Not Doing and Cannot-Not-Doing
不做与不得不做

Introduction

A man sits at his computer with the report up on the screen. His manager wants him to recalculate one of the figures on a different basis. On the new basis the number looks good, and his review at the end of the year will look good too; the only thing is that, with the basis changed, a problem now developing on another team's side gets covered up. The report goes only to management; the other team will not see it, and the problem will not surface on its own until next quarter. He knows this. He also knows that if he hands in the report without changing the basis, he will have a hard year of it.

He has a decision to make. This series is about the person who stands in this position: a matter is in front of him, no one will decide it for him, no ready-made rule will decide it for him, and he does not want to hand himself over. I do not choose for him. What I write about is what this position itself is like, and what the person standing here is doing.

The person this series speaks of takes himself as an end and takes others as ends too; as the Prequel said, this marks out the domain of this series, and is not a demand made of the reader. A person standing on this premise I call a person who lives by his own law. The term one's own law is one this system has used all along; it means the law a person lays down for himself, which governs only himself. Another person's law, even if it is the same in content, is one he arrived at himself; it was not given by me (A16 §1).

What this paper sets out is the sentence form of the whole series. Everything the later papers say, they say in this form.

The question this paper answers is: when a person makes a decision on a matter, what is he doing? The answer comes in six steps. §1 and §2 describe the two negations he performs, not-doing and cannot-not-doing, and what happens when two negations collide; §3 describes what is left after a judgment; §4 says what these cannot-nots rest on, and why they are not written as ought; §5 says who draws the line, that is, judgment; §6 says how reasons become action. Each section sets out one or two propositions, seven in all (D1–D7), listed together at the end. The text names no outside theorists and cites no historical events; the conclusions of the earlier papers this series builds on are restated only in plain words, with the source given in parentheses; how this paper connects with existing theories is left to Paper 0.

§1 Not Doing

The first thing he gets clear is this: he will not change this number.

To see what this step is, first see what it is not. It is not that he first set himself a goal called "honesty" and then deduced from that goal that the number could not be changed. He made no such deduction. As soon as he saw the matter, he knew it was something he would not do. The goal is something he could only put into words later. If someone asked him why, he might say it was for honesty's sake; he might not be able to say; or the reason he gave might strike even him as wrong the next day. He can get the reason wrong, but the line was really drawn.

This is the first kind of thing a person who lives by his own law does: he negates doing something. I call this not-doing.

When a judgment settles a matter, this series calls that a drawing of a line (the term is borrowed from M-00E §1.1). Striking out doing is not-doing; striking out not doing is cannot-not-doing, which the next section takes up.

What he strikes out is in fact more than a thing to be done. Before not changing the number there was an earlier stroke: he does not regard the other team as people it would be all right to keep in the dark. What this stroke strikes out is not an act; it is a way of regarding people. Whether to change the number is this time's business; not regarding people as all right to keep in the dark is how he treats people in this matter. These two kinds of line fall on two layers, which §4 returns to.

D1 Not-doing is a person's negation of doing something; what he strikes out may also be a way of regarding a person or a thing. It does not need a goal affirmed beforehand.

Not-doing is not passivity. The things a person does not do in a lifetime far outnumber the things he does, and the great majority of them he has never thought about; those do not count as not-doing, they simply did not happen. Not-doing has been thought about: it is striking a thing out when it is set before him. Striking out is an act. No one sees it, and no report records it, but it is the very first thing this person does in the matter.

Nor is not-doing the same as not being allowed. "Not allowed" is what others say to me, or what an institution says to me; not-doing is what I say to myself. When an institution does not allow me to do something and I do not do it either, that is two things lying one on top of the other, not one thing. If the institution withdraws the rule, the not-doing is still there.

§2 Cannot-Not-Doing

He leaves the number unchanged. Then he discovers a second thing.

It is still the case, even after he leaves the number unchanged, that the other team does not see the report, and the problem will not surface on its own until next quarter. He could simply leave the number alone and then say nothing, hand in the report as usual, and let others find out or not. Nothing forbids him to do that. He sits there and finds that he cannot. He cannot not go and tell the people on the other team.

This is the second kind of thing a person who lives by his own law does: he negates not doing something. I call this cannot-not-doing.

D2 Cannot-not-doing is a person's negation of not doing something; it, too, does not need a goal affirmed beforehand.

It is worth being exact about the words cannot not. It is not that someone is forcing him. His manager did not ask him to tell anyone; quite the opposite. Nor is it his conscience forcing him, if conscience means something standing outside him. It is his own law speaking: in this position, the option "not tell them" has been struck out by him; what is struck out is not telling, and so telling becomes something he cannot not do. The cannot not here is his own; it is not given from outside.

D3 Not-doing and cannot-not-doing are two acts, and neither follows from the other. When two lines drawn on one matter collide, the one not followed is not cancelled by that, nor is "there was no choice" used to sign for it after the fact.

This proposition is the place in the whole series most easily misread, so it gets a few more words.

One line of thought goes: strike out everything that cannot be done, and what is left will naturally be the thing one cannot get out of doing. Tried on the report, this idea fails at once. He struck out "change the number," and what was left was not "go and tell the other team"; what was left was a wide field, including saying nothing, including quitting, including talking to his manager once more, including writing an anonymous letter. Striking one thing out does not automatically turn another into a cannot-not. Cannot-not-doing is a separate drawing of a line, and what it strikes out is the option "not do this." It is a second act, not what is left over from the first.

The same holds the other way round. A great many things have not been struck out, and that does not make them things he cannot not do. Not being struck out is not yet the same as being open either way: some of these he is still weighing, and some he has not touched at all this time; §3 comes back to this. What may be done or left undone are the things he has already judged that way; the difference between these and what he cannot not do is plain within the person, even though, from outside, what he does may be the same act.

One more situation is worth setting out here in advance: two lines drawn sometimes collide. Suppose that if he tells the other team, the young colleague in his group who made the first draft the way the manager wanted will be pushed forward to take the blame, and he equally cannot not protect her, cannot let her take the fall for others. Both lines are his own, and this time he can take only one of them first. The Conflict Treatise expressly left this cell to Action Theory (Conflict Treatise §1): a conflict in which both items can be reduced to values already avowed has been dealt with by Value Theory; what is taken up here is the part that remains.

Where the line about protecting her comes from is worth making clear first. It is not deduced from some value he has avowed: he has not necessarily avowed anything like "don't let a subordinate lose out," and there is no call for him to have avowed it first; as was said above, cannot-not-doing does not need a goal affirmed beforehand. It rests on three things. The fact: she really will be pushed forward to take the blame for others. The position he stands in: he does not regard her as someone who can be made to take the blame for others, and this is a matter of the same layer as his not regarding the other team as people it would be all right to keep in the dark. And his judgment this time: he is able to protect her. With all three in place, this line holds; if one of them changes, for instance if the manager backs down and drops the matter, this line changes along with it.

He picks one, and the one not taken is not voided by that: its conditions are still there, the young colleague may still be pushed forward to take the blame, and so that line still stands in his name. Nor does he close its account with "I had no choice," as if that line had never mattered. Which line to take first this time is his judgment; whether the other line still stands turns on its own conditions, and that is a separate matter. If part of protecting her is exactly a value he avowed earlier, there is a further proposition for that part: a yielding not accompanied by revision or exit does not change the standing of what was yielded, as this system established in Value Theory (Value Theory P5 V65).

§3 The Stretch in Between, and What This Judgment Leaves

After the two lines are drawn, the matter of the report stands like this for him: changing the number, struck out; not telling the other team, struck out. And what remains?

What remains is of three kinds. What has nothing to do with the matter was never in this judgment to begin with, and does not count as remaining.

One kind is what may be done or not: he has already judged it, and this way or that way are both fine. How to tell the other team, in person or by email; whom to tell first; whether to propose to his manager, at the same time, a basis both sides can live with. None of these has been struck out, and none has been marked as a cannot-not. This stretch is where his freedom lies. The differences between one person and another mostly show here: two people may both leave the number unchanged and both be unable not to speak, and the way they speak can be entirely different. Choices in this stretch rest on the person's judgment and his aesthetic sense; §5 comes back to this.

Another kind is what he is still weighing and cannot yet judge. Tell them today, or wait until the report is in? Speak too early, and the report may be held back and redone, and the other team still will not see it; speak too late, and the other team loses a few more days. This question is already within his consideration this time; it simply has not been decided. It is not a remainder; it is still inside this judgment, waiting for him to judge it. This judgment runs until he makes his decision under it; whatever he is still weighing at that point is not renamed a remainder either, but goes on into the next judgment.

The other kind is what this judgment did not take up. For example: whether he still wants to remain at this company. This is connected with the report; the matter of the report brought it before him. But this judgment did not take it up: it neither struck it out, nor marked it as a cannot-not, nor put it on the scales. It lies beyond what this judgment covers.

This system calls such a thing a remainder (M-00E §1.1). An item that bears on what this drawing of a line is doing, yet lies outside what this drawing of a line handles, is a remainder of this drawing of a line. Whether it is a remainder does not turn on whether anyone has spotted it: something no one has thought of can be one too (M-00E §1.5). Here, he has already seen that this item has not been dealt with, and for now he lets it stay; "stays" describes how he treats it once he has recognized it.

D4 After a judgment, three kinds of thing remain: what has been judged open either way, doing or not doing; what has entered consideration but cannot yet be judged; and what bears on what this judgment is doing but lies outside what it handles this time. Only the third kind is the remainder of this judgment; it stays, counted neither as settled nor as something that may be cleared away.

"Stays" is the exact word. There is no call for him to decide today whether to quit, but he cannot pretend the question does not exist either, still less take it as settled because he did not decide it today. After a while it may come before him again, and then it will be another judgment. The remainder of this judgment is often the starting point of the next.

One more thing is stated flatly here. A remainder gives possibility, not a push. The question "do I still want to remain here?" sitting there will not by itself push him toward quitting, nor by itself push him toward staying on. The only things that push him are his own judgment and choice. In this system there is no motor called the remainder (Value Theory P4 V45; Meaning Theory P1 Y12).

§4 Why Cannot-Not, Not Ought

Someone may ask him: on what grounds do you say you cannot not tell the other team?

This series does not answer that question with ought. Its premise is that each person legislates only for himself, and "you ought" is a way of legislating for someone else. Someone may say that "I ought" is said to oneself and does not legislate for anyone else. But ought, as a way of speaking, is itself a command, issued to a person from outside; even when the one issuing the command is his own reason, the sentence still has the shape of a command (Reason Cannot Not §2.1). And there is something "I ought" cannot keep out: a rule poured in from outside can also be voiced as "I ought," so this way of speaking cannot by itself rule out colonization from outside. Cannot not does not state a command; it states that something cannot be cancelled in this position, and it lies closer to his own law. Yet it too does not guarantee that a given line is sure to issue from himself; how colonization is recognized is taken up in Paper 2 and Paper 3.

Then what does cannot not rest on?

It rests on a position: a person still lives by his own law, still takes himself as an end and takes others as ends. For someone standing in this position, some things cannot be cancelled. The Conflict Treatise has used a statement of this shape: in a conflict that has already passed the gates, a person cannot, while still claiming to occupy the same position, cancel any one of the items in question (Conflict Treatise §11). That section covers only that conflict; this series applies the shape to action in general, and answers for that step on its own account. But what cannot be cancelled comes in two layers, and they are looked at separately.

One layer is how people are regarded. He does not regard the other team as people it would be all right to keep in the dark; at this layer, no one who stands in this position can cancel that. This layer is shared because everyone standing here has recognized that a person is an end. Where judgment can have something in common with other people's, it mostly falls on matters that involve people. Arithmetic and physics are another matter: there, reasoning things through brings people to the same place; on matters that involve people, shared judgment rests on everyone having recognized that a person is an end.

The other layer is what to do this time. Going to tell the other team is his judgment on this matter: based on his own law, on the facts, on timing, and on whether there is any other way. This layer can be questioned: others can ask him why, and he cannot turn the question away with nothing more than "this is my judgment"; nor does what he cannot yet say clearly count as having been made clear. This layer can also be revised: if, say, he later learns that the other team knew all along, telling them is no longer something he cannot not do. And the doing may be obstructed: his manager gets to him first, or he falls ill. Two people who both treat people as ends may well judge differently at this layer: someone else might go to the manager first, rather than to the other team. That they judge differently does not show that one of them has left this position; nor does not managing it this time.

D5 A cannot-not always hangs on conditions, and has two layers. The shared layer is mostly about how people are regarded, and cannot be cancelled while one stands in the position of taking oneself and others as ends; a person can leave this position in a given matter, but cannot stand in it and cancel it at once. The case layer is mostly about what to do this time; it is the person's own judgment, can be questioned, can be revised, and may be obstructed in the doing; to judge differently from others at this layer, or not to manage it this time, is not to have left the position.

The parts of this proposition that speak of leaving matter most, because they are often read backwards.

It does not say he has no way out. He has one. He can leave this position in this one matter: for the sake of his own review this year, regard the other team as people it would be all right to keep in the dark, change the number, and say nothing. Leaving means leaving that way of regarding people; whether he changed the number or spoke up does not settle it on its own: changing the number and not speaking might also be fear, or being held down, or wanting to act and not managing to, and that is for Paper 4 to discuss. If he really does leave in this way, this series does not judge him; in this one matter, this cannot-not is no longer there for him, and that is all; this series still speaks to him. Leaving goes matter by matter: having left in this one matter does not mean he is from then on no longer a person who treats people as ends. He can also, standing in this position, stop: leave the matter unhandled for now, and leave the company first. Stopping does not prove that cannot-not wrong; it only means he is not handling it this time. Only one thing is blocked: saying "I am a person who treats people as ends" while also saying that it is all right to keep the other team in the dark. That cannot be done, not because anyone forbids it, but because those two statements cannot sit together in one person.

This is the kind of guidance this series gives: it does not choose for the reader; it only makes clear what a person standing here cannot cancel, and each of these is one the reader can leave in a given matter.

§5 Judgment: Rules Cannot Do It; Only a Person Can

Every line drawn above was drawn by him. Who tells him where to draw it?

No rule can draw it for him. This system established that long ago: the step of applying a rule to the matter in front of one cannot itself be given by a rule (The First Critique, ch. 4; Value Theory P4 V45). Rules can come in, facts can come in, reasons and feelings can all come in; they are all material for his judgment. But which rule to apply, which one to take, and, once the judgment is made, taking it up and carrying it: only he can do that. This series calls this judging judgment. Judgment is not a special kind of cleverness; it is simply a person's capacity, when a matter is set before him, to mark off what he does not do, to mark off what he cannot not do, and to let the rest stay. Any set of rules that tries to replace it still comes down, in the end, to a person judging whether it applies to the case in front of him.

Then what does he judge by?

He judges by himself, and that has a concrete content. When a person does something, it feels smooth to him, or it feels awkward. This system, on the side of aesthetics, established a theorem about the thing itself: the stronger the control imposed from outside, unrelated to the thing's own logic, the weaker the beauty; the thing's own constraints do not diminish its beauty but let it unfold more fully (Judgment and Aesthetics, Theorem 3). The smoothness or awkwardness a person feels in doing something is a different thing: it is his feeling. That theorem cannot be run backwards to say that feeling smooth means he is following his own logic. What the feeling can do is make him go back and ask: is this his own law unfolding, or is some other measure deciding in his place? In the matter of the report, changing the basis felt awkward to him; not changing it and going to speak felt smooth, even though the consequences would be hard. This smoothness or its absence gives his judgment a direction: an opening for going back and asking, which does not answer the question for him.

D6 Judgment cannot be done by rules in anyone's place: rules, facts, reasons and feelings can all enter a judgment, yet applying, weighing and taking up remain the work of the one who judges. The feeling of whether something goes smoothly gives judgment a direction; it is a clue, fallible, liable to be bent by habit; it does not prove that something issues from oneself, still less can it be used to judge others.

Two remarks follow at once.

First: this feeling is a clue, not a proof. Smoothness does not prove that the thing issues from his own law; what a person is used to doing also goes smoothly. Awkwardness does not prove that someone is controlling him either; new things feel awkward to do in any case. This feeling can be bent by habit, and still less can it be used to judge others: that something feels smooth to him does not show that someone who does it the other way is wrong.

Second: he can get it wrong; this series does not guarantee that he judges right (Conflict Treatise §3). Perhaps the other team knew all along, and his telling them was not needed at all. This kind of mistake can be checked now, without waiting for later: a look at that team's recent records will show it. He may also be stumped by a question: a colleague asks him whether he is going to tell them for the other team's sake, or so that he can feel all right with himself; in answering, he may find he has not thought it through. Checking the facts and being questioned are two openings for correction that can be used right away. There is another kind of mistake that cannot be checked right away: he may have taken an unease that came from somewhere else for his own law. That waits for looking back after the fact, which Paper 3 writes about.

So what judgment gives is not being right. What it gives is this: this time it is he himself who judges; if he judges wrong, it is he himself who can go back and check, and go back and change it. Whether that judgment issued from his own law may not be something he can tell at the time.

§6 Reasons and Intention

He has made up his mind: before he leaves work today, he will tell them in person. How this decision came about is worth one more look.

He has a pile of reasons in hand. The later the other team finds the problem, the harder it will be to clean up; his own review for the year will look bad; the young colleague in his group may be pushed forward to take the blame; the report is due today, and if he speaks too early it may be held back and redone; he has always valued honesty, something he avowed long ago. These are all considerations he can use as grounds at this moment. This system, in Meaning Theory, gave them a name: the field of reasons (Meaning Theory P1 Y12). It is the reasons a person can draw on at a given time as grounds for choice, revision and action, taken as a whole. It is not how many reasons there are, and it does not judge whether these reasons hold up; it only records which ones he actually uses as grounds. Whether these reasons hold up normatively, this series does not answer with ought; what this series can say is the kind of cannot-not described in §4, which hangs on conditions.

Reasons do not turn into action by themselves. This sentence is easy to overstate, so it is worth putting exactly. Reasons can support a judgment, and they can also count against it: that the later the other team finds out, the harder things will be to clean up, supports speaking early; that the report may be held back counts against speaking early. Sometimes reasons can even yield a judgment directly: if he finds out that the other team knew all along, going to tell them is no longer called for, and that step is one the facts work out for him. But in the matter of the report, this pile of reasons, laid side by side, does not point him to any one course of action. The same reasons can support speaking today, and can support waiting until the report is in; they can support telling them in person, and can support writing an email. This system wrote the same thing in Meaning Theory: one relationship can hold several ways of acting that are not equivalent, and one and the same way of acting, as seen from outside, can issue from wholly different relationships (Meaning Theory P4, §5).

Making the judgment into this time's line is an act he performs. He has weighed these reasons, and then he draws the line: before he leaves work today, in person. At the moment he draws it, he has an intention: what he goes and does next is this. This intention grows out of that pile of reasons, but it is not derived from them; the reasons gave it material, and the one who drew the line was he. Meaning Theory has said that it supplies a field of reasons, and does not supply rules for calculating from reasons to action (Meaning Theory P4, §5); this series does not supply them either. There is no table into which reasons are entered and out of whose other end the action comes. This series is not a decision procedure; the reader cannot take away from it a method that works out for him which thing to do. What he can take away is a clear view of what he himself is doing.

Among this pile of reasons, one thing is still missing: a common scale. The other team's losses, his own review, whether the young colleague will be pushed forward, speaking today or tomorrow: these come from different sources and involve different people; they cannot be converted into one kind of number, added up, compared by size, with the largest picked out. Why they cannot be converted into one kind of number: on the side of value standing there is no quantity to begin with; what makes things seem higher or lower is mostly whether a window will close and whether a harm can be undone, and these vary with the situation; they are not a scale (Value Theory P5, §3). Without any such master scale across persons and across sources, the matter still lies before him this afternoon, waiting for him to decide. When he decides, he goes by the concrete things in the matter in front of him: who is involved; where the edge lies that cannot be crossed; which harm, once done, cannot be undone; which step, once taken, can still be withdrawn; which things can stay and be dealt with next time. This system has said, in Value Theory, that local decisions on action rest on exactly these (Value Theory P1, §8).

This does not mean he compares nothing. He compares all the same: which side carries more risk, when it is better to speak, whether he is able to do this at all. The prudence that can be spoken of on this afternoon lies in these comparisons; Value Theory left it to this series (Value Theory P5 V70). Only, the comparing is done within this one matter, and once it is done, the line is still his to draw; comparison does not grow by itself into a scale that settles the case for him.

D7 Reasons do not turn into action by themselves: reasons can support or oppose a judgment, and under limited conditions can also entail one; making the judgment into this time's line is an act done by the one who judges. Drawing a line does not presuppose a common master scale across persons and across sources; this does not rule out comparing risk, timing and feasibility within one matter.

The two colliding lines of §2 are ordered in the same way. Whether to protect the young colleague first or to tell the other team first: he judged an order. This order is the order of his action this time, not a ranking of the two lines: the one taken first is not heavier for that, and the one not taken is not lighter for that. In a different situation he might take the other one first; the order would change, and the weight of neither line would have moved because of it (Value Theory P1, §8; Value Theory P5, §3).

What he has avowed, honesty for instance, is also in this pile of reasons. It comes in as a reason, and is weighed along with the other reasons; it does not draw the line for him automatically. How something avowed earlier becomes the act of this moment is the entry point of this series (Value Theory P4, §7); Paper 3 continues from there.

Nor is a judgment done once and for all. Today's line is drawn, and the things that stay are still there: whether he will remain at this company, when it comes back before him some day, will be another judgment, still his to make. Whether all the lines he drew today issued from his own law may not be something he can tell at the time; that is what Paper 3 writes about.

Conclusion

This paper has set out seven propositions. Taken together they say one thing: for a person who takes both himself and others as ends, judgment and choice in a matter are acts of negation; the negation issues from himself, hangs on conditions, and is judged by himself; no set of rules, and no master scale, can judge in his place.

What this paper has not said is also recorded here. It does not say he judges right: getting it wrong is possible. It does not say the lines he draws all issue from his own law: that is Paper 3's business. It does not say that once a line is drawn he manages to do it: how body, emotion, capacity, habit and hesitation come in is Paper 4's business. It does not give the reader a method for deciding. What it gives is a sentence form: not-doing, cannot-not-doing, open either way, cannot yet be judged, remainder. The later papers all speak in this sentence form.

The Seven Propositions of This Paper

D1 Not-doing is a person's negation of doing something; what he strikes out may also be a way of regarding a person or a thing. It does not need a goal affirmed beforehand.

D2 Cannot-not-doing is a person's negation of not doing something; it, too, does not need a goal affirmed beforehand.

D3 Not-doing and cannot-not-doing are two acts, and neither follows from the other. When two lines drawn on one matter collide, the one not followed is not cancelled by that, nor is "there was no choice" used to sign for it after the fact.

D4 After a judgment, three kinds of thing remain: what has been judged open either way, doing or not doing; what has entered consideration but cannot yet be judged; and what bears on what this judgment is doing but lies outside what it handles this time. Only the third kind is the remainder of this judgment; it stays, counted neither as settled nor as something that may be cleared away.

D5 A cannot-not always hangs on conditions, and has two layers. The shared layer is mostly about how people are regarded, and cannot be cancelled while one stands in the position of taking oneself and others as ends; a person can leave this position in a given matter, but cannot stand in it and cancel it at once. The case layer is mostly about what to do this time; it is the person's own judgment, can be questioned, can be revised, and may be obstructed in the doing; to judge differently from others at this layer, or not to manage it this time, is not to have left the position.

D6 Judgment cannot be done by rules in anyone's place: rules, facts, reasons and feelings can all enter a judgment, yet applying, weighing and taking up remain the work of the one who judges. The feeling of whether something goes smoothly gives judgment a direction; it is a clue, fallible, liable to be bent by habit; it does not prove that something issues from oneself, still less can it be used to judge others.

D7 Reasons do not turn into action by themselves: reasons can support or oppose a judgment, and under limited conditions can also entail one; making the judgment into this time's line is an act done by the one who judges. Drawing a line does not presuppose a common master scale across persons and across sources; this does not rule out comparing risk, timing and feasibility within one matter.

---

Acknowledgments

This paper is the first in the Action Theory branch of the SAE (Self-as-an-End) philosophical system, and is signed by Han Qin alone. All philosophical argument and all decisions between alternatives were finally adjudicated by the author, who bears sole responsibility for the final text.

Drafting assistance and review were carried out with the help of large language models. Drafting and the internal precision audit were undertaken by Claude (Anthropic), and structural gatekeeping with pre-publication signature by ChatGPT (OpenAI); the first two review rounds of the series outline also included ontological-limit and contradiction detection and boundary testing, undertaken by Gemini (Google DeepMind) and Grok (xAI) respectively. The ten texts of the series received final review from the internal precision audit and the structural gate, followed by a check of every revised sentence. Thanks are due to these systems and to the research and engineering teams behind them. Their comments were adjudicated by the author item by item, and both what was adopted and what was rejected are on record.

This paper continues the standing product of an upstream-paraphrase audit table, comparing item by item what the paper says of upstream texts against their published wording. The record of revisions and withdrawals across the rounds of review is kept in the series outline and the review returns, and is not reproduced here. This English edition is an independent rewrite of the Chinese final text; a separate parity audit compared it with the Chinese paragraph by paragraph. The review markedly improved the precision and consistency of this paper, and thanks are due for it.

---

References

All works below are archived on Zenodo, with Chinese and English versions under the same record. The DOIs cited are concept DOIs, always pointing to the latest version; version numbers and access dates are therefore not given separately. All works are single-authored (Han Qin). Drafting and review assistance is described in the Acknowledgments.

The text restates upstream work in plain words, with the paper and proposition number, or the title of an unnumbered section, noted in parentheses at the place of use. Concept DOIs always point to the latest version, so proposition numbers may drift as upstream is renumbered. This paper's restatements are compared item by item with the upstream originals in its transcription audit table.

I · Works actually called upon in this text

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

Qin, H. (2026). On the Dynamics of Value (SAE Value Theory, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22050204

Qin, H. (2026). On the Conflict of Value (SAE Value Theory, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22122819

Qin, H. (2026). What Kind of Relation Meaning Is (SAE Meaning Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22652134

Qin, H. (2026). The Carrier's Boundary (SAE Meaning Theory, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22738709

Qin, H. (2026). Remainder: A Portable Method (SAE Methodology, M-00E) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22849616

Qin, H. (2026). SAE Judgment and Aesthetics (SAE Foundations) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19296710

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

Qin, H. (2026). SAE Conflict Treatise: How Subjecthood Is Not Swallowed by Conflict Itself [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22313628

Qin, H. (2026). Reason Cannot Not: An SAE Refinement of Kant’s Four Formulations of the Categorical Imperative [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19751512

II · Background works

Qin, H. (2026). The Finite Bearer's Carrying-Forward (SAE Meaning Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22713651

III · This series

Qin, H. (2026). What Action Is Not (SAE Action Theory, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22970747

Qin, H. (2026). Anti-Colonization (SAE Action Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22970758

Qin, H. (2026). Is It Mine? (SAE Action Theory, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22970762

Qin, H. (2026). Can and Cannot (SAE Action Theory, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22970769

引言

一个人坐在电脑前，报表就在屏幕上。上司要他把一项指标换个口径重算。换过之后数字好看，年底的评价也会好看；只是换了口径，另一个团队那边正在出的问题就被盖住了。这份报表只送上级，那个团队看不到它，要到下个季度，问题才会自己冒出来。他知道这件事。他也知道，不换口径，这份报表交上去，他自己这一年会难过。

他要做一个决定。这个系列写的就是站在这个位置上的人：一件事摆在面前，没有人替他决定，也没有一条现成的规则替他决定，而他又不想把自己交出去。我不替他选，我写的是这个位置本身是怎样的，站在这里的人在做一件什么事。

这个系列说的人，把自己当目的，也把别人当目的；前置篇说过，这是这个系列的定义域，不是对读者的要求。站在这个前提上的人，我叫他行自己的法的人。法（one's own law）这个词是这个体系一直用的，说的是一个人给自己立的、只管自己的那条法。别人的法哪怕内容一样，也是他自己到达的，不是我给的（A16 一）。

这一篇要立的是全系列的句式。往后各篇说的每一件事，都用这个句式说。

这一篇要回答的是：一个人在一件事上作决定的时候，他做的是什么。回答分六步。第一、二节说他做的两种否定，不做与不得不做，以及两次否定撞在一起时怎样；第三节说一次判断之后剩下的东西；第四节说这些不得不凭什么，为什么不写成应当；第五节说线由谁来划，也就是判断力；第六节说理由怎样变成行动。各节立一到两条命题，全篇七条（D1 至 D7），末尾合列。正文不点名外部理论家，不援引历史事件；上游各篇的结论只用平实话重述，括注篇次；这一篇与已有理论怎样对接，留给 Paper 0。

第一节　不做

他先想清楚的是：这个数字他不改。

要看清这一步是什么，先看它不是什么。它不是他先立了一个叫「诚实」的目标，然后从目标推出这个数字不能改。他没有做那个推导。他一看见这件事，就知道这件事他不做。目标是后来才说得出来的，如果有人问他为什么，他也许会说是为了诚实，也许说不出来，也许说出来的理由第二天自己也觉得不对。理由可以说错，这一划却是实实在在划下了。

这就是行自己的法的人做的第一种事：否定去做某件事。我把它叫作不做。

一次判断把一件事定下来，这个系列叫它一次划线（这个词借自 M-00E 1.1）。划掉做，是不做；划掉不做，是不得不做，下一节说。

他划掉的，其实不只是一件事。在不改数字之前，还有更早的一划：他不把那个团队看成瞒过去也无所谓的人。这一划划掉的不是一个动作，是一种看待人的方式。改不改数字，是这一次的事；不把人看成瞒过去也无所谓的，是他在这件事上怎样对待人。这两种划线落在两层，第四节再说。

D1　不做，是一个人对做某件事的否定；他划掉的，也可以是一种看待人或事的方式。它不需要先有一个被肯定的目标。

不做不是消极。一个人一辈子不做的事，比他做的事多得多，而其中绝大部分他从来没有想过，那些不算不做，那只是没有发生。不做是想过的，是在一件事摆到面前的时候，把它划掉。划掉是一个动作。这个动作没有人看见，也没有报表记录它，但它是这个人在这件事上最先做的那一件。

不做也不等于不准。不准是别人对我说的话，或者制度对我说的话；不做是我对自己说的。同一件事，制度不准我做而我也不做，是两件事叠在一起，不是一件事。制度撤了那条规定，不做还在那里。

第二节　不得不做

他把数字留着不改。接下来他发现了第二件事。

那个团队看不到这份报表，问题照样要到下个季度才会自己冒出来；这一点在他不改数字之后仍然在那里。他可以只不改数字，然后什么也不说，报表照交，让别人去发现或者不发现。这样做没有什么禁止他。他坐在那里，发现自己做不到。他没法不去告诉那个团队里的人。

这是行自己的法的人做的第二种事：否定不去做某件事。我把它叫作不得不做。

D2　不得不做，是一个人对不做某件事的否定；它同样不需要先有一个被肯定的目标。

「不得不」（cannot not）三个字要说准。它不是有人逼他。上司没有要他去告诉别人，恰恰相反。它也不是良心在逼他，如果良心是一个站在他外面的东西的话。它是他自己的法说话：在这个位置上，「不去说」这一项被他划掉了；划掉的是不说，于是说，成了他不得不做的事。这里的「不得不」是他自己的，不是外面给的。

D3　不做与不得不做是两个动作，互不推出。两次划线在一件事上相撞时，没有走的那一条不因此取消，也不拿「别无选择」去替它补签。

这一条是全系列最容易被读错的一处，要多说两句。

有一种想法是：把不能做的都划掉，剩下的自然就是非做不可的。这个想法在报表这件事上一试就不成立。他划掉了「改数字」，剩下的并不是「去告诉那个团队」；剩下的是一大片，包括什么也不说，包括辞职，包括跟上司再谈一次，包括写一封匿名信。划掉一件事，不会自动把另一件事变成不得不。不得不做是另一次划线，划的是「不去做这件事」这个选项。它是第二个动作，不是第一个动作的余数。

反过来也一样。没有被划掉的事多得很，它们不因此就成了不得不做的。没有被划掉，也还不等于做与不做都可以：有的他还在掂量，有的这一次根本没去碰，第三节再说。可做可不做的，是他已经这样判过的那些；它们和不得不做的事之间的差别，在一个人身上是分明的，尽管从外面看，他做出来的可能是同一个动作。

还有一种情形要先说在这里：两次划线有时会撞在一起。假如去告诉那个团队，组里那个照上司意思做出初稿的年轻同事就会被推出来顶罪，而他同样不得不护住她，不让她替别人担这一口锅。两条都是他自己的，这一次只能先走一条。冲突论把这一格明文留给了行动论（冲突论 一）：两项都能还原为已认领价值的冲突，价值论已经处理；这里接的是剩下的那一部分。

护住她这一条从哪里来，要先说清。它不是从他认过的哪一项价值里推出来的：他未必认过「不让下属吃亏」这样一条，也用不着先认过，前面说了，不得不做不需要先有一个被肯定的目标。它立在三样东西上：事实，她真会被推出来替别人担；他站的位置，他不把她看成可以替别人担一口锅的人，这和他不把那个团队看成瞒过去也无所谓的人，是同一层的事；还有他这一次的判断，他护得住她。三样都在，这一条就在；其中一样变了，比如上司改口，不再追究，这一条也跟着变。

他选了一条，没走的那一条并不因此作废：它的条件还在，那个年轻同事仍可能被推出来顶罪，那一条就还在他名下。他也不拿「我别无选择」替它结账，好像那一条从来不重要。这一次先走哪条，是他的判断；那一条还在不在，看它自己的条件，是另一件事。倘若护住她这件事里，有一部分正是他早先认下的一项价值，那一部分另有一条可用：没有跟着修订或退出的让，不改变被让的那一项的地位，这个体系在价值论里立过（价值论 P5 V65）。

第三节　中间那一片，和这一次判断剩下的东西

两次划线之后，报表这件事在他这里是这样的：改数字，划掉了；不去告诉那个团队，划掉了。剩下的呢。

剩下的分三种。和这件事无关的，本来不在这一次判断里，不算剩下的。

一种是可做可不做的，他已经判过，这样做那样做都行。怎么告诉那个团队，当面说还是写邮件；先告诉谁；要不要一并给上司提一个两边都过得去的口径。这些事没有一件被划掉，也没有一件被划成不得不。这一片是他的自由所在。人和人的差别，多半在这一片上显出来：同样是不改数字，同样是不得不去说，说的样子可以完全不同。这一片里的选择，靠的是这个人的判断和他的审美，后面第五节说。

一种是他还在掂量、一时判不下来的。今天就说，还是等报表交了再说：说早了，报表可能被压下来重做，那个团队照样看不到；说晚了，那个团队又多耽误几天。这件事已经在他这一次的考虑里，只是还没判定。它不是余项，它还在这一次判断里，等他判。这一次判断，到他照它做出决定为止；到那时还在掂量的，也不改叫余项，跟着进下一次判断。

另一种是这一次判断没有去处理的。比如：这家公司他还要不要待下去。这件事跟报表有关，报表这件事把它带到了他面前；但这一次判断没有去处理它，既没有把它划掉，也没有把它划成不得不，也没有把它放进掂量。它是这一次判断管不到的。

这个体系把这种东西叫余项（remainder；M-00E 1.1）。一项内容跟这一次划线要做的事有关，却不在这一次划线的处理之内，它就是这一次划线的余项。它是不是余项，不看有没有人认出它：没被想到的，也可能是（M-00E 1.5）。这里，他已经看见这一项还没处理，先把它留着；留着，说的是他认出以后怎样对待它。

D4　一次判断之后剩下的有三种：已经判过做与不做都可以的；进入了考虑而还判不下的；与这次判断要做的事相关、却不在这次处理之内的。只有第三种是这次判断的余项；它们留着，不算已决，也不算可以清掉。

留着，是一个准确的词。他不需要今天决定要不要辞职，但他也不能装作这件事不存在，更不能因为今天没有决定就认为它已经解决了。过一段时间，这件事可能再回到他面前，那时候是另一次判断。这一次判断的余项，常常是下一次判断的起点。

还有一件事要在这里说死。余项给的是可能，不是推力。「还要不要待下去」这个问题留在那里，不会自己把他推向辞职，也不会自己把他推向留下。推他的只有他自己的判断和选择。这个体系里没有一台叫余项的马达（价值论 P4 V45；意义论 P1 Y12）。

第四节　为什么是不得不，不是应当

有人会问他：你凭什么说你不得不去告诉那个团队。

这个系列不用「应当」回答这一问：这个系列的前提是每个人只给自己立法，而「你应当」是替别人立法的说法。有人会说，「我应当」是对自己说的，并不替别人立法。可「应当」这种说法本身是一道命令，从外面向人发出；哪怕发出命令的是他自己的理性，句子的样子也还是命令（《理性不得不》二·一）。「我应当」还有一处挡不住：一条从外面灌进来的规矩，也可以用「我应当」说出口，这个说法本身排除不了外面的殖民（colonization）。「不得不」说的不是命令，是一件事在这个位置上取消不了，离他自己的法更近；可它也不担保那一条一定出自他自己，殖民怎样认出来，第二篇与第三篇说。

那么「不得不」凭什么。

它凭的是一个位置：一个人还行自己的法，还把自己当目的、把别人当目的。站在这个位置上，有些事就取消不了。这个说法的形状，冲突论用过：在一场已经过门的冲突里，一个人不能在仍然宣称占据同一位置的时候，取消其中任何一项（冲突论 十一）。那一节只管那一场冲突；这个系列把它用到一般的行动上，这一步由这个系列自己担。不过取消不了的东西分两层，分开来看。

一层是怎样看待人。他不把那个团队看成瞒过去也无所谓的人；这一层，凡是站在这个位置上的人都取消不了。这一层之所以是共同的，是因为站在这里的人都认了人是目的。判断里能和别人有共同之处的，多半落在和人有关的事上。算术和物理是另一回事，讲道理就能讲到一处；和人有关的事能有共同的判断，靠的是大家都认了人是目的。

另一层是这一次做什么。去告诉那个团队，是他在这件事上的判断：依据他自己的法，依据事实，依据时机，也依据还有没有别的办法。这一层可以被问：别人可以问他为什么，他不能只凭一句「这是我的判断」把问题挡回去；一时说不准的地方，也不能算已经说清。这一层也可以修订：比如他后来得知那个团队早就知道了，去说就不再是他不得不做的事。做的时候还可能受阻：上司先一步找他谈了，或者他病了一场。两个同样把人当目的的人，在这一层完全可以判得不一样：另一个人也许会先去找上司，而不是先去找那个团队。判得不一样，不说明其中一个人离开了这个位置；这一次没做到，也不说明。

D5　一条不得不总是挂着条件，并且分两层。共同的一层多半是怎样看待人，站在把自己和别人都当目的的位置上就取消不了；一个人可以在一件事上离开这个位置，不能一边站着一边取消它。个案的一层多半是这一次做什么，是本人的判断，可以被问，可以修订，做的时候可能受阻；在这一层和别人判得不同，或者这一次没做到，都不等于离开了这个位置。

这一条里讲离开的那几句最要紧，因为它们常常被读反。

它不是说他没有退路。他有。他可以在这一件事上离开这个位置：为了自己这一年的评价，把那个团队看成瞒过去也无所谓的人，改了数字，什么也不说。离开，是离开那一种看待；单看他改没改数字、说没说，还定不了：改了数字没去说，也可能是怕，是被压着，是想做而没做到，那是第四篇要说的事。他若真这样离开了，这个系列不判他；在这一件事上，这一条不得不对他不在了，仅此而已，这个系列仍然对他说话。离开是按件的：这一件事上离开了，不等于他从此不再是一个把人当目的的人。他也可以在这个位置上停下来，先不处理，先离开这家公司再说。停下不证明那条不得不错了，只是他这一次不处理了。被挡住的只有一件事：一边说我是把人当目的的人，一边说那个团队瞒过去也无所谓。那一件做不成，不是因为有谁不准，是因为那两句话在同一个人身上放不到一起。

这个系列给的指引就是这样的：不替读者选择；它只说清站在这里的人取消不了什么，而每一条，读者都可以在一件事上离开。

第五节　判断力：规则代做不了，只有人

上面说的每一次划线，都是他划的。谁告诉他线划在哪里。

没有规则能替他划。这一点这个体系早已立下：把规则用到眼前这件事上，这一步不能再由规则给出（《SAE第一批判》第四章；价值论 P4 V45）。规则可以进来，事实可以进来，理由和感受都可以进来，它们都是他判断的材料；可是用哪一条、取哪一样、判下来之后担起来，只能由他自己完成。这个系列把这个判叫判断力（judgment）。判断力不是一种特殊的聪明，它就是一个人在一件事摆到面前的时候，划出不做、划出不得不做、把余下的留着的那个能力。任何一套试图代替它的规则，最后还是要由一个人来判眼前这个情况适用不适用。

那么他靠什么判。

他靠他自己，而这话有一个具体的内容。一个人做一件事的时候，会感到顺，或者别扭。这个体系在美学那边立过一条定理，说的是一件事本身：外加的、与这件事自身道理无关的控制越强，美越弱；这件事自身的约束不减损美，反而让它展开得更充分（《判断力与美学》定理三）。人做事时的顺与别扭，是另一样东西，是他的感受；不能照那条定理倒推回去，说顺了就是在按自己的道理走。它能做的，是让他回头问：这是自己的法在展开，还是有别的尺度在替他作主？报表这件事上，改口径让他别扭；不改而去说，他觉得顺，哪怕后果是难过的。这个顺不顺，给了他的判断一个方向：一个回头去问的入口，不替这一问作答。

D6　判断力不能由规则代做：规则、事实、理由与感受都可以进入判断，而适用、取舍与承接仍须由判断的人完成。做得顺不顺的感受给判断一个方向；它是线索，可错，可能被习惯养歪，不证明一样东西出自自己，更不能拿来判定别人。

要立刻补两句。

第一句：这感觉是线索，不是证明。顺，不证明这件事出自他自己的法；一个人做惯了的事，也会做得顺。别扭，也不证明有人在控制他；新的事做起来本来就别扭。这感觉会被习惯养歪，更不能拿来判定别人：他觉得顺，不说明别人那样做就错了。

第二句：判错也是可能的，这个系列不担保他判得对（冲突论 三）。也许那个团队早就知道了，根本不需要他说；这一种错现在就能核，不必等以后，查一查那边最近的记录就知道。他也可能被人问住：同事问他，你去说，是为了那个团队，还是为了自己心里过得去；他答的时候，也许会发现自己没想清楚。核事实和被人问，这两样是当下就能用的纠错的口。还有一种错，当下核不出来：他也许把别处来的不安当成了自己的法。那要等事后回看，第三篇写。

所以判断力给的不是对。它给的是：这一次是他自己在判；判错了，是他自己可以回头去查、回头去改的。那一判是不是出自他自己的法，当下未必分得清。

第六节　理由与意向

他定下来了：今天下班前，当面去说。这个决定是怎么来的，还要再看一眼。

他手上有一堆理由。那个团队越晚发现问题，越难收拾；他自己这一年的评价会难看；组里那个年轻同事可能被推出来顶罪；报表今天就要交，说早了，它可能被压下来重做；他一向看重诚实，这是他早就认下的一件事。这些都是他此刻能拿来作根据的考虑。这个体系在意义论里给它们起过一个名字，叫理由场（field of reasons；意义论 P1 Y12）：一个人在一个时候，能拿来作选择、修订与行动根据的那些理由，合起来的样子。它不是理由的多少，也不替这些理由判站不站得住，只记下他实际拿来作根据的是哪些。这些理由在规范上站不站得住，这个系列不用「应当」来答；这个系列能说的，是第四节那种挂着条件的不得不。

理由不会自己变成行动。这句话容易说过头，要说准。理由可以支持一个判断，也可以反驳它：那个团队越晚越难收拾，支持早说；报表可能被压下来，反驳早说。有的时候，理由还能把一个判断直接推出来：他若查到那个团队早就知道了，去说这件事就不必做了，这一步是事实替他推出来的。可是在报表这件事上，这一堆理由摆在一起，并没有替他指出是哪一个做法。同样这些理由，撑得起今天说，也撑得起交了报表再说；撑得起当面说，也撑得起写一封邮件。这个体系在意义论里写过同一件事：同一段关系可以容纳好几个互不等价的做法，同一个外面看得见的做法，也可以出自完全不同的关系（意义论 P4 第五节）。

把判断做成这一次的划线，是他做的动作。他掂量过这些理由，然后划下去：今天下班前，当面说。划下去的那一刻，他有了一个意向（intention）：接下来去做的，就是这一件。这个意向从那一堆理由里长出来，却不是被那一堆理由推出来的；理由给了它材料，划线的是他。意义论说过，它提供理由场，不提供从理由到行动的演算规则（意义论 P4 第五节）；这个系列也不提供。没有一张表，把理由填进去，行动就从另一头出来；这个系列不是一个决策程序，读者从这里拿不走一个替他算出做哪一件的办法，能拿走的是看清自己在做什么。

这一堆理由之间，还缺一样东西：一把共同的尺。那个团队的损失，他自己的评价，年轻同事会不会被推出来，今天说还是明天说，这几样出自不同的来源，牵涉不同的人，不能折成同一种数，加起来比大小，再挑最大的那一个。它们为什么折不成一种数：价值地位那一侧本来没有量；让人觉得有高下的，多半是窗口会不会关、伤害收不收得回，这些随情境变，不是一把尺（价值论 P5 第三节）。没有这样一把跨人、跨来源的总尺，这件事照样在今天下午摆在他面前，等他定。他定下来，凭的是眼前这件事里具体的东西：牵涉到谁；哪里是越不过去的边；哪一种伤害一旦发生就收不回来；哪一步做了还能撤回；哪些事可以留着，下一次再处理。这个体系在价值论里说过，局部的行动裁断凭的就是这几样（价值论 P1 第八节）。

这也不是说他什么都不比较。他照样比：哪一边的风险更大，什么时候说更好，这件事他做不做得到。这个下午能讲的审慎，就在这些比较里；价值论把它留给了这个系列（价值论 P5 V70）。只是比较在这一件事里比，比完了，那一道线还是由他来划；比较不会自己长成一把替他定案的尺。

D7　理由不会自己变成行动：理由可以支持或反驳一个判断，在有限的条件下也能推出某个判断；把判断做成这一次的划线，是判断的人做的动作。划线不以一把跨人、跨来源的共同总尺为前提；这不排除在一件事里比较风险、时机与做不做得到。

第二节那两条撞在一起的划线，也是这样定先后的。先护住那个年轻同事，还是先去告诉那个团队，他判了一个先后。这个先后是他这一次动作的次序，不是给那两条排高下：先走的那一条不因此更重，没走的那一条也不因此更轻。换一个情形，他可能先走另一条；次序变了，哪一条的分量都没有因此动过（价值论 P1 第八节；价值论 P5 第三节）。

他认下过的东西，比如诚实，也在这一堆理由里。它进来，是作为理由进来的，和别的理由一起被掂量，不自动替他划线。一项早先认下的东西怎样成为这一刻的动作，是这个系列的入口（价值论 P4 第七节）；第三篇接着写。

判断也不是一劳永逸的。今天这一道线划下去，留着的那几件还在：这家公司还待不待下去，哪一天回到他面前，是另一次判断，仍由他来判。他今天划下的这些，是不是都出自他自己的法，当下未必分得清；第三篇写的就是这件事。

结语

这一篇立了七条。合起来说的是一件事：一个把自己和别人都当目的的人，在一件事上的判断与选择是否定性的；否定出自他自己，挂着条件，由他自己判；没有一套规则，也没有一把总尺，能替他判。

这一篇没有说的，也记在这里。它不说他判得对：判错是可能的。它不说他划下的线都出自他自己的法：那是第三篇的事。它不说划下了就做得到：身体、情绪、能力、习惯与迟疑怎样进来，是第四篇的事。它不给读者一套决定的办法。它给的是一个句式：不做，不得不做，可做可不做，还判不下，余项。往后各篇都用这个句式说话。

本篇立的七条

D1　不做，是一个人对做某件事的否定；他划掉的，也可以是一种看待人或事的方式。它不需要先有一个被肯定的目标。 D2　不得不做，是一个人对不做某件事的否定；它同样不需要先有一个被肯定的目标。 D3　不做与不得不做是两个动作，互不推出。两次划线在一件事上相撞时，没有走的那一条不因此取消，也不拿「别无选择」去替它补签。 D4　一次判断之后剩下的有三种：已经判过做与不做都可以的；进入了考虑而还判不下的；与这次判断要做的事相关、却不在这次处理之内的。只有第三种是这次判断的余项；它们留着，不算已决，也不算可以清掉。 D5　一条不得不总是挂着条件，并且分两层。共同的一层多半是怎样看待人，站在把自己和别人都当目的的位置上就取消不了；一个人可以在一件事上离开这个位置，不能一边站着一边取消它。个案的一层多半是这一次做什么，是本人的判断，可以被问，可以修订，做的时候可能受阻；在这一层和别人判得不同，或者这一次没做到，都不等于离开了这个位置。 D6　判断力不能由规则代做：规则、事实、理由与感受都可以进入判断，而适用、取舍与承接仍须由判断的人完成。做得顺不顺的感受给判断一个方向；它是线索，可错，可能被习惯养歪，不证明一样东西出自自己，更不能拿来判定别人。 D7　理由不会自己变成行动：理由可以支持或反驳一个判断，在有限的条件下也能推出某个判断；把判断做成这一次的划线，是判断的人做的动作。划线不以一把跨人、跨来源的共同总尺为前提；这不排除在一件事里比较风险、时机与做不做得到。

---

致谢

本篇为 SAE（Self-as-an-End）哲学体系行动论分支的第一篇，由秦汉独立署名。全部哲学论证与取舍由作者最终裁决，最终定稿与文责由作者独任。

起草辅助与评审均借助大型语言模型进行。起草与内部精度审计由 Claude（Anthropic）承担，结构闸门与发布前签署由 ChatGPT（OpenAI）承担；系列大纲的前两轮另有本体极限与矛盾检测、边界测试两方参加，分别由 Gemini（Google DeepMind）与 Grok（xAI）承担。十篇正文经内部精度审计与结构闸门两方终审，并经核句验收。谨向这些系统及其背后的研究与工程团队致谢。它们提出的意见由作者逐条裁定，采纳与驳回均记录在案。

本篇续行上游转述审计表这一常设产出，逐条比对本篇的转述与上游原句；各轮评审的改定与撤销记录存于本系列的大纲与交回件，不录于此。评审显著提升了本篇的精度与一致性，谨此致谢。

---

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）。起草与评审辅助已在致谢中说明。

正文对上游用平实话重述，句后就近括注篇次与命题号，或无编号节的节题。概念 DOI 始终指向最新版，故命题号可能随上游重编而漂移。本篇的转述与上游原句逐条比对，另记于本篇的转述审计表。

一 · 本篇正文实际调用的文献

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). 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). 论价值的动力学（SAE 价值论系列 Paper 4） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22050204

秦汉. (2026). 论价值的冲突（SAE 价值论系列 Paper 5） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22122819

秦汉. (2026). 意义是哪一种关系（SAE 意义论系列 Paper 1） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22652134

秦汉. (2026). 承载者边界（SAE 意义论系列 Paper 4） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22738709

秦汉. (2026). 余项：一个可携带的方法（SAE 方法论系列 M-00E） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22849616

秦汉. (2026). SAE 判断力与美学（SAE 基础论文） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19296710

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

秦汉. (2026). SAE 冲突论：主体性如何不被冲突本身吞噬 [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22313628

秦汉. (2026). 理性不得不：SAE对康德四定言律令的句式精细化 [预印本]. Zenodo. https://doi.org/10.5281/zenodo.19751512

二 · 背景文献

秦汉. (2026). 有限者的承接（SAE 意义论系列 Paper 2） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22713651

三 · 本系列

秦汉. (2026). 行动不是什么（SAE 行动论 前置篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22970747

秦汉. (2026). 反殖民（SAE 行动论 第二篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22970758

秦汉. (2026). 是不是我的（SAE 行动论 第三篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22970762

秦汉. (2026). 能与不能（SAE 行动论 第四篇） [预印本]. Zenodo. https://doi.org/10.5281/zenodo.22970769