C09-01 Check first open §OpeningIn a laboratory game, volunteers appointed to lead one to three followers could take money from them, enriching themselves while the group as a whole lost, and their testosterone was measured (figure moved to the data page at the edit pass of 2026-10-23, prose now says "a small team" and "the researchers")
- To check against
- CA-02 (Bendahan, Zehnder, Pralong and Antonakis 2015, Leadership Quarterly, "Leader corruption depends on power and testosterone", 162 appointed leaders, one to three followers, baseline endogenous testosterone). The desk, the cubicle and the single volunteer are staging of the design as CA-02 describes it, not a recorded session. Single laboratory study, place not given
- The verdict
- OPEN
C09-02 Check first open §OpeningThose with more power and fewer limits chose the corrupt option about six times in ten, those with less about four times in ten, and the difference was sharpest with high testosterone
- To check against
- CA-02 (predicted probability 0.61 against 0.39, "high power and low institutional constraint" against "lower power", effect "highly moderated" by testosterone). Rounded. CA-02 mixes power and constraint in the comparison, check the paper for which cells 0.61 and 0.39 belong to
- The verdict
- OPEN
C09-03 Check first open §OpeningA more famous biscuit experiment usually opens such chapters and is not in the library this book rests on
- To check against
- No research document describes it: CA-02 cites Keltner, Gruenfeld and Anderson 2003 only for the approach and inhibition theory, and a search of every report for cookie and biscuit finds nothing. The Preface's P-40 was reworded on 2026-10-23 to the money game (C09-01, C09-02) and no longer promises it
- The verdict
- OPEN
C09-07 Check first open §Four in every hundredIn the population at large, a few in every hundred at most, perhaps nearer one, on thin evidence
- To check against
- CA-01 (1 to 4.5 per cent by threshold, sourced to a vendor page, psychological-evaluations.com, its note 4, while its own summary says "approximately 1%" and Babiak's rate is "approximately four times higher than the general population baseline"). CA-02 says "less than 1%". This chapter owns the row that the Preface (P-14) and chapter 1 (C01-17) cite
- The verdict
- OPEN
C09-09 Check first open §Four in every hundredAbout four in every hundred of them were at or above the clinical threshold, several times the population rate (figure moved to the data page at the edit pass of 2026-10-23, prose now says "far above the rate" in place of "several times")
- To check against
- CA-01 (3.9 per cent at PCL-R 30 or above, closest to the paper, "approximately four times" the baseline). CA-02 gives "between 5.75% and 13.5%" for the same paper via Semantic Scholar. CA-06 gives about 3.5 per cent from Snakes in Suits (Babiak and Hare 2006) via Goodreads. "Several times" because the baseline itself is contested (C09-07)
- The verdict
- OPEN
C09-12 Check first open §Four in every hundredPooled studies: narcissism helps people get picked modestly and does almost nothing for how well they lead, middling levels best
- To check against
- CA-01 (Grijalva et al. 2015, Personnel Psychology 68(1):1 to 47, emergence rho 0.16, k 18, N 3,131, effectiveness rho 0.03, k 32, N 5,593, curvilinear). The printed interval .08 to .15 does not contain 0.16
- The verdict
- OPEN
C09-17 Check first open §Four in every hundredSmall groups solving a survival puzzle: both dominant and prestigious members steered the group, the prestigious were liked, the dominant disliked
- To check against
- PS-01 (Cheng et al. 2013, NASA Moon Survival task, with eye tracking). Statistics stripped in extraction, not checked
- The verdict
- OPEN
C09-20 Check first open §Four in every hundredPeople made to feel powerful were quicker to act, at a card table and elsewhere, and less held back by others
- To check against
- CA-02 (Galinsky, Gruenfeld and Magee 2003, blackjack, about 128 per study). Priming literature with replication concerns
- The verdict
- OPEN
C09-21 Check first open §Four in every hundredTestosterone pushes towards dominance mainly when cortisol is low, and high office keeps cortisol low
- To check against
- CA-02 (Mehta and Josephs 2010, Hormones and Behavior, N 82, dual-hormone hypothesis). That office lowers cortisol is CA-02's inference
- The verdict
- OPEN
C09-25 Check first open §The quiet door of moneyAbout two in five retiring members of Congress have gone straight into lobbying (figure moved to the data page at the edit pass of 2026-10-23, prose now says "a large share", the 42 per cent on the data page only)
- To check against
- CA-11 ("nearly 42%", Lessig 2011 Republic, Lost, via Wikipedia and an IRPP book review). Secondary source chain
- The verdict
- OPEN
C09-33 Check first open §The quiet door of moneyBy 1220 a handful of families had swallowed the independent chieftaincies
- To check against
- PS-01 (39 goðorð into five family dynasties by 1220, Sturlungs, Haukdælir, Oddaverjar, Ásbirningar, Vatnsfirðingar). The count is contested (three to six by document), chapter 5 owns it at C05-61 and says "a handful", which the prose now follows
- The verdict
- OPEN
C09-37 Check first open §Who keeps the bookIn June 1381 rebels burned court rolls at St Albans and Bury St Edmunds, and a chronicler wrote that they meant to destroy the records of their unfree status
- To check against
- ME-14 (Peasants' Revolt, Thomas Walsingham, via Cambridge chapters and a Substack essay). Secondary chain
- The verdict
- OPEN
C09-40 Check first open §Who keeps the bookBetween 1625 and 1627 a palace eunuch took the censorate over and turned it on reforming officials, hundreds driven out, tortured or killed
- To check against
- CA-12 (Wei Zhongxian and the Donglin, via Britannica and a dokumen.pub copy of a Donglin study). CA-12 also credits the secret police, so how far the censorate itself was the instrument should be checked
- The verdict
- OPEN
C09-41 Check first open §Who keeps the bookIn 1766 a Swedish faction, with a country priest from Ostrobothnia, passed a law giving public access to official records
- To check against
- CA-12 (Freedom of the Press Act, the Caps against the Hats, Anders Chydenius, via Wikipedia, chydenius.net and Facebook). Weak sources
- The verdict
- OPEN
C09-43 Check first open §Who keeps the bookIn 1297 William de Boochose carved extra notches, the halves did not match at the Exchequer, and he was jailed for a year and a day
- To check against
- ME-14 (Flegg, Numbers Through the Ages, via Scribd). Single weak source
- The verdict
- OPEN
C09-46 Check first open §Who keeps the bookInca communities kept two to four knot-keepers who cross-checked, and later villages unrolled their cords in open assembly (figure moved to the data page at the edit pass of 2026-10-23, prose now says "more than one knot-keeper")
- To check against
- ME-14 (Hyland 2022, Latin American Research Review, Salomon 2004, The Cord Keepers, Rapaz and Casta). Two to four is single source
- The verdict
- OPEN
C09-51 Check first open §Who keeps the bookA hoard of ostraka with Themistocles' name was written in a few hands, prepared in advance
- To check against
- PS-04 (via Wikipedia, Ostracism). Secondary
- The verdict
- OPEN
C09-52 Check first open §Who keeps the bookEuropean witch trials killed tens of thousands, mostly women, starting from neighbours' talk about illness or failed crops, made murderous by courts and torture (figure moved to the data page at the edit pass of 2026-10-23, prose now says "most of whose dead were women", the toll on the data page)
- To check against
- PS-04 (1450 to 1750, about 90,000 prosecutions, 45,000 to 50,000 executions, 75 to 85 per cent women, citing Levack and Behringer through a secondary chain)
- The verdict
- OPEN
C09-57 Check first open §The strongest case againstA much quoted finding on power and hypocrisy has among its authors a psychologist later found to have invented data
- To check against
- CA-02 relies on Lammers, Stapel and Galinsky 2010 without comment. That Diederik Stapel fabricated data is general knowledge, not in the documents. CA-02 also cites its hypocrisy replications (d 1.24) to "Power Increases Anchoring Effects", a citation mismatch
- The verdict
- OPEN
Whether power corrupts or the corrupt seek power matters because each asks for its own defence: limits on tenure and checks if power changes people, a narrower door if the corrupt select themselves
- To check against
- CA-02 (its four quadrants: rotation and audit for transformation, screening and lot for selection)
- The verdict
- OPEN
C09-05 open §Four in every hundredNarcissism is the need to be admired, Machiavellianism the habit of treating people as instruments, psychopathy the absence of conscience
- To check against
- Cites the Preface's row P-15 (CA-01)
- The verdict
- OPEN
C09-06 open §Four in every hundredPsychologists bundled the three traits as the dark triad in 2002
- To check against
- CA-01 (Paulhus and Williams 2002, Journal of Research in Personality)
- The verdict
- OPEN
C09-08 open §Four in every hundredPaul Babiak, Craig Neumann and Robert Hare assessed about two hundred managers picked by their companies for executive development, using Hare's checklist rather than self report (figure moved to the data page at the edit pass of 2026-10-23, prose now says "a group of managers", N 203 on the data page)
- To check against
- CA-01 (Babiak, Neumann and Hare 2010, Behavioral Sciences and the Law 28(2):174 to 193, N 203 high-potential executives and managers selected for management development programmes, 77.8 per cent male, mean age 45.8, PCL-R). Hare's credential as author of the PCL-R from CA-01
- The verdict
- OPEN
C09-10 open §Four in every hundredChecklist scores went with high in-house marks for charisma and presentation and low marks for responsibility and teamwork
- To check against
- CA-01 (Babiak et al. 2010)
- The verdict
- OPEN
C09-11 open §Four in every hundredThe claim that one chief executive in five is a psychopath came in 2016 from a paper by two Bond University researchers, retracted in 2018 after the senior author was found to have copied a student's dissertation (figure moved to the data page at the edit pass of 2026-10-23, prose now says "researchers at Bond University")
- To check against
- CA-01 (Brooks and Fritzon 2016, "21 percent", retraction reported by Retraction Watch, 12 January 2018). The figure is fallen
- The verdict
- OPEN
C09-13 open §Four in every hundredA later pooling of the triad across tens of thousands of people: small help with emergence, more for men than women, slight harm to performance (figure moved to the data page at the edit pass of 2026-10-23, prose now says "a later and much larger pooling")
- To check against
- CA-01 (Landay, Harms and Credé 2019, Journal of Applied Psychology 104(1):183 to 196, k 46, N 32,680, rho 0.07, men 0.10, women 0.04, effectiveness minus 0.03)
- The verdict
- OPEN
C09-14 open §Four in every hundredExperts rating election candidates worldwide scored autocrats higher on psychopathy and narcissism
- To check against
- CA-02 (Nai and Toros 2020, Political Research Exchange, 124 candidates, 57 elections, 1,030 experts, d 0.60 and 0.59)
- The verdict
- OPEN
C09-15 open §Four in every hundredNarcissists claim authority early, talk longer and sound certain, and observers mistake this for competence. An interview measures the ability to perform competence
- To check against
- CA-01 (halo effect and implicit leadership theories, and its own sentence "Interviews do not measure competence; they measure the ability to perform competence")
- The verdict
- OPEN
C09-16 open §Four in every hundredDominance is deference paid to someone feared, prestige is deference freely given to someone worth learning from
- To check against
- Cites chapter 1, C01-21 (PS-01, Henrich and Gil-White 2001)
- The verdict
- OPEN
C09-18 open §Four in every hundredThe schemer is the hardest to detect, passing as generous and self-sacrificing and hoarding information
- To check against
- PS-01 (its table: Machiavellianism, vulnerability to detection "Low", "Masquerades as prestigious and self-sacrificing", "Information hoarding")
- The verdict
- OPEN
C09-19 open §Four in every hundredDacher Keltner, a psychologist, argued with two colleagues in 2003 that power activates reward seeking and suppresses inhibition (figure moved to the data page at the edit pass of 2026-10-23, prose now says "with colleagues")
- To check against
- CA-02 (Keltner, Gruenfeld and Anderson 2003, Psychological Review, approach and inhibition theory). The first name Dacher is general knowledge and the chapter's plan, not in the documents, which give "Keltner, D."
- The verdict
- OPEN
C09-22 open §Four in every hundredBoth are true. Rotation and audit answer the first, lot and community refusal the second
- To check against
- CA-02 (conclusion "the corrupt do seek power, and power provides the biological and psychological safety required", four quadrants)
- The verdict
- OPEN
C09-23 open §The quiet door of moneyAn economist wrote in 1971 that regulation "is acquired by the industry and is designed and operated primarily for its benefit"
- To check against
- CA-11 (Stigler 1971, The Theory of Economic Regulation, quoted via a Chicago Unbound retrospective)
- The verdict
- OPEN
C09-24 open §The quiet door of moneyLater critics showed the capture theory too strong
- To check against
- CA-11 (Sunstein 2021, ProMarket: Clean Air Act and OSHA rules opposed by industry)
- The verdict
- OPEN
C09-26 open §The quiet door of moneyLocal government knows more and is easier to capture, because the poor depend on the local rich for work, credit and land
- To check against
- CA-11 (Bardhan and Mookherjee 1999, 2003, 2006, via World Bank papers)
- The verdict
- OPEN
C09-27 open §The quiet door of moneyA review of World Bank community programmes worth billions found participation often cosmetic and elite capture common (figure moved to the data page at the edit pass of 2026-10-23, prose now says "had spent so heavily")
- To check against
- CA-11 (Mansuri and Rao 2013, Localizing Development)
- The verdict
- OPEN
C09-28 open §The quiet door of moneyIndonesia: nearly six households in ten attended full community meetings, fewer than one in six where local leaders ran them
- To check against
- CA-11 (NBER working paper w18798, "Does Elite Capture Matter? Local Elites and Targeted Welfare", 59 per cent against 15 per cent). Authors not named in the documents. Rounded
- The verdict
- OPEN
C09-29 open §The quiet door of moneyFormal office holders and relatives were a few points more likely to receive cash, health cover and subsidised rice than equally poor neighbours, informal elites slightly less likely (figure moved to the data page at the edit pass of 2026-10-23, prose now says "somewhat more likely" and "not large differences")
- To check against
- CA-11 (same paper, about 5, 8 and 3 percentage points for BLT, Jamkesmas and Raskin, informal elites 6 to 7 points less likely)
- The verdict
- OPEN
C09-30 open §The quiet door of moneyThe patient chair who schedules decisive meetings at harvest and fills the quorum with tenants, debtors and cousins
- To check against
- CA-11 (its capture test for decentralised village governance, precedent Panchayati Raj before 1994). A design scenario, not an event
- The verdict
- OPEN
C09-31 open §The quiet door of moneyAthens made its richest pay publicly for warships and festivals, any man could challenge a richer neighbour to take the burden or swap estates, a speech from about 330 BC survives, and the rich hid "invisible property"
- To check against
- CA-11 (liturgy, trierarchy, antidosis, Demosthenes, Against Phaenippus, about 330 BC, aphanes ousia, via The Economy of Classical Athens)
- The verdict
- OPEN
C09-32 open §The quiet door of moneyIcelandic chieftains who owned churches kept half the tithe of 1096 without accounts (figure moved to the data page at the edit pass of 2026-10-23, prose now says "their own share", the half stays in chapter 5 at C05-60)
- To check against
- Cites chapter 5, C05-60 (NO-07), and PS-01 ("without fiscal accounting")
- The verdict
- OPEN
C09-34 open §The quiet door of moneyMoney's capture is personal in village and district and runs through campaign money and rule drafting from the town up
- To check against
- CA-11 (scale ledgers for decentralisation and for lobbying)
- The verdict
- OPEN
C09-35 open §Who keeps the bookJames C. Scott's legibility: states read people through surnames, standard measures and cadastral maps
- To check against
- CA-12, ME-14 (Scott 1998, Seeing Like a State). Scott introduced in chapter 3
- The verdict
- OPEN
C09-36 open §Who keeps the bookSoviet collectivisation and Tanzanian villagisation in 1973 and 1974 broke local knowledge
- To check against
- CA-12 (via Wikipedia and reviews of Scott)
- The verdict
- OPEN
C09-38 open §Who keeps the bookTenants used the same rolls to prove rents fixed time out of mind
- To check against
- ME-14 (citing Dyer 1994, Everyday Life in Medieval England)
- The verdict
- OPEN
C09-39 open §Who keeps the bookImperial China kept a censorate for more than a millennium to watch officials for corruption (figure moved to the data page at the edit pass of 2026-10-23, prose now says "through dynasty after dynasty")
- To check against
- CA-12 (yushitai, duchayuan, via Medium and chinaknowledge.de)
- The verdict
- OPEN
C09-42 open §Who keeps the bookA split tally was a hazel stick notched and split through the notches, each party keeping a half, the halves matched at settlement
- To check against
- CA-12, ME-14 (Fitznigel, Dialogus de Scaccario, about 1179, stock and foil)
- The verdict
- OPEN
C09-44 open §Who keeps the bookThe treasury used tallies until 1826, and in 1834 the burning of old tallies set fire to the Houses of Parliament
- To check against
- CA-12, ME-14
- The verdict
- OPEN
C09-45 open §Who keeps the bookA chirograph was written twice on one sheet with CHIROGRAPHUM across the gap and cut apart along a wavy line through the word
- To check against
- ME-14
- The verdict
- OPEN
C09-47 open §Who keeps the bookEvery record should be readable by the governed: the right to have one's entry read aloud and a yearly reading of the book (figure moved to the data page at the edit pass of 2026-10-23, prose now says "at its spring assembly", ME-14's wording)
- To check against
- ME-14 (village design: every adult may have an entry read aloud, cadaster read at the spring assembly), CA-12 (localised freedom of information for a village). Design
- The verdict
- OPEN
C09-48 open §Who keeps the bookGroups able to gossip and exclude cooperated far more, and knowing one could be gossiped about raised cooperation
- To check against
- PS-04 (Feinberg, Willer and Schultz 2014, Psychological Science). Statistics stripped in extraction
- The verdict
- OPEN
C09-49 open §Who keeps the bookAn anthropologist listening to Kalahari firelight talk found about a third of criticism driven by envy
- To check against
- PS-04 (Wiessner 2014, 44 years of fieldwork among the Ju/'hoansi)
- The verdict
- OPEN
C09-50 open §Who keeps the bookGossip stops reaching past about 150 people and a village splits into circles
- To check against
- PS-04 (Dunbar, organic gossip fails above about 150), CA-11 (Hutterite fission at 150 to 230 adults). The 150 is a shared number owned by chapter 11 at C11-41 (the Preface carries it at P-20)
- The verdict
- OPEN
C09-53 open §Who keeps the bookGossip works in household and hamlet, strains in the village, and with a court behind it is most dangerous at village and town scale
- To check against
- PS-04 (scale ledgers for organic gossip and mob shaming, "catastrophically effective" from 150 to 50,000)
- The verdict
- OPEN
C09-54 open §Who keeps the bookMoney and records are the two quiet doors
- To check against
- Inferred from CA-11, CA-12 and ME-14
- The verdict
- OPEN
C09-55 open §The strongest case againstPower posing rested on a small study and failed in a sample several times larger (figure moved to the data page at the edit pass of 2026-10-23, prose now says "a much larger sample")
- To check against
- CA-02 (Carney, Cuddy and Yap 2010, N 42, Ranehill et al. 2015, Psychological Science, N 200, 98 women and 102 men)
- The verdict
- OPEN
C09-56 open §The strongest case againstThe Stanford Prison Experiment was taken apart in 2019
- To check against
- CA-02 (Le Texier 2019, American Psychologist)
- The verdict
- OPEN
C09-58 open §The strongest case againstCritics argue two of the triad's parts cannot be told apart by questionnaire and the triad may be low agreeableness and conscientiousness renamed (figure moved to the data page at the edit pass of 2026-10-23, prose now says "the schemer and the one without a conscience", which CA-01 gives as Machiavellianism and psychopathy, Miller, Lynam and colleagues 2018)
- To check against
- CA-01 (Miller and Lynam, Vize et al. 2018, Miller et al. 2019)
- The verdict
- OPEN
C09-59 open §The strongest case againstMark Van Vugt, an evolutionary psychologist: in crisis a hard, unsympathetic person in charge may save the group, as Churchill in 1940, and a village screening out the aggressive may be conquered
- To check against
- CA-02
- The verdict
- OPEN
C09-60 open §The strongest case againstGuarding every door against capture can leave a society unable to build or to act in a crisis
- To check against
- CA-02 (Fukuyama, vetocracy)
- The verdict
- OPEN
C09-61 open §The strongest case againstIn parts of Indonesia elites seized development money and shared it fairly, for their standing
- To check against
- CA-11 (Dasgupta and Beard 2007, pro-poor elite capture, Fritzen 2007)
- The verdict
- OPEN
C09-62 open §The strongest case againstAn agricultural historian answers Scott that farming before the state was ravaged by famine and Tanzanian food production rose after villagisation
- To check against
- CA-12 (Tauger 1999, H-Net review, about ten per cent by 1975, export surplus by 1978 to 1979)
- The verdict
- OPEN
A village of 300 and a district of 3,000 need a chest, iron-gall ink, hazel, a calendar and three literate people
- To check against
- ME-14 (implementability: at least three literate and numerate people, iron-gall ink, hazel tallies, shared calendar), CA-11 (village of 300, district of 3,000). Brief rung sizes
- The verdict
- OPEN
Prefer lot from a list the assembly has not refused to any campaign
- To check against
- CA-02 (lot removes campaigning, community elder veto as screening), CA-01 (sortition against charm). Design
- The verdict
- OPEN
Ask those who worked under a candidate, since the confident talker is mistaken for the able by those above
- To check against
- CA-01 (Babiak et al. 2010 in-house ratings, halo). Design inferred
- The verdict
- OPEN
Ask whether influence survives absence and whether the candidate shares or keeps information
- To check against
- Cites chapter 1, C01-54 (PS-01, Maner and Mead 2010), and PS-01 (information hoarding). Design
- The verdict
- OPEN
No office holder handles the common purse alone, the Three Keys as in chapter 12
- To check against
- Chapter 12 defines the Three Keys at C12-11 (ME-14 Canon 70 of 1603, PS-01 triple-lock chest). Design
- The verdict
- OPEN
Post the accounts on the meeting-place wall in the village's own words
- To check against
- CA-11 (Ostrom's cheap monitoring, "physically posting budgets and allocation lists on village walls in the local dialect"). Design
- The verdict
- OPEN
Relief decided by the whole village, at a time the poorest can come and never at harvest
- To check against
- CA-11 (NBER w18798 attendance, capture test at harvest). Design
- The verdict
- OPEN
Richest households carry the heaviest public burdens, and any household may challenge another claiming to be poorer
- To check against
- CA-11 (village liturgy, antidosis). Design
- The verdict
- OPEN
An outside check at a time nobody chose belongs to the district, with the evidence in chapter 12
- To check against
- CA-11 (randomised external audits backed by an authority able to penalise). Chapter 12 owns the audit rows at C12-76 (MO-09, Olken)
- The verdict
- OPEN
Records in iron-gall ink, read aloud yearly, two copies with the district holding one delivered each spring, every adult may have their entry read in daylight before the keyholders, debts on split tallies (figure moved to the data page at the edit pass of 2026-10-23, prose now says "at the spring assembly" and "a twin in another place")
- To check against
- ME-14 (four village registers, iron-gall ink, Canon 70's yearly transcripts as precedent, district redundant register, reading rights)
- The verdict
- OPEN
Talk may start but not finish an inquiry: accused present, first-hand over hearsay, no shaming punishments
- To check against
- PS-04 (first-hand tracking resisted liars in simulations, Nussbaum 2004 and Whitman 1998 against shame punishments). Design
- The verdict
- OPEN
The rumour campaign: one person telling different stories to different circles of a village too big to compare notes
- To check against
- PS-04 (first failure for a village of 300)
- The verdict
- OPEN
The district holds second copies, writes water and boundary agreements as chirographs, and can overrule a captured village
- To check against
- ME-14 (district design: chirograph treaties, redundant register, appellate court). Design
- The verdict
- OPEN
The first failure is the treasurer nobody checks: altered boundaries, grown debts, fees for writing, stopped by weekly reading before the other keyholders and loss of office and land (figure moved to the data page at the edit pass of 2026-10-23, prose now says "the reading of every new entry", ME-14's weekly cadence on the data page)
- To check against
- ME-14 (first failure, the custody bottleneck of the clerk, weekly public recitation, removal and forfeiture of the land allotment), CA-12 (the clerk bribed or bullied to alter the register)
- The verdict
- OPEN