Non-Substitution
不顶替
The first eight papers make conceptual distinctions, analyze mechanisms and check institutional facts, and the norms they take over are marked where used. This paper changes its ground: it sets out together the normative conditions this series adds, conditional on "a human being is an end," addressed only to the constructs that write standards and the constructs that adopt standards as gates; how particular seats and procedures are set up is written beneath each proposition, as suggestion only. The constructs that write and adopt standards ought to give a standard's normative ledger a channel through which it can be questioned, a channel that meets Institution Theory's three conditions and is open before deployment (S46). The channel must connect to a seat that meets the four conditions (S47). The writing construct and the adopting construct may not use a review's conclusion to claim more than its scope of proof (S48). The state of a normative challenge having gone unanswered ought to be visible and registered on the record; those unable to put forward a complete alternative do not thereby lose the entry to questioning (S49). The line of a seat is drawn on its function and the division of authority: it checks whether premises have been set out clearly and whether responses have reached the matter; it does not use its own ordering of oughts to supplant the trade-offs the party under review is authorized to make (S50). Keeping to the division of labor lowers the risk that a seat's evaluation becomes tied to new premises; it does not remove it (S51). The seat is itself a construct, has a remainder, and can be questioned (S52). Standards do not judge in place of those present, and neither do seats (S53). Eight propositions in all (S46 to S53), stated at the end of each section and listed together at the end.
Writing Declaration: This paper was co-drafted with Claude (Anthropic). All intellectual decisions, framework design, and final editorial judgments were made by the author.
Introduction
First an illustration, not a case: the lender, the developer of the score model and the regulator below are all illustrative; they do not refer to the example in Paper 6, nor to any particular party. A person's loan application has been refused. What refused him was a cut-off: his credit score was a few points too low. What he wants to ask is not "was my score calculated wrongly," but "why is the line here."
He asks the lender. The lender says the line was set according to the rules, and the result came out of the system automatically. Paper 7 said that the one who deploys a system bears this judgment, and cannot say "the program decided it"; but bearing it is not the same as the line having been questioned. He asks the developer of the score model. The developer says: we only produce scores; the line is set by the adopter. Paper 6 said that the writing side and the adopting side are booked separately, and each must answer for its own share; but "booked separately" says how the accounts should be kept, not where he should go to ask. He asks the regulator. The regulator says this model passed validation. Paper 3 said that if what validation reviewed was the model's predictive performance, its conclusion taken to answer "why is the line here" is misrouting.
Having asked all the way round, at every place there is someone, and every place may have completed the work entrusted to it there, yet no place takes up his question. What the first eight papers described is exactly this round: the setting position is always occupied, the questioning position can be empty (Paper 2); the conclusions of reviews are sent to questions they did not handle (Paper 3); the arrangements for questioning normative premises can be advisory only, or set up and then stopped (Paper 4); the line goes on being used, and the using side treats it as absent (Paper 5); no single withdrawer can be found (Paper 6); the line is written into a program and the same trade-off is carried out again and again (Paper 7); even if a position comes to ask, its independence must be checked again (Paper 8).
This paper asks: what conditions ought the constructs that write standards and the constructs that adopt standards as gates to meet, so that a question like this has somewhere to be taken up.
This paper changes its ground. The conceptual distinctions and descriptions of the first eight papers can be tested by readers who do not accept "a human being is an end"; the norms they take over are marked where used. What this paper sets out are the normative conditions this series itself adds, conditional on the end postulate. How Is Subjecthood Possible sets up this postulate: "A human being is an end in itself." It says: "This postulate is a normative premise this paper openly undertakes. It is not derived from the operation of negation, and it does not pose as the result of any physical test." And: "Whoever refuses Postulate One domiciles a branch of his own, and that branch may be inhabited." (F4 §2.3) This paper's conditions hold for readers who accept this postulate; readers who refuse it can set up a branch of their own, and the distinctions and descriptions of the first eight papers remain usable.
To whom this paper is addressed. The ought is addressed to constructs: the construct that writes a standard, and the construct that adopts a standard as a gate. Seats, working groups and reviewers hold accountability. No ought is written for persons.
How this paper is written. The conditions are claims, at the normative layer. What a seat looks like, and how procedures run, are matters for the rule layer, written beneath each proposition and marked "suggestion"; a reader who takes another approach does not damage the claim.
The reading follows the Prequel. This paper concerns the second domain, standards that act on persons (Prequel §5). §1 speaks of channels; §2, of seats; §3, of the scope of proof; §4, of the state of going unanswered; §5, of the line; §6, of the risk the division of labor reduces; §7 says that the seat itself can be questioned; §8 says that judgment is not cancelled; §9 compares several forerunners; finally, where this series stops.
§1 Channels
The condition. The construct that writes a standard and the construct that adopts a standard as a gate ought to give the standard's normative ledger a channel through which it can be questioned. The channel meets three conditions, following Institution Theory: "it is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation" (Institution Theory Z34). The original sentence speaks of institutions; here it is extended to the constructs that write and adopt standards (own undertaking).
Open before deployment. A channel is not opened only once something has gone wrong. Paper 7 of The Second Enlightenment says of the judgment of deployment: "Beforehand, the remit is prevention" (The Second Enlightenment, Paper 7, W44). The original speaks of the one who decides on deployment; here it is extended to constructs: before a standard is written into an arrangement intended to act on persons, its normative ledger must have somewhere to be questioned (extension, own undertaking). The Prequel said that checking begins when the relevant choices are written into an arrangement intended to act on persons, without waiting for harm to occur; this proposition is that sentence's counterpart at the normative layer.
Who is inside. "Internal subjects can actually initiate it": the interior is drawn as Institution Theory draws it: "The interior is drawn by entanglement: those in the entanglement the construct handles, together with those its acts reach who have no way to avoid it by exiting, make up the interior" (Institution Theory Z4). It is checked by the two branches.
Those directly in the entanglement the construct handles fall in the first branch, and need not further prove that they cannot exit. The person classified is in this branch: a classification standard handles precisely him; he is not excluded because he was not kept out at the door. On the other side, being affected does not by itself admit one. Reach of influence is not reach of authority. To qualify under the second branch, a person must be reached by the construct's acts and have no way to avoid them by exiting.
Political Economy speaks of a chokepoint: "Those the door reaches and who cannot get around it are, by Institution Theory's reading that draws the interior by entanglement, in this door's interior, without needing to be its members" (Political Economy, Paper 5, E31). This is one reading of the second branch, not a third route beyond the two branches; the subject of the original sentence is degrees and certifications, and extending it to standards of the kind of a cut-off score is a step of this series (own undertaking).
By the illustration in the Introduction: the person applied to the lender, he is directly in the entanglement the lender is handling, and by the first branch he is inside; he is not a member of the lender, nor need he be. This relation does not pass automatically to the developer of the score model. For the developer's construct it must be checked separately: whether he is directly in the entanglement the developer handles; or whether the developer's acts reach him and he has no way to avoid them by exiting. If either branch holds, he is admitted by that branch; it cannot be presumed from his relation to the lender alone. It must be checked separately for which construct, and which actual arrangement.
The reason. That there must be a channel does not rest on "otherwise the premise cannot be changed." Paper 2 said that premises can be rewritten without any questioning position at all (in force but not occupied). The reason is that a premise acting on persons must have somewhere addressable, bound by these three conditions, where it can be questioned. This reason rests on the end postulate: for those who are inside by the two branches above, the construct must give the line that acts on them an addressable channel of questioning.
S46 Channels. The construct that writes a standard and the construct that adopts a standard as a gate ought to give the standard's normative ledger a channel through which it can be questioned; the channel meets three conditions: it is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation. The channel is open before deployment. "Inside" is checked by two branches: being directly in the entanglement the construct handles, or being reached by its acts with no way to avoid it by exiting; those falling directly in the first branch need not further prove that they cannot exit; the person classified is not excluded because he was not kept out at the door, nor is anyone admitted merely because he is affected; those the door reaches and who cannot get around it are inside by the second branch. The reason lies in being addressable and bound by the three conditions, not in "otherwise it cannot be changed."
§2 Seats
Where does the channel lead? To a seat that questions normative premises, a seat that meets the four conditions of Paper 2: jurisdiction, right of response, right of review, independence.
The reason is this: if the channel leads only to an inbox, a question is raised, and there is no position that can take up the premise, require a response or initiate an effective review; raising and questioning just sit there. Only when the channel connects to a position meeting the four conditions does an actual connection arise between the views raised and the premise questioned.
This proposition does not derive an institutional obligation once more from the indestructibility postulate. "Constructs have remainders" says that the remainder is there; that there must be a seat rests on the condition of the previous section and the connection spoken of in this one. Institution Theory, speaking of channels, wrote it the same way: "That there must be a channel rests on the stance and the three conditions, not on the sentence 'constructs have remainders.'" (Institution Theory Z35a)
S47 Seats. The seat that questions normative premises, to which the channel leads, meets the four conditions (Paper 2, S12). Reason: the channel must connect to a position that can take up the premise, require a response and initiate an effective review; only then do the views raised and the questioning have an actual connection. No institutional obligation is derived once more from the indestructibility postulate.
Suggestion: seats can take many forms. One approach: have external members; members serve terms and rotate on schedule; set statutory deadlines for responding to normative challenges; make the records of handling public. These are designs at the rule layer; a reader who takes another approach, so long as it meets the four conditions, does not damage this proposition.
§3 No Claiming Beyond the Scope of Proof
The construct that writes a standard and the construct that adopts it may not use a review's conclusion to claim more than its scope of proof. Paper 3 said that passing a review proves the part actually handled; consequences beyond the scope of proof are transfer of use (Paper 3, S17). This proposition turns that description into a requirement: when the writing side and the adopting side say a standard "has been reviewed" or "has been validated," they must state what was reviewed and how far the conclusion reaches.
Two judgments must also be kept apart. One is whether a label is accurate: whether a standard's statement that it "has been through evidence review" is itself true. The other is whether something has been answered: whether a challenge to that line on its normative ledger has been responded to. An accurate label is not an answer; and an answer does not turn a standard that contains normative choices into pure fact. A normative choice that has received a sufficient response is still a normative choice; it has only been questioned.
S48 No claiming beyond the scope of proof. The construct that writes a standard and the construct that adopts it may not use a review's conclusion to claim more than its scope of proof (Paper 3, S17). Whether a label is accurate and whether something has been answered are two judgments: even a sufficient response to a normative challenge does not make a standard that contains normative choices into pure fact.
§4 The State of Going Unanswered
That a normative challenge has gone unanswered: this state ought to be visible, and registered on the record.
Institution Theory says that an institution bears the burden of explanation for what it has itself produced, "and must answer the very matter it is questioned about" (Institution Theory Z34); and that "When something raised goes unanswered, the original item remains" (Institution Theory Z35a). The sentence is borrowed here with its subject changed from institutions to the constructs the seat checks (extension, own undertaking): once a normative challenge has been raised and not answered, the item is still on the ledger; it must be visible, and cannot, because it has not been answered, be treated as though it had never been asked.
Those affected often cannot put forward a technically complete alternative. A person knows that the line presses on him, but does not know where it ought to be drawn, and cannot build a new model. He does not thereby lose the entry to questioning. Asking "why is it here" is not conditional on first giving "where it ought to be."
The force of this proposition must be stated precisely. Whether a label goes beyond the scope of proof is checked separately as in the previous section. Whether an adoption is lawful, or should be suspended, this proposition does not settle on "unanswered" alone or on one correction of a label; that is checked separately by the corresponding authority and conditions. Conversely, this proposition neither prohibits continued adoption nor permits it: it makes no judgment on adoption itself, and requires only that the state of going unanswered be seen.
S49 The state of going unanswered. The state of a normative challenge having gone unanswered ought to be visible and registered on the record; those affected who cannot put forward a technically complete alternative do not thereby lose the entry to questioning. Whether a label goes beyond the scope of proof is checked separately as in S48; this proposition does not settle, on going unanswered or a correction of a label alone, whether an adoption is lawful or should be suspended; adoption, suspension and their review are checked separately by the corresponding authority and conditions; this proposition neither prohibits nor permits adoption itself.
Suggestion: a "core normative challenge" can be defined by procedure: raised through the channel, naming the specific parameter and the alternative. This is one threshold for entering further procedures such as evaluation of a particular alternative, technical testing or formal initiation of a project. Those who cannot put forward a complete alternative can still raise a challenge to a specific premise; whether it is taken up, which item the answer reached, and the reasons for not starting further procedures, all remain within arrangements that can be questioned.
§5 The Line
Once a seat that questions the ought has teeth, what is the problem most likely to arise? That it takes decisions in place of the discipline under review.
So a line must be drawn. A seat can check these things: whether premises have been set out clearly; whether responses have reached the matter questioned; whether relevant alternatives have been left out; whether claims about factual grounds and consequences have sources, and whether they are consistent with public evidence review. For this last item, what the seat checks is sources and consistency; it does not itself re-review the descriptive ledger, or it would become misrouting turned inside out. A seat can require the party under review to explain why it does not take up a relevant alternative; if, having explained, the party still does not adopt it, this is not registered as unanswered. The seat's judgment of what counts as relevant, and how far a response must reach, is also subject to the questioning of §7. It does not use its own ordering of oughts to supplant the trade-offs the discipline under review, or a legitimate public decision-maker, is authorized to make. Nor does it decide for the discipline by which route the content of the ought is reached: from principle, or from experience, either will do.
This is not value neutrality. Which premises are chosen to be set out, and how a consequence is described, themselves carry values; when a seat checks whether premises have been set out clearly, its own "clearly" carries a view. This proposition does not pretend that the seat has no standpoint. It places the limit on authority: the seat has the authority to require clear setting-out and answers that reach the matter; it has no authority to make trade-offs in place of the authorized party.
Several writers' accounts are close to this section. Alexandrova, discussing "mixed claims" in the science of well-being, that is, claims containing both empirical and evaluative components, proposed several procedures: dig out the value presuppositions in methods and measures; check whether measures and conclusions hold up under several competing value settings; where the choice of measure falls between conflicting values, have the relevant parties deliberate together ("mixed claims" in Alexandrova 2017; the three procedures in Alexandrova 2018, §8). This is very close to "whether premises have been set out clearly." This series goes on to check institutional arrangements item by item: whether they are standing, whether they have the right of response and the right of review; these conditions cannot be directly replaced by listing procedural principles. Pamuk discussed designs of the "science court" kind: initiated by citizens, with opposing experts examining each other, and a citizen jury forming views on scientific questions and the value judgments bound up with them; its decisions serve as reference for the legislature, the president or administrative agencies, and are not binding (Pamuk 2021, p.24, p.123); this is close to the line of this section: give advice, do not decide for the decision-maker. Pielke's "honest broker" expands, rather than narrows, the options available (Pielke 2007); this points the same way as "whether relevant alternatives have been left out."
The title "Non-Substitution" points both ways. One way: a standard ought not quietly stand in for the normative layer. The Prequel to The Second Enlightenment says: "The norm gives the two layers below their direction, and is also used to test them; the layers below cannot turn round and supplant the norm" (The Second Enlightenment, Prequel §6). A standard is a rule; if the lines on its normative ledger have no account at the normative layer, and no position requires one of them, the rule has come to stand in the place of the norm. The other way: a seat that questions the ought ought not stand in for the discipline under review either. It is there to ask, not to decide.
S50 The line is drawn on function and the division of authority. A seat can check: whether premises have been set out clearly, whether responses have reached the question, whether relevant alternatives have been left out, whether claims about factual grounds and consequences have sources and are consistent with public evidence review. If the party under review has explained why it does not take up a relevant alternative and still does not adopt it, this is not registered as unanswered. The seat does not use its own ordering of oughts to supplant the trade-offs the discipline under review or a legitimate public decision-maker is authorized to make; nor does it decide for the discipline by which route the content of the ought is reached, from principle or from experience. This is not value neutrality: which premises are chosen to be set out, and how consequences are described, themselves carry values; the limit is placed on authority.
§6 The Risk the Division of Labor Reduces
The division of labor in the previous section also reduces a risk whose source Paper 8 has already described.
If a seat takes part in setting premises and brings about a rewriting, its evaluation may become tied to the new premises: the "evaluation" spoken of in Paper 2 includes reputation, and a seat that made its name through a rewriting, when it questions that rewriting, is questioning its own reputation. Keeping to the division of labor of "checking, not deciding" lowers this risk: a seat that does not set premises is less likely to make its name by setting them. It does not remove the risk: the seat requires clear setting-out and answers that reach the matter, and premises rewritten after its questioning can still become tied to its reputation.
The relation of reputation must be graded by strength. Having endorsed a revision; having made one's name through the arguments about that revision; having one's credentials conditional on that revision not being denied: these three differ. The first two are signals to watch; for the third, by the conditions of dependence in Paper 8, §3, the corresponding condition of independence does not hold.
If a seat has taken part in setting premises, or its duties have changed, its independence is checked again with respect to the newly added role (Paper 8, S41). Whether it went beyond its authority is checked separately.
US science policy research has the notion of the "boundary organization": an organization answerable to both the side of science and the side of policy, which by this dual answerability stabilizes the work between the two (Guston 2001). It is a close forerunner of this paper's seat. This section says one step more: an organization answerable to both sides, once it has made its name on one side, must have its independence checked again with respect to that side.
S51 The risk the division of labor reduces. Through jointly setting premises or bringing about a rewriting, a seat's evaluation may become tied to new premises ("evaluation" among the four conditions includes reputation); keeping to the division of labor lowers this risk, and does not remove it. The relation of reputation is graded by strength: having endorsed a revision, having made one's name through the arguments, and having one's credentials conditional on the revision not being denied are three different things. When a seat takes part in setting premises or its duties change, its independence is checked again with respect to the newly added role; whether it went beyond its authority is checked separately.
§7 The Seat Itself Can Be Questioned
The seat that questions the ought is itself a construct, and it too has a remainder. A seat is an institutional construct; Institution Theory says that "so long as the institution is still dividing, it is still leaving unaccommodated residue" (Institution Theory Z35a). Paper 1, §9 already anticipated this (Paper 1, S8). It cannot treat its own boundary as an end: the seat's boundary too is a division, and to keep to it without questioning it any further is the parochialism of Paper 1, fallen on the seat itself.
So there is no single, ultimate, remainderless questioning of everything. Once a seat has questioned, its conclusion too is a division, and leaves its own unaccommodated residue. Theory of Government says: "Decisions can be final: a particular matter can be answered and closed. Positions cannot be final: no position may monopolize the power to shut the door"; and says that questioning needs no unquestionable end point: "A construct's configuration ought to place every position under some channel it cannot itself close." (Theory of Government G36) Institution Theory says that an institution "cannot acquire the standing of permanent exemption from questioning" (Institution Theory Z37).
These two, applied to the seat: matters the seat has checked can be closed; the seat's own position cannot be final. It too sits under a channel it cannot itself close. This proposition takes over the two upstream sources directly, with no further derivation.
Here the first and last papers form a pair. Paper 1 said that a discipline has a construct but does not acknowledge its remainder; this paper says that the seat that questions the ought is itself a construct, with a remainder of its own, which it must acknowledge.
S52 The seat itself can be questioned. The seat that questions the ought is also a construct, and also has a remainder (Institution Theory Z35a; anticipated earlier in Paper 1, S8); it cannot treat its own boundary as an end. There is no single, ultimate, remainderless questioning of everything; questioning needs no unquestionable end point.
§8 Judgment Is Not Cancelled
A last proposition, about those present.
Standards do not judge in place of those present. Nor do seats. Theory of Government says: "Rules cannot do judgment in anyone's place" (Theory of Government G3); Action Theory says: "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." (Action Theory, Paper 1, D6) Paper 7 of The Second Enlightenment says of the one who decides on deployment that he "cannot say 'the program decided it' or 'the standard told us to write it'" (W44).
This paper extends the same point to seats, and places it on constructs (own undertaking): constructs that establish or adopt standards and their review arrangements may not register "cleared by the seat" as meaning that the judgment and bearing on this occasion have been taken over. The seat's questioning lets the standard's normative ledger be questioned; it does not make, in place of the one applying this standard this time, this time's trade-off. The sentence in W44 is addressed to persons, and serves here only as an upstream interface; this paper sets no further requirement on persons.
So the normative conditions of this series are placed on constructs, not on persons. Channels, seats, the scope of proof, the state of going unanswered, are all the business of constructs; the judgment of the one present, at that moment, is still his.
S53 Judgment is not cancelled. Standards do not judge in place of those present, and neither do seats. Constructs that establish or adopt standards and their review arrangements may not register "cleared by the seat" as meaning that the judgment and bearing on this occasion have been taken over.
§9 With Several Forerunners
This paper's conditions have many near neighbors. Three are listed here as in-text contrasts; several others appear at the relevant points in §5. This paper does not claim that no one has made its synthesis before.
Guston. Boundary organizations are answerable to both science and policy (Guston 2001); see §6.
Kitcher. In Chapter 10 of Science, Truth, and Democracy he discusses "well-ordered science": how research agendas should be set, taking an ideal deliberation as the standard (Kitcher 2001, Chapter 10). What he asks is by whom, and by what, agendas are set; what this paper asks is whether the premises within the agenda have anywhere to be questioned.
Ulrich and Reynolds. Critical systems heuristics lists "those affected but not involved" as a party to be asked about, and asks where a system's motivation, power, knowledge and legitimacy come from (Ulrich & Reynolds 2010). "Who is inside" in §1 is closest to it.
Where This Series Stops
The forms of seats are suggestion only. External members, terms and rotation, statutory deadlines for response, public records: these are designs at the rule layer, which this series writes as suggestion and does not set up as propositions.
It sets no discipline's ought. This series does not say what the statistical threshold should be, how diagnoses should be divided, or where the cut-off score should be drawn. What it says is: the oughts these carry must have somewhere to be questioned.
It does not answer who sits there. A seat need only meet the four conditions; which discipline, which kind of people, which institution occupies it, this series does not answer. Paper 1, §9 has already said: the position that questions the ought is not assigned to philosophy.
Handed downstream.
- Theory of the University. Education and research as such: the standards of research evaluation, including the credit standard of mathematics, and authorship when AI takes part in proofs. Several major publishers and preprint platforms provide that AI may not be listed as an author, on the grounds that it cannot take responsibility for the work (Springer Nature; Elsevier; arXiv): this provision is one place where §8's "judgment is not cancelled" lands, in authorship. The gates of resources; degrees, certifications and admissions: when a degree becomes a door others cannot get around, it is what Political Economy calls a chokepoint (E31); the channel of §1 of this paper lands there on the university as one kind of construct.
- Specific theories of institutions. The forms of seats.
- Action Theory. The judgment of those present belongs to Action Theory (Prequel §10 lists it as upstream).
The remainder it leaves. The items left by the drawing of the two domains, animals, ecosystems, people not yet born, have for now no place in this series (Prequel §5); of the remainder of the seat that questions the ought, §7 says that it is there, not where it is; a complete four-step instance of parochialism, an individual case of pushing back, and the cell "occupied and retained" were none of them found in the public records examined.
The Eight Propositions of This Paper
- S46 Channels. The constructs that write and adopt standards ought to give the normative ledger a channel through which it can be questioned; three conditions; open before deployment; "inside" checked by two branches; those the door reaches and who cannot get around it are inside; the reason lies in being addressable, not in "otherwise it cannot be changed." (Normative; the end postulate; takes over Institution Theory Z34, Z35a, extended to the constructs of standards and of adopting standards, own undertaking; "beforehand" takes over The Second Enlightenment, Paper 7, W44, extended to constructs, own undertaking; the interior takes over Institution Theory Z4; E31 is a reading of the second branch, extended to standards of the kind of a cut-off score, own undertaking)
- S47 Seats. The channel connects to a seat meeting the four conditions; the reason lies in an actual connection; no institutional obligation is derived once more from the indestructibility postulate. Suggestion: external members, terms and rotation, statutory deadlines for response, public records. (Normative; the end postulate; takes over S46; the four conditions supplied by Paper 2, S12, this series' own undertaking; "not on 'constructs have remainders'" as in Institution Theory Z35a)
- S48 No claiming beyond the scope of proof. A review's conclusion may not be used to claim more than its scope of proof; whether a label is accurate and whether something has been answered are two judgments. (Normative; the end postulate)
- S49 The state of going unanswered. It ought to be visible and registered on the record; inability to put forward a complete alternative does not lose the entry; it alone does not settle the lawfulness or suspension of adoption; it neither prohibits nor permits adoption. Suggestion: a "core normative challenge" defined by procedure, serving only as a threshold for further procedures such as evaluation of a particular alternative, while the taking up of and response to general challenges remain open to questioning. (Normative; the end postulate; "must answer the very matter it is questioned about" borrowed from Institution Theory Z34, "the original item remains" from Z35a, with the subject changed from institutions to the constructs the seat checks, extension, own undertaking)
- S50 The line is drawn on function and the division of authority. The five things it can check (for facts and consequences, only sources and consistency); an explanation of why a relevant alternative is not taken up is not registered as unanswered; it does not supplant authorized trade-offs; it does not decide for the discipline by which route the ought is reached; not value neutrality, the limit placed on authority. (Normative; the end postulate)
- S51 The risk the division of labor reduces. Lowered, not removed; reputation graded by strength; checked again when roles change. (Descriptive; takes over Paper 2, S12; checking again joins Paper 8, S41, with dependence checked as in S42)
- S52 The seat itself can be questioned. It too is a construct, with a remainder; it cannot treat its own boundary as an end; there is no ultimate, remainderless questioning of everything. (Normative; takes over Theory of Government G36 and Institution Theory Z37 directly; the seat's remainder takes over Institution Theory Z35a)
- S53 Judgment is not cancelled. Standards do not, and neither do seats; a construct may not register "cleared by the seat" as meaning that the judgment and bearing on this occasion have been taken over. (The conceptual statement takes over The Second Enlightenment, Paper 7, W44, Theory of Government G3, Action Theory D6; the sentence on constructs: normative, the end postulate, extension, own undertaking)
References
All SAE works below are cited by concept DOI, which always resolves to the latest version.
I · This system
- Qin, H. (2026). How Is Subjecthood Possible: Symmetry, Negativity, and Subjecthood (SAE Foundational Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.18777363
- Qin, H. (2026). To Whom the Ought Is Addressed (SAE Institution Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23269635
- Qin, H. (2026). Open to Questioning (SAE Institution Theory, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23271324
- Qin, H. (2026). Judgment (SAE Theory of Government, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23412130
- Qin, H. (2026). Questioning and Bearing (SAE Theory of Government, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413511
- Qin, H. (2026). Growth (SAE Political Economy, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23874150
- Qin, H. (2026). At the Door (SAE The Second Enlightenment, Paper 7) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24030415
- Qin, H. (2026). What the Second Enlightenment Is Not (SAE The Second Enlightenment, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24030393
- Qin, H. (2026). Not Doing and Cannot-Not-Doing (SAE Action Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.22970752
- Qin, H. (2026). What a Discipline Is Not (SAE Theory of Disciplines, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177508
- Qin, H. (2026). Parochialism (SAE Theory of Disciplines, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177748
- Qin, H. (2026). Standard-Setting (SAE Theory of Disciplines, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177754
- Qin, H. (2026). Misrouting (SAE Theory of Disciplines, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177759
- Qin, H. (2026). Seats (SAE Theory of Disciplines, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177765
- Qin, H. (2026). Shelving: Treating Premises as Absent (SAE Theory of Disciplines, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177775
- Qin, H. (2026). Withdrawal (SAE Theory of Disciplines, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177783
- Qin, H. (2026). Writing Standards into Programs (SAE Theory of Disciplines, Paper 7) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177788
- Qin, H. (2026). Independence (SAE Theory of Disciplines, Paper 8) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.24177795
II · Institutional documents
- arXiv. Policy on AI tools and authorship (current version).
- Elsevier. Policy on generative AI and authorship (current version).
- Springer Nature. Editorial policy on large language models and authorship (current version).
III · External literature
- Alexandrova, A. (2017). A Philosophy for the Science of Well-Being. Oxford University Press.
- Alexandrova, A. (2018). Can the science of well-being be objective? The British Journal for the Philosophy of Science, 69(2), 421–445.
- Guston, D. H. (2001). Boundary organizations in environmental policy and science: An introduction. Science, Technology, & Human Values, 26(4), 399–408.
- Kitcher, P. (2001). Science, Truth, and Democracy. Oxford University Press. Chapter 10.
- Pamuk, Z. (2021). Politics and Expertise: How to Use Science in a Democratic Society. Princeton University Press.
- Pielke, R. A., Jr. (2007). The Honest Broker: Making Sense of Science in Policy and Politics. Cambridge University Press.
- Ulrich, W., & Reynolds, M. (2010). Critical systems heuristics. In M. Reynolds & S. Holwell (Eds.), Systems Approaches to Managing Change: A Practical Guide (pp. 243–292). Springer.
Interfaces
Main upstream: F4 §2.3 (the end postulate); Institution Theory Z4, Z34, Z35a, Z37; The Second Enlightenment, Prequel §6, Paper 7, W44; Theory of Government G3, G36; Action Theory D6; Political Economy, Paper 5, E31; this series S8, S12, S17, S41, S42.
Also used: Alexandrova 2017, 2018; Pamuk 2021; Guston 2001; Kitcher 2001; Ulrich & Reynolds 2010; Pielke 2007; the AI authorship policies of publishers and preprint platforms.
This series: S46 to S53; the whole series S1 to S53.
Handed downstream: Theory of the University; specific theories of institutions. The judgment of those present belongs to Action Theory.