Sources

Every figure and quotation on this site traces to the materials below. Statistics were checked against the originals in August 2026; where a number has since been revised, the revision is stated rather than the older figure quietly kept. A fourth section lists the works that cut against this argument, because a bibliography of friendly witnesses only is itself a letterhead — and an argument you cannot audit is nothing else.

Documents

Diogenes Laertius, Lives of Eminent Philosophers, Book VI (3rd c. CE). The life of Diogenes of Sinope, including the reply “I am a citizen of the world” (kosmopolitês), VI.63.

Michel de Montaigne, Essays (1580–1592). The invention of the essay as the anti-creed; the motto Que sais-je? Via Project Gutenberg (Cotton translation).

Baruch Spinoza, Letter to J. Louis Fabritius (1673). Declining the Heidelberg chair: “I do not know the limits, within which the freedom of my philosophical teaching would be confined…” Elwes translation, via the Internet Sacred Text Archive.

Voltaire, “Sect,” Philosophical Dictionary. “Every sect, in whatever sphere, is the rallying-point of doubt and error”; the schools’ labels — “Scotist, Thomist, Realist, Nominalist, Papist, Calvinist, Molinist, Jansenist” — are “only pseudonyms”; and “There are no sects in geometry; one does not speak of a Euclidian, an Archimedean.” H. I. Woolf’s translation (New York: Alfred A. Knopf, 1924), via Hanover College. Voltaire issued his alphabetical articles across several works, beginning with the Dictionnaire philosophique portatif (1764); later editors merged them into the single volume Woolf translated. This site cites the Woolf arrangement rather than assign the article a first-printing date it has not verified.

Voltaire, Candide (1759), ch. III. The battle of the Bulgars and the Abares: two armies destroyed in symmetry, after which “the two kings were causing Te Deum to be sung each in his own camp.” Philip Littell’s translation (Boni & Liveright, 1918), via Project Gutenberg.

George Washington, Farewell Address (1796). The departing president’s warning “in the most solemn manner against the baneful effects of the spirit of party generally” — party spirit which “agitates the community with ill-founded jealousies and false alarms, kindles the animosity of one part against another.” Text via the Avalon Project, Yale Law School.

Ralph Waldo Emerson, Self-Reliance (1841). “Whoso would be a man, must be a nonconformist,” and “Nothing is at last sacred but the integrity of your own mind.” Via Project Gutenberg (Essays, First Series).

Walt Whitman, Song of Myself, §51, from the 1891–92 “deathbed” edition. “Do I contradict myself? / Very well then I contradict myself, / (I am large, I contain multitudes.)” Quoted from that text; the 1855 first edition is differently punctuated and has no section numbers. Via the Walt Whitman Archive.

Henrik Ibsen, An Enemy of the People (1882), Act V. Dr. Stockmann, the compact majority, and “the strongest man in the world is he who stands most alone” — spoken in the final seconds of the play, in a wrecked house, to the family surrounding him. Farquharson Sharp translation via Project Gutenberg. Whether the play endorses the line is contested and this site should not pretend otherwise: Ibsen wrote to Hegel that “the Doctor is a more muddleheaded person than I am,” but defended the substance of Stockmann’s public-meeting speech to Brandes. Read the play as a caution about the temperament, not a retraction of the position. Stockmann’s poodle-men and cur-men speech in Act IV is where the curdling is visible.

F. Scott Fitzgerald, “The Crack-Up,” Esquire (February 1936). “…the test of a first-rate intelligence is the ability to hold two opposed ideas in the mind at the same time, and still retain the ability to function.” Quoted as first printed; the version in wide circulation drops the article and reads “in mind,” which is a misquotation. Collected in The Crack-Up, ed. Edmund Wilson (New Directions, 1945); text via PBS American Masters.

Simone Weil, On the Abolition of All Political Parties (written 1943; first published in La Table ronde, 1950; English translation by Simon Leys, NYRB Classics, 2013). The party as “a machine to generate collective passions,” designed “to exert collective pressure upon the minds of all its individual members,” whose “first objective and also the ultimate goal … is its own growth, without limit.”

Václav Havel, “The Power of the Powerless” (October 1978), §III. The greengrocer, his sign, and its real message: “I am obedient and therefore I have the right to be left in peace.” Paul Wilson’s translation. Three things a careful reader should carry, because the passage is borrowed carelessly more often than not. The message has an addressee — it points upward, to the greengrocer’s superior, and shields him from informers; this is a vertical relation to state power, not ambient peer pressure. The cost of refusal is specified: demotion, lost pay, and his children’s access to higher education. And Havel’s phrase is “living within the truth,” used some three dozen times in Wilson’s text; “living in truth” comes from the title of a later English anthology, not from this essay. Havel does license the westward extension — §VI asks whether the post-totalitarian grey is not “an inflated caricature of modern life in general” — but note that the direction of the analogy makes the coerced case the enlargement, not the equivalent.

George Orwell, Notes on Nationalism (1945). Nationalism defined as “the habit of identifying oneself with a single nation or other unit, placing it beyond good and evil and recognising no other duty than that of advancing its interests” — with Orwell’s insistence that the unit need not be a nation at all. Two qualifications Orwell makes that this site is obliged to carry: he distinguishes nationalism from patriotism, “devotion to a particular place and a particular way of life… but has no wish to force on other people”; and he observes that nationalist feeling can be “purely negative,” available to people who have joined nothing. Via the Orwell Foundation. (Orwell’s account of the Spanish press inventing battles — “I saw great battles reported where there had been no fighting” — is from Looking Back on the Spanish War (1943), not from Notes on Nationalism.)

Eric Hoffer, The True Believer: Thoughts on the Nature of Mass Movements (Harper & Brothers, 1951), §9. “The less justified a man is in claiming excellence for his own self, the more ready is he to claim all excellence for his nation, his religion, his race or his holy cause.”

Groucho Marx, Groucho and Me (Bernard Geis, 1959), ch. 26, “Foot in Mouth Disease.” The resignation telegram: “Please accept my resignation. I don’t want to belong to any club that will accept me as a member.” Wording per Marx’s own memoir, where the anecdote is staged at a humorously fictionalised “Delaney Club”; tradition identifies the real club as the Friars, Arthur Marx’s accounts have the actual resignations turning on unused dues, and newspaper precursors of the joke run back to 1891 (Quote Investigator traces the history). Note also what the joke is: self-deprecation about the speaker’s own unworthiness, not a verdict on clubs. This site repurposes it, and says so rather than pretending Groucho was making its argument.

Ray Oldenburg, The Great Good Place (Paragon House, 1989). The sociology of “third places” — the cafés, taverns, and halls where community happens without enrollment. Behind the repair proposed in The Empty Lodge.

Robert D. Putnam, “Bowling Alone: America’s Declining Social Capital,” Journal of Democracy 6, no. 1 (1995), expanded as Bowling Alone (Simon & Schuster, 2000). Source of the civic-membership declines cited on the Evidence page. Putnam reads the decline as loss, not liberation — a reading this site engages rather than hides.

The laboratory

Solomon E. Asch, “Effects of Group Pressure Upon the Modification and Distortion of Judgments” (1951); “Opinions and Social Pressure,” Scientific American 193, no. 5 (1955). Primary texts, both free. From the 1951 chapter: errors on “32 per cent of the total number of estimates”; “one-fourth of the critical subjects was completely independent,” hence about 75 percent conforming at least once; a truthful partner cutting pro-majority errors to 5.5 percent; a control mean under 1 percent; a group of eight, so seven confederates. Note that 32 percent is a share of answers, not of people — a distinction most secondary summaries garble. Asch also sorted the yielders three ways, and the ranking matters: distortion of perception (“a very few subjects”), distortion of judgment (“most submitting subjects belong to this category” — they doubted themselves and deferred), and distortion of action (compliance “with full awareness”). The 1955 article adds that all the yielders underestimated how often they had conformed. The popular version of Asch — the room rewiring the eye — describes his rarest category.

Rod Bond & Peter B. Smith, “Culture and Conformity: A Meta-Analysis of Studies Using Asch’s Line Judgment Task,” Psychological Bulletin 119, no. 1 (1996). 133 Asch-type studies across 17 countries; conformity higher in collectivist than individualist cultures. The paper’s other headline finding, which cuts against any use of Asch as a constant: “an analysis of U.S. studies found that conformity has declined since the 1950s.” See also Perrin & Spencer, “The Asch effect: a child of its time?” (1980), a UK replication that found almost none.

Henri Tajfel, “Experiments in Intergroup Discrimination,” Scientific American 223, no. 5 (1970). The minimal-group experiments. The Klee-versus-Kandinsky study is the one this site describes: reproductions shown unsigned so that boys could be assigned at random, after which they favored their arbitrary in-group and frequently preferred maximizing the difference between groups over maximizing their own group’s absolute gain. Two precisions: the boys knew each other well beforehand — Tajfel’s point being that the arbitrary sorting overrode that — and the reliable minimal-group result is in-group favouritism, not out-group harm, a distinction later work (Brewer, 1999) insists on and this site should not blur.

Dan M. Kahan, Ellen Peters, Erica Cantrell Dawson & Paul Slovic, “Motivated Numeracy and Enlightened Self-Government,” Behavioural Public Policy 1 (2017). Identical data read correctly as a skin-cream trial and tribally as a gun-control study; polarization greatest among the most numerate. Read the result precisely: in the uncongenial condition high-numeracy partisans “enjoyed no meaningful advantage over their low-numeracy counterparts” — the advantage vanished rather than inverting — and overall accuracy was low throughout. See the replication note, which this site regards as part of the citation rather than a footnote to it.

Cass R. Sunstein, “The Law of Group Polarization,” Journal of Political Philosophy 10, no. 2 (2002). Deliberation among the like-minded predictably moves the group toward a more extreme point than the members’ pre-deliberation median. (Linked text is the 1999 working-paper version; the published article is behind Wiley’s paywall.)

More in Common, The Perception Gap (2019). United States study: partisans imagine roughly twice as many “extreme” opponents as exist; the politically disengaged are about three times more accurate than the highly engaged; heavy news consumers are nearly three times more distorted; frequent political posters more distorted than the silent. Disclosures this site owes, since it runs a letterhead test on everyone else: More in Common is an advocacy organization whose mission is reducing polarization, the study is a single cross-sectional online survey of about 2,100 respondents fielded in November 2018, and cross-sectional data cannot establish the direction of the relationship it reports.

Douglas J. Ahler & Gaurav Sood, “The Parties in Our Heads: Misperceptions About Party Composition and Their Consequences,” Journal of Politics 80, no. 3 (2018). A peer-reviewed study of a neighbouring misperception — the demographic composition of the parties, overestimated by a mean of 342 percent — rather than a replication of the perception-gap figures above. Cited because it is the sturdier finding, not because it measures the same thing.

The ledgers

Gallup, “New High of 45% in U.S. Identify as Political Independents” (January 2026). 45 percent independent in 2025 — a record — against 27 percent for each party; trend series from 33 percent in 1988. Basis of the first chart. The same release reports that once leaners are allocated the split is 47 percent Democratic-leaning to 42 percent Republican-leaning, which is the fact that makes the 45 percent a measure of self-description rather than of behavior. See the contested section for the standard rebuttal.

Pew Research Center, “As Partisan Hostility Grows, Signs of Frustration With the Two-Party System” (2022). Very-unfavorable views of the opposing party: under 25 percent of each party in 1994; 62 percent of Republicans and 54 percent of Democrats by 2022. Majorities of each party call the other immoral and dishonest; 48 percent of independents say they often wish for more parties to choose from.

Ingrid van Biezen, Peter Mair & Thomas Poguntke, “Going, going, … gone? The decline of party membership in contemporary Europe,” European Journal of Political Research 51 (2012). Party membership averages 4.7 percent of European electorates; absolute membership nearly halved since 1980; UK, France, and Italy each lost 1–1.5 million members over three decades; declines of at least 25 percent in every established democracy. The underlying data are from 2007–2009, so “today” in the paper is not today; this site dates the figures rather than carrying the present tense forward. Summary at the LSE EUROPP blog.

Pew Research Center, “How the Global Religious Landscape Changed From 2010 to 2020” (2025). The religiously unaffiliated grew by 270 million to 1.9 billion — 24.2 percent of world population — driven principally by switching out of religion, against demographic headwinds.

Pew Research Center, Religious Landscape Study (2023–24). Religiously unaffiliated Americans: 29 percent, up from 23 percent (2014) and 16 percent (2007); Christians 62 percent, down from 78 percent (2007). Read the report’s title, which this site quotes on the Evidence page rather than hoping nobody clicks: Pew’s own headline finding is that the Christian share has been “relatively stable, hovering between 60% and 64%” since 2019. The decline is real and has, on this evidence, slowed.

Gallup, “U.S. Church Membership Falls Below Majority for First Time” (2021), with the continuing series at “Religious Americans”. Formal congregation membership 47 percent in 2020 — below half for the first time since measurement began in 1937 at 73 percent — versus 70 percent as recently as 1999. The series has since run 47 (2021), 46 (2022), 45 (2023); this site quotes the current figure, not the one that was current when the headline was written.

U.S. Bureau of Labor Statistics, “Union Members — 2025” (February 2026). Union membership rate 10.0 percent of wage and salary workers, versus 20.1 percent in 1983, the first year of comparable data. Two caveats from the release itself: the 2025 annual figures are eleven-month averages, October data having gone uncollected during the federal shutdown; and 2024 was 9.9 percent, so 2025 is a tick up rather than a further fall.

Gallup, “Labor Union Approval Relatively Steady at 68%” (August 2025), and Thomas A. Kochan, Duanyi Yang, William T. Kimball & Erin L. Kelly, “Worker Voice in America,” ILR Review 72, no. 1 (2019). Approval of unions at 68 percent, a fifth consecutive year near 70; nearly half of non-union workers saying they would vote for a union, against roughly a third in the two prior national surveys. Together these are the reason the Evidence page declines to read falling union density as a decision workers made.

Unite America Institute, “Not Invited to the Party Primary” (February 2024). Fifteen U.S. states hold closed congressional and state primaries, excluding 15.7 million registered independents (17.6 million voters including minor parties); 22 states close their presidential contests, excluding 23.5 million independents. Unite America’s later explainer gives revised counts (16 states / 16.6 million; 23 states / 24.4 million); this site quotes the report, and notes that both publishers are organizations campaigning for open primaries.

Office of the U.S. Surgeon General, Our Epidemic of Loneliness and Isolation (2023). About half of American adults reporting measurable loneliness. An advisory is a policy document rather than a finding, and the underlying trend is contested — meta-analytic work finds only a small secular increase — so this site should be read as citing a widely accepted description of the present, not a proven deterioration.

For the twelve questions

The empirical references invoked in Twelve Questions, Answered in Grey:

National Academies of Sciences, Engineering, and Medicine, The Economic and Fiscal Consequences of Immigration (2016). The consensus review behind “aggregate gains and contested local wage effects.”

Shanshan Li, Meir Stampfer, David Williams & Tyler VanderWeele, “Religious Service Attendance and Mortality Among Women,” JAMA Internal Medicine 176, no. 6 (2016). Among the studies behind “congregational life correlating with longevity.”

Caitlin Hughes & Alex Stevens, “What Can We Learn From the Portuguese Decriminalization of Illicit Drugs?” British Journal of Criminology 50, no. 6 (2010). The standard scholarly assessment of Portugal’s 2001 reform; an accessible overview is at Transform.

Where this argument is contested

A reader can only audit an argument if he is told where to dig. These are the works that cut against this site, listed because leaving them out would be the exact failure the site accuses everyone else of.

On the laboratory findings. Emil Persson, David Andersson, Lina Koppel, Daniel Västfjäll & Gustav Tinghög, “A preregistered replication of motivated numeracy,” Cognition 214 (2021): 104768 — motivated reasoning replicated, the numeracy interaction did not. Gordon Pennycook & David Rand, “Lazy, not biased,” Cognition 188 (2019), finding analytic thinking associated with better discernment regardless of congeniality. Ben Tappin, Pennycook & Rand, Current Opinion in Behavioral Sciences 34 (2020), on why these designs cannot cleanly separate motivated reasoning from ordinary Bayesian updating. Peter Ditto et al., “At Least Bias Is Bipartisan,” Perspectives on Psychological Science 14, no. 2 (2019) — the surviving, symmetrical version of the claim. And note that Kahan’s own prescription is trusted messengers within groups, which is not this site’s prescription.

On collective action — the strongest case against. Mancur Olson, The Logic of Collective Action (1965): public goods are under-supplied without selective incentives and sanction. Elinor Ostrom, Governing the Commons (1990): clearly defined boundaries are the first design principle of commons that actually work. Charles Tilly, Social Movements, 1768–2004: worthiness, unity, numbers and commitment. Doug McAdam, “Recruitment to High-Risk Activism: The Case of Freedom Summer,” American Journal of Sociology 92, no. 1 (1986): participation predicted by prior organizational embeddedness, not by conviction. Aldon Morris, The Origins of the Civil Rights Movement (1984), on the Black church as movement infrastructure. Adam Hochschild, Bury the Chains (2005), and Christopher Leslie Brown, Moral Capital (2006), on the Society for Effecting the Abolition of the Slave Trade, founded 1787 by nine Quakers and three Anglicans — subscribing membership, provincial chapters, mass petitions, a consumer boycott, and the Wedgwood medallion, which is to say a logo. Its Quaker founders belonged to a denomination that disowned members trading in slaves. These are answered as best this site can at objection 5, and the answer is a concession.

On exit. Albert O. Hirschman, Exit, Voice, and Loyalty (1970): exit and voice are substitutes, and loyalty is the mechanism that raises the cost of exit enough to force voice. The finding cuts against the unqualified reading of Article X and is met at objection 7. Paul Frymer, Uneasy Alliances (1999), on electoral capture, is the supporting half — a constituency with no credible exit gets taken for granted.

On precommitment. Thomas Schelling, The Strategy of Conflict (1960); Jon Elster, Ulysses and the Sirens (1979). Binding your future self can be an exercise of judgment rather than a surrender of it. Conceded at objection 4.

On independents. Bruce Keith et al., The Myth of the Independent Voter (1992); Samara Klar & Yanna Krupnikov, Independent Politics (2016). Most independents lean and vote like partisans; independent identification is substantially self-presentation and predicts disengagement. Philip Converse, “The Nature of Belief Systems in Mass Publics” (1964), and Donald Kinder & Nathan Kalmoe, Neither Liberal nor Conservative (2017): ideological constraint rises with sophistication, so clustered opinions are a marker of engagement as readily as of capture — which is why The Grey tests for retraceable reasoning rather than for scatter.

On what membership delivers. Tyler VanderWeele, “Religious Communities and Human Flourishing,” Current Directions in Psychological Science 26, no. 5 (2017): the health association attaches to attendance, not to private belief. Robert Putnam & David Campbell, American Grace (2010): the civic dividend runs through congregational friendship networks specifically. Raj Chetty et al., “Social Capital I & II,” Nature 608 (2022): economic connectedness is the strongest known correlate of upward mobility, and religious organizations are among the venues where cross-class friendship actually forms. Christopher Masi, Hsi-Yuan Chen, Louise Hawkley & John Cacioppo, “A Meta-Analysis of Interventions to Reduce Loneliness,” Personality and Social Psychology Review 15, no. 3 (2011): among randomised trials, “the most successful interventions addressed maladaptive social cognition” — not those that simply added social contact. All four are unfriendly to The Empty Lodge, which says so.

On the political consequences. Peter Mair, Ruling the Void (2013) — the co-author of the membership figures above, reading the same decline as democratic hollowing. Michael Hout & Claude Fischer, American Sociological Review 67, no. 2 (2002), on the American “nones” as a political reaction rather than a set of private conclusions. Hannah Arendt, The Origins of Totalitarianism (1951), on mass movements recruiting from the atomized. Sung Eun Kim & Yotam Margalit, “Informed Preferences? The Impact of Unions on Workers’ Policy Views,” American Journal of Political Science 61, no. 3 (2017): union membership causally shifts members’ views on the issues their union takes positions on, with self-selection explaining at most a quarter of the effect — which complicates the clean case of the union card that does not want your worldview. In fairness the other way: Sheri Berman, “Civil Society and the Collapse of the Weimar Republic,” World Politics 49, no. 3 (1997), and Satyanath, Voigtländer & Voth, “Bowling for Fascism,” Journal of Political Economy 125, no. 2 (2017), find dense associational life accelerating Nazi mobilization — the best evidence this site has, and it did not go looking for it.

On the philosophy. Alasdair MacIntyre, After Virtue (1981); Michael Sandel, Liberalism and the Limits of Justice (1982); Charles Taylor, Sources of the Self (1989). The case that the unencumbered self is incoherent — that judgment requires a tradition to judge from. John Hardwig, “Epistemic Dependence,” Journal of Philosophy 82, no. 7 (1985): rational belief for any non-specialist is necessarily testimonial, which means deference is not automatically a pathology and this site owes a line between good and bad kinds. It does not yet have one. Both are named as unanswered.

A note on method: single studies wobble, so this site prefers findings that have been replicated, meta-analyzed, or measured independently by unrelated institutions — and where a favoured finding has failed replication, as motivated numeracy did, it says so beside the citation rather than in silence. Where an authority cited here disagrees with this site’s conclusions — Putnam, Mair, Hirschman, Kahan — the disagreement is printed. The section above exists because the alternative, a bibliography that only calls friendly witnesses, would fail this site’s own letterhead test on the first page.