Moral development: Kohlberg's stages, Gilligan's critique, moral foundations theory

Intuition Beginner

How do we develop our sense of right and wrong? Lawrence Kohlberg posed moral dilemmas — like whether a man named Heinz should steal an expensive drug to save his dying wife. He found that moral reasoning develops in stages, each a more adequate way of thinking about justice.

Young children focus on punishment and reward — the preconventional level. Older children and adolescents care about rules, social approval, and law-and-order — the conventional level. Some adults reach a postconventional level, questioning whether laws are just and appealing to universal ethical principles.

Carol Gilligan argued that Kohlberg's theory was male-biased. It valued justice — abstract, rule-based reasoning more common in males' responses — over care and responsibility in relationships, which she found more often in females' reasoning. One voice, she warned, had been mistaken for the universal.

Jonathan Haidt later argued that morality is driven more by gut feelings than by reasoning. We have innate moral "taste buds" — for harm, fairness, loyalty, authority, and sanctity. Reason usually justifies judgments we have already made, like a rider atop an elephant that goes where it will.

Visual Beginner

The diagram captures the three frameworks of the unit. Kohlberg's stages appear as a staircase of increasingly principled reasoning, most adults halting at the conventional level. Gilligan's critique appears as a rebalanced scale adding care to justice. Moral foundations theory appears as six innate "taste buds" whose relative sensitivities sort the moral taste of liberals and conservatives.

Worked example Beginner

The Heinz dilemma

Kohlberg's diagnostic instrument was the moral dilemma. The best-known is the Heinz dilemma: a woman is near death from a disease for which one drug is a cure. The druggist who invented it charges ten times the cost to make it — far more than Heinz can raise. Heinz begs the druggist to sell it cheaper or let him pay later; the druggist refuses. Should Heinz steal the drug?

Kohlberg did not care whether the answer was yes or no. He cared about the reasoning. A Stage 2 subject says Heinz should steal it because he needs his wife, or should not because he might go to jail — reasoning in terms of self-interest and exchange. A Stage 4 subject says he should not steal because stealing is against the law and laws must hold for society. A Stage 6 subject says he should steal because the right to life outranks property rights — reasoning from a consistent universal principle.

Check your understanding Beginner

Formal definition Intermediate

The vocabulary of moral development is standardised across the anchor texts ^{[source pending]}. These terms name empirically dissociable constructs and mechanisms, not mere labels.

Kohlberg's six stages

Kohlberg's theory rests on the claim that moral reasoning develops through an invariant sequence of stages, each defined not by the conclusion reached but by the form of the reasoning. The stages group into three levels ^{[source pending]}.

Level 1 — preconventional. Morality is external to the self. Stage 1: obedience and punishment orientation. An act is wrong if it draws punishment; "good" is literal obedience to power. Stage 2: instrumental exchange / individualism. Right action serves one's own needs; reciprocity is transactional ("you scratch my back, I'll scratch yours"), and fairness is understood as equal exchange rather than impartial principle.

Level 2 — conventional. Morality is identified with conformity to social expectations and rules. Stage 3: interpersonal conformity ("good boy / good girl"). Right is living up to the expectations of those one cares about; goodness is being nice, helpful, approved of. Stage 4: law and order. Right is doing one's duty, respecting authority, and maintaining the social order for its own sake; laws hold because they are laws.

Level 3 — postconventional. Morality is distinguished from the authority, law, or convention that expresses it. Stage 5: social contract. Laws are understood as instruments of mutual benefit and adjustable when they fail to serve human welfare; procedural justice and basic rights take priority over specific statutes. Stage 6: universal ethical principles. Right is defined by self-chosen ethical principles — justice, equality, dignity — valid for all humanity and binding regardless of consequence or law.

Two qualifications matter. Kohlberg treated the sequence as invariant (passed in order) and aspirationally universal, but most empirical samples cluster at Stage 4, with Stages 5 and 6 reached by a minority and Stage 6 largely inferred rather than directly observed. Kohlberg later speculated about a Stage 7 (cosmic or religious orientation) and, in response to critique, refined scoring toward a hard versus soft stage distinction; the empirical status of the upper stages is the locus of much debate, taken up in Advanced results ^{[source pending]}.

Gilligan's ethics of care

Gilligan's In a Different Voice (1982) charged that Kohlberg's scale encoded a male norm. The original longitudinal sample on which the scoring was refined was predominantly male; women, on average, scored at Stage 3 while men scored at Stage 4, which Kohlberg read as women lagging in development. Gilligan reinterpreted the same pattern: women's moral reasoning is not deficient but differently voiced. It centres care — responsiveness to need, the avoidance of harm, and the maintenance of relationships and responsibility — rather than abstract justice built on rights, contracts, and impartial rules ^{[source pending]}.

The reanalysis turns the Heinz dilemma inside out. Where Kohlberg's scoring rewards the subject who reasons from universal principle, Gilligan attends to the subject who refuses to separate the parties into rights-holders and asks instead what response would preserve the web of relationships and least damage those involved. The empirical status of the gender difference is contested; see Advanced results.

Moral foundations theory

Jonathan Haidt and Craig Joseph's moral foundations theory holds that human morality is built on a small set of innate, modular "foundations" — each an evolved adaptation to a recurrent social challenge, each generating fast, intuitive moral responses, and each variably activated across individuals and cultures ^{[source pending]}. The original five (later joined by a sixth) are:

• Care / harm — sensitivity to suffering, derived from mammalian nurturance.
• Fairness / cheating — sensitivity to reciprocity, proportionality, and the detection of cheaters.
• Loyalty / betrayal — sensitivity to the in-group and collective commitment.
• Authority / subversion — sensitivity to rank, tradition, and legitimate leadership.
• Sanctity / degradation — sensitivity to purity, contamination, and the sacred.
• Liberty / oppression — sensitivity to domination and the resentment of coercion (added later).

A compact description records individual differences as a profile. Let $\mathbf{m}=(m_{\text{care}},m_{\text{fair}},m_{\text{loy}},m_{\text{auth}},m_{\text{sanct}},m_{\text{lib}})$ be an individual's sensitivity across the six foundations. Haidt explicitly resists reducing moral judgment to a single weighted aggregate of these coordinates; the profile is a descriptive summary of which foundations an individual registers as morally relevant, not a calculus from which judgments are computed. The empirical claim is that this vector varies systematically with culture and political orientation — liberals concentrate their moral sensitivity in care and fairness, while conservatives spread sensitivity more evenly across all six — a prediction borne out by the Moral Foundations Questionnaire ^{[source pending]}.

The social intuitionist model

Haidt's social intuitionist model (2001) displaces reasoning from the centre of moral judgment. On the model, moral judgment is typically produced by a fast, automatic intuition — a flash of approval or revulsion. Reasoning, when it occurs, is mostly post hoc, generated to justify an intuition already held and directed primarily at persuading others. The model predicts moral dumbfounding: subjects confronted with a harmless taboo act (consensual incest between adult siblings, carefully sterilised and concealed) often condemn it firmly but cannot produce a non-question-begging justification, iterating through candidate harms and rejecting each as the judgment holds ^{[source pending]}.

Key model Intermediate

Three models and how each was tested

Three models organise the field, and each rests on a characteristic operationalisation. Reviewing them side by side exposes what each takes moral maturity to be.

Kohlberg's stage model and the Moral Judgment Interview. Kohlberg's instrument is the Moral Judgment Interview (MJI): a subject is presented with a dilemma (Heinz, or others) and probed with follow-up questions that force the reasoning into the open. Trained scorers rate the reasoning against the stage criteria. The model's empirical claims — invariance, universality, the upper-stage hierarchy — stand or fall with the reliability and validity of this scoring. The longitudinal Colby study (see Advanced results) supplies the strongest support; reliability critiques and cross-cultural ceiling effects supply the strongest challenges ^{[source pending]}.

Rest's Defining Issues Test. James Rest converted Kohlberg's insight into a practical instrument, the Defining Issues Test (DIT) and its successor the DIT-2. Instead of open-ended interview and trained scoring, the DIT presents a dilemma and asks the subject to rate the importance of pre-written considerations keyed to different stages, then rank the most important. The instrument is cheap, reliable, and scalable; its central index, the P-score, measures the proportion of postconventional thinking. Rest and colleagues moved the theoretical frame from stages to schemas (personal interest, maintaining norms, postconventional) and produced the bulk of the cross-cultural data ^{[source pending]}.

The Moral Foundations Questionnaire and political orientation. Haidt, Graham, and Nosek operationalised foundations theory with the Moral Foundations Questionnaire (MFQ), which asks respondents how much each foundation matters to them morally and how relevant considerations drawn from each foundation are to specific judgments. The headline finding maps the $\mathbf{m}$-vector onto political orientation: self-identified liberals score high on care and fairness and low on the remaining foundations, while conservatives score the six foundations more nearly equally. Haidt argues this asymmetry explains why political coalitions fail to understand one another — each is responding to moral "flavours" the other barely registers ^{[source pending]}.

The trolley problem and the personal-impersonal distinction

The trolley problem supplies the experimental wedge for the modern intuitionist and dual-process work. In the standard switch case, a runaway trolley will kill five; flipping a switch to divert it onto a side track kills one instead, and most people judge diversion permissible. In the footbridge case, the only way to stop the trolley is to push a large stranger off a bridge onto the tracks, again trading one life for five; most people judge the push impermissible. The two cases are, by the lights of impartial cost-benefit reasoning, equivalent; the stark divergence in judgment is the explanandum. Greene and colleagues exploited this gap to map the neural substrates of deontological and utilitarian responding (see Advanced results) ^{[source pending]}.

Exercises Intermediate

Advanced results Master

Scoring, reliability, and the schema reframing

Kohlberg's stage claims depend on the MJI's scoring, and the scoring has been contested on two fronts. First, inter-rater reliability is high only with extensive training, and disagreements about upper-stage reasoning are common precisely where the theory's most distinctive claims (Stages 5 and 6) are located. Second, the content-form distinction — the claim that stages classify the form of reasoning independent of its conclusion — is hard to enforce cleanly in practice, since form and conclusion are correlated in ordinary speech ^{[source pending]}.

Rest's reframing into schema theory partly defuses these worries. The three schemas — personal interest, maintaining norms, postconventional — are broader bands than the six stages and admit of mixed reasoning within a subject, which fits the data better than strict stage-sequential ascent. The DIT-2 is a far more reliable instrument than the MJI and has generated the bulk of large-sample and cross-cultural data. The cost is that the fine-grained stage claims — the heart of Kohlberg's original theory — are softened into a looser developmental tendency ^{[source pending]}.

Cross-cultural and longitudinal evidence

John Snarey's (1985) cross-cultural review of 45 studies in 27 countries found the stage sequence broadly supported — no culture skipped stages or reversed their order — but with a striking ceiling effect: urban, middle-class, educated samples climbed higher than rural or traditional samples, and postconventional reasoning was largely absent in non-Western, non-individualist settings. Snarey argued this is not evidence that non-Western subjects are morally immature, but evidence that the upper stages encode a liberal-individualist conception of morality that is itself culture-bound ^{[source pending]}.

The Colby, Kohlberg and colleagues longitudinal study (1955-1977 follow-up) is the strongest single piece of support for the theory: moral reasoning in the same individuals moved upward over decades, no subject regressed to a lower stage, and stage transitions were consistent with the predicted sequence. But the sample was small (the original 58 were boys, later supplemented), almost entirely male and American, and the upper stages were sparsely populated even at the final assessment — so the strongest support is for the lower and middle stages, and the universalist upper-stage claim remains the weakest link ^{[source pending]}.

The Gilligan debate re-examined

Gilligan's gender-difference claim received sharp empirical scrutiny. Lawrence Walker's (1984) review of the available studies found no consistent sex difference in stage attainment once age, education, and occupation were controlled; the apparent female lag in Kohlberg's original sample was an artefact of sampling and scoring. Sara Jaffee and Janet Hyde's (2000) meta-analysis reached a similar conclusion for the broader care-versus-justice contrast: gender differences in moral orientation are small, context-dependent, and heavily modulated by the dilemma's content (relationships dilemmas elicit care reasoning in both sexes; impersonal dilemmas elicit justice reasoning in both sexes) ^{[source pending]}.

The resolution most consistent with the evidence treats care and justice as context-sensitive orientations available to both sexes, not as gendered moralities. Gilligan's lasting contribution is then not the empirical gender claim but the conceptual recovery of care as a moral category distinct from justice — a contribution that survives the failure of its original empirical vehicle ^{[source pending]}.

Greene's dual-process theory of moral judgment

Joshua Greene and colleagues used the trolley-versus-footbridge gap as an experimental probe for a dual-process theory of moral judgment. In an fMRI study (Greene et al. 2001), impersonal moral dilemmas (trolley switch) preferentially activated brain regions associated with cognitive calculation (including dorsolateral prefrontal cortex, DLPFC), while personal moral dilemmas (footbridge push) preferentially activated emotional regions (medial prefrontal cortex, posterior cingulate, and superior temporal sulcus / angular gyrus) ^{[source pending]}.

The interpretation: personal moral dilemmas — those involving direct, up-close harm to a identifiable victim — trigger strong emotional responses that issue deontological judgments ("do not push"). Impersonal dilemmas leave the emotional system comparatively quiet and permit utilitarian calculation ("divert the trolley"). The two response types thus reflect the relative dominance of two systems, not two distinct moral philosophies realised in the same mind ^{[source pending]}.

Michael Koenigs and colleagues (2007) supplied a striking dissociation: patients with focal lesions to the ventromedial prefrontal cortex (vmPFC) — a region critical for integrating emotional responses into decision-making — made disproportionately utilitarian choices on personal moral dilemmas, while performing normally on impersonal ones. The emotional signal that normally blocks the push is absent, and the cognitive calculation is left to decide. The result supports the causal claim that emotional input, not abstract commitment, generates the deontological pattern in personal dilemmas ^{[source pending]}.

Moral foundations and political orientation

The MFQ finding that liberals weight care and fairness while conservatives weight all five (later six) foundations nearly equally is the empirical backbone of Haidt's thesis in The Righteous Mind ^{[source pending]}. Haidt argues that the liberal moral matrix is literally narrower — it is built on fewer foundations — and that this explains why liberal political messaging often fails to persuade conservatives (it speaks in only two moral "tongues") and why conservative messaging can appeal across the divide (it can draw on all of them).

The thesis is not without challenge. The care/fairness items historically loaded on a partly distinct construct from the binding foundations, raising the question of whether the asymmetry reflects a genuine difference in moral breadth or a difference in what the questionnaire happens to measure. Cross-cultural work also finds substantial variation within the liberal-conservative dimension, and the WEIRD critique (below) complicates any inference from American samples to moral psychology in general ^{[source pending]}.

The WEIRD critique

Henrich, Heine, and Norenzayan's WEIRD argument — that participants in Western, Educated, Industrialised, Rich, Democratic societies are among the least representative populations one could study — bears directly on moral psychology. The WEIRD subject is, on average, highly individualistic, analytically rather than holistically oriented, and inclined to ground morality in justice and harm rather than in duty, sanctity, or in-group obligation. Haidt's foundations theory is partly a response to this bias: it began by taking the moral responses of non-WEIRD subjects seriously as moral responses, rather than as underdeveloped approximations to liberal-individualist morality. Even so, the MFQ itself is a WEIRD-built instrument, and the foundations it indexes are not guaranteed to exhaust the moral domain outside the populations on which it was normed ^{[source pending]}.

Moral emotions

Haidt (2003) sorted the moral emotions into three families by their relational function. Other-condemning emotions — contempt, anger, disgust — punish or motivate sanction against moral transgressors. Other-praising emotions — gratitude, moral elevation, awe — reward and reinforce moral exemplars and bind people to those who elicit them. Self-conscious emotions — guilt, shame, embarrassment — regulate the self's conformity to moral and social standards, with guilt oriented to a specific act and its repair and shame oriented to the global self ^{[source pending]}.

The disgust family deserves special attention because it demonstrates the intuitionist thesis in a strong form: feelings of physical disgust (elicited, for example, by a foul smell or a dirty desk in the experimental room) measurably intensify moral condemnation of unrelated acts. The contamination of moral judgment by visceral disgust is hard to reconcile with any view on which moral judgment is a clean output of principled reasoning ^{[source pending]}.

Moral behaviour and situationism

The link between moral reasoning and moral behaviour is weaker than stage theories imply. Hugh Hartshorne and Mark May's Studies in the Nature of Character (1928), a large-scale study of honesty in schoolchildren, found that honesty is largely situation-specific: a child who cheats on a test is only weakly more likely to cheat in a different setting (lying, stealing). The trait-like stability assumed by moral-character education does not survive cross-situational measurement ^{[source pending]}.

Dan Ariely and colleagues extended the finding with experimental paradigms in which subjects could cheat for modest gain. They found that cheating is widespread but bounded — most people cheat a little, few cheat a lot — and that small cues (an honour code signature at the top rather than the bottom of a form, the presence of mirrors) shift behaviour substantially. Moral behaviour, on this evidence, is shaped as much by local context and by self-conception maintenance as by a stable disposition ^{[source pending]}.

Evolutionary origins

Frans de Waal's primate work (Good Natured, 1996; Primates and Philosophers, 2006) documents empathy, consolation, inequity aversion, and reciprocity in chimpanzees, bonobos, and capuchin monkeys — building blocks of morality present without language, principle, or stage. The capuchin that refuses to work for a cucumber when a neighbour receives a grape for the same task displays a precursor to the fairness foundation ^{[source pending]}.

Christopher Boehm (Moral Origins, 2012) reconstructs the evolution of morality in hunter-gatherer moral communities: small egalitarian bands that actively suppressed upstart alpha males through gossip, ridicule, and collective sanction, thereby selecting over many generations for individuals disposed to internalise group norms, feel shame, and punish free-riders. On this account the binding foundations — loyalty, authority, sanctity — are not conservative cultural add-ons but deep evolutionary structures that predate the care and fairness foundations in any recognisable moral community ^{[source pending]}.

Connections Master

• Developmental psychology across the lifespan 29.06.01 is the direct prerequisite and parent unit. This unit presupposes its lifespan frame and treatment of attachment and temperament, and deepens the moral thread that the parent unit surveyed.

• Cognitive development 29.06.02 pending is the conceptual prerequisite. Kohlberg's stages presuppose Piagetian operational structures (Stage 6 universal principles requires formal-operational abstraction), and theory of mind is a precondition for judging intention and assigning responsibility — without which neither justice reasoning nor care reasoning can get traction.

• Adult development and aging 29.06.04 pending (proposed successor) carries the lifespan thread forward: postconventional reasoning, moral identity consolidation, and generativity are characteristic developmental tasks of adulthood, and the present unit supplies their cognitive-developmental foundation.

• Social psychology 29.07.01 connects through the social foundations of moral judgment — conformity (Asch), obedience (Milgram), and the bystander effect all bear on the situationist findings about moral behaviour reviewed above. Kohlberg's conventional level is, in part, a developmental gloss on conformity.

• Neuroscience 29.02.01 supplies the substrate for Greene's dual-process theory: the vmPFC, DLPFC, amygdala, and anterior insula (disgust) are the regions whose dissociations ground the personal-impersonal distinction and the moral-dumbfounding data.

• Motivation and emotion 29.11.01 connects through the moral-emotions family. Disgust, anger, guilt, shame, and elevation are studied both as emotions in their own right and as moral inputs, and the two literatures meet in the intuitionist claim that moral judgment is emotional cognition.

• Cross-cultural and indigenous psychology 29.12.01 is the testing ground for the universality claims of Kohlberg's stages and the cross-cultural generalisability of the MFQ. Snarey's cross-cultural review and the WEIRD critique both belong to the territory that unit covers in depth.

• Psychological disorders 29.09.01 connects through the antisocial-behaviour and psychopathy literatures, in which deficits in moral emotion (especially guilt and empathy) are central, and through the vmPFC-lesion patients who supply the dissociation evidence for dual-process theory.

Historical & philosophical context Master

Kohlberg's theory is a developmental extension of two older traditions. From John Dewey he inherited the pragmatist conviction that moral growth is progressive reconstruction of thought about social conflict, and from Jean Piaget — whose The Moral Judgment of the Child (1932) traced children's movement from heteronomous to autonomous morality — he inherited the stage method and the constructivist premise that the child actively organises moral experience. Kohlberg's distinctive contribution was to extend the stage sequence into adulthood and to anchor it in a Rawlsian conception of justice as the regulative ideal of mature moral thought. The intellectual ambition was large: a developmental psychology that doubled as a normative ethical theory, in which the upper stages were defensible as more adequate rather than merely later ^{[source pending]}.

Gilligan's challenge was both empirical and philosophical. Empirically, it pointed to the male sample on which the scoring was refined and to the systematic downgrading of women's reasoning. Philosophically, it reopened a fault line running back through the Western canon: the priority of impartial justice (Kant, Rawls, Kohlberg) over the ethics of care and relationship (an underdeveloped counter-tradition that Gilligan connected to the maternal voice in development). The challenge was received as a threat to the theory's universalism, and it reshaped the field — care became a standing category in moral-psychology research even after the specific gender-difference claim failed to replicate robustly ^{[source pending]}.

Haidt's work reopens an older fault line still. His elephant-and-rider image and his foundations theory revive David Hume's dictum that "reason is, and ought only to be the slave of the passions" against the rationalist lineage from Kant through Kohlberg and Rawls, in which moral judgment is the application of principle to case. By relocating the generative moment of moral judgment in fast intuition and treating principled reasoning as largely post-hoc justification, Haidt makes the rationalist tradition look like a description of how a WEIRD, highly educated minority talks about morality after the fact, rather than a description of how moral minds actually work. The dispute — reason first, or intuition first — is the modern face of a disagreement as old as moral philosophy itself, and the empirical work reviewed here (moral dumbfounding, the footbridge gap, the vmPFC dissociation) is what has finally put it on terrain where converging evidence can, in principle, adjudicate it ^{[source pending]}.

Bibliography Master

1. Kohlberg, L. — The Philosophy of Moral Development: Moral Stages and the Idea of Justice (Harper & Row, 1981).

2. Kohlberg, L. — The Psychology of Moral Development: The Nature and Validity of Moral Stages (Harper & Row, 1984).

3. Piaget, J. — The Moral Judgment of the Child (Kegan Paul, 1932; orig. 1932).

4. Colby, A., Kohlberg, L., Gibbs, J. & Lieberman, M. — "A longitudinal study of moral judgment", Monographs of the Society for Research in Child Development 48(1-2), 1-124 (1983).

5. Gilligan, C. — In a Different Voice: Psychological Theory and Women's Development (Harvard University Press, 1982).

6. Rest, J. R., Narvaez, D., Bebeau, M. J. & Thoma, S. J. — Postconventional Moral Thinking: A Neo-Kohlbergian Approach (Erlbaum, 1999).

7. Walker, L. J. — "Sex differences in the development of moral reasoning: A critical review", Child Development 55, 677-691 (1984).

8. Jaffee, S. & Hyde, J. S. — "Gender differences in moral orientation: A meta-analysis", Psychological Bulletin 126, 703-726 (2000).

9. Snarey, J. R. — "Cross-cultural universality of social-moral development: A critical review of Kohlbergian research", Psychological Bulletin 97, 202-232 (1985).

10. Haidt, J. — "The emotional dog and its rational tail: A social intuitionist approach to moral judgment", Psychological Review 108, 814-834 (2001).

11. Haidt, J. & Joseph, C. — "Intuitive ethics: How innately prepared intuitions generate culturally variable virtues", Daedalus 133(4), 55-66 (2004).

12. Haidt, J. — The Righteous Mind: Why Good People Are Divided by Politics and Religion (Pantheon, 2012).

13. Haidt, J. — "The moral emotions", in Davidson, R. J., Scherer, K. R. & Goldsmith, H. H. (eds.), Handbook of Affective Sciences (Oxford University Press, 2003), pp. 852-870.

14. Graham, J., Nosek, B. A., Haidt, J., Iyer, R., Koleva, S. & Ditto, P. H. — "Mapping the moral domain", Journal of Personality and Social Psychology 101, 366-385 (2011).

15. Greene, J. D., Sommerville, R. B., Nystrom, L. E., Darley, J. M. & Cohen, J. D. — "An fMRI investigation of emotional engagement in moral judgment", Science 293, 2105-2108 (2001).

16. Greene, J. D., Nystrom, L. E., Engell, A. D., Darley, J. M. & Cohen, J. D. — "The neural bases of cognitive conflict and control in moral judgment", Neuron 44, 389-400 (2004).

17. Koenigs, M., Young, L., Adolphs, R., Tranel, D., Cushman, F., Hauser, M. & Damasio, A. — "Damage to the prefrontal cortex increases utilitarian moral judgments", Nature 446, 908-911 (2007).

18. Hartshorne, H. & May, M. A. — Studies in the Nature of Character, Vol. I: Studies in Deceit (Macmillan, 1928).

19. Ariely, D. — The (Honest) Truth About Dishonesty: How We Lie to Everyone — Especially Ourselves (HarperCollins, 2012).

20. de Waal, F. — Good Natured: The Origins of Right and Wrong in Humans and Other Animals (Harvard University Press, 1996).

21. de Waal, F. — Primates and Philosophers: How Morality Evolved (Princeton University Press, 2006).

22. Boehm, C. — Moral Origins: The Evolution of Virtue, Altruism, and Shame (Basic Books, 2012).

23. Henrich, J., Heine, S. J. & Norenzayan, A. — "The weirdest people in the world?", Behavioral and Brain Sciences 33, 61-83 (2010).

24. Hume, D. — A Treatise of Human Nature (1739), Book II, Part III, Section III.

25. Rawls, J. — A Theory of Justice (Harvard University Press, 1971).

26. Myers, D. G. & DeWall, C. N. — Psychology, 13th ed. (Worth, 2021), Ch. 4.

27. Gleitman, H., Gross, J. & Reisberg, D. — Psychology, 8th ed. (Norton, 2011), Ch. 10.