On the Dynamics of Value
论价值的动力学

Introduction

P1 established the value-source, P2 the positions of occurrence, P3 the structure of those positions when they are occupied. Three papers in, the synchronic picture is there, the entry point of occurrence is there, the readings under violation are there, and time is not. Wherever the earlier papers met the question of how something changes over time, they handed it here: how accessibility is gained, how the center of gravity shifts, why a content once avowed fades and returns.

The object of this paper is the dynamics of value. Three things enter time: the accessibility of a source, the concrete content, and the center of gravity. A gained source-position is not among them; within the range in which the bearer's structure still stands, once gained it no longer alters.

What most needs guarding against here is not saying too little but building a machine. The grammar of dynamics naturally wants a source of motion, and this system has none. A source is a structural ground, not a supply end for contents; uptake and avowal are first-person, and no outside description reaches the point at which they complete. Explanations offered here therefore take this form throughout: which positions were open at the time, and which path was accessible. Conditions can be listed, and conditions are necessary without being sufficient — with all of them in place the event may still fail to occur. Where they fall short is not a debt owed by the explanation; it is exactly where the bearer stands.

Before reading further, hold one cut. Accumulation and temporal extension are not the same thing. The same running repeated many times is accumulation; a single constitutive act not yet complete is extension in time. Constitution is not reached by accumulation, and constitution can extend in time. Running is the opposite case: it lies on a spectrum, and accumulation in time can make it deeper or shallower.

This paper adds no position of occurrence, and it judges no position. Whether a relation is colonization is judged by P3's criterion, which is narrow; this paper does not widen it, nor open a second route to it from the side of dynamics. What this paper gives are readings in time, and those readings are underdetermined before the third person: one and the same outwardly visible showing corresponds, structurally, to several different readings. No procedure of determination is offered here.

How a bearer adjudicates when several avowed contents demand incompatible actions, why the value of another cannot be exhaustively read, how an avowed content moves toward action — each has its own paper. Normative questions, that is, whether such things ought to be brought about or ought to be discouraged, are not drawn on at all.

The body observes the discipline of the series: no external theorists are named, no historical events are cited, and a first-tier term carries its English gloss once, at its point of definition.

---

Section I · The object and the form of explanation

V44 A gained source-position does not alter; three readings

Within the range in which the bearer's structure still stands, once a value-source is gained, its presence no longer alters from that moment: not with the arising and passing of concrete contents, not with the rise and fall of accessibility, not with the shifting of the center of gravity. This does not say that a person has only one set of source-positions for a lifetime. What dimension-opening does is add a source-position on top of the old ones (V58); it adds, and does not alter what is already held; those already gained are, thereafter, not diminished. As for the state of a source after the bearer has genuinely departed, P2 records that as a boundary condition and makes no assertion about it; neither does this paper.

Because what is gained does not alter, the other three can be read as rise and fall rather than as that thing itself coming into and going out of being. A content fades while that row of source is still there, so it is a fading and not a vanishing. There is somewhere for accessibility to return to, so its return is called a gaining. Walking-back costs nothing, because the thing by which a cost would have to be measured has not moved at all. All three sentences rest on the same fact.

Presence, in this paper, is said of a source only; of a concrete content it is said that the content does or does not enter the candidate field. The two do not share one word, because one by definition does not move and the other by definition does.

What does not alter is the presence of a gained source-position, not its running. When a bearer actively voids a higher source whose conditions of application have in fact appeared, what is voided is the accessibility of that row; the row is still there, and so is the construction-history (V49). What can be sealed off is the road to them, not they themselves — a sentence P3 has already established, and which this paper carries over.

The three readings do not share a subject. Accessibility is read row by row: whether, in this domain, a new candidate issuing from that row can still surface. Concrete content is read content by content: whether this one is in the candidate field at the moment, and whether it holds the center of gravity. The center of gravity is read action by action: which row this action's organization is in fact drawn from. The three subjects do not lie at one level, and so the three together are not a state vector; they cannot be assigned values in arbitrary combinations as coordinates can, and there will be no transition rule carrying one set of values to the next.

Among them exactly one inference holds: if an action has a reading for its center of gravity, then on that row at least one avowed content is taking part in organizing the action. In every other direction nothing follows. In particular, accessibility does not follow from the center of gravity: a person may go on living by the one thing he avowed long ago while that row, at present, surfaces nothing new. What accessibility asks is whether a new candidate issuing from that row can still surface in this domain, whereas an avowed content taking part in organizing an action is a calling-up of what is already held, and does not count as evidence of that row's present accessibility.

V45 No motor

Why a content occurs at this moment and not at that one is not answered here in the form of a cause. It is not asked what pushed it; it is asked which positions were open at the time and which path was accessible.

The reason was fixed already when P1 established the source: a source is a structural ground, not a causal origin producing contents, and not a store or a supply end for them. A supply end can be asked how much it puts out, when it puts it out, and whether it does so quickly or slowly; a structural ground admits none of these questions. To write dynamics as a set of regularities of output is to have changed the source, on the ontological map, into something else.

Conditions can be listed. This is not to say that conditions do no work. An experience, a body, a relation, a work encountered — each can be a condition of an avowal or an uptake. Of these, the arrangement of a relation, a work, an event are already listed by P2 in its section on shaping; the body is registered here as a general condition of running. What is refused is two other things: a motor built into the source, and a transition law by which conditions yield a completed constitution.

Conditions are necessary and not sufficient: with the list complete the event may still fail to occur, and this paper does not treat that as a debt owed by the explanation. Nor is the gap there for some stronger mover to fill. It is that this structure, at this point, simply stands open. The bearer stands there.

This cannot be closed off by adding further conditions. Set a further condition governing the application of the conditions, and that one must itself be applied to the case; the gap has merely moved to the next line of the list. The list does not close by growing longer.

The First Critique established a proposition of the same form for the one who does the work: applying a rule to a case cannot itself be given by a rule, and each attempt to set a rule for the application of rules yields a rule that must in turn be applied to a case. There it is said of the incision made by the one working; here it is said of a bearer's uptake of a bridge of his own. The two are of one form and neither reduces to the other; this proposition is established here independently, is not made a species of that one, and does not take that one as a premise.

V46 Time confers no standing

However long it runs, accumulation does not bring a content into a position it has not gained in constitution. Habituation produces no ownership, long holding produces no avowal, and long interruption revokes none. P3 has established one special case of the same sentence for self-colonization: repeated claiming produces no source, and would produce none in ten thousand runs; the reason there is not that repetition is powerless but that what it points at lies on the side of constitution.

But constitution can extend in time. P2 has established that some avowals occur in an instant and some take shape gradually and settle only after years; a person often has to try living a while before it settles, and what happens in that stretch of time is not the gathering of evidence but the avowal itself taking shape. This does not conflict with the paragraph above, because accumulation and extension in time are two things: accumulation is the same running gone through many times, extension in time is a single constitutive act not yet complete.

The two look the same from outside. A person does the same thing for ten years. Those ten years may be one and the same running gone through many times, with that thing never having been avowed in his own law; or they may be a single avowal taking shape across the decade and settling somewhere along the way. What is seen from outside is the same ten years.

Running, by contrast, can accumulate, and ordinarily that is just how it changes. Accessibility can be impaired by degrees and regained by degrees. By the discipline of this series, what is said is the direction of change, not the location of the boundary.

Constitution and the recognition of constitution are two things. Constitution completes when it occurs; the recognition may accompany it, and may occur later, in the running and in retracing. Whichever comes first, recognition adds no standing to constitution; and a constitution that never meets an occasion of being called upon has not thereby failed to have occurred.

V47 The readings of this paper are underdetermined from outside

Every reading offered here is underdetermined before the third person: one and the same outwardly visible showing corresponds, structurally, to several different readings, and those readings cannot be told apart by behavior alone.

This is not a matter of observation being too coarse, nor of the materials being thin. It has two roots.

One lies in P2. Avowal, revision, and exit are first-person constitutive acts, and there is no set of external signatures for them that is both necessary and sufficient and can be carried over to another person. P2 has already ruled out the two things most often mistaken for such signatures: verbal declaration is not necessary, since many value contents are embodied and tacit; a lasting preference is not sufficient, since a liking may persist for thirty years and never be avowed. More generally, the third person cannot exhaustively verify whether an avowal has in fact occurred, and so external evidence does not close into a terminal criterion. P2 also states both sides of the line: what can be audited from outside is whether the candidate position has been occupied and whether the bearer's room for restructuring and revision is still there; what cannot be audited is any determination, on the bearer's behalf, of what he truly avows. Facts within the relational structure are therefore genuinely readable: whether the candidate field has been filled, whether a denial can be registered as a denial, whether the paths of revision and exit are open. And reading them is not the same as being able to judge — a sentence P3 has already established.

The other root lies in the discipline of this series. Even on the observable side, the variables of running lie on a spectrum, and there is no locatable point on that spectrum at which the boundary is to be found. Whether the boundary has been crossed cannot, at the margin, be read off by finer rules; it can only be judged case by case.

P3 has already established an upper bound from the side of the criterion: an audit that oversteps generates, structurally, the very relation that introjection requires. What is established here is of the same origin on the other side: there what is limited is how far an audit may go, here what is limited is the form an explanation may take.

From this follows a general rule for this paper: no procedure is given by which constitution could be sufficiently inferred from third-person evidence, and no terminal test for it. This is not modesty; it follows directly from the two roots above. Facts within the relational structure can be told apart, and telling those facts apart does not convert into determining constitution; between the two steps stands that sentence of P3's. Whoever makes, out of the readings offered here, a measure to be taken up and applied to another person has made something other than what is established here.

Section II · The gaining of accessibility

V48 How accessibility is read structurally

A path is accessible (可达) when, in a given content-domain, there is still a candidate issuing from one of the bearer's own rows of value-source that can surface out of his own operation and enter the candidate field.

One thing must be held down first, or the whole section will be read crooked. A source gives a shape, not a candidate. P1, in establishing the source, wrote that a structural position gives, by its structure, the form under which a class of value can come to rest there — as the shape of a socket decides what kind of thing can be fitted into it. To say that a candidate issues from a row of source is to say that it has fitted that shape; it is not to say that the row produced it. The surfacing itself is, from first to last, the bearer's own operation.

This reading sets no new standard. The candidate position is the structural socket P2 established; the candidate field is what is in that socket at a given moment; the division of labor between the two words is already upstream. And the question of whether something can be given is the counterpart, taken from the side of what can surface, of the boundary P3 draws when it judges occupation: occupation says that this domain has been sealed by an external end to the point where no candidate issuing from one of the bearer's own rows can be had, and accessibility says that one can. The two are of one origin on the question of surfacing; and inaccessibility does not arise only from being occupied. Occupation is that species of inaccessibility which has been sealed shut by an external end, and not the whole of it.

Accessibility is read row by row, not content by content. It asks whether a new candidate issuing from that row can still surface in this domain; it does not ask whether one has surfaced at this moment, nor whether the ones already avowed are still in use. That a path is accessible while nothing at all has surfaced is ordinary: a person may for a while have no thoughts about some matter, and that does not mean the road is cut. Conversely, that a path is inaccessible does not mean the domain stands empty; usually it is not empty, only that everything getting in issues from somewhere other than that row. Being able to surface and having surfaced are two things.

Accessibility is also not presence. The presence of a gained source-position is the thing that does not move (V44); accessibility is the thing that does. P3 has established that colonization acts on accessibility and never on presence. The road on one of a person's rows of source may be obstructed for a long time, and this domain may go a long while without a single candidate issuing from it appearing, while that row is still there.

Accessibility lies on the side of running, and so it lies on a spectrum: a path may be clear, or it may be obstructed. And by the discipline of this series, there is no locatable point on the spectrum at which the boundary between accessible and inaccessible is to be found. What can be said is the direction of change — that the accessibility of a row is being impaired by degrees, or regained by degrees. What cannot be said is where it has got to, nor may the boundary drawn in one case be carried over and applied to another. How the margin is to be counted is judged case by case.

V49 Inaccessibility sorted by what maintains it

That nothing issuing from one's own row can any longer surface in a domain is a state that by itself judges nothing. What is to be asked is what maintains it. And there can be more than one maintainer — one and the same sealing may be both propped up from outside and held in place by the bearer himself — so the question must be put in order. First ask whether anything outside is maintaining it; if so, it belongs to the externally maintained kind, and does not move out of that kind merely because the bearer is also holding it. If not, ask next whether he himself is maintaining it. If neither, then it is the kind in which the conditions no longer obtain. The three kinds come from this; they are not a classification set up alongside, but the three answers this one question yields in turn.

The three look the same from outside, and so they must be set side by side to be seen.

Take the domain at 15DD. Four people; from outside, none of them any longer does, toward those near him, the things one does only if the other is taken as an end.

The first has in fact no one near him. He has moved several times; the old ties have broken and the new place has not yet grown any. The actual other that this row requires is not present, and in this domain nothing issuing from himself can surface.

The second has someone near him, that person right in front of him; he still holds all that he has come by along his own way, and he has voided entirely this person's standing as an end.

The third has someone near him. And on the matter of how to treat people, every way of proceeding he can put into words issues from a set of rules taken over from elsewhere; the ones that surface out of his own operation cannot get into this domain's candidate field.

The fourth has, from very early on, had only the ways of treating people that others gave him; a candidate issuing from his own row has never surfaced in this domain.

The four are in the same state and differ in what maintains it.

In the first kind nothing maintains it. The conditions carrying the running of that row no longer obtain: the setting is gone, the people who did that thing together are no longer there, the bodily or temporal conditions the thing requires have changed. What is said here is that a whole row can surface nothing issuing from itself, not that some one concrete content no longer comes up — the latter is the rise and fall of contents and belongs to Section III. No external end is maintaining anything here, and the criterion of the colonization paper rightly fails to reach it. It is not a pathology, and not a defect; it is that a person's circumstances have changed in time.

In the second kind the bearer maintains it himself. He actively voids a higher source whose conditions of application have in fact appeared — the active shutting that P1 V11 cut apart from a neutral walking-back. That the conditions of application have in fact appeared is, in the second man's case, just that the concrete other is already in front of him; this is exactly where this kind differs from a plain walking-back, for in a walking-back that condition is not present. Though there is someone beside him, no external end is maintaining this voiding, and so it does not fall under the judgment of colonization. What it alters is the accessibility of that row, not the presence of that row: the construction-history cannot be erased, and what is voided is the running, not the fact that the row was once gained.

In the third kind the outside maintains it. The third man and the fourth are both here: something occupies this domain's candidate field, and that something is propped up from outside. This kind goes to the colonization paper to be judged as colonization or not, and the reading there is asymmetric, which this paper records without alteration. With the three signatures complete, the reading of external colonization in this domain is determinate. With an irrefutable attribution absent, that branch does not hold, and this is a determinate judgment. With the attribution present while occupation or locking is incomplete, the third person stops at cannot be determined, which is not a verdict of some weaker form, and not a verdict of absence either.

The three readings differ, and none of them moves the case out of this kind: what the outside props up does not disappear because the branch has been excluded, and does not disappear because nothing could be determined. This paper therefore does not widen the narrow criterion there, and sets up no separate name for those cases in which the judgment does not hold.

What is called higher here means order in the structure — presupposing more, later in the order of unfolding. It is not an order of value. A content falling in a source-domain later in the structural order is not thereby weightier, and the bearer who calls upon it does not thereby stand higher in value.

The three taken together yield this: inaccessibility is not uniformly a pathology. One and the same state is, in the kind nothing maintains, a change of circumstance; in the kind he maintains himself, his own shutting; and only in the kind the outside props up does it fall to the colonization paper to judge.

V50 Gaining is not retrieving

The object of this section is the gaining of accessibility, not recovery. Recovery is only one case within it, the case in which there was something there before.

There is another case with nothing prior to return to. Where the fourth man is concerned, nothing issuing from his own row has ever surfaced in this domain; and if one day the first such candidate takes a position there, that too is a gaining of accessibility, and it is not the coming back of anything. To call all of it recovery is to suggest that something original is waiting there to be retrieved — which is exactly what the colonization paper shut out when it fixed the sense of discovery on the spot: discovery means only that a content takes a position under one of one's own rows of source, and does not mean finding a thing that was already there.

In the case where there was something before, what is gained is likewise not standing. P3 has established that the belonging of the positions is permanent, and that what colonization can do is make avowal, revision, and exit inaccessible in running. So no one needs to give the standing of avowal back to the bearer, because no one ever took it away. What is regained is the running. This has the same grammar as P1's account of walking-back: what walks back is the running, not the construction-history. What is constitutive is not cancelled by relational acts, and so it is not returned by relational acts either.

V51 The forms in which the outside can bear

Whether the outside has any form in which it can bear on this does not depend on which position has been sealed; it follows the same question as V49, namely who maintains the sealing. Each of the three kinds has its own answer.

In the kind nothing maintains, there is nothing to release. Nothing in this domain surfaces for the first man because the conditions are gone, and that was never a pathology; there is nothing to speak of releasing.

In the kind he maintains himself, there is nothing that could be stopped. The voiding is the second man's own doing, and no one outside is holding it down. Others may still stand before him, that other may still be present, and that row will still surface nothing. The outside is not without anything to do: it can be present, it can ask, it can bring the condition of application to bear again and again. What it does not have is an act that could revoke this voiding.

In the kind the outside props up, something is being held in place, and so there is a form in which it can be stopped. What is being held in place differs from case to case.

One sealing closes off where candidates come from. A single set of rules has filled this domain's candidate field, so that whatever does not issue from it cannot get in. The outside maintains this, and stopping the filling is its release.

One sealing closes off the channel of rebuttal for a particular attribution. A community says that some content is a value this bearer has avowed, and his denial of that attribution cannot be registered as a denial within the relational structure. By the colonization paper, whether the attribution is present and whether a denial can be registered as a denial are facts within the relational structure, and the third person can read them; making it registrable again is likewise something in the relational structure.

One sealing closes off the whole path of reflection, restatement, revision, and exit. This differs from the one before it: that one closes off the rebuttal of a particular attribution, this one closes off whether he can still change anything.

All three are propped up from outside, and all three have a form in which they can be stopped. When all three stop together, that state no longer holds: by the colonization paper, once the attribution occupying the constitutive position is withdrawn and can be denied, the content falls back to the standing of a candidate and is no longer registered as constitution already completed. The reading on the bearer's side was indeed also occupied; and the colonization paper writes, in the same place, that this is a structural matter and is not to be read as his subjective assent. Being a structural matter, once the things propping it up are withdrawn it too ceases to hold.

The bearing of the outside stops here, and where it stops is not at this point. After the content has fallen back to being a candidate, whether it thereafter becomes his value is a further step: it must take a position under one of his own rows of source. Release does nothing toward that step; it only clears the place. And no external act can perform that step for him — the outside may still hand over candidates, may demonstrate, may clarify, may have the conditions of occurrence in readiness, while the taking of a position under one of his own rows of source is a step only he can walk.

One branch does not dissolve along with it. If there is also, in this case, the kind of thing in which an external warrant is made to pass for standing in one's own source-domain, that branch is not automatically dissolved by the release of the relational conditions: what props it up is not anyone within the relation but that standing which stops outside, and by the colonization paper the only tense that branch has is retracing. A change in the relational conditions may be the occasion for retracing, and cannot do the retracing. The two branches may stand together in one case, and neither follows from the other, nor dissolves when the other does.

The line can therefore be put in one sentence: the act of releasing a condition has an outcome that the one acting does not decide; the three things the colonization paper establishes have a sealed state that can stand up without any cooperation from the side that bears it. Fill the candidate field, fix a content as already constituted, void the revisable position — with those three standing, the sealing stands. They may come from one person's arrangement, or be propped up jointly by a stretch of relations no one is coordinating; what the colonization paper judges is that state, not whose handiwork it is. What is made to stand is a sealing, not a constitution: whether a content has in fact been avowed in the bearer's own law is something that sealing does not decide by a hair. Whereas to stop the filling, to withdraw the attribution, to give back the path of revision — with those done, nothing has been completed: whether anything issuing from himself can still surface in that domain, and whether he avows what surfaces, rests wholly with him.

Handing over a candidate is therefore not a bringing about of accessibility. It does not overstep: P2 has established that a demonstration, a clarification, a work encountered are ordinary shaping; they put candidates into the field of the one who receives them, while whether that content becomes his value content still rests on his own avowal. And handing over a candidate is not a bringing about of accessibility, again not because it ought not to be done, but because what it does is not this thing. Accessibility asks whether anything issuing from one of his own rows can still surface in this domain, and the one handed in did not surface out of his operation.

One asymmetry should be written out. The outside can give back the runnability of the revisable position, while the revision itself can only be done by the bearer in the first person. The former is clearing the road, the latter is walking it. Neither can stand in for the other: with the road cleared and no one walking, no revision has occurred; and if he would walk while the road is locked, there is nowhere for the step to land.

A last limit. The conditions listed above, complete, do not guarantee that the gaining occurs. The permanence of the position gives only a place belonging to the bearer; it gives no attainment, and no timetable. As for who is entitled to do any of this releasing, and to what extent, and whether the release ought to be brought about at all — these are normative questions, and are not drawn on here.

Section III · The rise and fall of contents

V52 Fading takes two cuts

That an avowed content has faded is a sentence that holds from outside, and underneath it there may be three different things. Two cuts tell them apart.

The first cut asks whether a first-person constitutive act occurred in this change. If none did, it is a rise and fall at the level of running: the content may no longer enter the current candidate field, or it may enter and not hold the center of gravity; either way its standing as this bearer's value content has not altered. If one did, the second cut follows.

The second cut asks whether the original avowal has ended. Not ended, it is revision. Ended, it is exit.

Two cuts rather than three items set out together, and not in order to display fewer. The three outcomes do not lie on one side to begin with: a rise and fall in running lies on the side of running, while revision and exit both lie on the side of constitution, which is the cut made in the introduction. Set the three out in parallel and they read as three options of one rank, which covers over that cut.

A man has hardly played the piano these last few years. The sentence holds from outside, and underneath it there may be three things; and of the three, the first stands on the far side of the first cut from the other two, since in it no constitutive act occurred at all and in each of the others one did.

He may still take playing as his own affair and simply not have played much these few years. The position is open, he has had nothing he wanted to change, and he has not played.

He may still avow it while it is no longer what it was. What it was is becoming someone who can go on stage; what it is now is playing every week with a few others. The old form no longer appears, and so from outside it looks like a fading, while the avowal has not ended and what has changed is the content itself. This is revision.

He may no longer avow it. If he sits down at the piano hereafter, that is another matter. This is exit.

The three look the same from outside. What decides is not that state but what occurred in this change, and whether the original avowal is still standing.

V53 Standing is not lost by disuse and not gained by resemblance

An avowed content does not lose its standing by being long out of running. Its standing alters only by revision or by exit, not by the passage of time and not by attention moving elsewhere.

This is the mirror, at the level of contents, of P2's decoupling of source from occurrence: there what is established is that the presence of a source does not lapse with the rise and fall of contents; here, that the standing of a content does not lapse with the rise and fall of its own running.

On the branch where the avowal has not ended, what comes back is its entering the candidate field again, or holding the center of gravity again, and nothing at all has happened at the constitutive position. What has been revised in the interval is on this branch too, since that avowal has not ended. The branch where it has ended is otherwise: what appears now is a candidate. Whether it resembles the earlier one gives it nothing. Resemblance does not inherit constitution, and it must pass through first-person avowal again before it is this person's value content.

Those three men all take up the piano again ten years on. With the first two what comes back is one and the same avowed content, and with the second it comes back in its revised form. With the third what appears is a candidate; however much it resembles what he avowed back then, it must still go once more through his own law. From outside they look alike: all three have taken up the piano again after ten years.

Maintenance and investment, handed to this paper by P2, are settled here. Both are facts on the side of running, not conditions on the side of constitution. Sustained investment changes how often this content enters the candidate field and whether it holds the center of gravity; it neither alters nor consolidates its standing, and long non-investment does not diminish it either. These two words can therefore be used here without anxiety: what they speak of was never a matter of constitution, and so they need not borrow the name of valuing.

No external criterion of identity for contents is set here — that is, whether what surfaces now counts as the same item is not settled by degree of resemblance. Whether it counts is settled by how far that avowal itself reaches, and that is a first-person matter: what falls within its reach was already his content and needs no fresh taking-up; what falls outside it is a candidate, however alike, and must pass through avowal again. On the branch where the avowal has ended the question does not arise at all, since that avowal is over and reaches nothing.

The difference between the two branches must be kept. What it blocks is the thought that a content once exited returns to its place of itself upon appearing again. Exit is real.

As to which branch a given occasion falls under, that depends on whether the avowal reaches it, and only he can answer. The answering may come at the time, and may have to wait until he looks back (V46).

V54 Neither change nor stability is a criterion

P3 has established a negation: revision and change are never in themselves evidence of colonization, and to take this person has changed as a sign of this person has been colonized is to read the ordinary as pathology. This paper carries that over.

The positive half has to be narrowed. A content fades, another surfaces, the center of gravity shifts — these are rises and falls at the level of running, and they do not prove the revisable position open: with the position locked, contents still rise and fall. What does carry proof is something else, namely a first-person revision or exit actually carried through; that proves the position runnable at this moment, and does not prove in reverse that it was never locked before.

Stability proves nothing either. A content that has not moved in ten years may be a position standing open with nothing he has wanted to change, or a position locked with nothing he can change. The two look the same from outside.

So neither change nor stability serves as a criterion of colonization. How much change there has been serves no better, since that would require locating a point on the spectrum, and there is no such point.

Section IV · The center of gravity

V55 The center of gravity is a per-action reading of what is drawn on

The center of gravity (重心) is which row of source the organization of a single action is in fact drawn from.

It is read per action, and it has a domain of definition: only an action organized by one or more of the bearer's avowed contents has a center of gravity to read. If an action is organized only by an external demand, by a habitual response, or by a script that has not gained standing in his own law, no reading of the center of gravity is given for it here. This limit is not idle. To force a reading is to certify something occupied as the running of one of his own rows of source, which is exactly what the whole of Section II is set against. Having no reading does not mean lower value, and does not mean the action does not count.

Within that domain, any single action can be asked which row it is drawn from, and this question needs no stretch of time and no window. The reading need not be single: one action may be drawn from several rows at once. Nor need it be stable: two successive actions drawn from different rows say nothing in themselves, and a person with no stable center of gravity has no defect.

To say that someone's center of gravity has lately been on a certain row is already to have a vague window in the sentence, and already to have gathered many readings together. What is gathered is a description, not the center of gravity as defined here; nor does it thereby become one more state belonging to the bearer.

This paper accordingly does not ask how long counts, does not ask how many times counts, and sets no unified measure for the center of gravity. The reason is not that it cannot be counted: how many of the last ten actions were drawn from a given row is countable. The reason is that what is counted that way is not the center of gravity; the center of gravity is read per action and done, and it does not accumulate, above those counts, into a further state. As for taking such a count and ranking with it, something else blocks that step: standing as an end is not ranked, and quantities computed along this axis have no place to sit in an order of value.

V56 The center of gravity converts into no ranking of value

Richness and the center of gravity lie at two levels. Richness lies at the level of presence and construction-history: it is the breadth, as regards the bearer himself, of the value dimensions that can be run, and it accumulates with the construction-history and cannot be erased. The center of gravity lies at the level of running: it can move freely, and by P1 a walking-back in running costs nothing in value. The two levels do not imply each other. That little is in use at present does not alter the breadth already gained in the construction-history; that much has been gained in the construction-history does not require that much be in use now.

Which row a single action's organization is drawn from therefore has nothing to do with the bearer's standing in value.

The question a reader most wants to ask by this point is whether he has regressed. The question rests on a presupposition that does not hold. It presupposes that a center of gravity moving toward rows earlier in the structural order can be a loss, and that cannot be a loss: order in the structure is not order in value, as V49 has already written, and walking-back costs nothing, as P1 has established. This is not a refusal to answer the question; it is that what the question asks after has no place in the structure.

A center of gravity moving away likewise comes in two kinds, and from outside they look the same. In one the row is intact and only the organization of actions, for a while, is not drawn from it. In the other the row is already inaccessible and nothing can be drawn out. The first says nothing at all; the second sends one back to Section II, to sort out what maintains it. The outward showing of the two can be identical.

One further thing at the same level runs parallel. After a particular value demand has been met, that domain may go on issuing new demands, and may issue none thereafter. Neither alters the source — a source does not grow more numerous because demands keep coming, and no fewer because they cease — and neither alters the bearer's standing in value. All that is said here is that both occur; nothing is said about which is better.

Section V · Output

V57 Output and position are two independent axes

What a person has made, what he has learned, what he has left behind — these are legitimate objects of dynamics. Speaking of them does not violate P3. What P3 refuses is taking them as a criterion, not speaking of them.

The two axes do not imply each other. Where the position stands open there may be nothing to show for years; a relation that falls under P3's criterion may be attended by considerable achievement. A person may make a great deal, and make it well, inside an evaluative grammar that is in no way avowed by him; a person may go a long while and finish nothing on the very row he has avowed.

The inference a reader will actually draw runs in one direction: there is output, therefore this is not colonization. That one has to be driven out, because it is the one that most resembles common sense. What is wrong with it is not that it is too wide but that it has changed axes: output speaks of what has come out of that position, while the criterion asks whose end is occupying that position. A relation can both have an external end occupying the constitutive position and in fact have brought this person to make a good deal; the two hold separately, neither diminishing nor proving the other.

The other directions are blocked along with it. No output does not show that the thing is not truly his value content; no output does not show that the position is sealed; a position standing open does not guarantee output; and a colonization that holds does not mean there is no development there.

To take output as a criterion is in fact to put something visible in place of something in principle invisible. Output is there for anyone to read, whereas a constitutive act has no external signature that is both necessary and sufficient (V47). Substituting what is easy to come by for what is hard is a move that often works elsewhere; here it does not: the reading so substituted does not establish the thing that had to be established.

Development, as used here, means an observable result in capacity, in works, or in institutions. It does not mean dimension-opening, and it does not mean a rise in standing in value. Neither of those lies on this axis at all.

Section VI · The occasion of dimension-opening

V58 Two conditions, one standing and one occasional

That an uptake occurs at this moment and not at that one has two conditions that can be listed.

One is a standing precondition: the remainder of the current construct has left a legitimate bridge. This one does not change with the moment; while it is there, it is there.

One is an occasional condition: the conditions of application of that row of source have in fact appeared. What answers why now and not then is this one. At 15DD it is that a concrete other has entered the relational field.

With both in place, that the uptake fails to occur is legitimate. Two people whose current constructs have both left the same bridge, and in each of whose relational fields a concrete other is already standing; one takes up the bridge and one does not. The standing precondition is the same and the occasional condition is the same, and the events differ. This is not a condition left off the list. Adding one would not close it either, since that one must itself be applied to the case, and the gap has merely moved to the next line (V45).

The word accessibility is not borrowed here. Accessibility asks whether anything issuing from an already held row of source can still surface in this domain, whereas a bridge points toward a source-position not yet gained; of a source one does not have, accessibility cannot be said. It follows that occupation does not reach a bridge either: the boundary P3 draws is set at whether a candidate issuing from one of one's own rows can be had in this domain, and a bridge is not a candidate under any row already held. A bridge may in fact never be met with; a person may go a whole life without meeting such circumstances; and not meeting with it is not colonization, just as the man with no one near him is not colonized. P3's criterion does not cover this; the address is marked and left at that, and no further boundary is opened.

An uptake can occur only from the bearer in the first person. The question P1 left when it established valuing — why a bearer takes up at just this moment — is answered here on the side of conditions, not on the side of causes.

Whether a dimension-opening has been completed has no external criterion independent of first-person uptake. Two words must be kept apart here, and what keeps them apart is not time but their objects. What awareness is of is what is happening at this moment: a person may know what he is doing at this moment and know that it is not slight for him. What determination is of is another thing: whether what happened at this moment is an uptake. The two may coincide, and may be far apart; there is no general order between them.

What an uptake gains is a source-position. And a source-position is not the whole class of contents that has already appeared; it is the shape that lets a whole class take position. So the whole class need not actually appear before that position can be recognized. If this present occasion has already shown a new shape, the bearer may know at once; if it has shown only one content whose reach is not yet clear, later contents will make it plain.

Subsequent running is its manifestation, not its constitution. A person may, at some moment, resolutely treat another as an end, and the thing is complete at that moment; if he thereafter never meets an occasion that calls on that row, that uptake has still occurred, and is not retroactively read as not having occurred on account of the later scarcity. So what determination waits on is manifestation, not constitution: what it waits for is that this already accomplished thing show what it is, not that it come about.

That a cannot-but is strong enough that the bearer himself cannot account for it is a first-person clue, not a criterion. It cannot settle a new source-position into being: every construct carries a remainder, and being unable to account for something may be only that remainder, or only that he has not yet made the boundaries of the old source clear. Conversely, a real uptake need not be unaccountable; the bearer may well say it out in the old vocabulary that lies to hand. A clue is a clue: it may let him know almost at once, and it does not thereby become a measure to judge by.

Whether recognized at once or only afterward, knowing does not constitute the uptake: the uptake occurred when it occurred, and a later determination neither makes it happen nor makes it count (V46).

V59 Uptake is not an obligation

A bearer with a legitimate bridge before him may decline to take it up, and stands no worse in value for that. A person may live throughout within the few rows of source he currently has, and this is entirely self-consistent. P1 has established this; it follows directly from the absence of ranking.

It is restated here because the shape of the foregoing invites the opposite reading. The preceding sections listed so many conditions that they read as an indication of where one ought to go. They are not. Conditions are for reading how something occurs, not for saying that it ought to occur. The remainder gives a shape and gives no compulsion; no stretch of remainder forces the next incision.

The First Critique established a proposition running the same way elsewhere: that something can be advanced is a capacity, not an obligation. The two run the same way and neither reduces to the other; this series does not make one a species of the other.

Section VII · The reach of what lies downstream

A person has several avowed value contents, and in a concrete situation they demand incompatible actions; how that is adjudicated belongs to the paper on conflict. What this paper gives is the reading of each of them in time, not the manner of settling them when they meet.

Why the value of another cannot be exhaustively read belongs to the paper on the other. The external underdetermination established here is the showing, on the side of dynamics, of that same limit; that paper establishes it head-on.

How an avowed value content moves toward action is the entry point of the theory of action. Judgment belongs there entire; it is not used here, and no verdict on it is passed here.

This paper gives readings in time and draws no normative force. Whether the gaining of accessibility ought to be brought about, who is entitled to release those conditions and to what extent, whether a person's actively voiding a higher source whose conditions of application have in fact appeared ought to be discouraged, whether a domain's going on issuing demands or ceasing to do so ought to be recommended either way, what place the reading of output has in practical appraisal — for all of these the address is marked and left at that. Most belong to the critique of practice; and among them, those that involve one person judging another's case and claiming assent have their own door, which is a matter for the critique of judgment.

Of the four, only one does not move. Within the range in which the bearer's structure still stands, once a value-source is gained its presence no longer alters: it is not diminished by a content's fading, not diminished by the center of gravity moving away, and not extinguished by an active voiding. Dimension-opening may add a source-position on top of the old ones, while those already gained are thereafter only added to and never reduced. What enters time is accessibility, concrete content, and the center of gravity; and the rises and falls of these three can be read as rises and falls precisely because those unmoving things are there.

Time has no motor here. Accessibility impaired or gained, contents fading or returning, the center of gravity moving about — these do indeed change with conditions, and there is no engine built into the source pushing them, and no transition law computing events out of conditions. They are what the opening and closing of positions, and the holding or failing of conditions, look like in time; and however complete the conditions, they do not yield a constitution, and where they fall short is exactly where the bearer stands. What time can do is make the states of running deeper or shallower, give room for a constitutive act not yet complete to complete, and let something already accomplished show what it accomplished. It cannot avow on anyone's behalf, cannot make a content someone's by long duration, and cannot make it no longer someone's by interruption.

And every one of these readings is underdetermined from outside. What this paper gives is a way of reading, not a measure.

---

Acknowledgements

This is the fourth paper in the Value Theory branch of the Self-as-an-End (SAE) system, written by the author alone. The drafting followed a four-way cross-review method: internal precision audit, ontological-limit and contradiction detection, boundary testing, and a structural gate with pre-publication signature. Three load-bearing points were settled through review. The presence of a source was changed from unmoving across all time to: a gained source-position does not alter from the moment it is gained, while dimension-opening adds a source-position without altering those already held. The subject of accessibility was moved back from the value-source to the bearer's own operation, a source giving a shape rather than a candidate. And the two branches of colonization were cut apart, the release of relational conditions dissolving the external branch without thereby touching the other. One standing product was established with this paper, an audit table of upstream paraphrase, setting each of this paper's restatements against the upstream sentence; that table is reviewed by the gate and not self-audited by the drafter. The four-way review markedly improved the precision and consistency of this paper, and is gratefully acknowledged; responsibility for the text is the author's alone.

---

References

All works below are archived on Zenodo, with Chinese and English versions in the same record. The DOIs cited are concept DOIs and always resolve to the latest version, so no version number or access date is given. All works are single-authored (Han Qin); collaborators are acknowledged in the acknowledgements. Works are listed in the order of the series.

Qin, H. (2026). Systems, emergence, and the conditions of personhood: A normative theory centered on self-as-an-end (SAE Foundational Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18528813

Qin, H. (2026). Internal colonization and the reconstruction of subjecthood (SAE Foundational Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18666645

Qin, H. (2026). The complete self-as-an-end framework (SAE Foundational Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18727327

Qin, H. (2026). One's own law: The SAE critique of ethics and morality (SAE Foundational Paper A16) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19037566

Qin, H. (2026). A methodological overview: The chisel-construct cycle (SAE Methodology) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18842449

Qin, H. (2026). Via rho: The way of the remainder (SAE Methodology Paper 00) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19657439

Qin, H. (2026). On the origin of power (SAE Power Theory Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20370225

Qin, H. (2026). On the severing of power (SAE Power Theory Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20647463

Qin, H. (2026). On the thickness and boundary of basic rights (SAE Rights Theory Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20769253

Qin, H. (2026). On the limitation and violation of basic rights (SAE Rights Theory Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21212595

Qin, H. (2026). On rights and entitlements (SAE Rights Theory Special Study) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21314907

Qin, H. (2026). What value is not (SAE Value Theory Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21362575

Qin, H. (2026). On the source of value (SAE Value Theory Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21405207

Qin, H. (2026). On the occurrence and recognition of value (SAE Value Theory Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21437038

Qin, H. (2026). On the colonization of value (SAE Value Theory Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21611154

Qin, H. (2026). The first critique (SAE) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.21439120

引言

P1 立了价值源，P2 立了发生的位置，P3 读了这组位置被占据时的结构。三篇下来，共时的图有了，发生的入口有了，被侵犯时的读数也有了，唯独没有时间。前三篇里凡遇到在时间中如何演变，一律手递到这里：可达如何取得，重心如何位移，一项已认领的内容为何淡了又回来。

本篇的对象是价值的动力学（dynamics）。进入时间的是三样东西：源的可达，具体内容，以及重心。已经取得的源位，其在场不在其列：在承担者结构仍在的范围内，它一经取得就不再变动。

这里最要防的不是说得太少，是造出一台机器。动力学的语法天然要一个动力来源，而本系统没有：源是结构根据，不是内容的供应端，而接续与认领是第一人称完成的，外部描述不到它完成的那一处。解释因此一律取这样的形式：哪些位置当时开着，哪条路当时可达。条件可以列，而条件必要不充分，齐备了事件也未必发生；不充分之处不是解释的欠账，正是承担者站的地方。

读全篇要先持住一刀：累积与历时不是同一件事。同一种运行反复多次，是累积；同一个构成动作在时间中尚未完成，是历时。构成不由累积达成，而构成可以历时。运行那一侧则相反，它成谱，时间中的累积可以使它变深变浅。

本篇不新增任何发生的位置，也不判位置。一段关系是不是殖民，由 P3 的判据判，那里的判据窄，本篇不放宽它，也不从动力学这一侧另开一道。本篇给的是时间中的读数，而这些读数在第三人称面前是欠定的：同一段外面看得见的表现，在结构上对应几个不同的读数。故本篇不给判定构成的程序。

多项已认领的内容要求不相容的行动时如何裁断，他者的价值为何不可穷尽读取，已认领的内容如何走向行动，这些各有其篇。规范的问题，即这些事当不当被促成，当不当被劝阻，一律不支取。

正文遵系列纪律：不点名外部理论家，不援引历史事件；一级术语首次出现附英文注。

---

第一节 对象与解释形式

V44 已取得的源位不动，三读数

在承担者结构仍在的范围内，一项价值源一旦取得，它的在场自取得之时起不再变动：不随具体内容的生灭，不随可达的起落，也不随重心的位移。至于承载者真正退场之后源的状态，P2 已把它列为边界条件而不作断言，本篇同样不作。这不是说一个人一生只有同一组源位。开维所做的是在旧源之上新增一个源位（V58），它增添，不改动已有的；已经取得的那些，此后不减。

正因为已取得的那些不动，别的三样才读得出是起落，而不是那东西本身在生灭。一项内容淡了而那一行源仍在，所以那是淡，不是没了。可达有个地方可回，所以那叫取得。回走无亏，是因为亏所要依据的那个东西根本没有动。三句都靠同一件事。

作为不动的那一件而说的在场，专指已取得的源位；与具体内容的起落相对而言时，一律说它进不进候选场。这两处不共用一个词，因为一个按定义不动，一个按定义要动。他处说某个人在场，某项归属在不在场，是日常的用法，不指这个不动的东西。

不动的是已取得的源位之在场，不是运行。承担者主动作废一个适用条件已经现实出现的高层源，所作废的是那一行的可达，那一行仍然在，构造史也仍在（V49）。可被封住的是通往它们的路，不是它们本身，这一句 P3 已立。

三样读数的主词各不相同。可达逐源行读，问的是这一域上还有没有出自那一行的候选浮得起来。具体内容逐内容读，问的是这一项此刻在不在候选场里，居不居重心。重心逐次行动读，问的是这一次行动的组织实际取自哪一行。三个主词不在同一层上，故这三样合起来不是一张状态向量，不能像坐标那样任意取值组合，也不会有一条从这一组值算到下一组值的转移规则。

它们之间只有一个方向的推论成立：一次行动若有重心读数，那一行上至少有一项已认领的内容正参与着行动的组织。此外各方向皆不成立。尤其不能由重心推出可达：一个人可以一直照着他早就认领下来的那一项过日子，而那一行此刻浮不起任何新的东西。可达所问的是这一域上还有没有出自那一行的新候选浮得起来，而一项已认领的内容参与组织行动，是既有的调用，不作那一行当前可达的证据。

V45 没有马达

一项内容为何在此刻发生而不在彼刻，本篇的回答不取「有什么推动了它」的形式，取「哪些位置当时开着，哪条路当时可达」的形式。这不是说条件不起作用。经历，身体，关系，一部作品被遇见，这些都可以是一次认可或一次接续的条件。其中关系的安排，一部作品，一场事件，P2 讲塑形时已经明列；身体则在本篇作为一般的运行条件登记。所拒的是另外两样：一台内置于源的发动机，以及一条由条件算出构成已经完成的转移律。

理由在 P1 立源时就已给定：源是结构根据，不是产生内容的因果来源，也不是内容的储存器或供应端。一个供应端可以被追问它输出多少，何时输出，输出得快还是慢；而结构根据没有这些问法。把动力学写成一套输出的规律，等于在本体图上把源改成了另一样东西。

条件可以列。而条件必要不充分：列齐了，事件仍然可以不发生，本篇不把这看作解释的欠账。缺口也不是留给一个更强的动因去填的，它是这套结构在这一点上本来就开着。承担者就站在那里。

这一点不能靠再补条件来消掉。为条件的应用再立一条条件，那一条自身仍要被用于个案，于是缺口只是搬到了清单的下一格。清单不会因为更长而闭合。

第一批判在作业者那里立过同形的一条：把规则用于个案这件事，不能再由规则给出；每一次为规则的应用再立规则的尝试，那一条规则自身仍要被用于个案。那里说的是作业者的落刀，这里说的是承担者对自己一座桥的接续，两处同形而不互相还原；本篇独立立出这一条，不作其下位种，也不以那一条为前提。

V46 时间不产生资格

无论多久，累积不使任何内容取得它在构成上未曾取得的位置。习惯化不产生所有权，长期的持有不产生认可，长期的中断也不撤销认可。P3 已就自我殖民立过同一句的一个特例：重复的认领不产生源，跑一万次也不产生；那里的理由不是重复无力，是它所指的那件事在构成那一侧。

但构成可以历时。P2 已立，有的认可发生于一瞬，有的渐次成形，经年才定；人常要试着活一段时间才定得下来，那段时间里发生的不是取证，是认可本身在渐次成形。这与上一段不冲突，因为累积与历时是两件事：累积是同一种运行反复了很多遍，历时是同一个构成动作还没有完成。

两者在外面看着一样。一个人做同一件事做了十年。这十年可以是同一种运行跑了很多遍，而那一项始终没有被他在自己的法里认领下来；也可以是一次认可在这十年里渐次成形，到某处定了。外面看到的是同样的十年。

运行那一侧则可以累积，而且通常就是累积着变的。可达可以渐次受损，也可以渐次恢复。按本系列的纪律，说的是变化的方向，不是那道界的位置。

构成与对构成的辨认是两件事。构成在它发生的时候完成；辨认可以与它同时，也可以在后来的运行与回溯里发生，这里没有一条普遍的时序。确定不构成：一个构成动作一旦完成，它就已经完成，事后的确定不使它发生，也不使它作数；确定来得早还是来得晚，都不改动已经完成的那件事，而始终没有等来一个调用的场合，也不使它没有发生过。

V47 本篇的读数在外部欠定

本篇所给的每一项读数，在第三人称面前都是欠定的：同一段外面看得见的表现，在结构上对应几个不同的读数，而这几个读数不能单凭行为分辨开。

这不是观察还不够细，也不是材料不足。它有两条根。

一条在 P2。认领，修订与退出是第一人称的构成动作，没有一组必要且充分而又可以拿到别人身上去用的外部签名。P2 已经排掉两样最常被当作签名的东西：言语表态不是必要的，因为许多价值内容是具身的，默会的；持久的偏好不是充分的，因为一种喜欢可以维持三十年而始终未被认领。更一般地说，第三人称不能穷尽核验一项认可是否真的发生，故外部的证据合不拢成一套终局的判据。P2 同时写明了这道界的两侧：外部可以审计的，是候选位有没有被占据，承担者的重构与修订空间还在不在；不能审计的，是替承担者判定他真正认可的究竟是什么。关系结构里的事实因此确实读得到：候选场有没有被占满，一项否认能不能作为否认被登记，修订与退出的路径通不通。而读到它们不等于判得出，这一句 P3 已立。

另一条在本系列的纪律。即便在可以观察的那一侧，运行的变量落在谱上，而谱上没有一个可指认的位置是那道界之所在。故过与未过在边缘处不能由更细的规则读出，只能逐案判断。

P3 已从判据那一侧立过一条上界：越界的审计在结构上生成内摄所需的那种关系。此处所立的与它同源而异侧，那里限的是审计能走多远，这里限的是解释可以取什么形式。

由此得一条通则：本篇不给一套能从第三人称的证据充分推出构成的判定程序，也不给终局的检验方法。这不是谦逊，是上面两条根的直接推论。关系结构里的事实可以分辨，而由分辨那些事实转不成判定构成，两步之间隔着 P3 那一句。凡从本篇的读数里做出一把可以拿去看别人的尺的，所做成的不是这里所立的东西。

第二节 可达的取得

V48 可达的结构读法

一条路可达（reachable），指的是在某一内容域上，还有出自承担者自己某一行价值源的候选，能够从他自己的运作里浮起来，进入候选场。

源给出的是形状，不是候选。P1 立源时写明，一个结构位以其结构给出一类价值可以在此落脚的形态，如同一个位置的形状决定了哪一类东西能嵌进来。故说一个候选出自某一行源，说的是它嵌进了那个形状，不是说那一行把它生产了出来；浮起来的那一下，自始至终是承担者自己的运作。这个读法不新立标准。候选位是 P2 立的那个结构上的格，候选场是那个格里此刻装着什么，两词的分工上游已有。而能不能给出这一问，用的是 P3 判占位时那道界在能出这一侧的对照：占位说的是这一域被外来目的封到取不到出自自己某一行源的候选，可达说的是取得到。两者在能出与否上同源，而不可达不止因被占而起：占位是不可达当中被外来目的封死的那一种，不是它的全部。

可达是逐源行的读数，不是逐内容的。它问的是这一域上还有没有出自那一行的新候选浮得起来，不问此刻有没有浮起来，也不问已经认领下来的那些还在不在用。一路可达而当下什么也没浮上来，这是常事，一个人有一阵子对某件事没什么想法，不等于那条路断了。反过来，一路不可达也不是说这一域从此空着，那里通常并不空，只是凡进得来的东西都不出自这一行。浮得起与浮起来了是两回事。

可达也不是在场。已取得的源位之在场是不动的那一件（V44），可达是会变的那一件。P3 已立：殖民一律作用于可达，从不作用于在场。一个人某一行源上的那条路可以长期受阻，这一域上久久不见一个出自它的候选，而那一行仍然在。

可达落在运行那一侧，故它成谱：这一路可以通畅，可以受阻。而按本系列的纪律，谱上没有一个可指认的位置是可达与不可达之间那道界的所在。可说的是变化的方向，说一行的可达在渐次受损或渐次恢复；不可说它到了何处，也不可把某一个案上的那道界搬到另一个案上去用。边缘处怎么算，逐案判断。

V49 不可达按维持者分

这一域上再没有出自自己的候选浮得起来，这个状态本身不判任何事。要问的是它由什么维持着。而维持者可以不止一个，同一道封可以既由外部撑着，又由承担者自己守着，故这一问要有先后：先问有没有外部在维持，有则归入外部维持的那一种，不因承担者也在守而移出；没有，再问是不是他自己在维持；仍然没有，才是条件不再具备的那一种。三样由此而来，它们不是另立的分类，是这一问依次落下的三个答案。

三格在外面看着一样，故须并排摆出来看。

以 15DD 那一域为例。有四个人，在外面看都不再对身边的人做那些只有把对方当作目的才做得出的事。

第一个人身边确实没有人了。搬过几次，旧的联系断了，新的地方还没有长出关系。这一行所要的那个现实的他者不在场，他这一域上浮不起出自自己的候选。

第二个人身边有人，那个人就在眼前，他也保有着自己一路走过来的那些，却把这个人作为目的的地位整个作废了。

第三个人身边有人。而在怎么待人这件事上，凡他说得出口的做法都出自一套从别处领来的规矩，他自己的运作里浮起来的那些进不了这一域的候选场。

第四个人从很早起，在怎么待人这件事上就只有一套别人给的做法，出自他自己那一行的候选从来没有在这一域浮起来过。

四个人的状态相同，维持者不同。

头一种没有人在维持。承载这一行运行的条件不再具备：所在的场没有了，共同做那件事的人不在了，做那件事所需的身体或时间条件变了。这里说的是整整一行都浮不起出自自己的候选，不是某一项具体内容不再浮上来，后者是内容的起落，归第三节。没有外来的目的在这里维持着什么，殖民篇的判据正确地够不着它。它不是病理，也不是缺陷，它是一个人的处境在时间中变了。

第二种由承担者自己维持。他主动作废一个适用条件已经现实出现的高层源，即 P1 V11 与中性回走切开的那一种主动的关闭。适用条件现实出现，在第二个人身上就是那个具体的他者已经在眼前；这一种与单纯的回走之别正在这里，回走时那个条件并不在场。他身边虽有人，却没有一个外来的目的在维持这次作废，故它不落殖民之判。它所改变的是那一行的可达，不是那一行的在场：构造史不可抹，作废的是运行，不是那一行曾经取得过这件事。

第三种由外部维持。第三个人与第四个人都在这里：有东西占着这一域的候选场，且那东西由外部撑着。这一格交殖民篇判是不是殖民，而那里的判读是不对称的：三项签名齐备，这一域上外部殖民的读数确定；一项不可反驳的归属不在场，这一支不成立，而这一判是确定的；归属在场而占位或锁定未齐，第三人称停在判不出，判不出不是判出了一个较弱的形态，也不据此判无。

三种读数不同，而它们都不把这一案移出这一种：外部撑着的那些东西不因这一支被排除而消失，也不因判不出而消失。本篇因此不放宽那里的窄判据，也不为判不成的那些另立一个名。

此处所谓高层，指的是结构位序，即预设更多，在展开的次序上更后。它不是价值序。一项内容落在结构位序更后的源域，不因此更重，调用它的承担者也不因此在价值地位上更高。

三样合起来给出的是：不可达不一概是病理。同一个状态，在没有人维持的那一种里是处境变了，在他自己维持的那一种里是他自己关的，只有外部撑着的那一种才轮到殖民篇去判。

V50 取得不是找回

本节的对象是可达的取得，而不是恢复。恢复只是其中一格，即先前有过的那一格。

那第四个人那里没有先前可回。他自己那一行的形状上，从来没有一个候选从他自己的运作里浮起来过；若有一天头一个这样的候选在这一域取得位置，这也是可达的取得，而它不是任何东西的回来。以恢复统称，等于暗示有一个原先的东西在那里等着被找回，而那正是殖民篇把发现就地限死时挡掉的：发现只指一项内容在自己某一行源之下取得位置，不指找到一件既存之物。

在先前有过的那一格上，所取得的也不是资格。P3 已立位置之归属恒在，殖民所能做的是使认领，修订与退出在运行上不可达。故没有人需要把认领的资格还给承担者，因为没有人拿走过。所复者是运行。这与 P1 讲回走时的语法相同：回走的是运行，不是构造史。构成性的东西不由关系性的动作取消，也就不由关系性的动作归还。

V51 外部的着力形式

外部在这件事上有没有着力的形式，不看被封的是哪一位，仍按 V49 那一问：那道封由谁维持。三格各有各的答案。

没有人维持的那一种，没有可释放的东西。第一个人这一域上浮不起候选是因为条件不在了，那本来也不是病理，无从谈解。

他自己维持的那一种，没有一个可以停下来的东西。作废是第二个人自己做的，没有谁在外面按着它。旁人可以还在他眼前，那个他者可以仍旧在场，而那一行照样浮不起候选。外部并非无事可做，它可以在场，可以发问，可以使那个适用条件一再现实地出现；它所没有的，是一个能撤销这次作废的动作。

外部维持的那一种有东西被撑着，故有停下来的形式。而被维持的是什么，各有不同。

一处封的是候选的出处。那套单一的规矩占满了这一域的候选场，使不出自它的东西进不来。它由外部在维持，停止占满即是解。

一处封的是某一项归属的反驳通道。社群把某项内容说成这个人已经认领的价值，而他对这项归属的否认在关系结构中不能作为否认被登记。按殖民篇，归属在不在场，否认能不能作为否认被登记，这些都是关系结构里的事实，第三人称读得到；使它重新登记得上，同样是关系结构里的事。

一处封的是反思，重述，修订与退出的全部路径。这一道与上一道不同：上一道封的只是某一项归属的反驳，这一道封的是他还能不能改。

三道都由外部撑着，三道都有停下来的形式。三道一齐停下来，那个状态就不再成立：按殖民篇，占着构成位的那项归属一旦撤回并可被否认，那项内容就退回候选的身份，不再被当作构成已经完成。承担者那一侧的读数确实也被占过，而殖民篇同处写明，那是结构上的事，不读作他主观上的赞同；既是结构上的事，撑着它的那些一撤，它也就跟着不成立了。

外部的着力到此为止，而止住的地方不在这里。那项内容退回候选之后，它此后成不成为他的价值，是另一步：要它在他自己的某一行源之下取得位置。解除不替这一步做任何事，它只是把位子腾出来。这一步没有哪个外部动作代行得了：外部仍可以递候选，可以示范，可以澄清，可以把发生的条件备齐，而在他自己的某一行源之下取得位置这一步，只能他自己走。

另有一支不随之解除。若这一案上还有那种以一份外部的担保冒充自己源域资格的情形，那一支不因关系条件的释放而自动解除：撑着它的不是关系里的谁，是那道停在外面的资格，而按殖民篇，那一支唯一的时态是回溯。关系条件的变化可以成为回溯的机缘，却替不了回溯。两支可以在同一案上并存，而任何一支都不由另一支推出，也不随另一支的解除而解除。

界线因此可以写成一句：释放条件这一动作，其结果不由施加者决定；殖民篇所立的那三样，其所封的状态不需要承受者那一侧配合就能立起来。占满候选场，把一项内容定作已构成，作废修订位，这三样立住了，那道封就立住了。它们可以出自一个人的安排，也可以由一片没有谁在统筹的关系共同撑着，殖民篇所判的是那个状态，不是谁的手笔。所立起的是封，不是构成：一项内容在承担者自己的法里究竟有没有被认领下来，那道封一分一毫也决定不了。而停止占满，撤回归属，还回修订的路径，这些做完之后什么也没有完成：那一域上还有没有出自他自己的候选浮得起来，浮起来的他认不认领，全在他自己那里。

由此，递一个候选进来不是使可达取得。它不越界：P2 已立，一次示范，一次澄清，一部作品被遇见，都是日常的塑形，它们把候选送进承受者的场，而那项内容成不成为他的价值内容，仍在他自己的认可上。递候选之所以不是使可达取得，也不是因为它不该做，是因为它做的不是这件事。可达问的是这一域上还有没有出自他自己那一行的候选浮得起来，而递进来的那一个不是从他的运作里浮起来的。

一处不对称须写明。外部能还回可修订位的可运行性，修订本身仍只能由承担者第一人称做。前者是把路让开，后者是走。两者不可互相顶替：路让开了而人不走，修订就没有发生；而人若要走，路却锁着，那一步就无处落。

最后一句限。上面所列的条件，齐备也不保证取得发生。位置的恒在只给出一个属于承担者的位置，不给达成，也不给时间表。至于谁有资格去做这些释放，做到什么程度，以及释放当不当被促成，都是规范的问题，本篇不支取。

第三节 内容的起落

V52 淡去的两刀

一项已认领的内容淡了，这句话在外面成立，而它下面可以是三件不同的事。分辨取两刀。

第一刀问：这一次变化里有没有发生一个第一人称的构成动作。没有，那是运行上的起落，它可以是这项内容不再进当前的候选场，也可以是它进了而不居重心；无论哪种，它作为这个承担者的价值内容，地位没有变。有，则进第二刀。

第二刀问：原来的那道认领有没有终止。没有终止，是修订。终止了，是退出。

两刀不是为了少摆几项。三个结果本来就不在同一侧：运行上的起落在运行那一侧，而修订与退出都在构成那一侧，这正是引言里那一刀。把三项平列摆出来，会把它们读成同级的三个选项，反倒把那一刀遮掉。

一个人这几年很少再弹琴了。这一句在外面成立，而它下面可以是三件事。三种之中，第一种与后两种之间隔着第一刀：前者什么构成动作也没有发生，后两种各有一个。

一种是他仍旧把弹琴当作自己的事，只是这几年没怎么弹。位置开着，他没有什么要改的，只是没有弹。

一种是他仍旧认领着它，而它已经不是原来那一项了。原来那一项是要成为一个能上台的人，如今是每周和几个人聚在一起弹。旧的那个形不再出现，所以在外面看是淡了；而认领并没有终止，改的是内容自身。这是修订。

一种是他不再认领它了。此后他若再坐到琴前，那是另一件事。这是退出。

三种在外面看着一样。判不在这个状态上，在这一次变化里发生了什么，以及原来那道认领还在不在。

V53 地位不因不用而失，也不因相似而得

一项已认领的内容不因长期不在运行而失去它的地位。地位的改变只由修订或退出发生，不由时间的经过，也不由注意的移开。

这是 P2 那条源发生对偶解耦在内容这一层的镜像：那里立的是源之在场不随内容的起落而失，这里立的是内容的地位不随它自己运行的起落而失。

未终止的那一支，回来的是它重新进候选场，或者重新居重心，构成位上什么也没有发生。其间经过修订的仍在这一支，因为那道认领没有终止。已退出的那一支不同：此刻出现的是一个候选。它与先前那一项像不像，不给它任何东西。相似不继承构成，它须重经第一人称的认可，才成为这个人的价值内容。

那三个人十年之后都又弹起了琴。前两个人那里回来的是同一项已认领的内容，第二个人那里回来的是它修订之后的形。第三个人那里出现的是一个候选，与他当年所认领的那一项再像，也仍要重走一遍他自己的法。外面看着一样，都是十年之后又弹起来了。

P2 手递到本篇的维持与投入，在这里结清。两者是运行那一侧的事实，不是构成那一侧的条件。持续的投入改变的是这项内容进候选场的频率与它居不居重心，不改变也不巩固它的地位；而长久的不投入同样不削减它。故这两个词可以放心地讲，它们讲的自始就不是构成的事，也就不必借珍视之名。

这里不设外部的同一性判准，即不按相似的程度去定此刻浮上来的算不算同一项。算不算，由那道认领自己覆盖到哪里定，而那是第一人称的事：落在它覆盖之内的，本来就是他的内容，不需要任何重新接手；落在它之外的，无论多像都是一个候选，须重经认可。已退出的那一支更不必问，那道认领已经终止，覆盖不到任何东西。

两支之别却须保留。它挡的是这样一个想法：退出过的内容会因为再次浮现而自动复位。退出是实的。

至于某一次究竟属哪一支，看那道认领覆盖不覆盖得到，而这只有他自己答得出。答得出的时候可以就在当下，也可以要等到回头看（V46）。

V54 变动与稳定都不是判据

P3 已立一条否定：修订与变动本身不是殖民的证据，把某人变了当作某人被殖民了的迹象，是把常态读成了病理。

而正面的那一半要收窄。一项内容淡了，另一项浮起来，重心挪了，这些是运行上的起落，它们不证明可修订位开着：位置锁着的时候，内容照样会起落。真有证明力的是另一件事，即一次第一人称的修订或退出实际做成了；那证明的是此刻那个位置可运行，而不能反过来证明它从前没有被锁过。

稳定同样什么都不证。一项内容十年没有动过，可以是位置开着而他一直没有什么要改的，也可以是位置被锁着而他改不动。两种在外面看着一样。

故变动与稳定都不作殖民的判据。变动的多少同样不作，因为那要求在谱上指认一个位置，而那样的位置没有。

第四节 重心

V55 重心是逐次行动的取源读数

重心（center of gravity）指的是一次行动的组织实际取自哪一行源。

它逐次读，而它有一个定义域：只有由承担者已认领的内容组织起来的行动，才有重心可读。一次行动若只由外部的要求，习惯的反应，或一套尚未在他自己的法里取得地位的脚本组织着，本篇不给它重心读数。这一限不是可有可无的。硬派一个读数，等于把一件被占着的东西追认成他自己某一行源的运行，而那正是第二节整节在挡的。无读数不表示价值低，也不表示那次行动不算数。

在这个定义域之内，任何一次行动都可以问它取自哪一行，而这一问不需要一段时间，也不需要一个窗口。它不必唯一，一次行动可以同时取自数行。它也不必稳定，前后两次取自不同的行，这本身不说明任何事；一个人没有一个稳定的重心，这不是缺陷。

说一个人近来重心在某一行，这句话里已经有一个模糊的窗，也已经把许多次读数聚在了一起。聚得出来的是一句描述，不是本篇所定义的重心；它也不因此成为承担者身上另外的一个状态。

本篇因此不问多久算数，不问几次算数，也不为重心设一把统一的度量。理由不是数不出来：最近十次行动有几次取自某一行，那是数得出来的。理由是那样数出来的东西不是重心，重心逐次读完就完了，它不在那些次数之上另外积成一个状态。至于把这样的数拿去比高下，那一步另有一条挡着：是目的这一地位不排序，这类量在价值地位上没有位置可放。

V56 重心不折算价值高下

丰度（richness）与重心落在两层上。丰度落在在场与构造史那一层，指的就承担者自己而言可运行的价值维度的广度，它随构造史累积而不可抹。重心落在运行那一层，可以自由移动，而按 P1，运行的回走没有价值上的亏损。两层不互推：当下在用的少，不改变构造史上已经取得的广度；构造史上取得得多，也不要求当下就用得多。

故一次行动的组织取自哪一行，与承担者的价值地位无关。

读者读到这里最想问的那个问题，是我是不是退步了。这个问题的预设不成立。它预设了重心移向结构位序在前的那些行可以是一种亏损，而那不能是一种亏损：结构位序不是价值序，这一点在 V49 已经写过；而回走无亏是 P1 已立的。这不是拒答这个问题，是这个问题问的那件事在结构上没有位置。

重心移开也有两种，在外面同样看着一样。一种是那一行完好，只是这一阵子行动的组织不取自它。另一种是那一行已经不可达，浮不起东西来。前者什么也不说明，后者要回到第二节去分它由什么维持着。两种的外部表现可以完全一致。

同一层上还有一件事同理。一项具体的价值要求被满足之后，那个域上可以继续出新的要求，也可以就此不再出。两者都不改变源，源不因要求不断而变多，也不因要求止息而变少；两者也都不改变承担者的价值地位。这里只说这两种都会发生，不说哪一种好。

第五节 产出

V57 产出与位置两轴独立

一个人做成了什么，学会了什么，留下了什么，这些是动力学的正当对象，讲它们不违 P3。P3 所拒的是拿它们当判据，不是拒绝谈它们。

两轴互不推出。位置开着的地方可以长年没有拿得出手的东西；一段落 P3 判据的关系，可以伴随可观的成就。一个人在一套完全不是他自己认领下来的评价语法里，可以做出很多东西，而且做得很好；一个人在自己认领的那一行上，可以很久什么也没有做成。

读者真正会做的推论只有一个方向：有产出，所以那不是殖民。这一个必须驳掉，因为它最像常识。它的错处不在它太宽，在它换了轴：产出说的是从那个位置上出来了什么，判据问的是那个位置上占着的是谁的目的。一段关系可以既在构成位上占着一个外来的目的，又确实使这个人做出了不少东西，两件事各自成立，互不减损，也互不证明。

另几个方向一并挡住。没有产出，不说明那不是他真正的价值内容；没有产出，不说明位置被封着；位置开着，不保证会有产出；一段殖民成立，也不意味着那里必定没有发展。

拿产出当判据，实际做的是拿一个看得见的东西去替一个原则上看不见的东西。产出摆在那里，谁都读得到；而构成动作没有必要且充分的外部签名（V47）。以易得者代难得者，这一步在别处常常管用，在这里不管用：替下来的那个读数，所证的不是要证的那件事。

此处所说的发展，指的是可观察的能力，作品或制度上的结果。它不指开维，也不指价值地位上的提高。这两样东西在本系列里根本不在这条轴上。

第六节 开维的时机

V58 条件有两项，一恒一时

一次开维之所以在此刻而非彼刻被接续，可列的条件有两项。

一项是结构前提：当前构的余项给出了一座合法的桥。这一项不随时刻变，它在的时候一直在。

一项是时点条件：那一行源的适用条件现实地出现了。回答为什么是此刻而不是彼刻的，是这一项。在 15DD 那一处，它就是一个具体的他者已经进入关系场。

两项齐备而接续不发生，是合法的。两个人当前的构都留下了同一座桥，各自的关系场里也都已经有一个具体的他者站在那里；一个接续了，一个没有。结构前提相同，时点条件相同，而事件不同。这不是少说了一项条件。补一项条件进去也不会闭合，因为那一项自身仍要被用于个案，缺口只是搬到了下一格（V45）。

这里不借可达一词。可达问的是这一域上还有没有出自某一行已有源的候选浮得起来，而桥指向的是尚未取得的源位；没有的源谈不上可达。顺此，占位也够不着桥：P3 的那道界定在这一域上取不到出自自己某一行源的候选，而桥不是任何已有源之下的候选。一座桥可以事实上遇不到，一个人可以一辈子没有遇到那样的处境；而遇不到不是殖民，正如那个身边没有人了的人不是被殖民。此处 P3 的判据不覆盖，标出地址即止，不另开一道界。

接续只能由承担者从第一人称发生。P1 立珍视时留下的那个问题，即承担者为什么在某一时刻恰好接续，这里给出的是条件那一侧的应答，不是原因那一侧的。

一次开维是否已经完成，没有独立于第一人称接续之外的外部判据。这里要把两个词分开，而分开它们的不是时间，是对象。觉知的对象是这一刻在发生什么：一个人可以知道自己此刻在做什么，知道这件事对他不轻。确定的对象是另一件：这一刻发生的那件事究竟是不是一次接续。两者可以同时，也可以隔开很久，这里没有一条普遍的时序。

接续所取得的是一个源位。而源位不是已经出现的那一整类内容，是使一整类能够取得位置的那个形状，故不必等一整类实际出现，那个位置才认得出。当下这一次若已经显出一种新的形状，承担者可以当下就知道；若当下只显出一项内容而它的射程还不清楚，那要靠此后的内容显明。

此后的运行是它的显现，不是它的构成。一个人可以在某一刻决然地把另一个人当作目的来待，这件事就在那一刻完成了；此后他若再没有遇到调用那一行的场合，那一次接续也已经发生过，不因后来的稀少而被追认为没有发生。故确定所等的是显现，不是构成：等的是这件已经成了的事显出它是什么，不是等它成。

一个不得不强到承担者自己也解释不通，这是一条第一人称的线索，不是判据。它判不出一个新的源位：任何一个构都带着余项，解释不通可以只是那片余项，也可以只是他还没有把旧源的边界说清楚。反过来，一次真的接续也未必解释不通，承担者尽可以用手边旧有的说法把它说出来。线索归线索，它可以使他几乎当下就知道，而它不因此成为一把可以拿来判的尺。

而无论当下就知道还是事后才认出来，知道都不构成：接续在它发生的时候就已发生，后来的确定不使它发生，也不使它作数（V46）。

V59 接续不是义务

一个承担者面前有一座合法的桥，它可以不去接续，不因此在价值地位上有任何亏损。一个人可以始终活在它当前的几行源里，这完全自洽。这一条 P1 已立，是不排序的直接推论。

这一条须在此重提，因为前面几节列了那么多条件，读起来像是在指出该往哪里走。它们不是。条件是用来读一件事怎么发生的，不是用来说这件事应当发生。余项给出形状，不给出强制；没有哪一片余项逼迫下一刀。

第一批判在另一处立过同向的一条：可推进是能力，不是义务。两处同向而不互相还原，本系列不作其下位种。

第七节 下游的射程

一个人有多项已认领的价值内容，而它们在具体情境中要求不相容的行动，如何裁断，是冲突篇的对象。本篇给出的是每一项各自在时间中的读数，不给它们相遇时的裁法。

他者的价值为何不可穷尽读取，是他者篇的对象。本篇所立的外部欠定，是那道限界在动力学这一侧的显形；那一篇正面立起它。

已认领的价值内容如何走向行动，是行动论的入口。判断力整个归那里，本篇不用它，也不对它下任何判。

本篇给出时间中的读数，不给规范的效力。可达的取得当不当被促成，谁有资格去释放那些条件，释放到什么程度，一个人主动作废一个适用条件已经现实出现的高层源当不当被劝阻，一个域上的要求继续出还是不再出当不当有所推荐，产出的读数在实践的评价里占什么位置，这些一律标出地址即止。它们多数是实践之批判的对象；而其中涉及一个人对另一个人的个案下判并要求同意的那些，其资格另有门牌，那是判断之批判的事。

四样东西之中，只有一样不动。在承担者结构仍在的范围内，一项价值源一经取得，它的在场就不再变动：不因内容的淡去而减，不因重心的移开而减，也不因一次主动的作废而灭；开维可以在旧源之上新增一个源位，而已经取得的那些，此后只增不减。进入时间的是可达，是具体内容，是重心，而这三样的起落之所以读得出是起落，正因为有那些不动的东西在。

时间在这里没有一台马达。可达受损或取得，内容淡去或回来，重心移来移去，这些确实随条件而变，而没有一台内置于源的发动机在推着它们，也没有一条从条件算出事件的转移律。它们是位置的开合与条件的成立与否在时间中的样子，而条件列得再齐也不推出构成，不齐的那一处正是承担者站着的地方。时间所能做的，是让运行的状态变深变浅，是让一个尚未完成的构成动作有余地完成，是让一件已经完成的事显出它完成了什么。它不能替任何人认领，不能凭长久使一项内容成为谁的，也不能凭中断使它不再是谁的。

而这一切读数在外面都是欠定的。故本篇给出的是读法，不是尺。

---

致谢

本篇为 SAE（Self-as-an-End）哲学体系价值论分支的第四篇，由作者独立完成。成稿过程采用四方交叉评审方法：内部精度审计，本体极限与矛盾检测，边界测试，结构闸门与发布前签署。本篇的三处承重经评审改定：源之在场由全时域不动改为已取得的源位自取得之时起不动，开维新增源位而不改动已有的；可达的主词由价值源移回承担者自己的运作，源所给出的是形状而不是候选；外部殖民与自我殖民两支切开，关系条件的释放解除前者而不自动触动后者。本篇另立一项常设产出，即上游转述审计表，逐条比对本篇的转述与上游原句；该表由闸门复核，不由起草者自审。四方评审显著提升了本篇的精度与一致性，谨此致谢；文责由作者独任。

---

参考文献

以下文献均存档于 Zenodo，中英双语版本同记录。所引 DOI 为概念 DOI，始终指向最新版本，故不另注版本号与访问日期。全部文献单一作者署名（秦汉），合作者于致谢致意。文献按系列次序排列。

秦汉. (2026). 系统、涌现与人的条件：以主体作为目的本身为核心的规范社会理论（SAE 基础论文 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.18528813

秦汉. (2026). Internal Colonization and the Reconstruction of Subjecthood（SAE 基础论文 Paper 2） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.18666645

秦汉. (2026). 三层二维统一结构：Self-as-an-End 的完整框架（SAE 基础论文 Paper 3） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.18727327

秦汉. (2026). 自己的法：SAE 对伦理与道德的批判（SAE 基础论文 A16） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19037566

秦汉. (2026). 方法论总论：凿构循环（SAE 方法论） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.18842449

秦汉. (2026). 余之道 · Via Rho（SAE 方法论 Paper 00） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19657439

秦汉. (2026). 论权力的起源（SAE 权力论系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20370225

秦汉. (2026). 论权力的凿削（SAE 权力论系列 Paper 6） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20647463

秦汉. (2026). 论基础权利的厚度与边界（SAE 权利论系列 Paper 2） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.20769253

秦汉. (2026). 论基础权利的限与侵（SAE 权利论系列 Paper 5） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21212595

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

秦汉. (2026). 价值不是什么（SAE 价值论系列 前置篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21362575

秦汉. (2026). 论价值的源（SAE 价值论系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21405207

秦汉. (2026). 论价值的发生与承认（SAE 价值论系列 Paper 2） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21437038

秦汉. (2026). 论价值的殖民（SAE 价值论系列 Paper 3） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21611154

秦汉. (2026). SAE 第一批判 [预印本]. Zenodo. https://doi.org/10.5281/zenodo.21439120