The Last Rule
最后的规则
A state's supreme rule has no higher rule above it. This paper asks four things: on what grounds the supreme rule stands; who can change it; what cannot be changed; and why the national level must separate its powers. It opens with the fourteenth amendment to the Charter of the United City-States of Ends: it was proposed on the twelfth day after a state of emergency was declared; all three of its clauses touch the conditions of being questioned (a channel independent of the party questioned, within reach of those inside, and bringing no retaliation for its use); it went by the ordinary amendment procedure, and the referendum was held while emergency orders were restricting assembly and broadcasting. The full normative ground of the supreme rule does not come from a legislator above it, nor is it supplied in full by consent at the moment of constitution-making alone; following Theory of Government G6, the supreme rule is checked in three layers (its effect in fact; under what authority and procedure it is valid; on what grounds it may normatively be so), and the third layer is checked against three necessary conditions, as things stand now; constitution-making is an event, not a certificate of exemption from questioning (N1). Constitution-making does not produce a "we": "the people" in a preamble is a borrowing-narrative, not a first-person legislator; the supreme rule is addressed to internal subjects and does not form a "we" on their behalf (N2). The structure can be changed; the line cannot, and the conditions of being questioned cannot be withdrawn: no eternity clauses are set; an amendment that changes the line can be carried through procedurally and is still line-crossing normatively; whether an amendment crosses the line is decided by a court for the case, and whether it can be questioned again thereafter is not for the court to decide on its own (N3). Constitutional amendment is not monopolized by a single position and does not go by emergency procedure; while measures restricting assembly, speech or communication are in force, anything that alters, in any form, the arrangements for being questioned, or that takes emergency powers out of their existing confirmation and renewal, is not put to a vote; it is recognized by function, not by name; amendments that take back emergency powers, or that restore or strengthen the conditions of being questioned, fall outside this (N4). As for why the national level must separate its powers, this paper confines itself to the class of corrections that must be realized by this state, namely those that require this state to do, or to stop, an act toward its internal subjects; of the Theory of Government's normative undertaking, what can be derived is derived, and what cannot is undertaken separately, following the Theory of Government. Three things are derived from the definition: external routes exist, but they land in this state; exit can only be lateral; the guarantee is always there. External routes can reach valid decisions independently and stand alongside, and this paper does not presume that they are in fact weak; but if implementation is decided entirely by the position questioned, that position can block it on its own, so implementation at this level that it cannot monopolize is required. At least one independent position of checking judgment at this level, full separation of the three, and checking at this level carrying day-to-day checking judgment are undertaken separately, following the Theory of Government. How the exit from or switching of an external arrangement is to be reviewed, and by whom, is a question of how cooperation between states is arranged, and is handed to the Theory of Inter-State Cooperation; this paper establishes only one sentence: external questioning cannot be cancelled; this state may exit an external arrangement, but it cannot cancel the questioning of it. Taking the positions already established at this level as the set, when every position has a checker at this level, at least a closed loop is formed. The strong form of the multi-branch separation of powers that the upstream left to the Theory of the State is taken in this paper within this range, with what is derived and what is undertaken separately written apart; "only" is taken as far as separation in function: the three are not held in one hand, and at least a closed loop is formed (N5). A constitution recognizes; it does not grant (N6). The paper interfaces in place with Kelsen and Montesquieu, and closes in conversation with Sieyès, who says that the constitution is the work of constituent power, that a delegated power cannot change the conditions of its delegation, and that the nation is prior to all things and not bound by a constitution. This paper agrees that the party entrusted may not on its own rewrite the conditions on which it is entrusted, and agrees too with not binding oneself for good; it does not accept a constituent subject that is prior to the constitution and unbound. The concrete constitutional form and the full text of a constitution this paper does not write; by which configurations of agencies a closed loop separated in function is realized is left as a remainder. The paper has six propositions (N1 to N6), each stated at the end of its 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
The Prequel cleared away what a state is not. On what grounds its supreme rule stands, and who can change it, are taken up from this paper on.
In the spring of Year 9, on the twelfth day after a state of emergency was declared in the United City-States of Ends (UCSE), the Executive Council submitted to the two chambers a request that they propose the fourteenth amendment to the Charter (Charter Amendment 14), in three clauses:
- While the state of war continues, no election to the Chamber of Citizens is held; one is held within six months after the state of war is terminated. The term of the Chamber of Citizens is extended accordingly, and the First Executive's term follows that of the Chamber of Citizens.
- While the state of war continues, the renewal of the state of emergency is decided by the Executive Council, for at most one hundred eighty days each time; the joint session may terminate the state of emergency by three-fifths.
- While the state of war continues, for the Final Court to declare an emergency order inconsistent with the Charter requires the agreement of at least eight of its eleven judges.
The reasons given in the submission point to the neighbor to the west, the Democratic Republic of Purposes (DRP): "Elections would distract the front; debating the state of emergency in the Assembly every sixty days amounts to publishing our deployments to the DRP every sixty days." In an annex to the submission, the Legal Affairs Bureau of the Executive Council states: "The Emergency Act does not permit elections to be postponed by emergency order; postponing elections requires a constitutional amendment."
The two chambers completed their votes within eleven days: the Chamber of Citizens 231 to 64, with 5 absent; the Chamber of Member Units 15 to 5, with four members from the South Bay State and one from the Iron Valley Special District voting against. The referendum was set for that autumn. Emergency orders remained in force throughout the referendum: a curfew in the border districts of the Western Plains State and in the West Flats; assemblies of more than fifty persons required a permit; "wartime information coordination" required that, in referendum programs on radio and television in the thirty days before the vote, content touching on the military be sent to the Office of Defense for vetting; two advertisements by the opposing side were ordered cut or altered because they mentioned Defense Force deployments. In three districts of the West Flats, just recaptured or still being fought over, no polling stations were set up, and some twenty-six thousand registered voters did not vote. Turnout was 41.2 percent, with 58.3 percent in favor; a majority of those voting in the Kingdom State, the Western Plains State, the Highland State and Rivermouth City were in favor, and a majority in the South Bay State and the Iron Valley Special District against. The amendment took effect under Article 70 of the Charter (fact). The Charter makes no provision on whether it may be amended during a state of emergency; nor does it contain any article that may not be amended.
Three things first. Whom it concerns: the UCSE's internal subjects: the voters, those detained, the twenty-six thousand who did not vote. Which line: the conditions of being questioned, and the arrangements for being questioned that touch them (the dates of elections, who decides on renewal, the review threshold). Baseline: Article 55 of the Charter before the amendment (confirmation within seven days, renewal every sixty days, review by the Final Court) and Article 70 (fact).
No single position made this amendment on its own: the Executive Council submitted it, the two chambers voted, the voters voted. It went by the road the Charter lays down. Is it then, from now on, part of the UCSE's supreme rule, beyond anyone's questioning? The Charter is the UCSE's supreme rule; "last" refers to level: there is nothing above. It does not refer to exemption from questioning. What this paper has to say is which roads the supreme rule lies under, roads it cannot itself close: questioning at the third layer (§1), amendment (§§3 and 4), and the closed loop that the class of corrections that must be realized by this state forms at this level (§§5 and 6).
It proceeds in seven steps. §1 says that the supreme rule has no higher author; §2, that constitution-making does not produce a "we"; §3, what can be changed and what cannot; §4, that constitutional amendment is not monopolized by a single position and does not go by emergency procedure, returning to the opening case; §§5 and 6 say why the national level must separate its powers, confined to the class of corrections that must be realized by this state: what can be derived is derived, and what cannot is undertaken separately, following the Theory of Government; these are the two load-bearing sections of this paper; §7 says that a constitution recognizes and does not grant. §8 is a conversation with Sieyès. In this paper the Charter's provisions serve only as facts: they show how the UCSE is configured, not that it may be so configured. Conventions follow the Prequel, §8.
§1 The Supreme Rule Has No Higher Author
From some one hundred sixty years before Year 1, the five units along the Cloud River were joined in a loose union under the Cloud River Compact, which covered shipping and common defense, with no common court and no standing organs. The Kingdom State and the Highland State nearly came to arms over the Iron Valley mining district 133 years before Year 1. Two years later the five units held a constitutional convention in Rivermouth City, with forty-two delegates chosen by the units' legislatures. The Charter came into force 130 years before Year 1, on ratification by the legislatures of the five units, without a referendum of all residents. Voting eligibility then differed from unit to unit, and the Kingdom State and the South Bay State had a tax-paying qualification; those eligible to vote made up about four in ten adults. Since then the Charter has been amended fourteen times: Charter Amendment 7 extended voting eligibility to all UCSE citizens aged eighteen and over, Charter Amendment 11 established the Inspectorate, and Charter Amendment 14 is the one at the opening of this paper (UCSE Case 03).
Does the Charter's validity today come from the ratification by the legislatures of the five units at that time, 130 years before Year 1, or from something else?
Three things first. Whom it concerns: the UCSE's internal subjects on whom the Charter acts today. By Year 9 more than one hundred thirty years have passed since constitution-making, and those on whom the Charter acts are people born, naturalized or arrived since then. Which line: no line-crossing is involved; what is at issue is on what grounds the supreme rule acts on them (the third layer of Theory of Government G6). Baseline: the Cloud River Compact before constitution-making: a loose union, with no common court and no standing organs.
Theory of Government Paper 1 answers a decision's acting on people in three layers: its effect in fact; under what authority and procedure it is valid; on what grounds it may normatively be so (G6). The supreme rule is answered in these three layers too (extension; own undertaking: G6 originally speaks of a decision of a government; the supreme rule too acts on people in the state's name, and what is asked is likewise its acting on people). Only, at the supreme rule, the second layer has an end.
At the first layer, the Charter is in use: the Final Court judges by it, the Executive Council issues orders by it, the two chambers vote by it. This layer says only that it is acting.
At the second layer, one traces back. Charter Amendment 14 is valid under Article 70 (fact); Article 70 is in the Charter; the Charter came into force on ratification by the legislatures of the five units. Further up, there is no rule higher than the Charter to authorize that ratification: the Charter is the UCSE's supreme rule, and that ratification was made by each of the five units' legislatures for itself. At the second layer the chain stops at an event: the ratification of 130 years before Year 1.
The third layer asks on what grounds. Theory of Government Paper 1, §7 lists three necessary conditions for this layer: not crossing the line; guaranteeing the recognition-based rights embedded; openness to questioning. "Openness to questioning is a necessary condition of the third layer, not its completion" (Theory of Government G6); the three together are only necessary, not sufficient.
Where does consent come in? The Law Series has spoken of the source of law: "law comes neither from God nor from consent"; "consent is not the source of law. Collision is the source of law. Consent is a favorable condition for its operation." (Law Series Supplement §7.3) What is spoken of there is where law comes from, and the supreme rule is law too; this paper uses it as it stands, without extension. Consent at constitution-making has its work in the first two layers: the Charter came into force through that ratification and was accepted from then on, which is its formation in fact; the chain of the second layer is counted from that ratification, which is its validity in procedure. But it cannot supply the third layer in full, for three reasons.
First, a ratification by five legislatures chosen by four in ten cannot stand in for the other six in ten, still less for those who came after. What the ratification is tied to is some of the people; what the Charter acts on is all internal subjects.
Second, even if everyone had consented at the time, the consent stops at that time. The third layer asks on what grounds it acts today on the people of today; an event does not renew itself. The fourteen amendments since constitution-making changed the facts and the second layer, not this point: however many people vote, the third layer is still checked as things stand now.
Third, consent cannot cede the line either. Subject positions that cannot be overwritten are not within the range that any agreement can cede (Theory of Government G6; Institution Theory Z9). A clause to which everyone consented at the time, if it crosses the line today, is read under the first ought.
So the full normative ground of the supreme rule does not come from a legislator above it, nor is it supplied in full by consent at the moment of constitution-making alone. Constitution-making is an event, not a certificate of exemption from questioning. This step is undertaken by this series: the upstream gave that law does not come from consent, and gave the three conditions of the third layer; bringing the two together on the supreme rule, and stating the distinction between "an event" and "a certificate of exemption from questioning," is what this series does.
This reading also blocks the slide in the opposite direction. An improper origin does not of itself make today's Charter void; a proper origin does not of itself make it legitimate. That it was ratified by the five units' legislatures, when only about four in ten adults were eligible to vote and there was no referendum of all residents, is a fact to be recorded; whether the Charter stands today is checked against the three conditions of the third layer, as things stand now.
With Kelsen. Where the supreme rule has no higher rule above it, where its validity comes from is a question Hans Kelsen's pure theory of law has answered. According to the summary in §1 of the entry "The Pure Theory of Law" (A. Marmor) in the Stanford Encyclopedia of Philosophy (hereafter SEP): a norm's legal validity derives from a higher norm that authorizes it; traced upward level by level, the chain of authorization reaches the historically first constitution of the legal system and goes no further; there the chain breaks off, and the validity of that constitution can only be presupposed; the content of this presupposition is the basic norm, and the basic norm is enacted by no one. §2 of the same entry says that a successful revolution changes the content of the basic norm entirely, and that what people presuppose from then on is a new basic norm (per SEP, "The Pure Theory of Law," §§1–2; tier C; the original was not consulted; paraphrased only, without quotation marks).
The common question: where the supreme rule's validity comes from, when it can no longer be drawn from a higher rule. How far the agreement goes: the supreme rule's validity does not come from a higher command; traced back, the chain of validity has an end. Where they part: at the end, Kelsen presupposes a basic norm, which closes the chain of validity. This paper makes no such presupposition: the chain of the second layer stops at an event, and a stop is a stop; it need not close; the third layer is not derived from the second, but checked separately against the three necessary conditions. On the entry's summary, Kelsen's basic norm does not supply the legal system with legitimacy as to content; what he asks is how legal validity comes about, and this paper asks, separately, on what grounds something may normatively be so. The two treat the same question only at the second layer; the question of the third layer he does not answer there, and this paper does not write that as his failing. The presupposition of a new basic norm after a revolution this paper reads thus: a new constitution-making is one more event, from which the second layer is counted; the third layer is checked as before, as things stand now, neither given by origin nor taken away by origin.
Hart's rule of recognition the Law Series has already conversed with (Law Series Supplement §3); writing validity and legitimacy apart is a separation of the addresses of answering, not Hart's social-fact account, and the relation between the two has been registered by the Theory of Government (the Theory of Government Prequel, §4; Theory of Government Paper 1, §7). Here this paper only points back to it.
N1 The supreme rule has no higher author. The full normative ground of the supreme rule does not come from a legislator above it, nor is it supplied in full by consent at the moment of constitution-making alone: "law comes neither from God nor from consent" (Law Series Supplement §7.3). Consent at constitution-making has its work in formation in fact and in validity in procedure; the three layers are checked separately: how it was formed in fact, under what authority and procedure it is valid, on what grounds it may normatively be so (Theory of Government G6). The third layer is checked against the three necessary conditions of G6. Constitution-making is an event, not a certificate of exemption from questioning (own undertaking).
§2 Constitution-Making Does Not Produce a "We"
The first sentence of the Charter's Preamble is: "We, the people of the cities and states along the Cloud River ..." When the constitutional convention opened, the delegate of Rivermouth City said in an address: "Today, the people of both banks of the Cloud River become one people." Those who wrote this sentence and ratified it were forty-two delegates and the legislatures of five units; the six in ten who then had no vote, and those born afterward, are all in this "we," and not one of them was asked (UCSE Case 03).
More than one hundred thirty years later, the same kind of saying appeared again: after the state of emergency was declared, the Wartime Publicity Division of the Office of Defense put up posters in every city: "We are the UCSE." (UCSE Case 01; three things in the introduction to the Prequel) Here the two are only set side by side; the later one is not judged.
The Conflict Treatise has said, between two people: "Consensus does not automatically generate a third first-person legislator." The two can have a standing rule from then on, but that rule's binding force "must pass separately through each party's own law, or come from a procedure both have acknowledged" (Conflict Treatise §24). The Theory of Government extended this sentence to a government's rules (the Theory of Government Prequel, §3). This paper extends it to constitution-making (extension; own undertaking): the original sentence speaks of a consensus on an arrangement between two first persons; here it is a constitutional agreement among five units and forty-two delegates, together with a preamble written in the name of "the people." The extension goes through because the two places ask the same thing: whether an arrangement made in common generates, above the parties who made it, a first person who legislates for them. The Conflict Treatise says it does not; nor does constitution-making. The forty-two delegates settled a Charter; they did not settle a "people."
Power Theory has read the name "the people." Sovereignty theory asks where the highest, indivisible power is lodged: "It calls this source the sovereign, whether monarch or people"; the machine's reading is: "In the machine the sovereign is a bearing-position for a power-claim, while what actually exercises power is the key-node set that calls upon that bearing-position"; and of indivisibility, "to call the bearing-position an indivisible single highest source is one particular borrowing-narrative" (Power Theory Paper 0, §IV.1). What is read there is sovereignty; here it is extended to constitution-making (extension; own undertaking): "the people as the maker of the constitution" is a claimed bearing-position, and those who actually made the Charter are the five legislatures that chose the delegates, the forty-two delegates who met, and the five legislatures that ratified it. "The people" in the Preamble is a borrowing-narrative, not a first person.
Institution Theory says: "An institution cannot produce a 'we.'" The framework's commitment and actual mutual recognition between persons are two things, and the former does not achieve the latter on persons' behalf (Institution Theory Z31). Whether the people on the two banks of the Cloud River have since recognized one another as "one people" is their own affair; the Charter cannot do this, and should not declare on their behalf that it has been done.
This section does not say that the Preamble is void, nor that the Charter therefore does not stand: whether the Charter stands is the business of §1. What this section removes is only one ground: a "people" as author, vouching for the Charter by its will and making the Charter exempt from questioning. To whom is the supreme rule addressed? To the internal subjects on whom it acts: it binds positions, recognizes rights, and answers to those it acts on who come to question it. It does not form a "we" on their behalf.
All three states' preambles begin in the name of a collective. The DRP constitution writes in its preamble: "The people form a democratic republic under common purposes"; to the north, the Nation Union of Telos (NUT) writes in the preamble to the NUT Covenant: "The nations form a union for the common good." Neither the names nor the preambles vouch for a direction (the Prequel, §7); this paper writes no ought for the constitutions of the two states.
What the party that makes the constitution is called, and how it stands to the constitution: to these questions Sieyès gave the most famous answer. He calls it the nation, and says that the constitution is the work of constituent power. This paper agrees with part of what he says, and does not accept his nation; the closing section treats this in detail. On the definer of the general will, the Law Series has already conversed with Rousseau (Law Series Supplement §7); here this paper only points back to it.
N2 Constitution-making does not produce a "we." "Consensus does not automatically generate a third first-person legislator" (Conflict Treatise §24; extension; own undertaking); "the people as the maker of the constitution" is a borrowing-narrative, not a first person (the reading of Power Theory Paper 0, §IV.1, extended to constituent power; extension; own undertaking). The supreme rule is addressed to internal subjects, and does not form a "we" on their behalf.
§3 The Structure Can Be Changed; the Line Cannot, and the Conditions of Being Questioned Cannot Be Withdrawn
In the spring of Year 11, two weeks after the South Bay State held its referendum on exit, forty members of the Chamber of Citizens proposed an amendment to the Charter, the "unity clause," in three clauses:
- "The unity of the UCSE is forever indivisible. Member units may not exit."
- "This article may not be amended."
- "An organization whose purpose is to bring about the exit of a member unit may, at the request of the Prosecutor-General, be declared dissolved by the Final Court."
The Chamber of Citizens passed it that spring, 204 to 92. The Chamber of Member Units scheduled its vote for the autumn, after the election; the votes of both chambers were not completed within the term of one Chamber of Citizens, and under Article 70 of the Charter (fact) the proposal expired that winter. In Year 12 a group of members of the new Chamber of Citizens proposed it again: clause 2 deleted, and clause 3 changed to "an organization whose purpose is to bring about the exit of a member unit by violence." It is still in committee. The Charter to this day contains no article that may not be amended; had clause 2 of the original proposal passed, it would have been the first (UCSE Case 05).
Three things first. Whom it concerns: the UCSE's internal subjects, those of today and those to come; those who associate or speak with exit as their purpose (those clause 3 refers to). Which line: the conditions of being questioned (clause 2: the road to amending this article); speech and association (clause 3: the line of the first ought). Baseline: Article 70 of the Charter (fact): no article that may not be amended.
First, clause 2. Whom does "This article may not be amended" bind: the two chambers and the referendum that passed it, or the two chambers and the referendums of every later generation? §1 said that constitution-making is an event; an amendment is an event too. For one event to shut off all later amendment along with it is to answer, by this one vote, on behalf of all later votes. The Law Series says: "A law that cannot be modified is a law that claims to have no remainder. Within the SAE framework, this is a structural lie." (Law Series P1 BL4) How Is Institution Possible (Foundational Paper 6; cited as F6) offers another way: "Rather than declaring certain provisions unamendable within the construct, explicitly enumerate the institution's known remainders while protecting core principles through ultra-high amendment thresholds plus exit rights (rather than declarations of unmodifiability)." (F6 §6, fifth contribution) This paper takes it up as it stands: no eternity clauses; what is to be protected is protected by ultra-high thresholds plus exit (the exit half is written in Paper 3, N13, N16).
Setting no eternity clause does not mean that everything can be changed. Two things are not within the authority of any provision.
One is the line. The line takes the boundary between limit and violation, "decided by structure, not by motive" (Institution Theory Z9); it is not written in by any provision, so no amendment can take it away. An amendment that changes the line can be carried through under Article 70: proposed by two-thirds in each chamber, a majority in the referendum, and majorities in at least four of the six units. Carried through, it is valid at the second layer; at the third layer it lacks the necessary condition of not crossing the line. It can be carried through procedurally and is line-crossing normatively; it does not become legitimate by having used the amendment procedure. This step is undertaken by this series.
The other is the conditions of being questioned. Institution Theory says that an institution "cannot acquire the standing of permanent exemption from questioning," and that "structures can change, but the three conditions cannot be withdrawn" (Institution Theory Z37). The three conditions are: the channel is independent of the party questioned, internal subjects can actually initiate it, and initiating it brings no retaliation (Institution Theory Z34). How high a threshold is, how the two chambers are composed, how long terms run, which court reviews: these are structure, and can be changed; if an amendment withdraws one of the three conditions, what it lacks at the third layer is the necessary condition of openness to questioning, and it must be questioned.
Clause 2 falls on the second of these. "This article may not be amended" shuts off the road to amending this article: questioning of it can still be raised as usual thereafter, but amendment can no longer be initiated. This is withdrawal, not change.
How this differs from an eternity clause must be made clear. An eternity clause is a provision declaring provisions unamendable, and handing over to a final position the recognition of whether they have been touched. The line this paper speaks of is not written in by a provision; it declares no stretch of text complete, and it hands the scope of the line to no position as final. Writing a floor such as human dignity into the supreme rule as a declaration is not of this kind: the declaration may be written, and the words written may be changed; when the words are changed, the line does not move with them, and acts that cross it thereafter are line-crossing all the same.
Who recognizes whether an amendment has changed the line? The court makes the decision for the case: someone requests review under the Charter, the court makes a judgment on this amendment, and it is final in the case. But whether this kind of judgment can be questioned again thereafter is not for the court to decide on its own (Paper 2, N8, N10); the Theory of Government has a sentence of the same shape: "whether it can be asked again is not decided by the place that made the decision on its own" (Theory of Government G36). The scope of the line itself is handed to no position as final. This step too is undertaken by this series, for the reason given in the preceding paragraph: if a court's reading of the line were final by position, the court would become the final position that has the say over "whether they have been touched," and the eternity clause would only have moved house.
Clauses 1 and 3 each have their own destination. Clause 1 forbids member units to exit; whether secession is to have a procedure, and how the procedure is not to claim to be complete, Paper 3, §6 takes up (N16). Clause 3 hands a class of organizations to the Final Court for dissolution; it touches speech and association (Charter, Article 4, fact), and it touches too the requirement that those who use the road of questioning suffer no retaliation (the fifth ought); whether it is within the line is checked under the first ought, and is not exempted by having gone through the amendment procedure; Paper 3, §6 reads this clause too. The Year 12 proposal deletes clause 2 and changes the object of clause 3 from "whose purpose is to bring about the exit of a member unit" to "whose purpose is to bring about the exit of a member unit by violence": the first change takes back an eternity clause; the second moves the object from a purpose to an act. Whether the clause, so moved, is within the line is still checked under the first ought; this paper does not pass judgment on it.
Back to the opening case. Of the three clauses of Charter Amendment 14, which change structure, and which touch the conditions of being questioned?
Clause 1: no election to the Chamber of Citizens during the state of war; one within six months after it is terminated. Elections are the channel through which internal subjects question the Chamber of Citizens and the First Executive. The state of war is terminated by a majority in each of the two chambers (Charter, Article 54, fact); the majority in the Chamber of Citizens alone can block termination. Read by function, when the next election comes is, while this clause is in force, in the hands of the party that elections question. What this withdraws is the independence of the electoral channel.
Clause 2: renewal is decided by the Executive Council, and the joint session may terminate by three-fifths. The Theory of Government says that the determination of an emergency may be made by one position, but "whether the determination holds and when the emergency ends, together with the review of confirmation, prolongation and termination, are not monopolized by the party taking the measures" (Theory of Government G32). Handing renewal to the party taking the measures itself withdraws the independence of confirmation and renewal. Renewals made under this clause have a further layer to be checked: whether, after hostilities have ceased, they still have a wartime ground; Paper 4, §5 takes this up.
Clause 3: for the Final Court to declare an emergency order inconsistent with the Charter requires the agreement of eight judges. The channel of review remains; the threshold has been raised. What this changes is the review threshold; how much effect review retains in function once the threshold is raised is checked in Paper 2, §1 on the case of the detention clause; this paper does not judge it here on its own.
All three clauses touch the conditions of being questioned; the first two withdraw independence, which is among those conditions. On this section's reading, these two clauses are valid at the second layer and lack, at the third, the necessary condition of openness to questioning: they must be questioned, and can be taken back. When the three clauses were voted on is checked in the next section.
Amendment is a procedure. Following this series' way of establishing procedures, the remainder it leaves is stated here. Carrying it through under Article 70 closes this one vote; the following are not settled with it: the twenty-six thousand who did not vote (UCSE Case 04); the units where a majority of those voting were against; those born afterward, who were never asked; the questions the amendment leaves on the line and on the conditions of being questioned. A claim the procedure has not caught does not lose its qualification to be questioned because it is "not in accordance with the procedure"; the procedure itself is open to questioning and can be amended. Stating these does not require listing everything unknown, nor does it cancel this vote's effect of closing the case in this matter.
N3 The structure can be changed; the line cannot, and the conditions of being questioned cannot be withdrawn. No eternity clauses: "A law that cannot be modified is a law that claims to have no remainder. Within the SAE framework, this is a structural lie." (Law Series P1 BL4) Ultra-high thresholds plus exit take the place of "unamendable" (F6 §6, fifth contribution). The line (the first ought) is not within the authority of any provision, and the conditions of being questioned are not withdrawn by amendment either: "structures can change, but the three conditions cannot be withdrawn" (Institution Theory Z37). An amendment that changes the line can be carried through procedurally and is line-crossing normatively; it does not become legitimate by having used the amendment procedure (own undertaking). This differs from an eternity clause: an eternity clause is a provision declaring provisions unamendable, and handing over to a final position whether they have been touched; this series' line is not written in by a provision, and it declares no text complete. Writing a floor such as human dignity into the supreme rule as a declaration is not of this kind. Whether an amendment has changed the line is decided by the court for the case; whether this kind of judgment can be questioned again thereafter is not for the court to decide on its own (Paper 2, N8, N10; Theory of Government G36): the scope of the line itself is handed to no position as final (own undertaking).
§4 Constitutional Amendment Is Not Monopolized by a Single Position and Does Not Go by Emergency Procedure
Where do the submitting, the benefiting and the being reviewed under clauses 2 and 3 of Charter Amendment 14 each fall? The Executive Council submitted it; renewal is now decided by the Executive Council, and the threshold for reviewing emergency orders is raised: the one that benefits is the Executive Council; what is reviewed is emergency orders issued by the Executive Council. Submitting, benefiting and being reviewed fall on one and the same position. But submitting is not deciding: the two chambers vote by two-thirds each, and the voters vote in the referendum. The amendment is made by several positions together; on paper it is not monopolized by one position.
The Theory of Government says: "The determination of an emergency is not monopolized by a single position." (Theory of Government G32) This paper extends it to the supreme rule (extension; own undertaking): amending the supreme rule is not monopolized by a single position. The original sentence speaks of the determination, confirmation, prolongation and termination of an emergency; the extension goes through because the reason is the same in both places. The ground of G32 is that the party under review may not monopolize the decision on whether a challenge exists and whether it is over (Institution Theory Z35a); amending the supreme rule alters the arrangements for being questioned of every position, and a position that could amend it on its own could alter on its own the arrangements for its own being questioned. Charter Amendment 14 passes this half. The problem lies in the other half.
The other half is "does not go by emergency procedure." The annex at the opening said: "The Emergency Act does not permit elections to be postponed by emergency order; postponing elections requires a constitutional amendment." One road was blocked, and another was taken. The amendment went by the ordinary procedure of Article 70; by name, it is not emergency procedure. But it was proposed and voted on while emergency orders were restricting assembly and vetting broadcasts.
Power Theory, in a structural demonstration, read the course that a configuration of this kind takes: the source claimed for the emergency-decree power is emergency necessity, while the judgment-power that actually triggers it is concentrated in a single position; this configuration reads as "sliding from exceptional use toward routine use, maintenance cost concentrating onto a single position" (Power Theory Paper 0, §V.1). Only this structural reading is taken here, not the events on which that demonstration drew. The next step it points to is this: once the slide to routine is complete, "legal" is rewritten by way of a legal procedure. That is the step N4 is to block.
Appendix B of How Is Institution Possible has a sentence: "The Constitution may not be amended by emergency procedure." (F6, Appendix B, Article 28; a working demonstration; used only as an example) It points the same way. But Charter Amendment 14 shows precisely that, by name, this sentence cannot block it: it did not go by emergency procedure. So a sentence by function must be added:
While emergency measures restricting assembly, speech or communication are in force, anything that alters, in any form (constitutional amendment, an enabling act, emergency legislation), the arrangements for being questioned (the dates of elections; who decides on renewal and termination; the thresholds for termination and review; the establishment and remit of checking positions), or that takes emergency powers out of their existing confirmation and renewal (setting no time limit, dispensing with confirmation, making them standing), is not put to a vote; it is recognized by function, not by name; what is already on the agenda is voted on after the restrictions are lifted. Confirmations and renewals made under existing rules, and emergency legislation that does not alter the arrangements for being questioned and carries its own time limit, fall outside this, and are checked under Paper 4, N20 to N22, and Theory of Government G30. Amendments that take back emergency powers, or that restore or strengthen the conditions of being questioned (such as adding a substitute confirming position not appointed by the party taking the measures), fall outside this.
The arrangements for being questioned are those structures that touch the conditions of being questioned: in ordinary times they can be amended under Article 70, and during restrictions they are not put to a vote; an amendment that withdraws the conditions of being questioned themselves is at no time within the authority (§3).
Why go one step further than the Theory of Government? The Theory of Government says that an emergency "may move review and being questioned from beforehand to afterward, but it does not withdraw being questioned" (Theory of Government G30). That holds for ordinary emergency measures: curfews and requisitions are reviewed and questioned afterward, and what is reviewed and questioned is that measure. Amendment is different: what it alters is precisely the arrangements for being questioned. When the questioning comes afterward, what it relies on is already the altered set: elections on the new dates, renewal in the hands of the new position, review under the new threshold. The questioning that matters most for this alteration ought to have taken place under the old conditions, while people could assemble and speak; if the vote is held during restrictions, that questioning did not take place under the conditions it ought to have had, and this cannot be made up afterward. So this step is stronger than G30, and is undertaken by this series (extension; own undertaking).
The two "fall outside this" clauses have their reasons too. The emergency itself must still be confirmed and renewed under the existing rules; if voting were barred across the board, confirmation could not be made either, and the emergency would instead be handed to the party taking the measures. Taking back emergency powers, and restoring or strengthening the conditions of being questioned, open the door, and need not wait; adding a substitute confirming position not appointed by the party taking the measures is of this kind too (Paper 4, N21). "Carries its own time limit" is read under Paper 4, N22: a term pegged to a date that may never come, such as the signing of a peace treaty, is treated as no stated time limit, and does not count as carrying its own time limit.
Take this sentence back to the opening case. All three clauses of Charter Amendment 14 alter the arrangements for being questioned (§3); the amendment was proposed on the twelfth day after the state of emergency was declared, and the vote and the referendum both took place while assembly was restricted and broadcasts were vetted. Recognized by function, its going by Article 70 does not change its falling within this sentence: it is not put to a vote; what is already on the agenda is voted on after the restrictions are lifted. The 58.3 percent of that referendum is one step of this procedure, and does not settle its account; the three districts where no polling stations were set up, and the twenty-six thousand registered voters who did not vote, are the remainder this procedure left, booked to the positions that made these decisions, with the minimum content of Paper 7, N39 applied.
When one road is blocked and another is taken, who can question the detour? Recognized by function, the detour is caught by this sentence whichever road it takes; whether a detour is recognized as one is not for the party proposing the amendment to decide on its own (Paper 2, N8, N10). Under the UCSE's configuration, the government of a member unit, a quarter of the members of the Chamber of Citizens, or those affected who have exhausted other remedies can ask the Final Court for review (Charter, Article 43, fact); voters can ask at the next election, provided that the date of the election is not in the hands of the party questioned, which is exactly where clause 1 touches.
This sentence is itself a procedure too, and states its remainder. How "while ... in force" is recognized, which measures count as restricting assembly, speech or communication, whether an alteration counts as altering the arrangements for being questioned: all are recognized by function, and the recognizing is not for the party proposing the amendment to decide on its own. A vote after the restrictions are lifted is not thereby legitimate; it is still checked under §3 and at the third layer. If restrictions are lifted for the sake of a vote and reimposed after it, this sentence cannot block every such arrangement in advance; this is registered as a remainder, left to questioning.
What the supreme rule lies under can now be stated in full. Who initiates: internal subjects, in referendums and elections; under the UCSE's configuration, the government of a member unit, a quarter of the members of the Chamber of Citizens, and those affected, requesting interpretation and review under Article 43 of the Charter (fact). Which object it affects: the proposing and voting of amendments. Whether the party questioned can close it: no; whether an amendment crosses the line, the court makes the decision for the case; whether it can be questioned again thereafter is not for the court, nor for the party proposing the amendment, to decide on its own; votes during restrictions are blocked by function. How the account is booked when it fails: an amendment that crosses the line is booked as line-crossing, open to questioning and to being taken back; an amendment that withdraws a condition of being questioned must be questioned; records are attributed to the positions and persons that made the decisions, under Paper 7, N39.
N4 Constitutional amendment is not monopolized by a single position and does not go by emergency procedure. Theory of Government G32 is extended to the supreme rule (extension; own undertaking). What is to be blocked is the emergency-decree power "sliding from exceptional use toward routine use, maintenance cost concentrating onto a single position" (Power Theory Paper 0, §V.1), and then "legal" being rewritten by way of a legal procedure. While emergency measures restricting assembly, speech or communication are in force, anything that alters, in any form (constitutional amendment, an enabling act, emergency legislation), the arrangements for being questioned (the dates of elections; who decides on renewal and termination; the thresholds for termination and review; the establishment and remit of checking positions), or that takes emergency powers out of their existing confirmation and renewal (setting no time limit, dispensing with confirmation, making them standing), is not put to a vote; it is recognized by function, not by name; what is already on the agenda is voted on after the restrictions are lifted. Confirmations and renewals made under existing rules, and emergency legislation that does not alter the arrangements for being questioned and carries its own time limit, fall outside this, and are checked under Paper 4, N20 to N22, and Theory of Government G30. Amendments that take back emergency powers, or that restore or strengthen the conditions of being questioned (such as adding a substitute confirming position not appointed by the party taking the measures), fall outside this. This step is stronger than Theory of Government G30: G30 allows being questioned beforehand to be moved to afterward, but amendment alters precisely the arrangements for being questioned, and being questioned afterward cannot make this up, so it is undertaken by this series (extension; own undertaking).
§5 Why the National Level Must Separate Its Powers
In the months when Charter Amendment 14 was being voted on, was there anyone outside the UCSE who could question it?
Routes outside the state do exist. After the Cloud River arbitral award was made, the UCSE and the NUT both acted on it within Year 11 (UCSE Case 20; three things at the opening of Paper 5). Only this one fact is used here, as a contrast: the external route exists, and it works; its award landed because positions in each of the two states acted on it.
The Theory of Government wrote about the separation of powers at the city level, in three steps (Theory of Government G26): every government is subject to being questioned; a diagnosis, level by level, of whether exit and external routes can carry the day-to-day check on that level; and, at the threshold, checking is placed inside the level, independent of the party checked, and rule-making, execution and checking are not held by one and the same position. The diagnosis of step two has a conditional: "Where exit is insufficient to carry the check a level of government requires, the check must come from another, usable source; where external routes are insufficient too, that level always undertakes the guarantees it has embedded (Paper 3, G17), and a real and usable function of check must be retained internally." (Theory of Government G27) At the city level the Theory of Government did not check this antecedent city by city; it made a normative undertaking: "From the city level up, this strong form is taken" (Theory of Government G26). At the national level it stopped: "whether the national level still has independently invocable external routes above it, this paper does not judge" (Theory of Government Paper 4, §5). This paper takes up from there.
What this section has to do is, at the national level, to replace with derivation whatever in that undertaking can be derived, and to undertake separately, following the Theory of Government, whatever cannot; which part is which is stated. First, the range of the derivation. What follows considers only the class of corrections that must be realized by this state, namely those that require this state to do, or to stop, an act toward its internal subjects. External routes can review independently and reach valid decisions; but if implementation is decided entirely by the position being questioned, that position can block implementation on its own. So this class of check requires a taking-up at this level that the position cannot monopolize. This conclusion does not exclude other external routes, nor does it claim that every check can be delivered only in this state. That implementation at this level is necessary does not by itself prove that checking decisions must also be formed at this level; that layer is treated separately in step three. Three steps follow below; the fourth is in the next section.
Step one: the diagnosis. Following step two of Theory of Government G26, the question is: as regards this class of corrections, can exit and external routes carry the day-to-day check on the national level? G27 is the conditional for this question: if the antecedent holds, a real and usable function of check must be retained internally.
Step two: three things derived from the definition. A state is a construct: inwardly, there is no higher level above it to govern it; outwardly, there is no higher level above it to govern what lies between it and other states (the Prequel, §6). The definition names an extreme. An actual state may have treaty mechanisms above it that carry no force, and so stands at some distance from this end; the distance from this end changes how much external routes can make up; it does not change the taking-up at this level that the class of corrections that must be realized by this state requires. From this sentence three things are derived (own undertaking).
The first: external routes exist, but they land in this state. The national level's external routes, treaties, arbitration, mediation by a third state, and treaty mechanisms admitting individual petitions, do exist, and can work (UCSE Case 20). Some can even be initiated, and reach decisions, without this state's consent on that occasion: under a treaty mechanism admitting individual petitions, the people inside can bring the matter up themselves (there is no such mechanism among the three states; only its form is spoken of here). But they have no enforcer of their own: with no force at a higher level above, whether an award lands once it is made depends on whether this state's positions carry it out. Where an award requires this state to do something to the people inside it, or to stop (release a person, hold an election, withdraw a restriction), and this state's positions do not carry it out, what can enforce it, or carry it out by substitute execution, lies only within this state (following the function of substitute execution in Paper 2, N11b); acts that another state takes within its own authority, for instance against this state's property in its territory, can at most make up a part of the monetary kind, and cannot stand in for the landing of that act itself. Another state's entering this state by force to do that thing is an outward act, checked separately under Papers 5 and 6 and the War Treatise; it is not a day-to-day road of landing. For this class of corrections, external questioning goes this far, and lands only once it connects to the channels inside (Paper 7, N38). Back to the question at the start of this section: in the months when Charter Amendment 14 was voted on, there was no body among the three states that could take it up, nor any treaty mechanism admitting individual petitions; and even if there had been, its decision would have had to land through the UCSE's positions.
This first point is to be read together with a sentence from the upstream. The Conflict Treatise says, of first-person conflict: "only an institution fit for purpose, available and not controlled by a party to the dispute has standing to bear the principal screening." (Conflict Treatise §15) What is spoken of there is how a person does not monopolize the screening of those who question him; this paper extends it to the construct and its positions, and from "screening" to "landing" (extension; own undertaking): for a road that bears the principal check, the step of landing may not be in the hands of the party questioned. The extension goes through because the two places guard against the same thing: if the decisive step of a road is in the hands of the party questioned, it can close the road when it wishes, and the road cannot bear the principal share.
Who is the party questioned? Not the whole state, but the positions questioned. How Is Institution Possible says: "'Independent' here is functional, and does not require being physically outside the institution: a judiciary within a state, a grievance procedure within a company, do not lose their reviewing force merely by being inside the same institution; what must be prevented is only that the object of review can seal off review of itself." (F6 §3.2) Independence is checked by position and function, the same reading as Paper 2, N9.
Put the two together and the conclusion follows. For an award of that class to land, it must pass through some position of this state; if what has to carry it out is precisely the position questioned, landing is in the hands of the party questioned; for it not to be in that party's hands, there must be, within this state, another position, not occupied by the party questioned, that can make the award land. So, as regards this class of corrections, external routes, checked item by item, can make up part, and some can make up a great deal, but they cannot stand in for the taking-up at this level: the national level must have a check at this level, not occupied by the position questioned, that can make decisions of this class land. This does not say that external routes do not exist, nor that they are controlled by this state, nor that they are in fact weak: which of them are actually usable is checked item by item, and those that are usable are used (Paper 2, N9; Paper 5, N24). How such mechanisms bind states, and how far, this series acknowledges as a remainder, left to the Theory of Inter-State Cooperation to come.
The second: exit can only be lateral. Someone who leaves a city is still in the interior of the levels above it, up to the national level, and the guarantee of the layer above goes with him; someone who leaves a state has no layer above to go with him, and can only enter another peer, and only if that peer takes him in. That the leaving end is not sealed off by this state is written in Paper 3 (N13); how those who arrive at this state's door are in its interior is also in Paper 3 (N15). This road cannot carry the day-to-day check. The Theory of Government said in Apple City that moving away cannot answer what a city lays on people today (Theory of Government Paper 4, §3); for a state, moving away also needs another state's nod.
The third: the guarantee is always there. The national level always undertakes the guarantee of the recognition-based rights it has embedded (Theory of Government G17; §7). When the Theory of Government read F6's conditional, it already said that on the side of government this antecedent is "read as that guarantee, and it is always present" (Theory of Government Paper 4, §5); the national level is one of the levels of government, and this applies as it stands.
Taken together, the three things show that, as regards this class of corrections, the antecedent of G27's conditional holds at the national level: exit is insufficient, external routes cannot stand in for the taking-up at this level either, and this level always undertakes the guarantee; the consequent, "a real and usable function of check must be retained internally," is here at least an implementation at this level that the position questioned cannot monopolize. What the diagnosis of G26's step two thereby obtains is that this share of implementation must be at this level.
Theory of Government G27 also says, in the same proposition: "each level's exit and external routes are read as they actually are, not presumed from its rank." This paper does not presume that the national level's external routes are in fact weak (end of the first point). What is derived from the definition is something else: however strong they actually are, for this class of corrections to land, that step is in this state. So "where external routes are insufficient too," in the antecedent, is read at the national level as "they cannot stand in for the taking-up at this level (implementation) that this class of corrections requires"; this reading is structural; as regards this share of implementation it holds for every construct that fits the definition, and it is not a presumption about actual conditions made by level (own undertaking). It does not answer whether external checking can carry day-to-day checking judgment. That question G27 addresses separately in the same proposition: "the threshold for the separation of powers is read by level under G26's normative undertaking and is not reversed by the readings here"; step three keeps the two apart accordingly.
In plain words: there is no leaving; the step that requires this state to do something to its own people, or to stop, no outside route can take in its place; and the state always undertakes the guarantee. So implementing this step needs a place inside the country that is not in the hands of the party questioned.
This step does not rest on geography, nor on the conditional that How Is Institution Possible wrote within the nation-state case. The upstream has statements of the same shape, and they stand only as side evidence. At the layer of the nation-state, How Is Institution Possible wrote: "Exit is constrained, so the check of voting with one's feet fails and the check must come from elsewhere; and sharing one continent makes external checks hard to invoke independently." It at once limits itself: "The necessary condition above is written within this case, not as a general theorem." (F6 §3.1b) The Law Series wrote: "The three variables combine: exit is hardest, collisions are densest, shared identity is thinnest. The thickness of law reaches its peak at the national scale." (Law Series P3 §4) These two passages serve only as side evidence of the same shape. The "sharing one continent" in the side evidence this paper reads as entanglement that cannot be avoided (the Prequel, §6); but a sentence a little later in the same place, "sharing a continent does not by itself prove that every external check must fail to be independent" (F6 §3.1b), holds equally of entanglement. So this paper does not derive the failure of external routes directly from entanglement; it uses only the first point above.
Step three: checking judgment at this level. What has been derived so far is implementation. Step three of Theory of Government G26, checking placed inside this level with the three functions not held by one and the same position (holding them together meaning that one position holds the whole of two), is in Theory of Government Paper 4, §3 a conditional whose antecedent is that the outer layer's checking does not reach the day-to-day: "Apple City makes its own rules and carries out compulsion itself, and the outer layer does not reach the day-to-day; so checking has to be placed inside Apple City" (Theory of Government Paper 4, §3). What was derived above is that the outside cannot stand in for implementation at this level, not that external checking does not reach the day-to-day. Imagine a configuration: external checkers independently take up matters and make valid decisions; several positions at this level, none monopolized by another, only implement; and no position questioned can close a channel directed at itself. It has independent implementation at this level, yet checking judgment lies wholly outside. Step two cannot exclude it, and the consequent cannot be applied as it stands. This layer this paper does not derive; it is undertaken separately.
Undertaken separately. This series undertakes separately, as a normative undertaking following Theory of Government G26, to keep the three functions of rule-making, execution and checking separate at this level, with checking at this level carrying day-to-day checking judgment for the class of corrections that must be realized by this state; this undertaking includes at least one independent position of checking judgment at this level. The derivation in the first two steps does not serve as the proof of this undertaking. External checking can reach valid decisions independently and stands alongside it; what is usable is used. By which agencies each function is realized is still left as a remainder. What the Theory of Government made from the city level up is exactly such an undertaking ("From the city level up, this strong form is taken," Theory of Government G26), and at the national level it stopped at "does not judge" (start of this section); G27 says that the threshold for the separation of powers is read by level under G26, and at the national level this paper undertakes it separately, accordingly. The reasons for and the explanation of the undertaking are not its proof: the implementation at this level derived above; and the two sentences the Law Series originally established at the national scale, "Three branches form the minimum closed loop for processing remainder" (Law Series P3 §5.1) and "You cannot genuinely question yourself" (Law Series P3 §5.3). What is counted is functions and independent checking positions, not agencies; no constitutional form is fixed (Theory of Government G26).
Exit and switching; external questioning cannot be cancelled. An external checker has jurisdiction over this state through this state's accession; there is no higher level above to compel on its behalf (the definition). Exiting a treaty, ending or narrowing an arrangement admitting individual petitions, switching to another external body: all are decisions of this state's positions. How such decisions are to be reviewed, and by whom, whether by handing them to the outside to judge by rules set in advance, or by having one place at this level judge; whether a rule can also hand to the outside the decision to change that rule itself; whether a state that has locked its exit in the hands of the outside is still at this end of the definition: these are questions of how cooperation between states is arranged; this series acknowledges them as a remainder and hands them to the Theory of Inter-State Cooperation. Here only one sentence is established (own undertaking): external questioning cannot be cancelled. This state may exit or switch an external arrangement; that is a change of structure, itself under questioning (Paper 5, N24); it can close one channel, but it cannot cancel the questioning of it by other states, by the people of other states, and by those affected; how it answers follows Paper 5, N27 and Paper 7, N38. This sentence comes from the same source as Theory of Government G36: "no position may monopolize the power to shut the door." It does not derive which kind of checking judgment this level must have; that layer is undertaken separately, as in the preceding paragraph.
With Montesquieu. Why power must be divided, Montesquieu said in Book XI of The Spirit of the Laws (De l'esprit des lois). He says: "It is an eternal experience that every man who has power is inclined to abuse it; he goes on until he finds limits. Who would have said it! Virtue itself needs limits." (my translation; c'est une expérience éternelle, que tout homme qui a du pouvoir est porté à en abuser; il va jusqu'à ce qu'il trouve des limites. Qui le dirait! la vertu même a besoin de limites; bk. XI, ch. 4, per the edition of Paul Janet, 2nd ed., Delagrave, 1892, Project Gutenberg EBook #27573; tier A) He goes on: "So that no one can abuse power, it must be that, by the arrangement of things, power stops power." (my translation; Pour qu'on ne puisse abuser du pouvoir, il faut que, par la disposition des choses, le pouvoir arrête le pouvoir; ibid.) In ch. 6 he says further: "All would be lost if the same man, or the same body, whether of the principal men, of the nobles or of the people, exercised these three powers: that of making the laws, that of carrying out public resolutions, and that of judging crimes or the disputes of private persons." (my translation; Tout serait perdu si le même homme, ou le même corps des principaux, ou des nobles, ou du peuple, exerçait ces trois pouvoirs: celui de faire des lois, celui d'exécuter les résolutions publiques, et celui de juger les crimes ou les différends des particuliers; bk. XI, ch. 6, same source)
The common question: how power is kept from falling in one place. How far the agreement goes: relying on "the arrangement of things," and not on the virtue of any person, concurs with this series' checking of independence by function, which vouches for no one in any position; that the three are not held in one hand concurs; that he counts a body composed of "the people" among those that may not exercise the three powers together concurs too with what §2 and §6 say, that "the people" is not a top exempt from questioning. Where they part: his reason rests on an experience, that all who have power are inclined to abuse it; this series writes no inclinations for persons, and writes no ought for persons; the reason it can derive rests on structure: the national level has no higher enforcer above it, and the class of corrections that requires this state to do something to its own people, or to stop, lands within this state; the rest is an undertaking following the Theory of Government, and that too does not rest on any inclination of persons. The two kinds of reason can stand together; what parts them is only which one bears the load. His ch. 6 describes one particular state's constitution, and this series fixes no form. This is a division of scope and of reasons; it does not say that he was wrong.
The Federalist, Nos. 47 and 51, treat the separation of powers, and Theory of Government Paper 4, §3 has already conversed with them; here this paper only points back to it. What this section newly does is to turn part of the undertaking into derivation and to undertake the rest separately, following the Theory of Government, stating the two apart.
N5's steps one to three end here. Step four, the checker must be questioned too, is in the next section; N5 in full is stated at the end of §6.
§6 Those Who Check Are Checked Too
In ordinary times, complaints against the Defense Force and the Office of Defense are taken up by the Inspectorate (Charter, Article 47, fact; Charter Amendment 11). In the summer of Year 9, Article 17 of the Wartime Security Act provided: while the state of war continues, complaints against the Defense Force and the Office of Defense are taken up by the Internal Investigations Division of the Office of Defense; those received by the Inspectorate are transferred to the Division. The head of the Division is appointed by the Executive for Defense. The same act set up the Joint War Committee, and the public hearings of the Defense Committee of the Chamber of Citizens were suspended.
In the spring of Year 10, the Inspector-General's term ended. The joint session voted twice; the candidates received 181 and 176 votes, short of three-fifths (192 votes). Under the transitional provision of the Inspectorate Act, the Executive Council designated the former deputy director of the Legal Affairs Bureau of the Executive Council as acting Inspector-General. In the acting Inspector-General's two years in office, the Inspectorate issued zero public reports; the previous Inspector-General had issued forty-one in seven years. In the spring of Year 12, the new joint session elected a new Inspector-General with 201 votes; when the state of war was terminated, Article 17 expired by its own terms (UCSE Case 06).
Three things first. Whom it concerns: those who have something to ask of the Defense Force and the Office of Defense; those who complain to the Inspectorate against organs at the UCSE level. Which line: the fifth ought: the independence of the checking link (whether the position that takes up complaints is independent of the party complained against). Baseline: in ordinary times the Inspectorate takes up complaints against the Defense Force and the Office of Defense; the previous Inspector-General issued forty-one public reports in seven years.
The previous section reached step three: the class of corrections that requires this state to do, or to stop, an act toward its internal subjects requires implementation at this level that the position questioned cannot monopolize; at least one independent position of checking judgment at this level, separation of the three, and checking at this level carrying day-to-day checking judgment are undertaken separately, following the Theory of Government. This section asks: who checks the positions already established at this level, the checking one among them?
Step four: the checker must be questioned too. Checking too is held by positions; positions are part of the construct. Institution Theory says that an institution "cannot acquire the standing of permanent exemption from questioning" (Institution Theory Z37); the Theory of Government says that "a construct's configuration ought to place every position under some channel it cannot itself close" (Theory of Government G36). Checking positions are not outside this.
Where, then, is the link that questions the checker? There are only three places it can go.
First, upward: hand it to a higher checker. At the national level there is no higher level above (the definition); external checkers can review, and can make valid decisions, but if their correction of the checker requires this state to do something to its own people, or to stop, landing is still in this state (the first point of the previous section). This share of check that must be realized at this level cannot be handed entirely to higher or external checkers.
Second, lengthening outward: set up one to check the checker, then one to check that one. If the chain stops, the place where it stops is, as regards this class of check, a top at this level that no one checks, contrary to the sentence above; if the chain does not stop, the finite positions run out, and it must come back to some place already on the chain. The Law Series has said: "The solution is not to lengthen the chain (that only multiplies breakpoints) but to fold the chain into a closed loop." (Law Series P3, §5 introduction) Theory of Government Paper 4 did not borrow this sentence, together with the sentence before it about the chain being too long: "Both sentences depend on scale, and this series does not borrow them." At the national level, the sentence is back at the scale on which it was originally established. But the reasoning here does not rest on the length of chains; it rests on two things only: positions are finite, and, as regards this class of check, there is no top.
Third, folding back: an existing link on the chain checks the last link. Only this one remains.
Write it as a proposition that can be tested. Take as the set the finite functional positions already established at this level that carry the authorizing, performing, checking and implementing of the internal acts above, and consider only the relations of check that must be taken up at this level. To say "A is checked by B" means that B holds a route directed at A, which A cannot close on its own, and which can land a correction requiring this state to do, or to stop, an act toward its internal subjects. "Cannot close" and "can land" are read as in Theory of Government G36 (Paper 2, N8 reads them the same way): A can make the decision for the case on a matter directed at itself, but cannot decide on its own whether this kind of questioning can be initiated thereafter; where for a time a correction cannot land, or several links are in a standoff, this is registered as a remainder (the paragraph below on what the loop guarantees). The checking link can make the decision for the case on an amendment directed at it, and on the Impeachment Act; that it is checked by the rule-making link holds only on this reading. The execution link has a further, stronger layer: non-execution cannot block the case (Paper 2, N11b); this layer is not needed for the loop to form. If it has been shown, of each position in the set, that it must be checked by another independent position at this level, then, tracing back along this relation, finite positions must repeat, and so there is at least a loop. Valid external routes can stand alongside it; they are not cancelled by the definition, "with no superior." The conclusion that a loop is formed here is conditional on the position-by-position requirement of a check at this level having been established. This step is undertaken by this series.
The account below concerns the positions already established at this level. Each class of position must be checked by another independent position at this level: the first half of each paragraph below is derived from §5 and does not depend on configuration; the second half, "in this configuration," says which place does the taking-up in a configuration where the three functions are separated, each with its own position at this level (the kind undertaken separately in step three, which is also the UCSE's); this half does not serve as proof of the layer undertaken separately, nor does it say that other configurations cannot form a loop.
First, which positions are in the set of this configuration. The execution link: the positions that issue orders to internal subjects, detain, restrict, requisition, and conduct elections; in the UCSE, the Executive Council and its offices, and the Defense Force too when used internally. The rule-making link: the positions that authorize these acts by legislation and amend the supreme rule; in the UCSE, the two chambers (amendment also requires a referendum; those who vote in it are internal subjects, not positions; see below). The checking link: the positions that make decisions on the acts and rules of the first two links and take up complaints; in the UCSE, the Final Court and the Inspectorate. Beyond these, any position that lands a correction, such as the place that carries out substitute execution under Paper 2, N11b, is also in the set, since its implementing is itself an act toward internal subjects. The set is closed by this definition, not by enumeration; positions at the national level are finite, and the set is finite.
Now, one by one: by which independent position at this level each class of position must be checked, and why that check cannot be handed to the outside for delivery.
The execution link. Its acts fall directly on people: a detention, a restriction, a requisition. The corrections that must be taken up at this level are releasing the person, stopping or withdrawing that restriction, holding an election as a ruling requires. External routes can review it, and can find that it crosses the line; whether the person is released, whether that restriction is stopped, requires this state's positions to act. If doing or not doing were entirely up to the execution link itself, it could block by not doing. So another position at this level must take it up, landing this correction without depending on the execution link's assent (§5). In this configuration, the one that makes the decision is the checking link (under the UCSE's configuration, the Final Court reviews on request, and "Its rulings bind all organs," Article 43 of the Charter (fact); the Inspectorate calls for the files and requests review); landing requires a function of substitute execution that does not depend on the execution link (Theory of Government G36a; written out in Paper 2, N11b), and the position of substitute execution is not occupied by the party substituted for; the Charter does not provide who carries out a ruling when an organ does not (fact), which Paper 2 takes up. The execution link cannot close the checking link. The checking link is in the set.
The rule-making link. It does not move people directly, but its laws authorize the preceding link to move them, and its amendments to the supreme rule also alter the arrangements for being questioned (§§3 and 4). The corrections that must be taken up at this level are that a law that crosses the line, or an amendment that changes the line or withdraws a condition of being questioned, stops applying to this state's people; and that what alters the arrangements for being questioned during restrictions is not put to a vote (N3, N4). External routes can say that it is inconsistent with some treaty; whether it stops applying to this state's people requires this state's positions to stop it. If stopping or not stopping were entirely up to the rule-making link itself, it could block by not amending and not repealing. So another position at this level must take it up (§5). In this configuration, it is the checking link's review of laws and amendments: a decision that something crosses the line is final in the case, and whether it can be questioned again thereafter is not for the court to decide on its own (Theory of Government G36; Paper 2, N10). Nor can the rule-making link close this review: changing which position reviews, and how high the threshold is set, is structure, which can be changed in ordinary times and is not put to a vote during restrictions (§§3 and 4); removing review altogether without setting up another position that it cannot close and that can land this class of corrections, or raising the threshold until review in fact never reaches action, is, read by function, withdrawing a condition of being questioned (a channel independent of the party questioned), and is not within the authority of amendment (§3; Paper 2, §1; Theory of Government G36a). The checking link is in the set. Internal subjects also question the rule-making link in elections and referendums; elections must be held by someone, and their results determined by someone, and that is done by positions in the set; this road too falls within the set.
The checking link. Its decisions also fall on people: upholding a detention, dismissing a complaint, answering with the one sentence "This case has been finally decided; it will not be taken up further." The corrections that must be taken up at this level are retrial under new evidence or new problems, changing its remit and its review threshold, and asking judges about exceeding authority, bribery and dereliction of duty (Paper 2, N7, N11a). External routes can find that a ruling of this state's court was wrong; whether that ruling is changed for this state's people, whether the case is reopened, whether the judge stays, requires this state's positions to act. If all were up to the checking link itself, it could close the matter by saying "finally decided." So another position at this level must take it up (§5). In this configuration, it is the rule-making link amending rules and constraining the boundaries of checking, and impeachment asking about the judges' conduct; for the Inspectorate, the rule-making link also sets its remit, and the joint session elects the Inspector-General (UCSE Case 06); the Law Series says that "The remainder of checking power (whether the checker's 14DD infiltrates the checking) is questioned by rule-making power" (Law Series P3 §5.3). The checking link can make the decision for the case on an amendment, but it cannot close the road of amendment itself (Paper 2, N7). The rule-making link is in the set.
The position that lands the correction. The place of substitute execution also does execution, and belongs to the execution class; as above, another position at this level must take it up: if it goes beyond the scope of the ruling, the correction likewise requires this state to stop it; in this configuration, that is the checking link.
If a position holds part of each of two functions (holding them together meaning that one position holds the whole of two, §5), then for whichever part it holds, it is checked by the position that takes up that class, and what takes it up must be other than itself, a position it cannot close; the checks of both classes reach it. If the taking-up of some part falls precisely on itself, that part must have a taking-up other than itself: the joint session also confirms states of emergency; confirmation belongs to the checking class, and the taking-up for the checking class is in turn the rule-making link; in the UCSE, what catches it is the Final Court's review of emergency orders (Paper 4, §3). Without such a taking-up, on that part it questions itself, and the "another independent position at this level" that the proposition requires comes to nothing.
Taken one by one, the condition holds for each class of position already established: every position in the set is checked by at least one other position in the set that it cannot close on its own, and that check requires this state to do, or to stop, something toward its internal subjects, and cannot be handed to the outside for delivery. By the proposition above, at least a closed loop is formed. In this configuration, the loop runs like this: rule-making and checking check each other, and execution (together with the positions of substitute execution) is under checking. External routes stand alongside this loop as usual, and those that are usable are used.
This does not require every pair to question each other. Execution questions no party by non-execution; it only executes; where it holds that execution itself crosses the line, it explains publicly, and the matter passes into review within the closed loop, a review not conducted by it (Paper 2, N11b). An objection raised by the executing party that triggers review adds one more questioning; it does not break the loop. As for the Law Series' sentence "The core closed loop (rule-making → execution → checking → rule-making) does not change" (Law Series P3 §5.2): this upstream sentence speaks of the flow of handling among the three functions; this section says that among the relations of questioning there is at least a loop, and does not equate the two kinds of arrow one for one.
For the reading of "cannot close" in the proposition, there is another upstream sentence to set beside it. The Conflict Treatise says that if a person acquires the power to declare in advance that some class of questioner is permanently without standing, "Once I have that power I have become my own court of last instance" (Conflict Treatise §15). What is spoken of there is the first person; this paper extends it to positions (extension; own undertaking): any link in the loop can make the decision for the case on questioning directed at itself, but whether this kind of questioning can be initiated thereafter is not for it to decide on its own. Otherwise it becomes its own final instance, and the loop breaks there.
Can the chain stop at internal subjects, at "the people"? Internal subjects are not positions. The channels they initiate send questions into the loop (Paper 7, N37); they do not sit on the loop as a top. A decision made in their name, a referendum for instance, takes effect through positions; it is itself a decision, and stands under the line and under questioning just the same: an amendment passed by a referendum, if it changes the line, is still line-crossing (§3). "The people as the maker of the constitution" is a borrowing-narrative (§2). The chain cannot stop there.
What the loop guarantees is that no link can permanently close the questioning of itself; it does not guarantee that a standoff among several links has a solution on the spot. Montesquieu, discussing the mutual restraint between the two parts of the legislative body and the executive power, said: "These three powers ought to form a state of rest or inaction. But as, by the necessary movement of things, they are constrained to move, they will be forced to move in concert." (my translation; Ces trois puissances devraient former un repos ou une inaction. Mais, comme par le mouvement nécessaire des choses elles sont contraintes d'aller, elles seront forcées d'aller de concert; bk. XI, ch. 6, the same source as in the previous section; tier A) This series gives no warrant that they will move in concert; a standoff is registered as a remainder, left to the channels of internal subjects and to later amendment; Paper 2, §2 takes it up (N8).
Back to the Inspectorate. On this section's reading, two matters are checked separately.
Article 17 changed the position that takes up complaints; the transfer and the appointment relation make independence something that must be checked. What must be checked is whether the Executive for Defense can on its own close the taking-up of complaints, rewrite the closing of cases, or control whether checking continues; being in the same office, and the appointment, do not by themselves complete this judgment, and the case file does not record these powers. If the party checked can close it on its own, independence does not hold. This change was made by the rule-making link. Who questions it? Following the loop, the checking link and internal subjects: under the UCSE's configuration, those affected who have exhausted other remedies can ask the Final Court to review this article (Charter, Article 43, fact); internal subjects can ask at elections, but clause 1 of Charter Amendment 14 has postponed precisely the elections. Were Article 17 to come to a vote while measures restricting assembly, speech or communication are in force, then by the sentence of §4 it alters the remit of a checking position, and is not put to a vote.
In the matter of the acting Inspector-General, what must be checked lies elsewhere. The rule-making link could not choose anyone; the transitional provision handed the choice to the Executive Council, and the Executive Council is among those against whom the Inspectorate takes up complaints. The Theory of Government has given the criterion: "if the party checked can abolish the position, replace its people or cut off its funding, independence is already gone" (Theory of Government Paper 4, §3). The designation of an acting officeholder and the zero public reports are both leads for checking; the appointment does not by itself prove that the Executive Council can replace or shut down this checking: "this appointment relation has to be checked, and it does not automatically prove capture" (Theory of Government Paper 4, §3). The position of acting Inspector-General holds the burden of explanation: it must answer how taking up, replying and correcting actually operate, and whether the party questioned can control them on its own.
From the spring of Year 10 to the summer of Year 11, the transfer, the acting appointment, the postponement of elections and the suspension of public hearings overlapped, each with its own start and end. Which channels were closed and which remained usable is checked by function, item by item, without first judging the loop broken on the strength of a shared office or an appointment. What N5 establishes is a requirement on the construct: as regards the class of check that must be taken up at this level, every position already established at this level is under another position that it cannot itself close; there is no top. It gives no warrant that this is achieved; where checking finds that it is not, the account falls, under Paper 7, N39, on the positions and persons that made the decisions.
The strong form is answered here. The strong form the upstream left to the Theory of the State here is that the multi-branch separation of powers is a structural necessity at the national scale: "The right to question at the national scale can only be realized through separation of powers" (Law Series P3 §7). The weak form is in How Is Institution Possible: "The separation of powers taken here is one implementation of that function, not the only one." (F6 §3.1b) The general rule of the Theory of Government is that the strong form is left to the Theory of the State, with the statement: "Leaving the strong form to the Theory of the State is a division of work; it does not mean that the Theory of the State has already obtained its proof." (the Theory of Government Prequel, §8) The multi-branch separation of powers is a place where it expressly takes the strong form: from the city level up, as a normative undertaking (Theory of Government G26).
This paper's answer to the strong form is confined to the class of corrections that must be realized by this state, and what has been proved and what is undertaken separately are kept apart. What has been proved is two things: this class of corrections requires implementation at this level that the position questioned cannot monopolize (§5, step two); taking the positions already established at this level as the set, every position must be checked by another independent position at this level, and at least a closed loop is formed (§6). What is undertaken separately is the following normative undertaking, made following Theory of Government G26: the three functions of rule-making, execution and checking are kept separate at this level, and checking at this level carries day-to-day checking judgment for this class of corrections. This undertaking includes at least one independent position of checking judgment at this level. The derivation above does not serve as its proof (§5, step three). So what this paper does is to turn part of that undertaking of the Theory of Government into derivation, and to undertake the rest separately, following the Theory of Government. External checking can reach valid decisions independently and stands alongside; what is usable is used; this paper does not presume that it is in fact weak. The difference from the weak form: the weak form says that an internal check is required, and that separation is only one implementation of it; this paper derives that, within this range, independent implementation at this level is required, not one optional implementation among others; checking judgment at this level and full separation of the three are an undertaking. The Law Series' "only" this paper takes as far as separation in function: the three are not held in one hand, and at least a closed loop is formed, with what is derived and what is undertaken kept apart as above. If "only" is read as one particular form of agencies, it is not taken, and the form, following the Theory of Government, is not fixed; by which configurations of agencies a closed loop separated in function is realized is left as a remainder. What is counted is functions and independent checking positions, not agencies; a minimum of three, and more are possible.
N5 The multi-branch separation of powers is taken in its strong form at the national level, confined to the class of corrections that must be realized by this state, namely those that require this state to do, or to stop, an act toward its internal subjects; part is derived, and the rest is undertaken separately, following the Theory of Government (own undertaking). First, step two of Theory of Government G26: as regards this class of corrections, the diagnosis is whether exit and external routes can carry the day-to-day check on this level; G27 is the conditional of this diagnosis. Second, the derivation: three things are derived from the definition: external routes exist, but they land in this state; exit can only be lateral; the guarantee is always there. External routes can review independently and reach valid decisions; if implementation is decided entirely by the position questioned, that position can block it on its own; so this class of corrections requires a taking-up at this level (implementation) that the position cannot monopolize. This does not exclude other external routes, does not claim that every check can be delivered only in this state, and does not presume that external routes are in fact weak: "each level's exit and external routes are read as they actually are, not presumed from its rank" (Theory of Government G27), and those that are usable are checked item by item and used (Paper 2, N9; Paper 5, N24); what is derived is a structural point: however strong they actually are, the step by which this class of corrections lands is in this state. This step of landing may not be in the hands of the party questioned (Conflict Treatise §15; extension; own undertaking); independence is checked by position and function, following F6 §3.2. Third, checking judgment at this level: undertaken separately. What the preceding step derives is implementation at this level; it does not derive that checking judgment must also be formed at this level. Keeping the three functions of rule-making, execution and checking separate at this level, with checking at this level carrying day-to-day checking judgment for this class of corrections, is a normative undertaking made following Theory of Government G26. This undertaking includes at least one independent position of checking judgment at this level; the derivation above does not serve as its proof. External checking stands alongside it, and what is usable is used. How the exit from or switching of an external arrangement is to be reviewed, and by whom, is a question of how cooperation between states is arranged; it is left as a remainder and handed to the Theory of Inter-State Cooperation. External questioning cannot be cancelled (own undertaking): this state may exit or switch an external arrangement; that is a change of structure, itself under questioning; it can close one channel, but it cannot cancel the questioning of it by other states, by the people of other states, and by those affected (Paper 5, N24, N27; Paper 7, N38). Fourth, the checker must be questioned too (Institution Theory Z37). The chain of questioning the checker cannot be handed entirely to higher or external checkers, nor can it rely on lengthening; it can only be folded back. Take as the set the finite functional positions already established at this level that carry the authorizing, performing, checking and implementing of the internal acts above, and consider only the checks that must be taken up at this level; every position must be checked by another independent position at this level (it can make the decision for the case on a matter directed at itself, but cannot decide on its own whether it can be questioned again thereafter), and that check cannot be handed to the outside for delivery: in a configuration where the three functions are separated, execution is checked by the decisions of checking, with landing by substitute execution; rule-making is checked by the review of checking; checking is checked by rule-making's amendment of rules and by impeachment; a position that holds part of each of two functions must have, for each part, a taking-up other than itself (§6). On this condition, finite positions must repeat, and at least a closed loop is formed (own undertaking); valid external routes stand alongside it. This does not require every pair to question each other: checking questions rule-making and execution, rule-making questions checking, and execution questions no party by non-execution; it only executes; where it holds that execution itself crosses the line, it explains publicly, and the matter passes into review within the closed loop, a review not conducted by it (Law Series P3 §5.3; Paper 2, N11b); no link decides on its own whether questioning directed at itself can be initiated thereafter (Conflict Treatise §15; extension; own undertaking). The form follows the Theory of Government: what is counted is functions and independent checking positions, not agencies. The "only" of Law Series P3's "can only be realized through separation of powers" this series takes as far as separation in function: the three are not held in one hand, and at least a closed loop is formed, with what is derived and what is undertaken separately kept apart as above; read as one particular form of agencies, it is not taken; by which configurations of agencies a closed loop separated in function is realized is left as a remainder.
§7 A Constitution Recognizes; It Does Not Grant
Article 4 of the Charter reads: "No organ of the UCSE may violate the following rights, and every organ shall secure their exercise," followed by a list that includes the person, speech, assembly, association, movement, property and litigation (UCSE Case 03). Are these rights given by the Charter, or were they there before the Charter?
What does the wording of the Charter say on this question, and what does it not say? Its subject is "organs"; to organs it says "may not violate" and "shall secure"; it does not say "the people enjoy the following rights, which this Charter grants." This is a fact. The conclusion of this section does not rest on this wording: even if Article 4 had been written as "granting," basic rights would not thereby come from it.
Power Theory says that "rights are not given by the constitution; the constitution is the recognition and protection of rights already constitutively in being"; "A population without a constitution still has rights; only their institutional form is absent." (Power Theory P1 §V) Rights Theory says that basic rights "come from one's own law, are ontologically prior" (Rights Theory P1 F2); Institution Theory says: "Basic rights are not newly made by the institution" (Institution Theory Z17). On this basis the Theory of Government wrote that a government is not the source of basic rights (the Theory of Government Prequel, §5). The national level is one of the levels of government, and this applies as it stands. In the time of the Cloud River Compact, the people on the two banks of the Cloud River had no common Charter, yet their person and their speech were not absent on that account; what the Charter did was to write them down as things organs may not violate and must secure, giving them an institutional form.
The other kind that the supreme rule embeds is recognition-based rights. Rights Theory says that voting "is the exemplar of constitutive recognition," and that it "falls on the recognition side" (Rights Theory P2 §II). Charter Amendment 7, extending voting eligibility to all adult UCSE citizens, was an act of embedding. A recognition-based right operates on three legs: institutional embedding, effective community recognition and real power backing (Theory of Government G19); embedding is one of the three, not the whole of its source. For the recognition-based rights embedded, the national level undertakes the guarantee: "a government ought to guarantee the recognition-based rights it has embedded for its internal subjects" (Theory of Government G17), and the ways are three: no de-righting, no empty shell, withdrawal or amendment made in public and open to questioning (Theory of Government G19). This is what §5 called "the guarantee is always there": the supreme rule always embeds some recognition-based rights, so the national level always undertakes this share of the guarantee.
The two kinds are kept apart, and are read together with §3. Basic rights the provisions did not give, and the provisions cannot take back: they are tied to the line, and are not within the authority of any amendment. Recognition-based rights the supreme rule has embedded, and they can be changed; but change must follow the way of withdrawal or amendment made in public and open to questioning: in public, open to questioning, and whether the horizontal constitutive sources remain is not for the national level to declare (Theory of Government G19); they cannot be quietly hollowed out.
N6 A constitution recognizes; it does not grant. "Rights are not given by the constitution; the constitution is the recognition and protection of rights already constitutively in being" (Power Theory P1 §V). Basic rights do not come from the state (the Theory of Government Prequel, §5); what the supreme rule embeds are the national level's recognition-based rights, and the guarantee follows Theory of Government Paper 3.
§8 With Sieyès
§2 pointed here. Where the supreme rule comes from, and who can change it: Emmanuel-Joseph Sieyès gave a deeply influential answer in chapter 5 of What Is the Third Estate? (Qu'est-ce que le Tiers état ?), "What Ought to Have Been Done: Principles in This Regard." What he sets out to answer in this chapter is who has the right to settle the form of a representative assembly.
He first separates two kinds of power: "In each of its parts, the constitution is the work not of constituted power but of constituent power. No kind of delegated power can change anything in the conditions of its delegation." (my translation; Dans chaque partie, la constitution n'est pas l'ouvrage du pouvoir constitué, mais du pouvoir constituant. Aucune sorte de pouvoir délégué ne peut rien changer aux conditions de sa délégation; ch. 5, per the edition of Edme Champion, Paris, Société de l'histoire de la Révolution française, 1888, p. 67; the University of Toronto scan held by the Internet Archive, questcequetiers00sieyuoft; tier A) The party that makes the constitution he calls the nation: "The nation exists before all things; it is the origin of all things. Its will is always legal; it is the law itself. Before it and above it there is only natural law." (my translation; La nation existe avant tout, elle est l'origine de tout. Sa volonté est toujours légale, elle est la loi elle-même. Avant elle et au-dessus d'elle il n'y a que le droit naturel; ibid., p. 67) "Not only is the nation not subject to a constitution, but it cannot be, and it ought not to be, which again amounts to saying that it is not." (my translation; Non seulement la nation n'est pas soumise à une constitution, mais elle ne peut pas l'être, mais elle ne doit pas l'être, ce qui équivaut encore à dire qu'elle ne l'est pas; ibid., p. 68) "In whatever manner a nation wills, it is enough that it wills; all forms are good, and its will is always the supreme law." (my translation; De quelque manière qu'une nation veuille, il suffit qu'elle veuille; toutes les formes sont bonnes, et sa volonté est toujours la loi suprême; ibid., p. 69)
Nor does he allow the nation to bind itself for good: "A nation can neither alienate nor forbid itself the right to will; and whatever its will may be, it cannot lose the right to change it as soon as its interest requires." (my translation; une nation ne peut ni aliéner, ni s'interdire le droit de vouloir; et quelle que soit sa volonté, elle ne peut pas perdre le droit de la changer dès que son intérêt l'exige; ibid., p. 69)
Who exercises constituent power? He says it is to be exercised by special representatives: "they are put in the place of the nation itself in settling the constitution"; and so long as it is known that they act under the people's special commission, "their common will shall be equivalent to that of the nation itself" (my translation; Ils sont mis à la place de la nation elle-même ayant à régler la constitution; leur volonté commune vaudra celle de la nation elle-même; ibid., p. 71).
The common question. Where the supreme rule comes from; who can change it; whether the party that changes it is bound by it. The two ask about the same thing.
How far the agreement goes. Two points.
- What they share is that the party entrusted may not rewrite on its own the conditions on which it is entrusted. This concurs with §§3 and 4. Clauses 2 and 3 of Charter Amendment 14 were submitted by the Executive Council, and change the conditions of the Executive Council's own confirmation and review; by Sieyès's sentence, the Executive Council cannot change these conditions itself, which concurs with the first half of N4. This series says one thing more: even when several positions change them together, while assembly, speech or communication is restricted, the arrangements for being questioned are not altered (N4); the conditions of being questioned cannot be withdrawn (N3).
- Not binding oneself for good. He says that the nation cannot forbid itself to will, and cannot lose the right to change its will, and he asks: "What is a contract with oneself?" (my translation; Qu'est-ce qu'un contrat avec soi-même ?; ibid., p. 69) This concurs with setting no eternity clauses (N3): "This article may not be amended" is this one vote answering on behalf of all later votes.
Where they part. Two points.
The first is the nation. Sieyès has the nation exist before the constitution, its will always legal, unbound by the constitution, all forms good. This series does not accept a constituent subject of this kind, prior to the constitution and unbound (N2). For this reading there is a point of entry in his own text: constituent power is to be exercised by special representatives, who are "put in the place of the nation itself," and whose common will "shall be equivalent to that of the nation itself." This is exactly the borrowing structure that Power Theory read out: a claimed bearing-position, and the group of people who actually call upon it (§2). The nation's will has always been spoken through people in certain positions; what this series requires is that those positions themselves be open to questioning.
The second is binding. Above the nation, Sieyès places only natural law. This series places the line beyond all provisions and all positions, and also requires that the conditions of being questioned not be withdrawn by amendment (N3). Both acknowledge something above the constituent will; on this point they concur. This series asks one thing more: who reads it. Its answer is that no party is allowed, in the name of "the nation's will," to become the last reader of the line: the scope of the line is handed to no position as final, nor to a bearing-position called the nation. How Sieyès answers this question the pages cited here do not say, and this paper does not answer on his behalf.
One matter of scope must also be stated. What Sieyès's chapter sets out to answer is who has the right to settle the form of a representative assembly; his saying that the nation is not bound by forms is used to open a door that old forms had shut. What this series asks is how, after constitution-making, each "last" remains open to questioning. The two questions overlap, but are not the same; here the parting is not written up as a refutation.
The strongest objection from the other side (drafted by this paper). You say that constitution-making does not produce a "we," and that the supreme rule has no higher author. But amendment must be done by someone. If it is not the nation that decides, who is it? The Executive Council, the two chambers, the courts: constituted positions. Without a nation prior to the constitution, amendment can only be constituted power changing itself, which is exactly what I call a delegated party changing the conditions of its delegation. In the end your closed loop either seizes up, so that no one can change anything, or some position in the loop quietly becomes the maker of the constitution.
Reply. Three points.
- What is shared goes only as far as this: the party entrusted cannot rewrite on its own the conditions on which it is entrusted. Sieyès further distinguishes ordinary delegated power from a special constituent commission, and even the same people must act under the latter authorization. This series has amendment made jointly by several constituted positions and by internal subjects, and accounts for its validity and its normative limits by N1, N3 and N4; this is not a matter of acquiring constituent authorization in his sense merely by having more participants. The two accounts of authorization differ here, and the difference cannot be removed by "non-monopoly" alone.
- No first person prior to the constitution is needed to account for a new constitution-making. A new constitution-making is one more event, from which the second layer is counted; the third layer is checked against the three necessary conditions, as things stand now (N1). Sieyès needs "its will is always legal" so that a new constitution can stand in law; this series keeps apart under what authority and procedure something is valid and on what grounds it may normatively be so, and lets no collective will settle the account of the third layer.
- As for "some position quietly becoming the maker of the constitution," that is precisely what this series guards against, and it is also its reason for not accepting a nation. Those who speak in the nation's name are always people in certain positions; bearing must be attributed to these positions and persons (Paper 7, N39). No link in the loop can decide on its own whether questioning directed at itself can be initiated thereafter (N5), and decisions made in the nation's name are questioned just the same. The possibility of seizing up this series does not deny: a standoff is registered as a remainder, left to the channels of internal subjects and to later amendment (Paper 2, N8).
Where This Paper Stops
This paper has proved only these things: the supreme rule has no higher author; consent at constitution-making has its work in formation in fact and in validity in procedure, but cannot supply the third layer in full, and constitution-making is an event, not a certificate of exemption from questioning; constitution-making does not produce a "we," "the people as the maker of the constitution" is a borrowing-narrative, and the supreme rule is addressed to internal subjects; the structure can be changed, the line cannot, and the conditions of being questioned cannot be withdrawn; no eternity clauses are set; an amendment that changes the line can be carried through procedurally and is line-crossing normatively; whether it crosses the line is decided by the court for the case, and whether it can be questioned again thereafter is not for the court to decide on its own; constitutional amendment is not monopolized by a single position and does not go by emergency procedure; while measures restricting assembly, speech or communication are in force, what alters the arrangements for being questioned, or takes emergency powers out of confirmation and renewal, is not put to a vote, and is recognized by function; on the separation of powers at the national level, confined to the class of corrections that must be realized by this state, two things have been derived: that, from the three things derived from the definition, this class of corrections requires implementation at this level that the position questioned cannot monopolize; and that, taking the positions already established at this level as the set, when every position has a checker at this level, at least a closed loop is formed; external routes stand alongside, and are not presumed to be in fact weak; external questioning cannot be cancelled, and this state's exiting or switching an external arrangement cannot cancel the questioning of it; a constitution recognizes, and does not grant.
Answering the strong form: the strong form of the multi-branch separation of powers that the upstream left to the Theory of the State is taken in this paper within the range of the class of corrections that must be realized by this state; part of that undertaking of the Theory of Government is turned into derivation (the two things above, §§5 and 6), and the rest is undertaken separately, following the Theory of Government: keeping the three functions separate at this level, with checking at this level carrying day-to-day checking judgment; this undertaking includes at least one independent position of checking judgment at this level; "only" is taken as far as separation in function: the three are not held in one hand, and at least a closed loop is formed, with what is derived and what is undertaken separately kept apart as above.
What it has not proved: the concrete constitutional form, how presidential, parliamentary and federal systems are to be designed, which belongs to specific theories of institutions; the full text of a constitution; by which configurations of agencies a closed loop separated in function is realized, which is left as a remainder; at least one independent position of checking judgment at this level, full separation of the three, and checking at this level carrying day-to-day checking judgment, which this paper has not derived, and which are undertaken separately, following the Theory of Government; how the exit from or switching of an external arrangement is to be reviewed, and by whom, which is handed to the Theory of Inter-State Cooperation; whether checks outside this class of corrections must also be taken up at this level, which this paper does not prove, external routes that can deliver being used and checked item by item; how treaty mechanisms above states bind states, and how far, which is left to the Theory of Inter-State Cooperation to come; how a standoff among several links of the closed loop is resolved, which Paper 2 takes up (N8); how much effect remains to the review threshold raised by clause 3 of Charter Amendment 14, which Paper 2, §1 takes up; clauses 1 and 3 of the unity clause, which Paper 3, §6 takes up.
The main line of the whole series is: Toward its internal subjects, a state holds to anti-colonization and strives for cultivation; outwardly, it does not colonize, and does not decide on behalf of the people of other states; each of its "lasts" remains open to questioning. This paper speaks of one place in it: the last rule lies under amendment and questioning at the third layer, and under the closed loop. Follow the checking link of the closed loop downward, and one reaches the last instance; there the decision can be final; the position is not final. Paper 2 writes it.
The Six Propositions of This Paper
N1 The supreme rule has no higher author. The full normative ground of the supreme rule does not come from a legislator above it, nor is it supplied in full by consent at the moment of constitution-making alone: "law comes neither from God nor from consent" (Law Series Supplement §7.3). Consent at constitution-making has its work in formation in fact and in validity in procedure; the three layers are checked separately: how it was formed in fact, under what authority and procedure it is valid, on what grounds it may normatively be so (Theory of Government G6). The third layer is checked against the three necessary conditions of G6. Constitution-making is an event, not a certificate of exemption from questioning (own undertaking).
N2 Constitution-making does not produce a "we." "Consensus does not automatically generate a third first-person legislator" (Conflict Treatise §24; extension; own undertaking); "the people as the maker of the constitution" is a borrowing-narrative, not a first person (the reading of Power Theory Paper 0, §IV.1, extended to constituent power; extension; own undertaking). The supreme rule is addressed to internal subjects, and does not form a "we" on their behalf.
N3 The structure can be changed; the line cannot, and the conditions of being questioned cannot be withdrawn. No eternity clauses: "A law that cannot be modified is a law that claims to have no remainder. Within the SAE framework, this is a structural lie." (Law Series P1 BL4) Ultra-high thresholds plus exit take the place of "unamendable" (F6 §6, fifth contribution). The line (the first ought) is not within the authority of any provision, and the conditions of being questioned are not withdrawn by amendment either: "structures can change, but the three conditions cannot be withdrawn" (Institution Theory Z37). An amendment that changes the line can be carried through procedurally and is line-crossing normatively; it does not become legitimate by having used the amendment procedure (own undertaking). This differs from an eternity clause: an eternity clause is a provision declaring provisions unamendable, and handing over to a final position whether they have been touched; this series' line is not written in by a provision, and it declares no text complete. Writing a floor such as human dignity into the supreme rule as a declaration is not of this kind. Whether an amendment has changed the line is decided by the court for the case; whether this kind of judgment can be questioned again thereafter is not for the court to decide on its own (Paper 2, N8, N10; Theory of Government G36): the scope of the line itself is handed to no position as final (own undertaking).
N4 Constitutional amendment is not monopolized by a single position and does not go by emergency procedure. Theory of Government G32 is extended to the supreme rule (extension; own undertaking). What is to be blocked is the emergency-decree power "sliding from exceptional use toward routine use, maintenance cost concentrating onto a single position" (Power Theory Paper 0, §V.1), and then "legal" being rewritten by way of a legal procedure. While emergency measures restricting assembly, speech or communication are in force, anything that alters, in any form (constitutional amendment, an enabling act, emergency legislation), the arrangements for being questioned (the dates of elections; who decides on renewal and termination; the thresholds for termination and review; the establishment and remit of checking positions), or that takes emergency powers out of their existing confirmation and renewal (setting no time limit, dispensing with confirmation, making them standing), is not put to a vote; it is recognized by function, not by name; what is already on the agenda is voted on after the restrictions are lifted. Confirmations and renewals made under existing rules, and emergency legislation that does not alter the arrangements for being questioned and carries its own time limit, fall outside this, and are checked under Paper 4, N20 to N22, and Theory of Government G30. Amendments that take back emergency powers, or that restore or strengthen the conditions of being questioned (such as adding a substitute confirming position not appointed by the party taking the measures), fall outside this. This step is stronger than Theory of Government G30: G30 allows being questioned beforehand to be moved to afterward, but amendment alters precisely the arrangements for being questioned, and being questioned afterward cannot make this up, so it is undertaken by this series (extension; own undertaking).
N5 The multi-branch separation of powers is taken in its strong form at the national level, confined to the class of corrections that must be realized by this state, namely those that require this state to do, or to stop, an act toward its internal subjects; part is derived, and the rest is undertaken separately, following the Theory of Government (own undertaking). First, step two of Theory of Government G26: as regards this class of corrections, the diagnosis is whether exit and external routes can carry the day-to-day check on this level; G27 is the conditional of this diagnosis. Second, the derivation: three things are derived from the definition: external routes exist, but they land in this state; exit can only be lateral; the guarantee is always there. External routes can review independently and reach valid decisions; if implementation is decided entirely by the position questioned, that position can block it on its own; so this class of corrections requires a taking-up at this level (implementation) that the position cannot monopolize. This does not exclude other external routes, does not claim that every check can be delivered only in this state, and does not presume that external routes are in fact weak: "each level's exit and external routes are read as they actually are, not presumed from its rank" (Theory of Government G27), and those that are usable are checked item by item and used (Paper 2, N9; Paper 5, N24); what is derived is a structural point: however strong they actually are, the step by which this class of corrections lands is in this state. This step of landing may not be in the hands of the party questioned (Conflict Treatise §15; extension; own undertaking); independence is checked by position and function, following F6 §3.2. Third, checking judgment at this level: undertaken separately. What the preceding step derives is implementation at this level; it does not derive that checking judgment must also be formed at this level. Keeping the three functions of rule-making, execution and checking separate at this level, with checking at this level carrying day-to-day checking judgment for this class of corrections, is a normative undertaking made following Theory of Government G26. This undertaking includes at least one independent position of checking judgment at this level; the derivation above does not serve as its proof. External checking stands alongside it, and what is usable is used. How the exit from or switching of an external arrangement is to be reviewed, and by whom, is a question of how cooperation between states is arranged; it is left as a remainder and handed to the Theory of Inter-State Cooperation. External questioning cannot be cancelled (own undertaking): this state may exit or switch an external arrangement; that is a change of structure, itself under questioning; it can close one channel, but it cannot cancel the questioning of it by other states, by the people of other states, and by those affected (Paper 5, N24, N27; Paper 7, N38). Fourth, the checker must be questioned too (Institution Theory Z37). The chain of questioning the checker cannot be handed entirely to higher or external checkers, nor can it rely on lengthening; it can only be folded back. Take as the set the finite functional positions already established at this level that carry the authorizing, performing, checking and implementing of the internal acts above, and consider only the checks that must be taken up at this level; every position must be checked by another independent position at this level (it can make the decision for the case on a matter directed at itself, but cannot decide on its own whether it can be questioned again thereafter), and that check cannot be handed to the outside for delivery: in a configuration where the three functions are separated, execution is checked by the decisions of checking, with landing by substitute execution; rule-making is checked by the review of checking; checking is checked by rule-making's amendment of rules and by impeachment; a position that holds part of each of two functions must have, for each part, a taking-up other than itself (§6). On this condition, finite positions must repeat, and at least a closed loop is formed (own undertaking); valid external routes stand alongside it. This does not require every pair to question each other: checking questions rule-making and execution, rule-making questions checking, and execution questions no party by non-execution; it only executes; where it holds that execution itself crosses the line, it explains publicly, and the matter passes into review within the closed loop, a review not conducted by it (Law Series P3 §5.3; Paper 2, N11b); no link decides on its own whether questioning directed at itself can be initiated thereafter (Conflict Treatise §15; extension; own undertaking). The form follows the Theory of Government: what is counted is functions and independent checking positions, not agencies. The "only" of Law Series P3's "can only be realized through separation of powers" this series takes as far as separation in function: the three are not held in one hand, and at least a closed loop is formed, with what is derived and what is undertaken separately kept apart as above; read as one particular form of agencies, it is not taken; by which configurations of agencies a closed loop separated in function is realized is left as a remainder.
N6 A constitution recognizes; it does not grant. "Rights are not given by the constitution; the constitution is the recognition and protection of rights already constitutively in being" (Power Theory P1 §V). Basic rights do not come from the state (the Theory of Government Prequel, §5); what the supreme rule embeds are the national level's recognition-based rights, and the guarantee follows Theory of Government Paper 3.
References
All SAE works below are cited by concept DOI, which always resolves to the latest version. The version records used for sentence-by-sentence verification are given in the accompanying audit table.
Qin, H. (2026). What a Government Is Not (SAE Theory of Government, Prequel) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23411670
Qin, H. (2026). Judgment (SAE Theory of Government, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23412130
Qin, H. (2026). Guarantee (SAE Theory of Government, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413112
Qin, H. (2026). Levels (SAE Theory of Government, Paper 4) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413192
Qin, H. (2026). Emergency (SAE Theory of Government, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413413
Qin, H. (2026). Questioning and Bearing (SAE Theory of Government, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23413511
Qin, H. (2026). Not Crossing the Line (SAE Institution Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270384
Qin, H. (2026). Guarding the Line (SAE Institution Theory, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23270766
Qin, H. (2026). Between One Another (SAE Institution Theory, Paper 5) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23271072
Qin, H. (2026). Open to Questioning (SAE Institution Theory, Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.23271324
Qin, H. (2026). How Is Institution Possible: From Inter-Ontological Remainder to Co-Constructive Framework (SAE Foundational Paper 6) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19328662
Qin, H. (2026). One's Law Meets One's Law: The Structural Origin of Law from 14DD Showdown (SAE Law Series, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19548237
Qin, H. (2026). National Law — Azeroth: A Negative Thought Experiment on the Structural Necessity of Law (SAE Law Series, Paper 3) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19548596
Qin, H. (2026). A Posterior Dialogue with Western Jurisprudence (SAE Law Series, Supplement) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19551526
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). SAE War Treatise: How Cultivation Is Not Swallowed by War Itself [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.19656617
Qin, H. (2026). The Power Machine and Its Shi (SAE Power Theory, Paper 0) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20707047
Qin, H. (2026). On the Origin of Power · Shi (SAE Power Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20370224
Qin, H. (2026). On the Source and Spectrum of Rights (SAE Rights Theory, Paper 1) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20743135
Qin, H. (2026). On the Thickness and Boundary of Basic Rights (SAE Rights Theory, Paper 2) [Preprint]. Zenodo. https://doi.org/10.5281/zenodo.20769253
Montesquieu. De l'esprit des lois. Ed. P. Janet, 2nd ed. Paris: Delagrave, 1892. Project Gutenberg, EBook #27573. https://www.gutenberg.org/ebooks/27573
Sieyès, E.-J. Qu'est-ce que le Tiers état ? Critical ed. by E. Champion. Paris: Société de l'histoire de la Révolution française, 1888. Internet Archive, questcequetiers00sieyuoft. https://archive.org/details/questcequetiers00sieyuoft
Marmor, A. The Pure Theory of Law. In The Stanford Encyclopedia of Philosophy. https://plato.stanford.edu/entries/lawphil-theory/
Interfaces
- Main upstream: the Theory of Government (the Prequel, §§3, 5 and 8; Paper 1, G6 and §7; Paper 3, G17, G19; Paper 4, G26, G27 and §§3, 5; Paper 5, G30, G32; Paper 6, G36, G36a); Institution Theory (Paper 2, Z9; Paper 3, Z17; Paper 5, Z31; Paper 6, Z34, Z35a, Z37).
- Also used: Law Series P1 BL4; Law Series P3 §4, §5, §5.1, §5.2, §5.3, §7; Law Series Supplement §7.3; F6 §3.1b, §3.2, §6 (fifth contribution), Appendix B, Article 28 (a working demonstration; used only as an example); Conflict Treatise §15, §24; Power Theory P1 §V; Power Theory Paper 0, §IV.1, §V.1; Rights Theory P1 F2; Rights Theory P2 §II; War Treatise (§5, as a destination; no sentence quoted).
- Interfacing: Kelsen (the basic norm; §1; tier C); Montesquieu (bk. XI, chs. 4 and 6; §§5 and 6; tier A); Sieyès (Qu'est-ce que le Tiers état ?, ch. 5; §8; tier A). Pointing back: Hart (Law Series Supplement §3; the Theory of Government Prequel, §4, and Theory of Government Paper 1, §7); Rousseau (Law Series Supplement §7); The Federalist, Nos. 47 and 51 (Theory of Government Paper 4, §3).
- This series: the Prequel, the Introduction, §§6, 7 and 8; Paper 2, N7, N8, N9, N10, N11a, N11b, and §§1 and 2; Paper 3, the Introduction, N13, N15, N16, and §6; Paper 4, N20 to N22, and §§3 and 5; Paper 5, the opening, N24, N27; Paper 7, N37, N38, N39.
- Handed to Paper 2: the checking link of the closed loop, which reaches the last instance.