The rule, exactly. Before an election each citizen chooses: to vote — or to take a payment and not vote in that election. The payment is a percentage of the median income, set by referendum; there is no fixed sum. One ballot is one vote, with no multiplier in the count; the vote is amplified only in that each ballot's share grows as others step out. The budget pays under law; a candidate never pays. Only a referendum of all citizens — simple majority, no quorum — introduces, changes or repeals the rule.

The protocol has been introduced nowhere and no pilot has been run: shares, turnout and outcome figures in the chapters are estimates, and the protocol promises nobody an election victory. If a chapter says otherwise, Exact Answers and the Charter are correct. For a candidate: ten questions and ten steps. For a citizen, a mayor, a finance officer, a donor, a journalist, a scholar, a lawyer: answers by role. Everything in force in one file: llms-full.txt.

40at. Achen and Bartels: Democracy for Realists — Five Arguments: Nobody Judges by Substance, Retrospection Is Blind, the Vote Is a Tribe, Referendums Are Decided by Cues, Do Not Burden the Citizen

Chapter: 06 — The Structural Hypocrisy of Criticism and the Arsenal of the Elites File: 06_040at · v1 · 16 October 2026 Source: the architect's assignment of 16 October 2026 ("Achen and Bartels, go"; "yes, write them into the repo, the rest of yours is right"). Christopher Achen and Larry Bartels, Democracy for Realists: Why Elections Do Not Produce Responsive Government (2016). The strongest empirical attack in the arsenal: it strikes not at the payment but at the premise that those who stay judge by substance. Three moves by the architect — §1, §2, §4 — and two tests of his thesis "simple and clear to everyone": on three people and on the four latest referendums (§4).

---

How to read. Achen and Bartels demolish the "folk theory of democracy" — that the voter has views and chooses by them. Their data on the average voter are right. The chapter shows that the protocol does not rest on the folk theory: it changes the composition of the hall by self-selection, and their averages are averages over a hall in which the burned stand in the "did not come" line.

0. What they claim

Most people have no coherent views (Converse's non-attitudes, 040ao §1) and adopt a party's position after choosing the party, not before. Retrospective voting is blind: incumbents are punished for shark attacks off New Jersey in 1916, for droughts and floods, and voters remember only the last six months of income growth before the election — "myopic retrospection". The vote expresses group identity — the party as a team — not policy. Referendums are no better: decided by cues and money. The prescription: do not burden the citizen; rely on parties, groups and institutions.

1. Nobody judges by substance — the educated are not more rational but more partisan; "the burned judge by substance" is the folk theory again

Argument. A direct blow at 1f §6b: no voter, the educated included, chooses on the merits; he chooses by identity and rationalises.

The assistant's answer. The protocol asks nobody to judge by substance. It changes the composition of the hall by self-selection: whoever stays is one for whom the vote is worth more than D — a revealed preference (036e §4, Samuelson), not a claim about his head; a filter by what a person gives up (036e §6, Mirrlees). The assistant proposed softening the wording of 1f §6b to "preferred the vote to the money".

The architect's move — the leaving is the proof. "Here again everything is already proven. They left politics knowing it matters, but they counted everything many times and saw the result of their visits, and precisely the fact that they know the price and left proves that they can count by substance. If they are apathetic, they will not return; if they are the burned who did not leave, no return is needed. The only question is how many will return. And that is simple: they are so angry and offended that even polls will give an exact result." Do not touch the wording of 1f §6b: the proposal to soften it is the same error — measuring the burned by the average of the hall. A person who cannot count by substance does not leave — he keeps going out of habit or fear (that is "the meek", 1f §2). Knowing the price and leaving on that knowledge is an operation Achen and Bartels did not find in the voter, because they looked among those who stayed in the hall, while those who performed it stand in their sample in the "did not come" line (1f §6b, 040aj §0). Their data do not refute the burned — they do not contain them. The states separate by themselves: the apathetic will not return, the burned who did not leave needs no return; what remains is to measure how many return.

The burned are the one audience with no reason to lie to a pollster — and how to ask them. The architect: "If you tell them: 'you do not go to vote, you did not take the money, you already know that everyone will leave with the money and your vote will be ×2, maybe ×3 — will you go and vote?' — will their answer most likely be accurate, or is there here too a socially rewarded answer 'like everyone', only everyone took the money?" A socially desirable answer exists where there is a norm "like everyone": turnout is overstated in surveys by 10–20 points because "I went" sounds respectable. This question breaks the norm in two places: "everyone took the money" makes "I will take it too" the normal answer — conformity is on the side of the money; "I will go with weight ×2" becomes the answer against the current, and only one who thinks so answers against the current. The burned one answers not by the norm but by calculation, and the angry and offended has no reason to lie — this is his chance to be heard. What remains is hypothetical bias: "I would go" is easier than going, and "yes" is flattering. So the direct question gives the upper bound of the return, and a second question — by price — the lower bound: "would you give $20 so that this becomes the rule?" (040ap §4). Flattery is not answered with money. The gap between "I would go" and "I will give $20" is the honest estimate of the return, and among the burned it is smaller than in any group, because for them both answers are one calculation. A side property: the question itself is campaigning. One who was asked "everyone will leave, your vote ×2, will you go?" has already calculated, and the calculation cannot be recalled, like the number (059g §4); the $10,000 poll (040ar §4) is at once a measurement and the first of the two touches the burned one needs (040aq §4). The pair of questions goes into the questionnaire of 048l §4.

Residual. None.

2. Retrospection is blind and myopic: six months of memory, punishment for sharks; "the thermostat with memory" is a myth

Argument. Their data: voters respond to the income of the last six months and to events beyond the government's control; the memory the protocol counts on (040aj item 4) does not exist in the electorate.

The assistant's answer. The six-month window is an average over a hall in which the majority has a three-day horizon (059i §3). The protocol removes them from the count; those who stay chose the vote over money now, that is, performed an act with a long horizon. Forecast: the retrospection window of those who stay is longer than six months. The sharks, by the way, were contested: Fowler and Hall (2018) found no effect.

The architect's move — a horizon of at least one cycle is proven, not forecast. "It is already proven that the horizon is at least one cycle. Not six months, not a year, not even two. And it is proven: they know they will not get the money for one more cycle for certain. Retrospection is blind on the hospital average, but remove the mass of the apathetic from the average — and you will be surprised how much people remember." The stayer's horizon is bounded below by the length of the cycle by construction: by staying, he knowingly gave up D until the next election (048m art. 5) and knows he will not get that money for the whole cycle — two years, four, but not six months. It is not a preference to be measured but a decision with a known price, already taken. Achen and Bartels' six-month window is an averaging artefact: an average over a hall where the mass with a three-day horizon drags the window to zero; the same as with the polled "centre" (040ao §1) — an average over different states describes none of them. The pilot is left to measure only how much longer than the cycle the stayers' horizon is.

Residual. None.

3. The vote expresses a tribe; paying for exit does not change identity; the most tribal will stay, and the protocol purifies tribalism

Argument. If people vote by identity, those who refuse the money are the hardest partisans, and the electorate becomes purer by tribe, not by mind.

Answer. Dalton's cognitive partisans (040aj §0) stay — yes; but so do the apartisans, the burned, while the ritual partisans — the core on salary (039 §88.9) — take B. The tribe is the apathetic's cheapest mode (040ak §7.1), and a tribe without money and fear behind it is thin: an identity that survived an offer of money is no longer a tribe but a conviction, and convictions are exactly what the contest of ideas needs (040ah §4).

Residual. The share of hard tribals among those who stay — a pilot measurement.

4. Referendums are decided by cues, not substance; the protocol will be decided by "bribery" against "free money" — and the architect's move: this is a different referendum

Argument. Their referendum data: low-information voters vote by elite cues and campaign money; a referendum on the protocol is no exception.

The assistant's answer. The cue here is the substance: $800 in your name (059l §3); cue-voting works for the protocol — the reverse Luntz (040an §1); the referendum is last and on a simple question (040ae).

The architect's move. "Here one must take the general view. The referendum in essence decides only one thing: where 1 % of the budget goes. Now it goes to obscure, murky projects, and it will go to the citizens. The rest changes nothing in the present situation: you may not vote for free, you may vote for free. And unlike ordinary referendums, where cues are needed, here everything is simple and clear to everyone. So this is a different referendum — from those taken as the model of an ordinary referendum on complex topics incomprehensible to the people." Achen and Bartels studied referendums on the complex; this one is about distribution, and the only one in which the beneficiary and the voter coincide completely; a cue is needed where the object is incomprehensible, and here the object is one's own cheque. Both rights stay (048m art. 16).

The test on three people. The architect's assignment: "take someone who got the vote yesterday, a woman of 40 and a grandfather of 70 — how clear is the referendum question to them?" The text of the line by 048m art. 15.

Who	Understood at first reading	Did not understand or feared	Closed by
got the vote yesterday, 18	"$800 if I do not go, or my vote weighs more if I go; every election" — the lottery with a tip for a ticket (040ap §4) is native to him	"is it a scam? will they take the right away?"	the phrase "both rights stay" — already in the line; the formula p × M he skips like small print
a woman of 40, a family, near the median	the family calculation in ten seconds: four adults — $3,200; "can I change my mind next time?" — yes	where does the money come from, will taxes rise, will the school lose; "will my husband know?"	"from the 1–2 % of the budget now going to murky projects" (059l §4); "taxes do not rise, only a referendum changes p" (art. 14); "the choice is secret" (art. 9.3). The formula is noise, "a month of the median wage" is clear
a grandfather of 70, voted all his life	"they will give money" — at once; he is a pension recipient	the most dangerous reading: "they pay me not to go — so they do not need my vote" — an offence, not a benefit; "will they cut my pension?"; "forever — we know your forever"	the first line "nothing is taken away: neither the vote nor the pension; D is added"; his vote is a duty, and he stays with ×2; do not say "forever" to him — a politician's word (040an §6, rule 3)

The question "take or vote with weight" is clear to all three without cues. Words are needed not in the question but in three answers "what does not change": rights, pension, taxes — three lines, the same for all three (entered in 1d). The youngest understands best: the eighteen-year-old reads the protocol like an app, the grandfather like an insult; the first line for the old is "nothing is taken away", not "$800". The formula on the ballot helped none of the three — confirmation of the Luntz audit (040an §6, rules 2 and 10); by the architect's decision the charter is not touched.

Comparison with the four latest referendums. The architect's assignment: "compare the difficulty of understanding the protocol with the latest big referendums".

Referendum	What had to be understood	Size	Is the voter the beneficiary?	Decided by	Result
Chile, September 2022	a whole new constitution: 388 articles, 178 pages	a book	no	"did not read — against"; turnout 86 % under compulsory voting	rejected, 62 % (the second draft — December 2023, 56 % against)
Australia, October 2023, "the Voice"	an amendment creating an advisory body of indigenous peoples with powers Parliament would define later	a paragraph of indeterminate content	no, 97 % of the population unaffected	the slogan "if you don't know, vote No"	rejected, 60 %
Ireland, March 2024	two amendments: "family" extended to "durable relationships", "care" reworded	two lines of legal language	no	confusion over "durable relationships"	rejected, 68 % and 74 % — a record "no"
Switzerland, March 2024, the 13th pension	one extra pension payment a year	one line	yes, for pensioners	the only dispute "who pays"	passed, 58 %
The protocol	take D or vote with weight; both rights stay	one line and one number	yes, for every adult, and the choice is his own	the only dispute "where do the 1–2 % of the budget come from"	—

Three conclusions. The comprehension load of the protocol is lower than that of the only successful one: the Swiss pension is clear to pensioners, the protocol to everyone, because everyone is a beneficiary and everyone chooses himself; the three failures share one thing — the voter received nothing personally and could not check the object, and Achen and Bartels are right exactly about such referendums. The Australian cue reverses: "if you don't know, vote No" killed the Voice because the not-knowing was honest; against a number in your own name it does not work — "if you don't know, take the $800" is also a "yes"; the only referendum where the cue "I don't know" works for "yes". The one open question is the Swiss one: who pays; the protocol has the answer ready in advance (059l §4): the counter-audit, 1–2 %, the holes by name. On the three people: the Chilean constitution none of them would have read, the Australian body none would have understood, the Irish "durable relationships" would have confused all three alike, the Swiss pension the grandfather would have understood, the protocol all three.

Residual. None.

5. The realists' prescription: do not burden the citizen, rely on parties and groups — the protocol burdens him with one more choice

Argument. Reforms that demand more decisions of the citizen (primaries, referendums) make democracy worse; the protocol is one more decision.

Answer. The protocol reduces the burden to one question a person can actually decide: do I need the money. It is the most realist reform possible: it stops pretending that the apathetic decides on policy and gives him a decision he can take (040ag item 2: choice B too is a decision). "The group that holds accountability" in their prescription is the burned with weight. Achen and Bartels are closer to allies: their book demolishes the folk theory invoked by the protocol's opponents with their "sacred vote" (059c §2, 040as §5).

Residual. None.

6. What they give the protocol

The empirical base for 1f §4 — the apathetic has no views — and for the mirror of 059i: a voter without views chooses by the face. Converse's non-attitudes as the explanation of the "centre by motion" (040ao §1). And two measurements for the pilot: the horizon of those who stay (how much longer than the cycle) and the pair of questions "would you go / will you give $20" for the burned (048l §4).

7. Key and Fiorina: the classic answer to the realists — "voters are not fools"

Added 16 October 2026 from the architect's list. V. O. Key, The Responsible Electorate (1966): "voters are not fools" — the electorate as a whole responds to the results of government sensibly, and the "switchers" between elections move by substance, not by mood. Morris Fiorina, Retrospective Voting in American National Elections (1981): the voter keeps a "running tally" of results and votes by it; retrospection is not myopia but a cheap and rational way to judge without knowing the programmes. This is the direct answer to Achen and Bartels from inside their own school, and for the protocol it matters in three ways. First: Key described those the protocol keeps — the switchers by substance, that is the cross-pressured (040ao §1), and showed that they decide the outcome, not the cores. Second: Fiorina's running tally is a horizon of no less than a cycle (§2), shown on data before Achen and Bartels; six months is an averaging artefact of a hall in which not everyone keeps the tally. Third: retrospection as a cheap cue (040ay §2) breaks where it is bought — the brought-in vote by the campaign's cue, not by the tally; the protocol removes the bought retrospection and keeps Fiorina's. Key and Fiorina have no blow against the protocol: they are allies, and the burned of 1f §6b are their "responsible electorate" driven to leaving. What remains: Key wrote on 1936–1960, when the hall was not yet bought on today's scale; his "not fools" is about the hall before bringing-in.

Summary

Argument	What is true	Answer	What remains
nobody judges by substance	for the hall on average	the leaving is the proof of counting by substance; their sample does not contain the burned; the burned are the one audience with no reason to lie to a pollster — ask with the pair "would you go / will you give $20" (the architect's moves)	—
retrospection is blind	on the hall average	the stayer's horizon ≥ a cycle by construction — proven, not forecast; six months is an averaging artefact (the architect's move)	how much longer than the cycle — the pilot
the vote is a tribe	for ritual partisans	they take B; an identity that survived money is a conviction	the share of hard tribals — the pilot
referendums are decided by cues	for complex referendums	this is a different referendum: where 1 % of the budget goes, beneficiary = voter; clear at 18, 40 and 70; simpler than the only successful one of the four latest (the architect's move)	—
do not burden the citizen	yes	the protocol is one question the citizen can decide; the realists are closer to allies	—

Weak point

A reconstruction from the book; their data are the US and a few democracies, not the whole world. The sharks of 1916 are contested (Fowler and Hall, 2018) — cite with a caveat. The test on three people is reasoning, not a focus group; the comparison of referendums is by results and wordings, without a measurement of voters' comprehension; the figures for Chile, Australia, Ireland and Switzerland are from standard summaries. The share of tribals among those who stay is open until the pilot.

---

Related: 1f §2, §4, §6b · 040aj §0, item 4 · 040ao §1 · 040ak §7.1 · 040an §1, §6 · 040ap §4 · 040ar §4 · 040as §5 · 040ae · 036e §4, §6 · 059i §3 · 059l §3, §4 · 001d · 048l §4 · 048m arts. 5, 9.3, 14, 15, 16