what will theory says from what it acknowledges — it says choice, and acknowledges something else. And that something else has a name in SAE.

A will theorist might try to save the case, saying the sleeper has rights because on waking it can exercise choice. But this rescue at once recoils on will theory itself. For a child too can exercise choice once grown, and the comatose too once cured; if potential exercise were enough for holding a right, then children and the comatose would both hold rights — and the exclusion of children is will theory's most emblematic position. So facing sleep, will theory has only two roads, both losing. If it holds to present exercise, the sleeper has no right — which it will not itself accept. If it retreats to potential exercise, children and the comatose both have rights — and its own emblem collapses. And whether the potential exercise can finally occur — the criterion for that is precisely whether the load-bearing structure has maintained this subject's holding-position (F19) [6]. Once will theory retreats to potential exercise, what it acknowledges is already standing, not present choice; only it still wears will theory's name and excludes children, and so contradicts itself.

SAE reads the case thus: sleep is a temporary interruption of the bandwidth of exercise and of operation, not an exit. The subject is still maintained as a recoverable 13DD+ load-bearer; standing has not moved at all (P5 F19) [6]. P5 has already fixed the criterion: it is not whether one is at this moment awake, speaking, or actually reflecting, but whether the load-bearing structure still maintains this subject's 13DD+ holding-position. Rights are grounded in standing; standing is there, the right is there — the sleeper's rights are intact. That something else will theory acknowledges is precisely standing; it says choice with its mouth and acknowledges standing with its hand. Standing is not the capacity for choice; it is the position a subject as an end holds — a subject falls asleep, and its position as an end does not vanish with it.

The child is another face of the same thing. Will theory excludes the child because the child cannot exercise that choice — but what the child lacks is not standing; it is the bandwidth of exercise, yet to unfold. Standing and bandwidth are two things (P2 F6) [3]: standing is the position a subject as an end holds, bandwidth is how far it can at present exercise. The child's bandwidth is still unfolding; its standing is already equal. This series calls an existence like the child a quasi-subject (subject-to-be): a future subject, holding the same philosophical standing as a present subject, holding its own basic rights (P1) [2]. A quasi-subject is not a subject holding fewer rights; it is the same subject as an end, in another configuration of rights along the way of its coming-to-be. Its full theory is left to a dedicated paper; here, only at the point of engaging will theory, the boundary between standing and bandwidth is brought out.

This boundary between standing and bandwidth has a contemporary cold example in the Convention on the Rights of Persons with Disabilities [29]. The Convention turns the way of seeing disability from a medical model of individual deficit toward a human-rights model, and its core insight is that disability comes largely from the exclusions of social arrangement and environment, not from the individual's own deficit. On SAE's reading this is precisely the boundary between standing and bandwidth: a subject's 13DD+ standing is not lowered because some physiological bandwidth of it is constrained; the reasonable accommodation the Convention asks for is the making-reachable-again of a bandwidth the environment has blocked, not the supplying to the subject of a standing it already equally holds. The paper takes from this only the structure, adjudicating no particular accessibility arrangement.

Sleep and the child, used together, each strike will theory at one point. The child is an exclusion will theory can grit its teeth and accept, and it shows the parting of SAE from will theory on the ground of grounding rights — one grounds in choice, the other in standing. Sleep is an exclusion will theory will not accept yet by its letter must, and it shows will theory's internal contradiction — the letter demands exclusion, the adherents do not. The first draws clear the fork of the two roads; the second lights the fire inside will theory. The two examples together give the point that rights are grounded not in choice but in standing both the contrast of the parting and the reductio from within.

Sleep also confirms, in passing, one thing from §I. A sleeper, doing nothing, asserting nothing, exercising nothing toward the whole world — a subject purely at rest — has its basic rights intact all the same. This shows that basic rights hang on no present act, assertion, or choice of the subject; they are anchored in standing and ρ, ontologically prior (P1) [2]. §I said basic rights are pre-relational, that a subject alone still has its basic rights — that was a thought-experiment; the sleeper is the real-world version of that experiment, needing no supposition that the universe holds only one person: a single sleeper is already a subject at rest from all act and relation, with rights intact.

This also meets another tradition in rights theory. Feinberg hangs the value of a right on the activity of claiming [21]: to have a right is to be able to claim, and the act of claiming itself gives the holder a dignity. SAE reads that claiming as questioning — a subject, on the Shi spine, brings forward a position not yet registered and demands acknowledgment; and whether this road settles into a recognition-based right hangs on whether it passes T5 registration, not on the act of claiming itself. Basic rights await no claiming at all: a sleeping subject claims nothing, and its basic rights are intact all the same. So the activity of claiming is, in SAE, the road of questioning that leads to a recognition-based right, not the condition of holding a right; the weight Feinberg grasps in claiming is relocated onto the structure of questioning and registration. The paper takes from this only its structural position.

Here one must cut cleanly from exit. Sleep is a recoverable interruption of bandwidth, standing intact; exit is the load-bearer's falling below 13DD, standing lost (P5 F19) [6] — the two differ. Permanent incapacity is a harder case: if a subject is still maintained by the load-bearing structure as 13DD+, its standing is there, its right is there; if it falls below 13DD, that is an exit, the right ending for want of a load-bearer — but that is no-one holding, not someone denied a right for want of will, another matter than will theory's exclusion. This layer the paper reads only structurally, as a subject's structural situation; which dispositions ought or ought not to hold lies on another axis, and the paper does not adjudicate it.

Will theory hangs rights on choice; sleep and the child together show that rights are grounded in standing, not choice. Interest theory takes another road: it hangs rights on a protected interest, includes children, yet reads the right-holder as a passive beneficiary of the interest.

III. Interest Theory Grounds Rights in a Protected Interest

> M3. Interest theory grounds rights in a protected interest. Interest theory (Raz, MacCormick, Kramer; ascending to Bentham) [15][16][17][26] grounds rights in a protected interest: X has a right if and only if some aspect of X's well-being, an interest, is a sufficient reason to hold others under a duty (Raz 1986) [15]; it can bring children into the class of right-holders (MacCormick took children as a test-case for will theory), unlike will theory's exclusion of children. It faces three challenges. First, it grounds rights in interest, but interest is a 14DD legal-range concept, and the 15DD one's-own-law range has no such concept; the source of a right is in standing, not in interest. Second, it reads the right-holder as a passive beneficiary, while SAE's quasi-subject is an active future subject, holding its own basic rights, standing equal, bandwidth yet to unfold — not a beneficiary. Third, interest is broad — artifacts and works of art too have interests one can speak of — so interest theory must add a threshold like ultimate value to keep trees and artifacts out of the class of right-holders (Raz) [15]; SAE needs no such patch: an animal has no right yet has an entitlement, entitlement lying in the 15DD range and corresponding to another's giving, its receiving end needing no subjecthood, while interest lies in the 14DD legal range — the distinction of ranges together with the position of entitlement dissolves interest theory's animal-and-tree difficulty. The capabilities approach (Sen, Nussbaum) [22][23] is the chief neighbor on the quasi-subject question; the picture of thick-layer unfolding it draws neighbors SAE's thick layer plus bandwidth of exercise, but it presupposes that picture as a flourishing endpoint to be reached, a teleology that parts ways with SAE's open endpoint (X2, F24) [7] — here is a boundary-drawing, not an elimination.

Will theory hangs rights on the exercise of choice; interest theory takes another road. It grounds rights in a protected interest: a right is the protection of an interest important enough to place others under a duty (Raz 1986) [15]. This road has a strength will theory lacks — it can bring children in. MacCormick took children precisely as a test-case for will theory: children evidently have rights, and will theory by its letter would exclude them, so interest theory says children have interests worth protecting, and therefore have rights. On this point of including children, interest theory stands more steadily than will theory.

The challenge interest theory faces turns on the very word it grounds rights in: interest. In SAE, interest is a concept of the 14DD legal range, not of the 15DD one's-own-law range. The three positions SAE Rights Theory speaks of — basic rights, entitlement, recognition-based rights — all lie in the 15DD one's-own-law range, and none is called interest, because interest is not a thing of this range; it belongs to 14DD law, the layer law can take in and protect. Interest theory grounds rights in interest, taking a concept of the 14DD legal range to ground a thing that ought to be grounded in the 15DD one's-own-law range. SAE reads the source of a right as lying in standing, not in interest: a subject as an end holds its basic rights because of the standing it, as an end, occupies — not because it has an interest worth protecting.

There is a further, inner reason interest belongs to 14DD: investment. Interest, in the SAE frame, is a concept of input for return — you invest a cost and exchange it for a benefit, and this structure of input-for-return is precisely the logic of the 14DD layer of collision and exchange, where the two-party law of the Jurisprudence series argues. And of the three positions in the 15DD one's-own-law range, none is input-for-return: a basic right is by nature there, a subject as an end occupying standing without investing to hold it; an entitlement corresponds to another's giving, its holder making no investment; a recognition-based right corresponds to power's demand, a demand that is likewise not a return bought by investment. So interest and the three positions of rights theory are divided by the yardstick of investment: interest is input-for-return, and those three are not. Two things must be set down here, lest they be read askew. First, interest belonging to 14DD is a division of labor, not a demotion: input-for-return is a legitimate and real structure of that 14DD layer, only not handled in the 15DD range of rights theory — as, with the traditions, it is honoring what they grasp and relocating their source, so with 14DD it is acknowledging it as the legitimate structure of that layer without dragging it into this one. Second, the series argues from the 15DD view, but does not forbid a subject from doing 14DD things: a subject as an end may invest, exchange, and gain all the same — investing in shares for a return, say, is beyond reproach. So the challenge interest theory faces is not that a subject pursues interest — of course it may — but that interest theory takes a 14DD concept of exchange to ground a right that belongs to the 15DD one's-own-law range: it has grounded the source of the right at the wrong layer.

Interest theory also reads the right-holder into a passive position. In its picture the right-holder is the one who enjoys a protected interest, and a right is what it passively enjoys or is spared the loss of. Applied to the child, this reads the child as a passive beneficiary, an object having an interest worth protecting and therefore protected by others' duties. SAE reads the child otherwise. SAE's quasi-subject is an active future subject; it holds its own basic rights, its standing equal to a present subject's, only its bandwidth of exercise yet unfolding (P2 F6; P1) [3][2]. The boundary of standing and bandwidth (§II) works again here: the child is not an object holding a lesser interest and so lesser protection; it is the same subject as an end, its bandwidth yet unfolding along the way of its coming-to-be. A passive beneficiary and an active future subject are two readings; interest theory stops at the former, SAE gives the latter. Interest theory's most awkward point was touched already by §I's unborn beneficiary of a trust: an unborn beneficiary, with no existing body and therefore no existing interest, gives interest theory almost nothing to work on, and yet it already holds a position — a position that, in SAE, is entitlement, corresponding to the settlor's giving, not a protected interest.

The place interest theory strains hardest is its boundary. Interest is broad: an artifact, a plant, a work of art can all be said to have some interest, some state better or worse for it. If to hold an interest is to hold a right, then artifacts and trees too should be right-holders — which interest theorists themselves do not accept. To bar this consequence, interest theory must add a threshold — Raz uses ultimate value, Kramer a like criterion: only a being whose well-being is of ultimate value, or an artificial legal person, counts as a right-holder. This threshold is added from without; it does not grow from interest itself, but is patched on to keep trees and artifacts outside the door.

The same difficulty SAE dissolves not by a patch but by the distinction of ranges together with the position of entitlement. Does an animal have rights? SAE says no: an animal does not occupy 13DD+ standing and holds no basic right, and it is not in a relation of acknowledging power and holds no recognition-based right. But an animal, in the 15DD context, has an entitlement, corresponding to another's giving: a 15DD subject's care for and non-injury of an animal is the subject's giving, and the animal holds the entitlement-end that giving points toward. The receiving end of an entitlement need not be a subject, so an animal can hold an entitlement-end without being a subject; a stretch of sea is the same — the fishery-and-ecology protection of that stretch, the catch limits, the closed seasons, the fostering of fish stocks, is the giving of the subjects who protect, and that stretch of sea, as a non-subject, non-living whole, holds the entitlement-end that giving points toward. Here one takes only this structure, that a non-subject whole can be the receiving end of an entitlement, because what makes this an entitlement is the giving on the subject's side, not the standing of the receiving end; as for whether some stretch of sea ought to be protected, or how some fishery policy ought to be set, that lies on another axis, and the paper does not adjudicate it. So what interest theory must handle with an ultimate-value patch — the tree and the artifact — falls clearly here in SAE: non-subjects such as animals and stretches of sea hold no rights, yet may hold entitlements, while interest itself belongs to the 14DD legal range, taken in by law, needing no crowding into this layer of rights.

On the quasi-subject question there is, beyond interest theory, a nearer neighbor: the capabilities approach (Sen, Nussbaum) [22][23]. It is worth facing directly, for what it draws neighbors SAE. It asks what a subject can really do and be; Nussbaum lists a set of central capabilities — life, bodily health, bodily integrity, senses and thought, emotion, practical reason, affiliation with others, and so on — as a floor of human dignity. This set neighbors SAE's thick layer plus bandwidth of exercise: both speak of how far a subject can operate, and neither reduces a subject's operation to a quantity of resources. In drawing out a subject's operation, the capabilities approach sees finely. Its parting from SAE is at the endpoint. The capabilities approach sets the full unfolding of that set of capabilities as a flourishing state to be reached, an endpoint-picture with a content and a form — an Aristotelian teleology. SAE's additive path has a direction but no such presupposed endpoint; the unfolding of the spectrum is necessary in direction and open in endpoint (X2, F24) [7], and SAE sets no flourishing-picture as a constituting condition of rights. So here is a boundary-drawing, not an elimination: the operation the capabilities approach draws out SAE grants is truly drawn; it only cuts the flourishing-endpoint presupposition apart from SAE's open endpoint. This parting is chiefly on Nussbaum's side: her list of central capabilities is a relatively fixed flourishing-picture, in greatest tension with X2's open endpoint, and it is at just this fixed list that the teleological boundary-drawing aims. Sen's side is in fact nearer SAE: he explicitly opposes locking to a fixed list of capabilities, his capability-space is open, more a neighbor of the bandwidth-of-exercise and real-opportunity layer; this boundary-drawing is light on Sen and heavy only on Nussbaum. Neither side supplies SAE's source-criterion; the source of a right still lies in standing, not in a list of capabilities.

Read this far, the two readings of the quasi-subject phenomenon are clear. The capabilities approach reads a quasi-subject as an existence whose capabilities have not reached the flourishing threshold; interest theory reads it as a beneficiary with an interest worth protecting; SAE reads it as an active future subject with standing already in place and bandwidth yet to unfold. Three readings point at one phenomenon, their criteria different; SAE's criterion falls only on the structure of standing and bandwidth, not on a flourishing endpoint, and not on the enjoyment of an interest. Interest theory and the capabilities approach have each grasped one real facet of this phenomenon — the ground of protection, and the unfolding of operation — and SAE relocates each facet onto its structure, while leaving the source of a right in standing. The traditional classification of negative and positive rights, §IV handles.

IV. The Negative-Positive Distinction Cuts by Duty-Form; SAE Redraws by Component

> M4. The negative-positive distinction cuts by duty-form; SAE redraws by component. The traditional distinction of negative and positive rights cuts by duty-form: a negative right asks others' non-action, a positive right asks others' action, the first-generation civil-political rights read as negative, the second-generation economic-social rights as positive. What this grasps is a real difference in the form of power's demand, convergent versus supplying. The challenge it faces is that this classification cuts by duty-form, while SAE has no duty, so a classification by duty-form does not stand in SAE. SAE redraws along the source-constitution axis (P3 F12) [4] together with limit and violation (P5): whether a right falls in the basic region or the recognition region is fixed by the standalone test, binary, not by whether it asks action or non-action. A positive right such as housing or education has its basic component judged by the standalone test — not "resource-important, therefore basic" (P3 F12 has fixed this: livelihood and bare subsistence and such time-varying resource-names cannot define the basic) — and its recognition component corresponds to an institutional power (P4 F15) [5], the government being the bearer of that power, so that the so-called government duty is that power's demand, not a duty; the side of it that asks another's giving without engaging power may fall to entitlement. A negative right such as expression is a basic component plus a reading of limit and violation. The second-generation economic-social rights are read here as a synchronic division of components; their diachronic emergence is read in §V — the same instance, on different axes.

The classification of negative and positive rights is a much-used cut in rights theory. A negative right asks others' non-action — not interfering with your speech, not interfering with your association; a positive right asks others' action — giving you healthcare, giving you education. The first-generation civil-political rights are mostly read as negative, the second-generation economic-social as positive; after the UDHR [27] the two Covenants set these two out separately, the Covenant on Civil and Political Rights and the Covenant on Economic, Social and Cultural Rights [28]. This classification grasps a real difference: the form of power's demand differs — one is convergent, asking the other to stay its hand, the other supplying, asking the other to give.

The challenge it faces turns on what it cuts by. This classification cuts by duty-form: a negative right corresponds to a duty of non-action, a positive right to a duty of action, and the root of the classification is the form of duty. And in SAE there is no duty (§I), so a classification by duty-form has nowhere to land. So SAE does not cut along the negative-positive axis — not to say that the difference this classification sees is unreal, but that the yardstick it cuts with, duty-form, is not in SAE's picture.

SAE redraws along another axis, the source-constitution axis (P3 F12) [4]. Whether a right falls in the basic region or the recognition region is fixed by the standalone test — withdraw all in-presence others, and if this determinate claim still stands it is basic, if not it is recognition, the belonging binary, turning at the threshold. This axis asks where a right comes from, in-generated or constituted, not whether it asks action or non-action. Together with limit and violation (P5), this redrawing can set what the negative-positive classification saw back into SAE's structure.

A positive right must be taken apart along this axis. Housing and education, for example, are often read whole as positive rights, while in SAE such a right is to be split by component. Its basic component is judged by the standalone test, not by how important a resource it involves — P3 F12 has fixed this point: one cannot take livelihood and bare subsistence and such time-varying resource-names to define the basic, because belonging to the basic region is fixed by the standalone test and the type-criterion, which do not move with the times. So if education has within it some dimension that touches the unfolding of a subject as an end, the basic component there stands by the standalone test; and education as an institutional provision is a recognition component, corresponding to an institutional power (P4 F15) [5], the government standing as bearer of that institutional power, in the position invoked and converged by this right. Where the tradition says the government owes a duty to this right, SAE says that is the demand of the institutional power this recognition-based right corresponds to, the government being the bearer of that power, not a duty-bearer. This is not to say the demand is unreal or unenforced — it may be very real, very enforced; only its structural position is the demand of a power, not a horizontal duty. And within this provision, the side of it that asks another's giving without engaging the acknowledgment of the other's power falls to entitlement, corresponding to another's giving. A negative right such as expression is a basic component plus a reading of limit and violation: the dimension of expression it touches is a basic component (P2) [3], and what acts on it is read by limit and violation (P5) [6].

So redrawn, the negative-positive classification is taken apart into several SAE things: a right's basic component and recognition component, the institutional power the recognition component corresponds to, the entitlement that asks another's giving, and the reading of limit and violation. A single "positive right" of the tradition may in SAE hold at once a basic component, a recognition component corresponding to an institutional power, and an entitlement; it is not the establishment of a new kind of duty but the coexistence of these several structural components. The emergence of the second-generation economic-social rights is read here as this synchronic division of components; how they emerge along the axis of time is another matter, handled in §V — the same set of second-generation economic-social rights, read here for its component-constitution and there for its diachronic emergence, two readings on two different axes, not laid into one line.

V. The Diachronic Unfolding of the Rights Spectrum Is the Necessary Dynamics of Any Subject-Society

> M5. The diachronic unfolding of the rights spectrum is the necessary dynamics of any subject-society. The unfolding of the rights spectrum in time (P6 F23) [7] is no accident of human history; it is the necessary dynamics of any subject-society — any subject-society in which there are multiple subjects, a shared field, chains of questioning, and recognition-positions, where individuals as ends open new dimensions along the additive path and their historical aggregation is rights-expansion, its spectrum not stalling into a static list. The direction is necessary while the endpoint is open (P6 F24; X2) [7]. Vasak's three generations [24] — first civil-political, second economic-social, third collective-solidarity — and the postwar human-rights regime (the UDHR and the two Covenants) are a human instance of these universal dynamics, not the dynamics themselves. The generations framework reads the expansion as a march of progress across three generations, a latent progress-teleology, which X2 bars; that the generations do not replace one another is precisely the confirmation that the spectrum is two-way at the level of particular rights while the additive path's direction is one-way and not historically monotone. The emergence of the second-generation economic-social rights is read here for its diachronic emergence, the recognition component newly added and the basic component made manifest — two readings, on two different axes, from §IV's reading of its synchronic component-division.

The four preceding sections were the structural view, asking on a single time-slice for the constitution of rights — how the three positions divide, why a duty has no place, what will theory and interest theory hang rights on, how the negative and positive are redrawn, all synchronic structural analysis. This section turns to the dynamic view, asking how this structure moves in time. This is not a change from one topic to another but a turn from a view that looks at structure to a view that looks at evolution; and the dynamic view still looks at the same structure's unfolding in time — what emerges in time is the recognition-based right as a structure, what opens a new dimension is the subject as an end as a structure, not the time of some other thing apart from structure.

The rights spectrum is not a static list; it unfolds in time, with new positions emerging and old ones receding. In Western rights theory the most-used framework for describing this diachronic unfolding is Vasak's three generations: the first, civil and political rights; the second, economic, social, and cultural rights; the third, collective and solidarity rights — the postwar human-rights regime spread along this line, with the Covenant on Civil and Political Rights and the Covenant on Economic, Social and Cultural Rights set out separately after the UDHR, and later various group and development rights. This framework grasps something real: the rights spectrum does expand in time, and is not fixed once and for all.

The challenge it faces turns on what it reads that expansion as. The word "generations" reads the three kinds of rights as a step-by-step ascending history — the second after the first, the third after the second, like a history of rights upgrading generation by generation. Hidden in this reading is a teleology, as if the history of rights moves toward an ever-fuller state of completion. SAE's anti-teleology gate (X2, F24) [7] is precisely what bars this reading.

SAE reads this diachronic unfolding as the dynamics of rights-expansion (P6 F23) [7]. A subject as an end, along the additive step of the additive path, opens a new dimension — this is its structural tendency as an end — and this opening, on the Shi spine, brings forward to the other a remainder not yet registered, which through questioning, through registration, through the three-fold interlock, settles into a workable recognition-based right; and the many such openings of many subjects, aggregating in time, are rights-expansion. So rights-expansion is not that list growing of itself; it is the aggregation in history of countless individual openings through chains of questioning. These dynamics are not peculiar to human history; they hang on the structure of the subject as an end, and wherever there are subjects as ends, there are this opening and this aggregation.

And so this section's proposition is universal: the diachronic unfolding of the rights spectrum is the necessary dynamics of any subject-society — any subject-society in which there are multiple subjects, a shared field, chains of questioning, and recognition-positions, its rights spectrum will not stall into a static list. But this necessity must be read rightly. It is a necessity on the side of direction, not on the side of history. The necessity of direction is that a subject as an end cannot-not open toward new dimensions, and this direction cannot be sealed into a terminal list — this is not revocable. History is otherwise: a subject-society may perfectly well, in some period, press particular rights back; what one age opens, another may press back (P3, P5 have fixed this two-wayness at the level of particular rights) [4][6]. So that any subject-society must develop says that its direction cannot be pressed into a terminal list, not that it in fact grows ever fuller and fuller monotonically. X2 constrains right here: direction necessary, endpoint open — that the spectrum has a direction does not mean the history of rights has a purpose.

The postwar three generations are read at just this position. They are a concrete unfolding, in human history, of these universal dynamics, an instance, not the dynamics themselves, still less a history of progress. That the generations do not replace one another — the second comes and the first does not withdraw, the third comes and the first two are not voided — a point critics have long made, is precisely the confirmation of SAE's non-monotone spectrum: the three generations are interlaced stratified labels in history, not a step-by-step ascending stair. The emergence of the second-generation economic-social rights is read, in this section, for its diachronic emergence: a batch of recognition-positions newly added in the postwar period, some basic components made manifest, made concrete, backed by institutions. This is another matter than §IV's reading of the same batch of second-generation rights — §IV reads its present component-constitution, how the basic and recognition components divide; this section reads how it emerges in time. One is the synchronic component, the other the diachronic emergence; the same batch of rights, two readings on two different axes, not laid into one line.

This postwar diachronic unfolding also brings out something that is not a newly-added right but a structural position: the plight of the stateless. A person who loses a state's recognition, fallen into statelessness, has recognition-based rights and workable remedies largely hollowed out, while its basic rights are not thereby lost — basic rights are ontologically prior, hanging on no community's recognition (§I). The right to have rights that Arendt pointed to [25] is, on SAE's reading, not some particular basic right but a meta-recognition position, the qualification to enter the recognition-network and obtain an assertable position. What the postwar human-rights regime tried to catch is just this meta-recognition position; what it makes manifest is not one more right but the very entrance a rights spectrum needs in order to unfold toward a subject at all. This too is part of the diachronic unfolding: what unfolds is not only one recognition-based right after another, but also such structural positions made manifest. The paper takes from this only the structure, adjudicating no particular state's refugee arrangement.

VI. Conclusion

The paper argues at the level of the subject as an end, taking up a proposition close to an axiom in Western rights theory — correlativity, that every right corresponds to a duty. The answer it gives is not that this right lacks that duty; it is that SAE Rights Theory has no concept of a duty toward others, only one inward duty, the subject's answerability to its own law (the termination lemma; SAE's sole axiom is Negativa). The horizontal duty the correlativity axiom asks for sits in none of the three positions SAE Rights Theory can name — basic rights, entitlement, recognition-based rights — and so the correlativity axiom has no object (M1). With this settled, the two finer traditions on the subject-side are each placed: will theory hangs rights on the exercise of choice and faces the challenge of sleep and the child, while SAE grounds rights in standing (M2); interest theory hangs rights on a protected interest, but interest is a 14DD concept of that market layer of input-for-return, and SAE grounds rights in standing, dissolving its animal-and-tree difficulty with entitlement and the distinction of ranges (M3). On the division-side, the classification of negative and positive that cuts by duty-form is, in a SAE that has no duty, redrawn along the source-constitution axis, and a government's duty is read as the demand of the institutional power a recognition-based right corresponds to (M4). On the time-side, the generations and postwar human rights are read as a human instance of the dynamics of rights-expansion, direction necessary and endpoint open (M5).

These five steps together are not a one-by-one refutation of these traditions; they take the real facet each has grasped and move it off the position of the source of rights, each to its own place. Hohfeld grasped the grammar of the legal-relations field; will theory grasped the exercisable side of a right; interest theory grasped the protecting-a-subject side of a right; the negative-positive grasped the form of power's demand; Vasak grasped the diachronic unfolding of rights — these are all real, only none of them the source of a right. SAE leaves the source of a right at the standing a subject as an end occupies, gives the correlate of a recognition-based right to power, leaves the kind that asks another's giving to entitlement — and what is called duty appears only in the tradition's language, read here as a violation-reading, the demand of an institutional power, or a restorative act, never made a primitive.

There is a tradition neighboring this paper's basic rights, worth a mention here: Dworkin's rights as trumps [20]. It says a right is an individual's trump against collective utilitarian goals, not to be sacrificed for the collective's good. This neighbors the inalienability of SAE's basic rights: a basic right too is not to be traded away for a collective interest. But the two ground differently: Dworkin's trump grounds in equal concern and respect, a political morality, while SAE's basic right grounds in a subject's own law, in ρ and encapsulation, the structure of a subject as an end. So the trump may serve as a neighboring theory against policy's annexation, and not as a source-theory of rights — consonant with this paper's moving of every tradition off the position of the source.

There is another tradition neighboring this paper's reading of the limit: Nozick's side-constraints [19]. It takes individual rights as side-constraints on others' action, not to be crossed for a better overall outcome, and this neighbors SAE's reading of a crossing of the encapsulation boundary as a violation: a boundary's being a boundary is not measured by how much good crossing it could bring. But the two still part: Nozick's side-constraints grow within a libertarian frame, where rights serve almost only as negative non-crossables, while SAE's basic rights have, beyond, the dimension of cultivation — a subject as an end is not only un-crossed but also opens new dimensions and grows in being questioned; so SAE is not libertarianism, and the side-constraint may serve as a neighboring reading, and not as the whole of SAE. The paper takes from this only that neighboring, adjudicating no state theory of Nozick's.

This paper stands outside the main line of SAE Rights Theory. The main line — from the source and spectrum of rights, to the thickness and boundary of basic rights, to the constitution, withdrawal, and transformation of recognition-based rights, to the limit and violation of basic rights, to the cultivation and diachronic unfolding of rights — P1 through P6 has already built, and the internal structure of the rights spectrum closes there. This is the meta-paper; it adds no new intra-spectral structure, only sets this structure against the terrain of Western rights theory, showing where each tradition falls in SAE's coordinates and where it oversteps the source. Two things, too, are left outside this paper. One is the full dynamics of entitlement — whether it can be violated, whether it can be withdrawn, its relation to T5 registration; the paper only sets entitlement as the third position in the rights-theory architecture, its full unfolding left to its own place. The other is the coming-to-be of the quasi-subject — its gate, its topology, its holding; a child as a quasi-subject is a future subject, of equal philosophical standing with a present subject, and its arrangements of care and the topology of its gate of coming-to-be are left to a dedicated paper.

One sentence to close the paper's source. SAE inherits Kant, and also develops Kant, unfolding Kant's Kingdom of Ends, in the twenty-first century, with a finer structural analysis. Kant said the human being is an end and set up a regulative ideal; SAE gives that ideal a structural depiction, the checking-network of the 15DD transparency-state, where each subject is an end to the others at 15DD and its own law already looks out over the whole field and includes the other. In such a Kingdom of Ends the constraint is already internal to each member's inward law, and no further layer of duty toward others is needed; the horizontal duty the correlativity axiom asks for then has no place. Rights presuppose no duty, because at the level of the subject as an end a duty has, from the start, nowhere to sit.

Acknowledgments

This is the meta-paper of the rights-theory branch of the SAE (Self-as-an-End) philosophical system, completed independently by the author. The drafting used a method of four-fold cross-review: an internal precision audit, a corpus-consistency audit, an ontological-limit-and-contradiction probe, and a structural gate with pre-release sign-off. The four-fold review markedly improved the paper's precision and consistency; the author records his thanks here, and bears sole responsibility for the text.

References

[1] Qin, Han. On What Rights Are Not. SAE Rights Theory, Prequel. Zenodo. https://doi.org/10.5281/zenodo.20726894 [2] Qin, Han. On the Source and Spectrum of Rights. SAE Rights Theory, Paper 1. Zenodo. https://doi.org/10.5281/zenodo.20743135 [3] Qin, Han. On the Thickness and Boundary of Basic Rights. SAE Rights Theory, Paper 2. Zenodo. https://doi.org/10.5281/zenodo.20769253 [4] Qin, Han. On the Constitution and Withdrawal of Recognition-Based Rights. SAE Rights Theory, Paper 3. Zenodo. https://doi.org/10.5281/zenodo.20993658 [5] Qin, Han. On the Interface and Transformation Between Recognition-Based Rights and Power. SAE Rights Theory, Paper 4. Zenodo. https://doi.org/10.5281/zenodo.21093939 [6] Qin, Han. On the Limitation and Violation of Basic Rights. SAE Rights Theory, Paper 5. Zenodo. https://doi.org/10.5281/zenodo.21212595 [7] Qin, Han. On the Cultivation and Diachronic Unfolding of Rights. SAE Rights Theory, Paper 6. Zenodo. https://doi.org/10.5281/zenodo.21252863 [8] Qin, Han. One's Own Law: SAE's Critique of Ethics and Morality. SAE Foundational Paper A16. Zenodo. https://doi.org/10.5281/zenodo.19037566 [9] Qin, Han. The Four Foundational Theorems of the Moral Law. SAE Moral Law, Paper 1. Zenodo. https://doi.org/10.5281/zenodo.20011018 [10] Qin, Han. Positive-Sum: The Ontological Ground of 15DD Mutual Chiseling. SAE Moral Law, Paper 5. Zenodo. (DOI forthcoming) [11] Qin, Han. Mixed Reality: The Intermediate State of 14DD Growing Toward 15DD. SAE Moral Law, Paper 6. Zenodo. (DOI forthcoming) [12] Qin, Han. A Posterior Dialogue with Western Jurisprudence. SAE Jurisprudence supplement. Zenodo. (DOI forthcoming) [13] Hohfeld, W. N. (1919). Fundamental Legal Conceptions as Applied in Judicial Reasoning. Yale University Press. [14] Hart, H. L. A. (1982). Legal Rights. In Essays on Bentham. Clarendon Press. [15] Raz, J. (1986). The Morality of Freedom. Clarendon Press. [16] MacCormick, N. (1976). Children's Rights: A Test-Case for Theories of Right. Archiv für Rechts- und Sozialphilosophie, 62(3). [17] Kramer, M. H. (1998). Rights Without Trimmings. In A Debate over Rights. Oxford University Press. [18] Kant, I. (1785). Grundlegung zur Metaphysik der Sitten. [19] Nozick, R. (1974). Anarchy, State, and Utopia. Basic Books. [20] Dworkin, R. (1977). Taking Rights Seriously. Harvard University Press. [21] Feinberg, J. (1970). The Nature and Value of Rights. Journal of Value Inquiry, 4(4). [22] Sen, A. (1999). Development as Freedom. Oxford University Press. [23] Nussbaum, M. C. (2000). Women and Human Development. Cambridge University Press. [24] Vasak, K. (1977). Human Rights: A Thirty-Year Struggle. UNESCO Courier, 30(11). [25] Arendt, H. (1951). The Origins of Totalitarianism. Harcourt Brace. [26] Bentham, J. (1843). Anarchical Fallacies. In The Works of Jeremy Bentham. [27] United Nations (1948). Universal Declaration of Human Rights. [28] United Nations (1966). International Covenant on Civil and Political Rights; International Covenant on Economic, Social and Cultural Rights. [29] United Nations (2006). Convention on the Rights of Persons with Disabilities.

关键词

SAE Self-as-an-End 权利论 相关性公理 权利不预设义务 无义务 权益 他者付出 基础权利 承认性权利 意志论 利益论 睡眠归谬 准主体 动物权益 消极积极权利 世代 历时展开 结构学 动力学 康德 目的王国 15DD 透明态 份位 行使带宽 射程 市场功利主义 Negativa Hohfeld Hart Raz MacCormick Kramer Nozick Dworkin Feinberg Sen Nussbaum Vasak

0 导言

SAE 权利论不设义务这个范畴。一个主体没有对他者的义务 它只有一条朝内的义务 对自己的法负责 那是道德律域里的自律（终止引理 而 SAE 的唯一公理是 Negativa 否定方法 自律与道德律诸定理皆由它推出的引理与定理）[8]。所谓对他者的义务 是把一个主体内在的律外推成对他人的要求 那是伦理 SAE 判其为结构上的越界而删去。所以本篇标题里的不预设义务 不是说权利不需要某一个义务 是说义务这个词在 SAE 权利论里没有一个可预设的对象 相关性公理要的那样东西 在这里没有宾语。

本篇立在一个坐标之别上 请先记下它。当代关于权利的讨论 多在一个市场功利主义式的坐标里进行 投入与回报 成本与收益的权衡 几乎成了看待一切的默认。本篇不在这个坐标里 它的源头是康德 主体作为目的 目的王国。本篇有些结论 在市场功利主义的坐标里会显得反直觉 正因为它立在另一个坐标 另一个源头。这里要说清 市场功利主义是市场那一层正当而强大的逻辑 本篇不批判它 只指出它是一个特定的坐标 而非唯一的坐标 一个主体作为目的 照样可以投入 交换 逐利 本篇只是说 权利的来源不在这个坐标 而在份位。

本篇是 SAE 权利论的元篇。主干 P1 到 P6 已经把权利谱的内部结构立完 本篇走出主干 正面对接西方权利理论 处理主系列一路避开的那条相关性公理。作为元篇 它不再受主系列不点名理论家 不指认具体制度与历史事件的约束 它会点名 Hohfeld Hart Raz MacCormick Kant Nozick Dworkin Sen Nussbaum 会引 UDHR 两公约 Vasak 的世代 会举二战后的现实例子。但这个豁免只放开点名与举例 不放开冷结构本身 例子只作结构的演示 显示一项主张在 SAE 结构里落在哪一处 是基础还是承认分量 是限还是侵 对应权力还是无对应项 本篇不裁一项制度或一个政策的对错 不选边 遇到当下最有争议的价值议题 一律避开。SAE 法学系列有一篇与西方法哲学的后验对照 本篇是它在权利理论侧的孪生篇 它对话法哲学 本篇对话权利理论 重叠的理论家各做各自不同的著作。

本篇立五命题 分两个角度。M1 到 M4 从结构学的角度看权利的本体构造 在一个时间截面上问权利是什么结构 来源在哪 怎么分 M1 处理相关性公理 权利不预设义务 M2 处理意志论 它以选择立权利 M3 处理利益论 它以受保护的利益立权利 M4 处理消极与积极那道分类 它沿义务形态切。M5 从动力学的角度看权利谱的历时展开 权利谱的展开是任何主体社会的必然动力学。动力学看的是同一套结构在时间里怎么动 不是脱离结构另起一套 M5 历时涌现的是承认性权利这个结构 开维的是主体作为目的这个结构。五条各对接一支传统 从来源到分类到历时 把权利与义务的关系 从相关性公理手里 移到 SAE 的结构上。

I 权利不预设义务

> M1 权利不预设义务。权利不预设义务 因 SAE 没有对他者的义务这个概念。SAE 以自己的法替换伦理[8]横向的义务是把内在律外推成对他人的要求 结构上是越界 删。SAE 唯一的义务朝内 是 A16 的终止引理 不得不只自己给自己立法 主体对自己的法负责（SAE 的唯一公理是 Negativa 否定方法 自我立法与四定理皆由它推出的引理与定理 非公理）。本篇立论在主体作为目的的层 即 15DD 透明态的核验网络 康德目的王国的结构刻画：每个主体在 15DD 互为目的 其自己的法本就看向全场 含他者（§V 道德律第一定理）[2][3][9]故不得不反对任何主体余项被压制是其法的题中之义 约束朝内而朝内含他者。横向义务在目的王国里本就多余 相关性公理（correlativity claim-duty）要的横向 duty 没有宾语。Hohfeld 四对 权能与责任（power-liability）SAE 认领 承认性权利对应权力[5]受效者持责任 责任非义务。权利与义务（claim-duty）不适用 SAE 无横向义务。基础权利前关系 单主体即成立 不待另一方 锚 ρ≠∅ 出 Hohfeld 整张关系网 与豁免（immunity）顶多同源不同构 豁免预设另一方 基础权利不待另一方。SAE 权利论本体图还有第三处 权益 对应他者付出 非义务非权力 关系性而不承认权力（承认权力即升为承认性权利）授予端须主体 被授予端不必 准主体 动物 乃至非生物整体皆可。三处 基础权利无对应项 权益对应他者付出 承认性权利对应权力 皆不落义务。利益不在此图 属 14DD 法射程 非 15DD 自己的法射程。约束不由义务承载 由侵读数及其后果承载 越界即侵（F21）[6]后果是修复施为 承认层归位 权力支持撤回 非履行义务。

西方权利理论有一条几近公理的命题 相关性（correlativity）：每一项权利对应一项义务 二者互为蕴含 我有一项权利 意味某人对我负有一项义务 一项权利被侵 意味一项义务被违反。Hohfeld 把它做成法律关系的微观语法 一项 claim-right 与一项 duty 是一对相关项 有此即有彼[13]。本系列从 P1 到 P6 一路未正面处理这条公理 本篇处理它。

处理它要先说清本篇立论的层。SAE 继承康德 也发展康德 用更细致的结构分析 在二十一世纪展开康德。康德说人是目的 立起一个规范理想 目的王国（Reich der Zwecke）[18]：所有理性存在者互为目的 彼此立法又服从共同立法。SAE 给这个规范理想一个结构刻画 就是 15DD 透明态的核验网络。一个主体的权利展到谱顶 进入一种必然向全场敞开的透明态（§V）[2]在无中心的分布式核验里 每个节点检验各方是否仍以 15DD 位置互相承认对方为目的[3]。每个主体都在 15DD 互为目的 这就是目的王国 只是从规范理想成了结构描述。本篇谈权利 谈的是主体作为目的的权利结构 立论就在这个层 正如康德谈人是目的 谈的是人作为理性存在者的规范结构 不为它另加一句经验的预设（道德律系列也是这样分层 前几篇在纯 15DD 的理想型里立论 14DD 向 15DD 的混合现实另作处理）[11]。

在这个层上 义务只有一条 且朝内。SAE 以自己的法（one's own law）替换伦理与道德[8]。道德是一个主体在自己 DD 层级上内生的律 只对自己有效。伦理是把这内在律外推成对他人的要求 结构上是非法的越界。所以 SAE 没有横向的义务 没有一个主体对他者负有的义务。SAE 有的是一条纵向的义务 A16 的终止引理 不得不只自己给自己立法 主体对自己的法负责（SAE 的唯一公理是 Negativa 其余皆引理与定理）。这条义务朝内 是自律。

朝内 却不因此把他者关在外面。因为在目的王国里 一个主体作为目的 它的法本就看向全场 含他者。15DD 透明态的核验网络里 每个节点以 15DD 位承认他者为目的 这承认写在它自己的法里 不是外加的。道德律第一定理已把这条立定 一个 15DD 主体不得不承认具体他者为目的 且这承认是对称的 承认他者的同时承认自己（道德律第一定理 从对称的一侧看 它止步于他者的基础权利之前）[9][3]。两个 15DD 立法主体的互凿是共同探索物自体 各自尊重对方的余项而不消除它 双方的法因此都活跃 这是承认结构对他者作为目的的本体显现[10]。所以一个主体不得不反对任何主体的余项被压制 这条律是给自己的 是它作为目的的自我立法的题中之义 因为压制他者的余项 违背的正是它自己的法里那条含他者的承认。这里要分清一件事 一个主体在某个时刻不承认具体他者为目的 不是违反了一条义务的 15DD 主体 是在那个时刻不在 15DD 运作的主体[9]克制不是可违之责 是在不在目的王国里运作之别。约束朝内 而朝内的法已经含他者。这不是矛盾 是目的王国里成员既立法又服从的那个结构 说得更细。

于是相关性公理要的那样东西 在这里是多余的。它要一项横向义务 我对你的 duty 作我权利的对应项。但在目的王国里 约束已内在于每个成员朝内的法 每个成员服从的是自己作为目的立的 含他者的法 不需要再架一层对他者的义务。横向义务不是被 SAE 删掉的 是目的王国里本就不需要的一层。康德的目的王国已经不靠成员间彼此的外加义务运转 靠的是每个成员的自我立法本就普遍 SAE 把这一点用 15DD 的结构讲明。所以相关性公理要的横向 duty 没有宾语 SAE 的义务全都朝内 配不上任何他者的 claim。

这不是说 SAE 与 Hohfeld 全盘对立。Hohfeld 的四对里 SAE 认领一对 权能与责任（power-liability）。承认性权利对应权力[5]一个持承认性权利者 相对的是受权力收敛者持有的责任 受制于那权力的要求 不退出则承担（F18）[5]而责任恰恰不是义务 Hohfeld 自己就把二者分开。SAE 不认领的是权利与义务（claim-duty）那一对 因为 SAE 这一侧没有横向义务。至于基础权利 它根本不在 Hohfeld 的网里。Hohfeld 四对全是两方之间的关系位 而基础权利是前关系的 一个主体独在 宇宙里只有它一个 它的基础权利照样成立 因为核的不可夺来自 ρ≠∅ 这个操作事实 不来自另一方没有权力[2]。它与 Hohfeld 的豁免（immunity）顶多同源不同构 豁免是关系性的 我有豁免 当且仅当你有相应的无能力 预设了另一方 而基础权利不待另一方。所以基础权利连豁免都不是 它出 Hohfeld 整张关系网。这一点法学系列的对照篇也从另一侧印证 它把豁免映射到法的射程之外（§6）[12]。

这里还要把 SAE 权利论认可的另一个概念放进来 权益。权利分两层 基础权利与承认性权利[2]权益与权利相关 却不是这两层里的任何一层 它是另一个位置。权益不是基础权利 基础权利前关系 单主体即成立 而权益要另一方给予 没有他者就没有这个权益 它是关系性的。权益也不是承认性权利 承认性权利对应权力 持有者承认了对方那个可收敛它的权力 而权益不承认对方的任何权力 一旦被授予端承认了授予端的权力 权益就升格为承认性权利 那是另一件事。权益对应的是他者付出 一个主体作出付出 另一端持有那个指向付出的位置。人有获得他人知识的权益就是一例 没有他者就没有他人的知识 故非基础 而获取知识不必承认对方的权力 故非承认性 它对应的是他者知识的付出。权益的授予端必须是主体 被授予端却不必 它可以是主体 也可以是准主体 动物 乃至一片海域这样的非主体非生物的整体 只要有一个主体在付出。这里要显标一点 被授予端不是一个拥有权益的主体 是主体的付出所指向的那一端 权益这个结构锚在授予端的付出 不锚在被授予端 所以说海域持权益端 不是说海域有了主体性 只是说有一个主体对它付出 它是那付出的指向。而这一格对应的也是他者付出 不是他者的义务 付出是他者的施为 不是他者欠谁的债 他者可以付出 也可以不付出。至此 SAE 权利论本体图里可以指认的位置齐了 基础权利无对应项 承认性权利对应权力 权益对应他者付出 义务在这三处里没有任何一处容它 相关性公理要的那个横向义务 在整张图里都没有位置。这三处是目前 SAE 权利论本体图里能指认的位置 本篇不主张它们穷尽了一切可能 未来未必不会发现新的位置。但有一点可以断定 无论本体图往后添不添位置 义务都不会是其中任何一处 因为义务是伦理概念 属于被自己的法替换掉的那一层 不在 15DD 自己的法的射程里。至于利益 那是另一个射程的概念 属 14DD 的法 不属 15DD 的自己的法 不在这张图里 对接利益论时再说。权益的完整动力学 它能否被侵 能否撤回 与 T5 登记的关系 留给权益自己的展开 本篇只把它放进这张图 补全义务无位置这一点。

这三处 在具体的案例里看得更清。基础权利的本体先在 纽伦堡审判是一个冷例。纽伦堡拒绝了当时国内法合法 只是执行命令这一类抗辩 它断言某些侵犯 即便在当时的实证法下合法 仍是错。在 SAE 的读法里 这不是诉诸一部更高的自然法 而是说被侵的那些基础权利本体先在[1] 锚在 ρ 与封装 不由实证法造出 也不待任何在场他者的义务 所以实证法认它合法 并不使它不是侵。这走的是一条结构的第三路 既不把权利还原为实证法的设定 也不把它奠基于一个自然法的天赋（承 §7）[12]。承认性权利对应权力这一面 恶债是一个冷例。一个独裁政权倒台之后 新的政权与人民可以宣布前政权所借的恶债无效 这里起作用的不是要债权人负什么义务 而是一种变更系统法律状态的权力 承认性权利在这里就是这个权力 债权人相对的是一个责任 不是它对谁负有的义务。本篇只取这两个例子的结构位置 不裁纽伦堡审判的每一处 也不裁某一笔债务的具体归属 那是另一根轴上的事。

权益升格为承认性权利的那个相变 信托法里的未出世受益人是一个冷例。一份信托可以为一个尚未出世 甚至尚未受孕的未来受益人保留一个财产位置 在受益人出世之前 它没有当下的意志 也没有既存的肉身因而没有既存的利益 意志论与利益论在这里都很别扭 而它已经持有一个位置。这个位置此刻是权益 它对应的是设立者与遵嘱的信托体系的付出 受益人这一端并无投入。受益人出世之后 它有了否定门的位置 而信托常附带条件 信托体系遵从遗嘱 遗嘱可以附条件 这个附带的条件就是一次权力的收敛 受益人要满足条件才享有。到这里有两条路。受益人不以否定门回绝这个收敛 接受这份受益 那个权益就升格为一项承认性权利 它对应的正是信托体系那个附条件的收敛权力。受益人也可以拒绝 附带的条件既是权力的收敛 受益人拒绝这个权力 不接受条件 它就退出这个关系 不再享有这项权利 这正是退出权（F18）[5]受效者持一个责任 受制于附条件的权力要求 不退出则承担 退出则不再持有。整条链里没有一处是横向义务 受益人不对信托负有义务 信托体系也不对受益人负有义务 有的是权益 权力的收敛 与退出权。这一个案例 把权利论本体图的三处动态地串了起来 一个位置怎么从权益 即对应他者付出 相变到承认性权利 即对应权力 又怎么经退出权退出。本篇只取这个结构 未出世受益人这个位置 不是关于未生者道德地位的任何主张。

没有横向义务 不等于没有对他者的约束。约束不由义务承载 由结构承载。一个主体越过他者的封装边界 不是违反了一条对他者的义务 是触发了一个侵的读数（F21）[6]同时 在它自己的法这一侧 这一越界违背了它那条含他者的对内律。侵有它的结构后果 修复施为 把被改画的封装维度恢复到可登记的状态 承认层的归位 权力支持的撤回 这些不是履行义务 是侵读数的后果。传统用义务说的每一件事 都可以在结构里找到它的对应。传统说 A 对 B 有不干涉的义务 SAE 说 B 的封装边界 A 越即侵。传统说 A 对 B 有给付的义务 SAE 说 B 持一项承认性权利 对应一项制度性权力 A 是那权力的承载 不给付则那权力的要求落空。传统说 A 违反了对 B 的义务 SAE 说触发侵读数 或承认层失效。传统说侵后 A 须补偿 SAE 说修复施为。义务这个词能说的 结构语都能说 且说得更准 这正是 SAE 的一贯做法 拆掉一个传统概念 从不只做减法 总在结构里给出替换。

法学系列有一篇与西方法哲学的后验对照[12]它在讨论 Hohfeld 时 曾用 Hohfeld 的语言写过一句 我有权利不被压制 你有义务不压制。这句与本篇不冲突。那篇讲的是法 在两人法的层面 用 Hohfeld 的语言给出法律关系的对应 它自己说明这是对应 不是用 SAE 重写（§6）[12]。用 SAE 本体的话 你有义务不压制 就是 你越界即侵 加上 施为者违背了自己那条含他者的对内律。两篇是一个结构的两面 法学侧用 Hohfeld 的语言作对应 权利论侧讲明 SAE 的本体只有朝内的义务。有人会问 法学 supplement 既然用了义务这个词 和本篇的无义务岂不矛盾。不矛盾。那是在法学层用 Hohfeld 的语言作对应描述 用了 Hohfeld 的词 不代表 SAE 的本体里有义务这个原始件 两篇是同一结构的两种语言 不是两个相争的主张。本篇引它 是接着它讲 不是与它相争。

有一个反问值得先接住 既然 SAE 认领承认性权利对应权力 那权力的收敛不就是义务吗。不是。权力是条件性的收敛 只在实际收敛时 只收敛特定的受效者（F15）[5]义务是普遍的映射 一项 claim 对应所有人的一项 duty。一个是有条件的 落在特定关系上的 一个是无条件的 铺向所有人的 这个分别是结构的 不是措辞的。还有一层在时间上 权利本体先在[2]它先于任何救济而在。传统容易把侵之后救济的对称性 误当成权利产生时的本体对称性 好像权利与义务一同诞生 互为对方的根。但在 SAE 里 权利先在 侵是后来的事 救济是更后来的事 没有一个横向义务与权利同生。

相关性公理找不到它要的横向义务 因为 SAE 的义务全朝内。这一条立住 相关性在主体侧的两个更精细的传统对手就该处理了。它们不直接主张相关性 却各自把权利的持有 系在一个 SAE 要重置的东西上 意志论系在选择的行使 利益论系在受保护的利益。先看意志论。

II 意志论以选择立权利

> M2 意志论以选择立权利。意志论（will theory Hart 及 Steiner Sumner）[14]以受保护的选择行使立权利 权利持有人可主张或放弃对应义务 是小主权者 其面临的挑战 儿童 失能者 植物人按此判据落在持权之外。这抓住的是承认性权利可行使可放弃的能力（对应权力的行使）却把它误当一切权利的持有条件。SAE 以份位（13DD+）立权利 不由选择能力定 选择的行使属行使带宽 份位≠带宽（F6）[3]。睡眠是这一误当的内部归谬 一个睡着的成年人无以行使选择 按意志论字面即不持权利 而意志论者自己不敢接受睡着者可被随意处置 睡眠是行使带宽的暂时中断 不是退场 份位维持 判据不在当下是否清醒发声反思 在承载结构是否维持该主体的 13DD+ 持有位置（F19）[6]。儿童同理 份位平等 带宽待展开 是未来主体（准主体 subject-to-be 完整理论留专论）。睡着者对全世界不施为不主张不行使而基础权利完整 顺证基础权利前关系 不系于当下任何施为主张选择 锚份位与 ρ 本体先在。

意志论（will theory 或 choice theory Hart 及 Steiner Sumner）[14]把权利立在选择的行使上。有一项权利 就是有一个受保护的选择 权利持有人可以主张或放弃那个作为权利对应项的义务 是一个小规模的主权者[14]。这个图像抓住了一件真实的事 但抓住的是承认性权利那一侧。

它抓住的真实是 承认性权利确实带着一种可行使 可放弃的能力。一个持有合同权利的人 可以主张它 也可以放弃它 可以启动追问 可以撤回主张 这是对应权力的行使 是承认性权利在运作层的真实样子。意志论把这个能力看得很准。

意志论面临的挑战 出在这个能力被放到哪个位置。它抓住的可行使可放弃 是承认性权利在运作层的能力 一旦这个能力被放到一切权利的持有条件的位置上 挑战就来了 不能行使选择的就落在权利之外。这个位置把儿童 严重失能者 植物人 都排在持有权利之外 因为它们无以行使那个选择。意志论者中有人接受这个结论 宁可让理论排除儿童 也不改立权的判据 这是意志论为它的选择判据付出的代价。

但有一个情形 意志论者自己也不敢接受。一个正常的成年人睡着了 他此刻无以行使任何选择 按意志论的字面 他睡着时就不持有权利。没有一个意志论者会接受由此而来的结论 睡着的人可以被随意处置。这不是外面来的质疑 是意志论内部的自相矛盾 它的字面要求排除睡着的人 它的信奉者却绝不排除。睡眠把意志论说的和认的劈开了 说的是选择 认的是别的东西。那个别的东西 SAE 有名字。

SAE 读这个情形 睡眠是行使带宽与运作的暂时中断 不是退场。主体仍被维持为可恢复的 13DD+ 承载者 份位分毫未动（F19）[6]。P5 已把判据写死 它不在当下是否清醒 是否发声 是否实际反思 在承载结构是否仍维持这个主体的 13DD+ 持有位置。权利以份位立 份位在 权利就在 睡着的人权利完整。意志论认的那个别的东西 正是份位 它嘴上说选择 手上认份位。份位不是选择的能力 是主体作为目的所持的位置 一个主体睡着了 它作为目的的位置没有随之消失。

意志论者或许会救 说睡着的人有权利 因为他醒来之后能行使选择。但这一救 立刻反噬意志论自己。因为儿童长大之后也能行使选择 植物人治好之后也能行使选择 若潜在的行使就足以持有权利 那么儿童与植物人也都持有权利 而排除儿童正是意志论最招牌的立场。所以意志论面对睡眠 只有两条路 都是输。它若坚持当下的行使 睡着的人就无权利 这它自己不敢接受。它若退到潜在的行使 儿童与植物人就都有权利 这它自己的招牌就塌了。而潜在的行使能不能最终发生 判据正是承载结构有没有维持这个主体的持有位置（F19）[6] 意志论一旦退到潜在的行使 它认的其实已经是份位 不是当下的选择 只是它还挂着意志论的名字排除儿童 于是自相矛盾。

儿童是同一件事的另一面。意志论排除儿童 因为儿童无以行使那个选择 但儿童丢的不是份位 是尚未展开的行使带宽。份位与带宽是两回事（F6）[3]份位是主体作为目的所持的位置 带宽是它当下能行使到哪。儿童的带宽尚在展开 它的份位已经平等。本系列称儿童这样的存在为准主体（subject-to-be）它是未来主体 享有与当下主体同等的哲学地位 自持它的基础权利[2]。准主体不是持有更少权利的主体 是同一个作为目的的主体 在它生成途中的另一种权利配置。它的完整理论留给专论 本篇只在对接意志论处 把份位与带宽这道分界显出来。

份位与带宽这道分界 残疾人权利公约[29]是一个当代的冷例。这份公约把看待障碍的方式 从个体缺损的医疗模式 转到一个人权模式 它的核心洞见是 障碍很大程度来自社会安排与环境的排斥 不来自个体本身的缺损。在 SAE 的读法里 这正是份位与带宽的分界 一个主体的 13DD+ 份位 不因它的某种生理带宽受限而降格 公约要的合理便利 是让被环境卡住的行使带宽重新可及 而不是给这个主体补上一份它本就平等据有的份位。本篇只取这个结构 不裁某一项具体的无障碍安排该如何设。

睡眠与儿童并用 各打意志论一处。儿童是意志论敢咬牙接受的排除 它显出 SAE 与意志论在立权基础上的分野 一个立在选择 一个立在份位。睡眠是意志论不敢接受 却按字面必须接受的排除 它显出意志论的内部自相矛盾 字面要排除 信奉者不排除。前者划清两条路的分岔 后者在意志论内部点火。两个例子合起来 立权不在选择而在份位这一点 就既有分野的对照 又有内部的归谬。

睡眠还顺带印证 §I 的一件事。一个睡着的人 对全世界不施为 不主张 不行使 是一个纯粹什么都不做的主体 他的基础权利照样完整。这说明基础权利不系于主体当下的任何施为 主张或选择 它锚在份位与 ρ 的本体先在[2]。§I 说基础权利是前关系的 一个主体独在其基础权利照样成立 那是一个思想实验 睡着的人是这个思想实验的现实版 不必假设宇宙里只剩一个人 一个睡着的人 已经是一个暂时退出一切施为与关系 而权利完整的主体。

这里正好和权利理论里另一个传统对上 Feinberg 把权利的价值系在主张这个活动上[21] 有权利意味着能够主张 而主张这个行为本身给了持有者一种尊严。SAE 读这个主张 读作追问 一个主体在势脊上把一个尚未被登记的位置提出来 请求承认 而这一路能不能落成一项承认性权利 系于它是否经 T5 登记 不系于主张这个行为本身。基础权利更不待主张 一个睡着的主体什么都不主张 它的基础权利照样完整。所以主张这个活动 在 SAE 这里是追问这条通往承认性权利的路 不是持有权利的条件 Feinberg 抓住的主张的分量 被重置到追问与登记的结构上。本篇只取这个结构位置。

这里要与退场切干净。睡眠是可恢复的带宽中断 份位在 退场是承载者跌破 13DD 份位失（F19）[6]二者不同。永久的失能是一个更难的情形 若一个主体仍被承载结构维持为 13DD+ 它的份位在 权利在 若它跌破 13DD 那是退场 权利因无承载者而终止 但那是无人持有 不是有人却因无意志而无权利 与意志论的排除是两回事。这一层本篇只读结构 一个主体处于哪种结构处境 至于哪些处置应当或不应当 在另一根轴上 不由本篇裁。

意志论把权利系在选择上 睡眠与儿童一起显出立权在份位不在选择。利益论走另一条路 它把权利系在受保护的利益上 含儿童 却把权利持有人读成利益的被动承载者。

---

III 利益论以受保护利益立权利

> M3 利益论以受保护利益立权利。利益论（interest theory）[15][16][17][26]以受保护的利益立权利 X 有权利当且仅当 X 福祉的某方面 即一项利益 足以构成让他人负一项义务的充分理由[15]它能把儿童纳入权利持有者（MacCormick 以儿童为意志论的 test-case）不像意志论那样把儿童排除。它面临的挑战有三。其一 它把权利立在利益上 而利益是 14DD 法射程的概念 15DD 自己的法射程里没有利益这个概念 权利的来源在份位不在利益。其二 它把权利持有人读作被动受益者 而 SAE 的准主体是主动的未来主体 自持基础权利 份位平等 带宽待展开 不是受益者。其三 利益宽泛 器物与艺术品也有可说的利益 利益论须外加 ultimate value 一类门槛才能把树与器物挡在权利持有者之外[15]SAE 不需要这个补丁 动物无权利而有权益 权益在 15DD 射程 对应他者付出 被授予端不必是主体 利益则在 14DD 法射程 射程之别加权益这一格 化解利益论的动物与树难题。能力进路（capabilities Sen Nussbaum）[22][23]是准主体问题的主要邻居 它描出的厚层展开图像与 SAE 的厚层加行使带宽相邻 但它把这个图像预设为一个应抵达的繁荣终点 这个目的论与 SAE 的终点开放（X2 F24）[7]分岔 此处是划界 不是淘汰。

意志论把权利系在选择的行使上 利益论走的是另一条路。它以受保护的利益立权利 一项权利 就是对一项足够重要的利益的保护 这项利益足以构成让他人负一项义务的充分理由[15]。这条路有一个意志论没有的长处 它能把儿童纳进来。MacCormick 正是拿儿童作意志论的 test-case 儿童显然有权利 而意志论按其字面要把儿童排除 利益论说 儿童有值得保护的利益 所以有权利。在纳入儿童这一点上 利益论比意志论走得稳。

利益论面临的挑战 出在它立权利所用的那个词 利益。在 SAE 这里 利益是 14DD 法射程的概念 不是 15DD 自己的法射程的概念。SAE 权利论谈的三处 基础权利 权益 承认性权利 都在 15DD 自己的法射程里 没有一处叫利益 因为利益不是这个射程的东西 它属 14DD 的法 是法可以收纳与保护的那一层。利益论把权利立在利益上 是拿一个 14DD 法射程的概念 去立一个本该在 15DD 自己的法射程里立的东西。SAE 读权利的来源 在份位 不在利益 一个主体作为目的持有它的基础权利 是因为它作为目的所据的份位 不是因为它有一项值得保护的利益。

利益归 14DD 有一个更内的理由 投入。利益在 SAE 框架下是一个有投入得回报的概念 你投入成本 换取好处 这个投入换回报的交换结构 正是 14DD 这个碰撞与交换的层的逻辑 法学系列的两人法就在这一层立论。而 15DD 自己的法射程里的三处 没有一处是投入换回报 基础权利是天然的 一个主体作为目的据份位 不需要投入就持有 权益对应他者付出 持权益的一方自己并无投入 承认性权利对应权力的要求 那要求也不是一笔投入换来的回报。所以利益与权利论的三处 由投入这把尺分开 利益是投入换回报 那三处都不是。这里要摆两条 免得读歪。其一 利益归 14DD 是分工 不是贬低 投入换回报是 14DD 那一层正当而真实的结构 只是不归权利论的 15DD 射程处理 正如对传统理论是荣誉其所抓住而重置其来源 对 14DD 也是认它是那一层的正当结构而不把它拉进这一层。其二 本系列从 15DD 视角立论 并不禁止一个主体去做 14DD 的事 一个主体作为目的 照样投入 交换 获利 譬如投资股票而得利益 无可厚非。所以利益论面临的挑战 不在一个主体追求利益 主体当然可以 而在利益论拿一个 14DD 的交换概念 去立一个本属 15DD 自己的法射程的权利 是立权利的来源用错了层。

利益论还把权利持有人读成了一个被动的位置。在它的图像里 权利持有人是那个享有受保护利益的一方 权利是他被动地享有或不受损的东西。这一读法用在儿童身上 就把儿童读成一个被动受益者 一个有值得保护的利益 因而被他人的义务所保护的对象。SAE 读儿童不是这样。SAE 的准主体是主动的未来主体 它自持它的基础权利 份位与当下主体平等 只是行使带宽尚在展开（F6）[3][2]。份位与带宽的分界（§II）在这里再次起作用 儿童不是持有较少利益因而受较少保护的对象 是同一个作为目的的主体 在生成途中带宽待展开。被动受益者与主动未来主体 是两个不同的读法 利益论停在前者 SAE 给的是后者。利益论最别扭的地方 §I 的信托未出世受益人已经点到 一个尚未出世的受益人 没有既存的肉身 因而没有既存的利益 利益论在它身上几乎无从下手 而它已经持有一个位置 那个位置在 SAE 这里是权益 对应设立者的付出 不是一份受保护的利益。

利益论最吃力的地方 是它的边界。利益是宽泛的 一件器物 一株植物 一件艺术品 都可以说有某种利益 有对它有利或不利的状态。若持有利益就持有权利 那么器物与树也该是权利持有者 而这是利益论者自己不接受的。为挡住这个后果 利益论须外加一道门槛 Raz 用 ultimate value Kramer 用类似的判准 只有其福祉具有 ultimate value 的存在者 或人造的法律人格 才算权利持有者。这道门槛是外加的 它不从利益本身长出来 是为了把树与器物挡在门外而补上的。

同一个难题 SAE 不靠补丁化解 靠射程之别与权益这一格。动物有没有权利 SAE 说没有 动物不据 13DD+ 的份位 不持基础权利 也不在承认权力的关系里 不持承认性权利。但动物在 15DD 语境下有权益 权益对应他者付出 一个 15DD 主体对动物的护育与不伤害 是主体的付出 动物持那个指向付出的权益端。权益的被授予端不必是主体 所以动物能持权益端而不必是主体 一片海域也一样 对这片海域的渔业生态保护 限捕 禁渔期 护育鱼群 是作出保护的主体的付出 这片海域作为一个非主体非生物的整体 持那个指向付出的权益端。这里只取这个结构 一个非主体的整体能作权益的被授予端 因为使这成为权益的 是主体那一侧的付出 不是被授予端的资格 至于某片海域该不该保护 某项渔业政策该怎么定 那是另一根轴上的事 本篇不裁。于是利益论要靠 ultimate value 补丁处理的树与器物 在 SAE 这里落得很清楚 动物与海域这类非主体 不持权利 而可持权益 利益本身则属 14DD 法射程 由法去收纳 不必挤进权利这一层。

准主体这个问题上 利益论之外还有一个更近的邻居 能力进路[22][23]。它值得正面对 因为它描出的东西与 SAE 相邻。它问一个主体真实能够做什么 成为什么 Nussbaum 列出一组核心能力 生命 身体健康 身体完整 感知与思维 情感 实践理性 与他者的亲和 等等 作为人的尊严底线。这组能力与 SAE 的厚层加行使带宽是相邻的 都在说一个主体能运作到哪 都不把主体的运作还原为资源的多寡。在描出主体运作这一层 能力进路看得很细。它与 SAE 的分岔 在终点。能力进路把那组能力的充分展开 设为一个应当抵达的繁荣状态 一个有内容 有形态的终点图像 这是一个亚里士多德式的目的论。SAE 的加法路径有方向 却没有这样一个预设的终点 谱的展开方向必然而终点开放（X2 F24）[7]SAE 不把任何一张繁荣图像设为权利的构成条件。所以这里是划界 不是淘汰 能力进路描出的主体运作 SAE 认它描得真 只是把它的繁荣终点这一预设 与 SAE 的终点开放分开。这道分岔主要在 Nussbaum 一侧 她的核心能力清单是一张较固定的繁荣图像 与 X2 的终点开放张力最大 目的论的划界主要针对的是这张固定清单。Sen 一侧其实离 SAE 更近 他明确反对锁定一张固定的能力清单 他的能力空间是开放的 更像行使带宽与真实机会那一层的邻居 这道划界对 Sen 是轻的 对 Nussbaum 才是实的。两侧都不提供 SAE 的来源判据 权利的来源仍在份位 不在能力清单。

读到这里 准主体现象的两种读法就清楚了。能力进路读一个准主体 读它是能力尚未达到繁荣阈值的存在 利益论读它 读它是有值得保护的利益的受益者 SAE 读它 读它是份位已在 带宽待展开的主动未来主体。三种读法指向同一个现象 判据各不相同 SAE 的判据只落在份位与带宽这个结构上 不落在繁荣的终点 也不落在利益的享有。利益论与能力进路各自抓住了这个现象的一个真实侧面 保护的理由 与运作的展开 SAE 把这两个侧面各自重置到它的结构里 而把权利的来源留在份位。消极与积极权利的那道传统分类 是另一道切法。

IV 消极积极之分沿义务形态 SAE 沿分量重画

> M4 消极积极之分沿义务形态 SAE 沿分量重画。传统消极与积极权利之分沿义务形态切 消极权利要他人不作为 积极权利要他人作为 一代公民政治权利被读作消极 二代经济社会权利被读作积极。这抓住的是权力要求形态的一个真实差异 收敛式与供给式。它面临的挑战是 这道分类沿义务形态切 而 SAE 没有义务 义务形态的分类在 SAE 不成立。SAE 沿来源构成轴（F12）[4]与限侵[6]重画 一项权利落基础区域还是承认区域 由独存检验定 二值 不由它要人作为还是不作为定。积极权利如住房 受教育 其基础分量按独存检验判 不是资源重要就算基础（F12 明立 不能用生计最低生存这类随时代变的资源名目定义基础）[4]其承认分量对应制度性权力（F15）[5]政府是那权力的承载 所谓政府义务是那权力的要求 不是义务 其中要他者付出而不涉权力的那一面 可落权益。消极权利如表达 是基础分量加限侵读。二代经社权在本节读的是共时的分量之分 其历时涌现读法归 §V 同源例不同轴。

消极权利与积极权利的分类 是权利理论里一道常用的切法。消极权利要他人不作为 不干涉你说话 不干涉你结社 积极权利要他人作为 给你医疗 给你教育。一代的公民政治权利多被读作消极 二代的经济社会权利多被读作积极 UDHR[27] 之后 两公约把这两类分列 公民政治权利公约与经济社会文化权利公约[28]。这道分类抓住了一个真实的差异 权力要求的形态不同 一种是要对方停手的收敛式 一种是要对方给予的供给式。

它面临的挑战 在它凭什么切。这道分类是沿义务形态切的 消极权利对应一项不作为的义务 积极权利对应一项作为的义务 分类的根是义务的形态。而在 SAE 这里没有义务（§I）义务形态的分类也就没有落脚的地方。所以 SAE 不沿消极积极这道轴切 不是说这道分类看到的差异不真 而是说它用来切的那把尺 义务形态 在 SAE 的图里没有。

SAE 沿另一根轴重画 来源构成轴（F12）[4]。一项权利落在基础区域还是承认区域 由独存检验定 撤去当下在场的所有他者 这项确定主张仍在 它属基础 不在 它属承认 归属是二值的 过阈即突转。这根轴问的是权利从哪里来 内生还是被构成 不问它要人作为还是不作为。配合限与侵[6]这道重画能把消极积极那道分类切中的东西 放到 SAE 的结构里重新摆一遍。

积极权利在这根轴上要拆开看。以住房 受教育为例 它们常被整块读作积极权利 而在 SAE 这里 一项这样的权利要按分量拆。它的基础分量 由独存检验判 而不是由它涉及的资源有多重要判 P3 F12 已把这点立死 不能拿生计 最低生存这类随时代变的资源名目去定义基础 因为基础区域的归属由独存检验与类型判据定 这些判据不随时代移动。所以受教育里若有触及一个主体作为目的之展开的维度 那一处的基础分量按独存检验成立 而受教育作为一项制度供给 是承认分量 它对应一项制度性权力（F15）[5]政府作为这项制度权力的承载 处在被这项权利调用与收敛的位置。传统说政府对这项权利负有义务 SAE 说 那是这项承认性权利对应的制度性权力的要求 政府是那权力的承载 不是一个义务人。这不是说这项要求不真 不强制 它可以非常真 非常强制 只是它的结构位置是权力的要求 不是一个横向义务。这项供给里 若有要他者付出而并不涉及承认对方权力的那一面 它落的是权益 对应他者付出。消极权利如表达 则是基础分量加限侵读 它触及的表达维度是基础分量[3]对它的作用按限与侵读[6]。

这样重画之后 消极积极那道分类就被拆成了 SAE 的几样东西 一项权利的基础分量与承认分量 承认分量对应的制度性权力 以及要他者付出的权益 加上限与侵这一读法。传统一句积极权利 在 SAE 这里可能同时含一个基础分量 一个对应制度权力的承认分量 与一处权益 它不是一类新义务的确立 是这几样结构成分的并存。二代经济社会权利的涌现 在本节读的是这个共时的分量之分 它在时间轴上如何涌现 是另一回事 属历时那一轴 同一组二代经社权 在这里读它的分量构成 在那里读它的历时涌现 是两根不同的轴上的两次读 不叠成一条线。

---

V 权利谱的历时展开是任何主体社会的必然动力学

> M5 权利谱的历时展开是任何主体社会的必然动力学。权利谱在时间中展开（F23）[7]不是人类史的偶然 是任何主体社会的必然动力学 凡存在多个主体 共享场域 追问链与承认性位置的主体社会 个体作为目的沿加法路径开维 历史地聚合即扩权 谱不会停成静态清单。方向必然而终点开放（F24 X2）[7]。Vasak 的三世代 一代公民政治 二代经济社会 三代集体连带 与战后的人权体制（UDHR 两公约）是这一普遍动力学的人类实例 不是它本身。世代框架把扩展读成三代的历史进步 隐一个进步目的论 X2 拦住这个读法 世代不相互替代 正是谱在具体权利层双向 加法路径方向单向而非历史单调的印证。二代经社权的涌现 本节读的是它的历时涌现 承认分量新增而基础分量显明 与 §IV 读它的共时分量之分 是两根轴上的两次读。

前面四节是结构学 在一个时间截面上问权利的构造 权利的三处怎么分 义务为什么无位置 意志论利益论把权利系在什么上 消极积极怎么重画 都是共时的结构分析。这一节转到动力学 问这套结构在时间里怎么动。这不是从一个话题换到另一个话题 是从一门看结构的角度 转到一门看演化的角度 而动力学看的仍是同一套结构在时间里的展开 历时涌现的是承认性权利这个结构 开维的是主体作为目的这个结构 不是脱离结构另起一种东西的时间。权利谱不是一张静态的清单 它在时间里展开 有新的位置涌现 有旧的位置退隐。西方权利理论里 描述这个历时展开最常用的框架 是 Vasak 的三世代[24] 第一代公民政治权利 第二代经济社会文化权利 第三代集体与连带权利 战后的人权体制沿这条线铺开 UDHR 之后 公民政治权利公约与经济社会文化权利公约分列 后来又有各类群体与发展的权利。这个框架抓住了一个真实的东西 权利谱确实在历时中扩展 不是一成不变的。

它面临的挑战 在它把这个扩展读成了什么。世代这个词 把三类权利读成一部拾级而上的历史 第一代之后是第二代 第二代之后是第三代 像一部权利逐代升级的进步史。这个读法藏着一个目的论 仿佛权利史朝着一个越来越全的完满态推进。SAE 的反目的论命门（X2 F24）[7]正是要拦住这个读法。

SAE 读这个历时展开 读的是扩权的历时动力（F23）[7]。一个主体作为目的 沿加法路径的加法步 向新的方向开维 这是它作为目的的结构倾向 这个开维在势脊上向他者显出一个尚未被登记的余项 经追问 经登记 经三支咬合 落成一项可运行的承认性权利 许多主体的许多次这样的开维 在时间中聚合 就是扩权。所以扩权不是那张清单自己在长 是无数个体开维经追问链在历史中的聚合。这个动力不是人类史特有的 它系于主体作为目的这一结构 凡有主体作为目的的地方 就有这个开维与聚合。

于是这一节的命题是普遍的 权利谱的历时展开是任何主体社会的必然动力学 凡存在多个主体 共享场域 追问链与承认性位置的主体社会 其权利谱都不会停成一张静态清单。但这个必然要读准 它是方向侧的必然 不是历史侧的必然。方向侧的必然是 主体作为目的不得不向新方向开维 这个方向不能被封成一张终态的清单 这一点不可撤。历史侧则不然 一个主体社会完全可以在某个时期把具体的权利压回去 一个时代扩开的 另一个时代可以压回（具体权利这一层双向）[4][6]。所以任何主体社会必经发展 说的是它的方向压不成终态清单 不是说它事实上单调地越来越全。X2 在这里当场约束 方向必然 而终点开放 谱有方向 不等于权利史有一个目的。

战后的三世代 就在这个位置上被读。它是这个普遍动力学在人类历史里的一次具体展开 一个实例 不是这个动力学本身 更不是一部进步史。世代不相互替代 第二代来了第一代不退场 第三代来了前两代不作废 这一点批评者早已指出 而它恰恰印证 SAE 的谱不单调增 三代是历史里交错的分层标签 不是一部拾级而上的阶梯。二代经社权的涌现 在这一节读的是它的历时涌现 一批承认性位置在战后被新增 一些基础分量在战后被显明 具体化 制度背靠。这与 §IV 读同一批二代经社权是两回事 §IV 读的是它此刻的分量构成 基础分量与承认分量怎么分 这一节读的是它在时间里怎么涌现 一个是共时的分量 一个是历时的涌现 同一批权利 两根轴上的两次读 不叠成一条线。

战后这段历时展开里 还显出一个不是新增权利 而是结构位置的东西 无国籍者的处境。一个人失去了国家的承认 沦为无国籍 他的承认性权利与可运行的救济会被大幅掏空 而他的基础权利并不因此失去 基础权利本体先在 不系于任何共同体的承认（§I）。Arendt 由此指出的有权利的权利[25] 在 SAE 的读法里 不是某一项基础权利 而是一个元承认位置 一个进入承认网络 取得可主张位置的资格。战后的人权体制试图接住的 正是这个元承认位置 它显明的不是又一项权利 而是权利谱得以对一个主体展开所需的那个入口。这也是历时展开的一部分 展开的不只是一项项新的承认性权利 也有这样的结构位置被显明。本篇只取这个结构 不裁某一国的具体难民安排。

VI 结语

本篇立在主体作为目的的层 处理西方权利理论里一条几近公理的命题 相关性 每一项权利对应一项义务。它给出的回答不是权利这里没有那个义务 是 SAE 权利论里没有对他者的义务这个概念 只有一条朝内的义务 主体对自己的法负责（终止引理 而 SAE 的唯一公理是 Negativa）[8]。相关性公理要的横向义务 在 SAE 权利论可以指认的三处 基础权利 权益 承认性权利 没有任何一处容它 相关性公理因此没有宾语（M1）。这一条立住之后 相关性在主体侧的两个更精细的传统 意志论与利益论 各被安置 意志论把权利系在选择的行使上 面临睡眠与儿童的挑战 SAE 以份位立权利（M2）利益论把权利系在受保护的利益上 而利益是 14DD 那个投入换回报的市场层的概念 SAE 以份位立权利 以权益与射程之别化解它的动物与树难题（M3）。分类侧 消极与积极那道沿义务形态切的分类 在无义务的 SAE 里改沿来源构成轴重画 政府义务被读为承认性权利对应的制度权力的要求（M4）。历时侧 世代与战后人权被读为扩权历时动力的人类实例 方向必然而终点开放（M5）。

这五步合起来 不是把这些传统一一驳倒 是把它们各自抓住的真实侧面 从权利来源这个位置上挪开 各归其位。Hohfeld 抓住了法律关系场的语法 意志论抓住了权利的可行使一面 利益论抓住了权利保护主体的一面 消极积极抓住了权力要求的形态 Vasak 抓住了权利的历时展开 这些都是真的 只是都不是权利的来源。SAE 把权利的来源留在主体作为目的所据的份位 把承认性权利的对应留给权力 把要他者付出的那一类留给权益 而所谓义务 只在传统的语言里出现 在 SAE 这里被读成侵读数 制度权力的要求 或修复施为 从不作一个原始件。

有一个传统与本篇的基础权利相邻 值得在此一提 Dworkin 的权利作王牌[20]。它说权利是个人对抗集体功利目标的王牌 不可为集体的好处而牺牲。这与 SAE 基础权利的不可让渡有一处相邻 基础权利也不可为集体利益交易掉。但两者接地不同 Dworkin 的王牌接地在平等的关切与尊重 一种政治道德 SAE 的基础权利接地在主体自己的法 在 ρ 与封装 是主体作为目的的结构。所以王牌可作一个反政策吞并的相邻理论 而不作权利的来源论 这与本篇把一切传统从来源位置挪开是一致的。

另有一个传统 与本篇对限的读法相邻 Nozick 的边界约束[19]。它把个人权利看作对他人行动的边界约束 不可为了更好的总体结果而穿越 这与 SAE 把越过封装边界读为侵有一处相邻 边界之为边界 不由穿越它能带来多少好处来衡量。但两者仍要分开 Nozick 的边界约束长在一个自由至上的框架里 权利几乎只作消极的不可穿越 而 SAE 的基础权利之外还有涵育这一维 主体作为目的不只是不被穿越 也在被追问中开维成长 所以 SAE 不是自由至上 边界约束可作一个相邻的读法 而不作 SAE 的全部。本篇只取这处相邻 不裁 Nozick 的国家理论。

本篇是 SAE 权利论主干之外的一篇。主干从权利的来源与谱 到基础权利的厚度与边界 到承认性权利的构成撤回与转化 到基础权利的限与侵 到权利的涵育与历时展开 P1 到 P6 已经立完 权利谱的内部结构在那里闭合。本篇是元篇 它不加新的谱内结构 只把这套结构放到西方权利理论的地形上 显示传统各家在 SAE 坐标里的位置与它们越出来源的那一处。还有两件留在本篇之外 一件是权益的完整动力学 它能否被侵 能否撤回 与 T5 登记的关系 本篇只把权益立为权利论本体图里的第三处 完整的展开留给它自己的位置 另一件是准主体的生成 它的门 它的拓扑 它的持有 儿童这样的准主体是未来主体 与当下主体哲学地位平等 它的照护安排与生成门的拓扑 留给一篇专论。

SAE 继承康德 也发展康德 用更细致的结构分析 在二十一世纪展开康德的目的王国。康德说人是目的 立起一个规范理想 SAE 给这个理想一个结构刻画 15DD 透明态的核验网络 每个主体在 15DD 互为目的 其自己的法本就看向全场 含他者。在这样一个目的王国里 约束已内在于每个成员朝内的法 不需要一层外加的对他者的义务 相关性公理要的那个横向义务 于是没有位置。权利不预设义务 因为在主体作为目的的层 义务本就无处可容。

致谢

本篇是 SAE 自我作为目的哲学体系权利论支线的元篇 由作者独立完成。起草用了四家交叉复核之法 内部精度审 语料一致性审 本体极限与矛盾探测 结构闸门与发布签字。四家复核显著提升了本篇的精度与一致性 作者在此致谢 文责自负。

参考文献

[1] 秦汉《权利不是什么》SAE 权利论 前置篇. Zenodo. https://doi.org/10.5281/zenodo.20726894 [2] 秦汉《论权利的来源与谱》SAE 权利论 Paper 1. Zenodo. https://doi.org/10.5281/zenodo.20743135 [3] 秦汉《论基础权利的厚度与边界》SAE 权利论 Paper 2. Zenodo. https://doi.org/10.5281/zenodo.20769253 [4] 秦汉《论承认性权利的构成与撤回》SAE 权利论 Paper 3. Zenodo. https://doi.org/10.5281/zenodo.20993658 [5] 秦汉《论承认性权利与权力的界面与转化》SAE 权利论 Paper 4. Zenodo. https://doi.org/10.5281/zenodo.21093939 [6] 秦汉《论基础权利的限与侵》SAE 权利论 Paper 5. Zenodo. https://doi.org/10.5281/zenodo.21212595 [7] 秦汉《论权利的涵育与历时展开》SAE 权利论 Paper 6. Zenodo. https://doi.org/10.5281/zenodo.21252863 [8] 秦汉《自己的法：SAE 对伦理与道德的批判》SAE 基础论文 A16. Zenodo. https://doi.org/10.5281/zenodo.19037566 [9] 秦汉《道德律四定理》SAE 道德律 Paper 1. Zenodo. https://doi.org/10.5281/zenodo.20011018 [10] 秦汉《正和：15DD 相互凿构的本体论根据》SAE 道德律 Paper 5. Zenodo.（DOI 待补） [11] 秦汉《混合现实：14DD 向 15DD 成长的中间态》SAE 道德律 Paper 6. Zenodo.（DOI 待补） [12] 秦汉《与西方法理学的后验对照》SAE 法学 supplement. Zenodo.（DOI 待补） [13] Hohfeld, W. N. (1919). Fundamental Legal Conceptions as Applied in Judicial Reasoning. Yale University Press. [14] Hart, H. L. A. (1982). Legal Rights. In Essays on Bentham. Clarendon Press. [15] Raz, J. (1986). The Morality of Freedom. Clarendon Press. [16] MacCormick, N. (1976). Children's Rights: A Test-Case for Theories of Right. Archiv für Rechts- und Sozialphilosophie, 62(3). [17] Kramer, M. H. (1998). Rights Without Trimmings. In A Debate over Rights. Oxford University Press. [18] Kant, I. (1785). Grundlegung zur Metaphysik der Sitten. [19] Nozick, R. (1974). Anarchy, State, and Utopia. Basic Books. [20] Dworkin, R. (1977). Taking Rights Seriously. Harvard University Press. [21] Feinberg, J. (1970). The Nature and Value of Rights. Journal of Value Inquiry, 4(4). [22] Sen, A. (1999). Development as Freedom. Oxford University Press. [23] Nussbaum, M. C. (2000). Women and Human Development. Cambridge University Press. [24] Vasak, K. (1977). Human Rights: A Thirty-Year Struggle. UNESCO Courier, 30(11). [25] Arendt, H. (1951). The Origins of Totalitarianism. Harcourt Brace. [26] Bentham, J. (1843). Anarchical Fallacies. In The Works of Jeremy Bentham. [27] United Nations (1948). Universal Declaration of Human Rights. [28] United Nations (1966). International Covenant on Civil and Political Rights; International Covenant on Economic, Social and Cultural Rights. [29] United Nations (2006). Convention on the Rights of Persons with Disabilities.