NOTE-132: Moral disciplining: foundations of puritanical morality

Contribution

Existing accounts treat puritanical morality as an exception to morality’s cooperative function. Moral Foundations Theory explains it by a disgust-based Purity foundation. Self-interest models explain it by mate-guarding or reproductive strategy. The paper proposes Moral Disciplining Theory (MDT) instead: puritanical judgements come from the same cooperation-sensitive moral system as the rest of morality, fed by folk-psychological beliefs about self-control. For each part of the “puritanical constellation” it derives three shared predictions (§4) and states a falsification condition. It also offers an explanation of the decline of puritanism in rich, WEIRD societies (§5).

Key insight

The paper’s key point is that the harm lies in the causal story, not in the act. If people believe that indulgence weakens the self-control needed to resist future temptations to cheat (adultery, violence, free-riding), a victimless pleasure becomes an indirect, probabilistic social harm. A harm/fairness-based moral system will then condemn it pre-emptively. Cultural variation in moral values need not imply multiple moral systems: one system fed different causal beliefs gives different outputs (§3.3).

Assumptions

• A1 (ultimate). Reciprocal and reputation-based cooperation objectively require delaying gratification. Defecting pays now and cooperating pays later (iterated PD; Roberts 2020; Lie-Panis and André 2022, in whose model cooperators are those with a sufficiently low discount rate) (§3.1.1).
• A2. People perceive that cooperation requires self-control (§3.1.3).
• A3. People hold lay theories of state self-control modifiers (intoxicants; cue exposure, hence immodesty) and of trait modifiers (overindulgence breeds addiction; discipline and ritual train self-control) (§3.2, Table 1).
• A4 (cognitive architecture). A unitary, cooperation-sensitive moral system, harm-based or fairness-based, is triggered not only by cheating but by behaviour perceived as “indirectly and probabilistically” leading to social harm (§3.3).
• A5 (cultural evolution). Norms spread through “subjective selection”: they are retained because people perceive them as effective, whether or not they are (§3.4; Singh 2022).
• Scope. “Puritanism” is defined to exclude the pathogen and contamination part of the Purity construct (p. 5). The theory does not cover the whole of Purity.

Key results

The paper is a theoretical review, so its “results” are the evidence it assembles:

• Against disgust accounts (§2.1). Classic purity vignettes are pathogen-laden and unlike real puritanical targets. Pathogen-free “spiritual purity” violations do not produce the facial expression of disgust, nausea or “grossed-out” reports. A meta-analysis (Landy and Goodwin 2015) and replications find that disgust does not amplify moral condemnation, and correlations between disgust sensitivity and condemnation vanish once perceived harm is controlled for (Schein et al. 2016).
• Against mate-guarding accounts (§2.2). Those accounts cannot explain the condemnation of gluttony, idleness, marital-sex excess or masturbation, or the moralisation of male desire. They are treated as complementary, not refuted.
• Mediation, bodily pleasures (§4.1.3). In the authors’ vignette studies, perceived change in self-control mediated 100% of the effect of indulgence versus restraint on perceived cooperativeness.
• Mediation, alcohol (§4.2). The same design gave full mediation for cooperativeness and 71% mediation for perceived moral character (p. 33).
• Moral Foundations Questionnaire data (§4.1.1). Across four studies (N > 3,000), the moralisation of victimless excesses is predicted by Loyalty and Authority about as strongly as by Purity (Mooijman et al. 2018).
• Heroin stigma, UK (§4.2). Fear of violence explains 23.5% of the variance and “bad character” 12.4%, against 0.9% for contagion risk (Mushtaq et al. 2015). In the 1996 GSS, 70.9% (alcohol) and 87.3% (drugs) expect addicts to be violent.
• Modesty (§4.4.1). Stereotyping men as unable to control sexual urges predicts the moralisation of immodesty, controlling for reproductive strategy, religiosity, conservatism and sexism (Moon et al. 2021).
• Hygiene (§6.1.1). Across about 20,000 participants in 56 countries, how much a society values self-control predicts the restrictiveness of its hygiene norms better than pathogen threat does (Eriksson et al. 2021).
• Decline (§5). Puritanical values fall with wealth (WVS; Inglehart). Individuals with puritanical values have lower trust (claim uncited, p. 39). State-level life-history strategy predicts purity values better than pathogen prevalence does (Van Leeuwen et al. 2014).

Claims

id	claim	strength	support
C1	Puritanical morality “does target cooperation, and is reducible to concerns for harm or fairness” (p. 9).	moderate	Theoretical synthesis. Direct tests exist only for some behaviours; §4.2 and §4.3 predictions are stated as untested
C2	Reciprocal and reputation-based cooperation objectively require delaying gratification.	moderate	Formal models cited (Axelrod; Roberts 2020; Lie-Panis and André 2022) plus correlational evidence. Economic-game meta-analysis finds no association (Thielmann et al. 2020), which the authors explain away by a moderation argument
C3	Perceived self-control fully mediates the perceived effect of indulgence on cooperativeness.	moderate	Authors’ own vignette studies, N > 1,100 (Fitouchi et al. 2022). These measure third-person perceptions of a hypothetical target
C4	Disgust does not explain puritanical condemnation.	moderate	Review of replications, a meta-analysis and facial-expression studies (§2.1)
C5	Modesty norms are moralised because of perceived risk to male self-control.	moderate	One direct study (Moon et al. 2021) plus design-fit evidence (eye-tracking; exemptions for elderly women)
C6	Ritual observance is praised because it is perceived to train self-control.	weak	Ethnographic and textual quotations and trust-perception studies. The authors state that the direct prediction is untested (p. 34)
C7	Puritanism declines where others are perceived as self-controlled and trustworthy.	weak–moderate	Cross-national correlations and analogy with Nettle and Saxe (2021). One key WVS figure is uncited
C8	The theory does not require puritanical norms to work.	n/a (a scope statement)	pp. 10, 26. It makes the theory robust to evidence that self-control training is ineffective, but harder to test
C9	Explaining puritanism “does not require adding mechanisms unrelated to cooperation” (abstract).	weak	Parsimony argument. Contamination Purity is excluded by definition (p. 5), so the claim is narrower than the abstract suggests

Concepts

• Puritanical morality / puritanism — “the ascetic, austere moralization of apparently victimless pleasures” (p. 3). It is a five-part constellation: bodily pleasures; valorisation of temperance and self-discipline; intoxicants, gambling, music and dance; modesty; piety and ritual. It excludes contamination-type Purity.
• Temptations to cheat / moral self-control — impulses for immediate reward that conflict with prosocial obligations, and the resistance to them (p. 16, after Hofmann et al. 2018).
• Modifiers of state / trait self-control — behaviours believed to alter momentary self-control (intoxicants, cue exposure) or dispositional self-control (overindulgence, discipline, ritual) (§3.2).
• Moral disciplining — pre-emptive moralisation of behaviour perceived to affect moral self-control, and the cultural “behavioural technologies” built on it (§3.4).
• Subjective selection — cultural retention of norms because they are perceived as effective (Singh 2022).

Connections

MDT is set against Moral Foundations Theory (Haidt, Graham) and the Reproductive Religiosity Model (Weeden and Kurzban), and aligned with unitary theories: the dyadic-harm theory of Gray and Schein, and the fairness and mutualistic morality of Baumard and André. It extends Nettle and Saxe’s (2021) account of authoritarian values to puritanism, and applies Singh’s subjective-selection view of cultural evolution. No held nucleation work treats moral psychology or Moral Foundations Theory. A grep of literature.d found none.

Ethics. The work presupposes a descriptive question: why do people moralise victimless pleasures? Its answer is that such behaviours are (perceived as) indirect threats to cooperation. It addresses no question of what is owed, permitted or good. It neither endorses nor debunks puritanical norms, and it brackets whether they work. The only moral-philosophical relevance is indirect: the claim that morality has a single cooperative function bears on moral-domain pluralism, and a reader could build an evolutionary debunking argument from it. The paper does neither. The ethics tag is not justified by the tag’s own blurb (moral philosophy). The work belongs under social-science (psychology, anthropology, cultural evolution) and cognition. religion would be justifiably appropriate too, since ritual, piety and world-religious moral codes are a large part of the subject.

Bearing on the record

The paper carries nothing for ML practice. The suggested primary topic is social-science, with cognition kept. The reasoning is that the paper is evolutionary and cultural psychology argued largely from surveys, experiments, history and ethnography, which is the social-science blurb (“psychology, anthropology”), not moral philosophy. religion could be added.

Limitations

• The central mediation evidence is the authors’ own vignette studies about hypothetical third parties. It shows that people believe indulgence erodes self-control and cooperativeness, not that such beliefs cause puritanical norms historically.
• The historical and ethnographic evidence consists largely of selected quotations, open to selection bias. The authors themselves call for “systematic, quantitative documentation” (p. 42).
• Contamination Purity is excluded by definition, so the case against disgust-based accounts covers only part of the Purity domain.
• Agnosticism about efficacy (C8) makes the cultural-evolution claim hard to falsify at the population level.
• The trust/WVS figure on p. 39 is uncited, and one subsection is misnumbered (“4.3.2” inside §4.4). The published version may fix both; that is unverified.
• The BBS commentaries and the authors’ response were not read, so the objections they raised are unrecorded here.

Open questions

• Direct tests of the §4.2 (intoxicants) and §4.3 (ritual) mediation predictions, which the authors list as future work.
• Does the perceived effect on self-control predict moralisation cross-culturally and outside vignette paradigms, e.g. in field or historical-quantitative data?
• Why are many small-scale societies not puritanical (§6.2)? Does MDT’s trust and self-control mechanism predict the observed distribution?
• What did the BBS commentators object to, and did the response narrow the theory?

Corrections to the seeded skim

• ethics tag: not justified — the work is descriptive, evolutionary and cognitive science of moral judgement. It explains why people judge victimless pleasures wrong and takes no position on whether they are wrong. It is explicitly “agnostic as to whether puritanical norms are objectively effective” (p. 10), and it makes no debunking or vindicating argument. A reader browsing moral philosophy would not be right to expect it.
• primary topic: social-science
• The skim framed the work as possibly “a debunking or vindicating genealogy”. The full text contains neither argument. Normative or metaethical implications are never drawn, and the only evaluative remarks are asides, e.g. that some temperance crusades may have reduced violence (p. 26).
• The skim says §4 gives “falsifiable predictions for each behaviour” and suggests they are supported. The support is uneven. For bodily pleasures (§4.1) the key mediation evidence is the authors’ own vignette studies (“Fitouchi, André, Baumard & Nettle, 2022”). For intoxicants the authors write “Future studies could test these predictions” (p. 31). For piety (§4.3) the prediction is “Future research could test” (p. 34). Only modesty (§4.4) cites a direct independent test (Moon et al. 2021).
• The skim describes §5’s explanation of decline as a “cost-benefit shift”. More exactly, the mechanism is perceived self-control and trustworthiness of others. It is modelled on Nettle and Saxe’s (2021) account of authoritarian values (pp. 37–39), and rich environments are also said to produce more self-controlled psychologies (p. 38).
• The skim omits §6.1, which extends the theory to hygiene norms (Eriksson et al. 2021, 56 countries) and to the moralisation of “impure thoughts” (pp. 40–41). It also omits §6.2’s point that small-scale societies are often not puritanical (pp. 41–42).
• One headline figure is uncited. The claim that “recent analyses of the World Value Survey (N > 200,000)” show lower trust among people with puritanical values (p. 39) has no reference in this manuscript.