Kohlberg’s Stages of Moral Development Explained

Lawrence Kohlberg proposed that the way people think about right and wrong changes in a predictable sequence as they grow. His theory describes three levels — preconventional, conventional, and postconventional — each with two stages, six in total. The stages are not about which choices a person makes, or how good a person is, but about the reasons behind a moral choice: punishment, self-interest, social approval, the law, shared agreements, or universal principles.

Kohlberg’s stages remain one of the most taught ideas in developmental psychology, and also one of the most criticized. This article explains the six stages, how Kohlberg studied them, what the research found, and where the theory falls short — patterns found across many people in many studies, not rules for any one person.

What Moral Development Means

Moral development is the study of how people come to understand right and wrong, fairness, rules, and obligations to others. A young child may judge an action by whether it is punished, an older child by whether it breaks a rule everyone should follow, and an adult by whether it respects a principle such as justice. Kohlberg’s theory describes that progression.

Kohlberg built directly on Jean Piaget, who had argued that children’s moral thinking moves from a heteronomous stage — rules seen as fixed, handed down by authority — toward an autonomous stage, in which rules are understood as agreements people make together and can change. Kohlberg agreed that moral thinking develops in stages tied to cognitive growth, but he believed Piaget’s two stages were too broad. In his 1958 doctoral dissertation at the University of Chicago, he proposed six stages, and he spent the rest of his career testing them (Kohlberg, 1958). The fullest statements appear in his two volumes of Essays on Moral Development (Kohlberg, 1981, 1984). For Piaget’s own stage theory, see our guide to Piaget’s stages of cognitive development.

A key idea runs through Kohlberg’s work: moral reasoning is a form of thinking, and it develops through the same kind of reorganization that cognitive psychology describes. Each later stage can handle moral conflicts the earlier stages cannot resolve.

The Heinz Dilemma: How Kohlberg Studied Moral Reasoning

Kohlberg did not study moral development by watching what children did. He studied it by asking them to think out loud about moral dilemmas — stories in which two values conflict and there is no obviously correct answer. The most famous is the Heinz dilemma:

> A woman is dying of a rare form of cancer. A druggist in her town has discovered a drug that could save her, but he is charging ten times what it cost him to make it — $2,000 for a drug that cost $200. Heinz, her husband, has borrowed all he can and raised only $1,000. He asks the druggist to sell it for less or let him pay the rest later, and the druggist refuses. That night, Heinz breaks into the druggist’s shop and steals the drug. Should Heinz have done that?

The point is not the yes-or-no answer: people at very different stages might give the same answer. What differs is the reasoning behind it — Stage 1 might say Heinz should not steal because he will be caught and punished, Stage 3 that he should steal because a good husband would do anything for his wife, and Stage 5 that the question turns on whether the law, as a social agreement, should override a person’s right to life.

For his dissertation, Kohlberg interviewed 72 boys aged 10, 13, and 16 from middle- and lower-class families in Chicago, presenting dilemmas like Heinz’s and probing their answers: Why do you think that? What if the druggist were a stranger? The follow-ups mattered as much as the first answer, because they revealed how far the reasoning held up when the details changed. Later studies followed the same boys into adulthood, and other researchers applied the method in many countries.

The method’s strength is that it makes reasoning visible; its limit is that an interview dilemma is not a real-life one — answering calmly about a fictional druggist differs from deciding under fear, anger, or pressure, a gap this article returns to below.

How Moral Reasoning Was Scored

Kohlberg’s Moral Judgment Interview was a semi-structured interview, not a questionnaire. A trained interviewer presented several dilemmas, asked for a decision and a justification, then probed to clarify the reasoning. The interview was recorded and transcribed, and trained scorers classified each response.

Scoring did not work by matching answers to a list of “correct” responses. Scorers looked for the underlying structure of the reasoning — the consideration the person treated as most important. A response appealing to avoiding punishment was scored at Stage 1 even if it sounded confident, while one appealing to maintaining the social order was scored at Stage 4 even if it was hesitant. Responses were typically mixed, so scorers assigned a global stage based on the person’s most common and most advanced reasoning, using the standard scoring manual published by Colby and Kohlberg (1987).

Because the interview is slow and requires trained scorers, James Rest and his colleagues developed a faster alternative, the Defining Issues Test: participants read dilemmas and rate and rank considerations, yielding a score for the proportion of postconventional reasoning a person uses. Rest and his colleagues later presented this as a neo-Kohlbergian approach — keeping the idea of stages while revising some of Kohlberg’s claims about how they work (Rest, Narvaez, Bebeau, & Thoma, 1999). The test made large-scale studies possible, but it measures recognition of postconventional reasoning rather than its spontaneous production, which is not quite the same thing.

The Six Stages at a Glance

The table summarizes the three levels and six stages. Read it as a map, not as boxes people fit into: most people use reasoning from more than one stage, and Kohlberg did not tie stages to exact ages.

LevelStageWhat reasoning sounds likeExample
Preconventional1. Obedience and punishment“It’s wrong because I’ll get punished.”A child does not take a biscuit because they will be told off.
Preconventional2. Self-interest and exchange“It’s fair if I get something back.”A child shares a toy so the other child will share later.
Conventional3. Good relationships“A good person would do this.”A teenager helps a friend because that is what a loyal friend does.
Conventional4. Law and social order“Rules keep society working.”An adult pays taxes because everyone must follow the law.
Postconventional5. Social contract“Laws are agreements we can change.”An adult argues a law should be reformed because it no longer serves everyone.
Postconventional6. Universal principles“It violates a principle that applies to everyone.”A person refuses an unjust order on the grounds of human dignity.

Level 1: Preconventional Morality

At the preconventional level, reasoning is organized around consequences for the self and fairness as exchange. “Preconventional” does not mean the person has no morals — only that their reasoning does not yet take social conventions as its reference point. It is typical of young children, though elements appear in adults too.

Stage 1: Obedience and Punishment

At Stage 1, an action is right if it avoids punishment and wrong if it leads to punishment. Authority is obeyed because authority has power, not because rules are understood as fair or useful. In the Heinz dilemma, a Stage 1 reasoner might say Heinz should not steal because he will go to prison — the sentence, not the harm to his wife, being decisive.

Stage 2: Self-Interest and Exchange

At Stage 2, reasoning becomes more flexible but remains organized around the self. Right is what serves one’s own interests, and fairness means exchange: you scratch my back and I will scratch yours. A Stage 2 reasoner can recognize that other people have interests too, but treats them as separate bargains rather than as part of a shared system.

Level 2: Conventional Morality

At the conventional level, reasoning takes social norms as its reference point: what would a good member of the group do, or what do society’s rules require? This is the level at which most adults reason most of the time.

Stage 3: Good Relationships

At Stage 3, right means living up to what others expect of a good person in your roles: a good son, a loyal friend, a caring partner. A Stage 3 reasoner might say Heinz should steal the drug because a good husband does everything he can for his wife. The reasoning now takes other people’s perspectives seriously, but it is organized around relationships and reputation rather than abstract rules or principles.

Stage 4: Law and Social Order

At Stage 4, the reference point widens from particular relationships to society as a whole. Right means following the law and fulfilling one’s duties, because the law and social institutions keep order and protect everyone. A Stage 4 reasoner might say Heinz should not steal because if everyone broke the law when they thought they had a good reason, society would fall apart. It is also where most adults settle: Stage 4 is the most common stage of adult reasoning.

Level 3: Postconventional Morality

At the postconventional level, reasoning goes beyond existing laws and conventions and asks whether they are justified. Laws are treated as human agreements that can be evaluated against more general principles — and a law can be wrong. Postconventional reasoning is rare, and Stage 6 has been the most contested part of the theory.

Stage 5: Social Contract

At Stage 5, laws are understood as social contracts — agreements made by people to protect everyone’s rights, which can be changed when they no longer serve that purpose. A Stage 5 reasoner might say that Heinz’s case turns on a conflict between two rights — the druggist’s property rights and his wife’s right to life — and that the right to life should take priority. Stage 5 reasoning is uncommon and becomes more frequent with higher education.

Stage 6: Universal Principles

At Stage 6, reasoning appeals to universal ethical principles — justice, equality, and respect for human dignity — meant to apply to everyone, in every society, regardless of what the law says. A Stage 6 reasoner might argue that Heinz should steal the drug because every person has an equal right to life, and that this principle would still hold if the roles were reversed. Kohlberg’s test was reversibility: would you still endorse the principle in the other person’s position?

Stage 6, however, is where the theory’s empirical support runs thinnest. In the long-term studies it was so rarely found that Kohlberg and his colleagues dropped it from the scoring manual: the revised manual scores reasoning only up to Stage 5, treating Stage 6 as a philosophical ideal rather than an empirically observed stage (Colby & Kohlberg, 1987). Textbooks still list six stages, and this article does too — but the sixth stage has a different evidential status from the first five.

How the Stages Develop

Kohlberg claimed that the stages form an invariant sequence: people move through them in order, without skipping a stage, and genuine regression is rare. Each stage builds on the one before it. Movement is driven, in Kohlberg’s account, by cognitive conflict: when a person’s current way of reasoning cannot resolve a moral problem, the resulting disequilibrium pushes them toward a more adequate way of thinking.

The long-term evidence supports the broad shape of this claim but not every detail. The twenty-year study by Colby, Kohlberg, Gibbs, and Lieberman (1983) followed 58 boys from the original sample, interviewing them six times over twenty years, and found that moral reasoning rose in order with age, that stage skipping was not observed, and that regression was uncommon. But development proceeds slowly: many adults remained at Stages 3 and 4, Stage 5 was reached by only a minority, and progress often depended on education and on social roles that required considering other perspectives.

Two cautions follow. First, the stages are not a ladder everyone climbs at the same speed, or at all — remaining at Stage 4 for a lifetime is normal, not a failure. Second, a stage score describes a person’s reasoning on dilemmas in an interview, not their character: a person who reasons at Stage 5 can still act selfishly, and a person who reasons at Stage 3 can act with great kindness.

What the Research Found

The strongest evidence comes from the longitudinal study just described and from cross-cultural research. John Snarey (1985) reviewed 45 studies in 27 countries and found support for the core claims: the early stages appeared in every culture studied, the sequence was the same, and stage skipping and regression were rare. But the higher stages were not equally distributed: Stages 1 to 4 were found widely, while Stage 5 was extremely rare outside urban, industrialized, Western-influenced samples, and no tribal or folk society in his review showed postconventional reasoning. Snarey concluded that the theory’s claims held best for the lower stages.

Research using the Defining Issues Test has supported the developmental picture differently: scores rise with age and education. Rest and his colleagues (1999) argued for a revised, neo-Kohlbergian view — moral development as gradual shifts in reasoning across schemas, rather than clean jumps from one stage to the next. That revision keeps Kohlberg’s central insight while dropping his strictest claim, that development is a staircase of discrete stages.

Criticisms and Limits

The best-known criticism came from Carol Gilligan (1982), in In a Different Voice. Gilligan argued that Kohlberg’s theory, built from interviews with boys and men and scored for reasoning about justice, rights, and rules, undervalued a different moral orientation — an ethic of care, organized around relationships, responsibility, and avoiding harm. Women in Gilligan’s interviews, many of them considering whether to have an abortion, reasoned in terms of care and connection rather than abstract rights. Her point was that a theory built on one moral voice could mistake a different voice for a lower stage.

The empirical response was more complicated than the popular debate suggests. Lawrence Walker (1984) reviewed 79 studies comparing men’s and women’s moral reasoning and found non-significant sex differences overall: few consistent differences appeared in childhood and adolescence, and adult differences were confounded with education and occupation. But that does not settle Gilligan’s deeper point — a theory can still be incomplete if it treats justice reasoning as the whole of morality and care reasoning as a lesser form of it. Many researchers now treat justice and care as two complementary orientations rather than two points on a single scale.

A second criticism concerns culture. As Snarey’s (1985) review showed, the higher stages are rare outside Western, urban, educated samples. Critics argue that this reflects Kohlberg’s dilemmas and scoring categories, which draw on Western liberal philosophy — Kant and Rawls on rights, contracts, and universal principles — and treat it as the endpoint of development. A tradition emphasizing duty, community harmony, or religious obligation may organize moral thinking differently without being less developed.

A third criticism concerns the gap between reasoning and behaviour. Kohlberg’s theory is about moral judgment, not moral action, and research consistently finds only a modest connection between stage of reasoning and actual behaviour. People reason well about hypothetical dilemmas and still act selfishly or dishonestly when real stakes, emotions, and social pressures are involved. Kohlberg acknowledged the gap, but the stage theory as usually taught says little about why people fail to act on their own best reasoning.

A fourth criticism comes from moral domain theory, associated with Elliot Turiel (1983). Turiel argued that children distinguish early between moral rules — about harm, fairness, and rights — and conventional rules, such as dress codes, which depend on social agreement. In his experiments, even young children judged that hitting another child would still be wrong if a teacher said it was allowed, while wearing the wrong clothes would be acceptable if the rule changed. If Turiel is right, moral understanding is not a single sequence from punishment to principles.

Key Takeaways

  • Kohlberg’s theory describes six stages of moral reasoning across three levels — preconventional (1–2), conventional (3–4), and postconventional (5–6) — and the stages are about the reasons people give, not the choices they make.
  • The Heinz dilemma is a research tool: a dying woman, a drug priced at $2,000 that cost $200 to make, and a husband who steals it.
  • Reasoning was scored from interviews in which trained scorers classified the structure of a person’s justifications; Rest’s Defining Issues Test offered a faster alternative.
  • The stage sequence is broadly supported — people move through the stages in order, without skipping — but most adults reason at the conventional level, and Stage 5 is uncommon.
  • Stage 6 was dropped from the scoring manual because it was almost never found in interviews; it remains a philosophical ideal, not an observed stage.
  • Gilligan argued that Kohlberg’s theory undervalued an ethic of care; Walker’s review of 79 studies found no consistent sex difference.
  • Cross-cultural evidence supports the early stages widely but finds the higher stages rare outside Western, urban, educated samples.
  • Turiel’s moral domain theory offers an alternative: children distinguish moral rules from social conventions early, suggesting moral understanding develops in domains, not on a single ladder.

References

  • Colby, A., & Kohlberg, L. (1987). The Measurement of Moral Judgment: Vol. 1. Theoretical Foundations and Research Validation. Cambridge University Press.
  • Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study of moral judgment. Monographs of the Society for Research in Child Development, 48(1–2), 1–124. https://doi.org/10.2307/1165935
  • Gilligan, C. (1982). In a Different Voice: Psychological Theory and Women’s Development. Harvard University Press.
  • Kohlberg, L. (1958). The Development of Modes of Moral Thinking and Choice in the Years 10 to 16 (Doctoral dissertation, University of Chicago).
  • Kohlberg, L. (1981). Essays on Moral Development: Vol. 1. The Philosophy of Moral Development. Harper & Row.
  • Kohlberg, L. (1984). Essays on Moral Development: Vol. 2. The Psychology of Moral Development. Harper & Row.
  • Piaget, J. (1932). The Moral Judgment of the Child. Kegan Paul.
  • Rest, J., Narvaez, D., Bebeau, M. J., & Thoma, S. J. (1999). Postconventional Moral Thinking: A Neo-Kohlbergian Approach. Lawrence Erlbaum Associates. https://doi.org/10.4324/9781410603913
  • Snarey, J. R. (1985). Cross-cultural universality of social-moral development: A critical review of Kohlbergian research. Psychological Bulletin, 97(2), 202–232. https://doi.org/10.1037/0033-2909.97.2.202
  • Turiel, E. (1983). The Development of Social Knowledge: Morality and Convention. Cambridge University Press.
  • Walker, L. J. (1984). Sex differences in the development of moral reasoning: A critical review. Child Development, 55(3), 677–691. https://doi.org/10.2307/1130121

Keep Exploring

What are Kohlberg’s six stages of moral development?

Kohlberg’s six stages are grouped into three levels. At the preconventional level, Stage 1 is obedience and punishment and Stage 2 is self-interest and exchange. At the conventional level, Stage 3 is good relationships and Stage 4 is law and social order. At the postconventional level, Stage 5 is the social contract and Stage 6 is universal principles. The stages describe the reasons people give for moral choices, not the choices themselves.

What is the Heinz dilemma?

The Heinz dilemma is a story Kohlberg used to study moral reasoning. A woman is dying of cancer, and a druggist is charging $2,000 for a drug that cost $200 to make. Her husband Heinz raises only $1,000, is refused a lower price or more time to pay, and steals the drug. Kohlberg asked whether Heinz should have done that — and, more importantly, why. The reasoning behind the answer is what revealed a person’s stage.

What level of moral reasoning do most adults reach?

Most adults reason at the conventional level, typically Stage 3 or Stage 4. In the twenty-year longitudinal study by Colby and colleagues (1983), most adults remained at Stages 3 and 4, Stage 5 was reached by only a minority, and Stage 6 was so rarely found that it was dropped from the scoring manual. Remaining at the conventional level is normal, not a failure.

What did Carol Gilligan criticize about Kohlberg’s theory?

Gilligan argued in In a Different Voice (1982) that Kohlberg’s theory, built from interviews with boys and men and scored for reasoning about justice and rights, undervalued an ethic of care — reasoning organized around relationships, responsibility, and avoiding harm. A review of 79 studies by Walker (1984) found no consistent sex difference in stage scores, but many researchers now treat justice and care as two complementary moral orientations rather than two points on one scale.

Does a higher stage of moral reasoning mean a person behaves better?

Not necessarily. Kohlberg’s theory is about moral judgment — how people think about moral problems — not about moral action. Research finds only a modest connection between stage of reasoning and actual behaviour, because real decisions also involve emotions, habits, social pressure, and practical constraints. A person can reason at a high stage and still act selfishly, and a person reasoning at a lower stage can act with great kindness.

W

Waqar holds a Master’s degree in Applied Psychology and works as an SEO content writer. He writes clear, research-informed content that helps readers understand psychology and related topics.

Leave a Comment