CHAPTER SEVEN

KOHLBERG'S STAGES OF MORAL DEVELOPMENT

BIOGRAPHICAL INTRODUCTION

An outstanding example of research in the Piagetian tradition is the work of Lawrence Kohlberg. Kohlberg has focused on moral development and has proposed a stage theory of moral thinking which goes well beyond Piaget's initial formulations.

Kohlberg, who was born in 1927, grew up in Bronxville, New York, and attended the Andover Academy in Massachusetts, a private high school for bright and usually wealthy students. He did not go immediately to college, but instead went to help the Israeli cause, in which he was made the Second Engineer on an old freighter carrying refugees from parts of Europe to Israel. After this, in 1948, he enrolled at the University of Chicago, where he scored so high on admission tests that he had to take only a few courses to earn his bachelor's degree. This he did in one year. He stayed on at Chicago for graduate work in psychology, at first thinking he would become a clinical psychologist. However, he soon became interested in Piaget and began interviewing children and adolescents on moral issues. The result was his doctoral dissertation (1958a), the first rendition of his new stage theory.

Kohlberg is an informal, unassuming man who also is a true scholar; he has thought long and deeply about a wide range of issues in both psychology and philosophy and has done much to help others appreciate the wisdom of many of the "old psychologists," such as Rousseau, John Dewey, and James Mark Baldwin. Kohlberg has taught at the University of Chicago (1962-1968) and, since 1968, has been at Harvard University.

PIAGET'S STAGES OF MORAL JUDGMENT

Piaget studied many aspects of moral judgment, but most of his findings fit into a two-stage theory. Children younger than 10 or 11 years think about moral dilemmas one way; older children consider them differently. As we have seen, younger children regard rules as fixed and absolute. They believe that rules are handed down by adults or by God and that one cannot change them. The older child's view is more relativistic. He or she understands that it is permissible to change rules if everyone agrees. Rules are not sacred and absolute but are devices which humans use to get along cooperatively.

At approximately the same time--10 or 11 years--children's moral thinking undergoes other shifts. In particular, younger children base their moral judgments more on consequences, whereas older children base their judgments on intentions. When, for example, the young child hears about one boy who broke 15 cups trying to help his mother and another boy who broke only one cup trying to steal cookies, the young child thinks that the first boy did worse. The child primarily considers the amount of damage--the consequences--whereas the older child is more likely to judge wrongness in terms of the motives underlying the act (Piaget, 1932, p. 137).

There are many more details to Piaget's work on moral judgment, but he essentially found a series of changes that occur between the ages of 10 and 12, just when the child begins to enter the general stage of formal operations.

Intellectual development, however, does not stop at this point. This is just the beginning of formal operations, which continue to develop at least until age 16. Accordingly, one might expect thinking about moral issues to continue to develop throughout adolescence. Kohlberg therefore interviewed both children and adolescents about moral dilemmas, and he did find stages that go well beyond Piaget's. He uncovered six stages, only the first three of which share many features with Piaget's stages.

KOHLBERG'S METHOD

Kohlberg's (1958a) core sample was comprised of 72 boys, from both middle- and lower-class families in Chicago. They were ages 10, 13, and 16. He later added to his sample younger children, delinquents, and boys and girls from other American cities and from other countries (1963, 1970).

The basic interview consists of a series of dilemmas such as the following:

Heinz Steals the Drug

In Europe, a woman was near death from a special kind of cancer. There was one drug that the doctors thought might save her. It was a form of radium that a druggist in the same town had recently discovered. The drug was expensive to make, but the druggist was charging ten times what the drug cost him to make. He paid $200 for the radium and charged $2,000 for a small dose of the drug. The sick woman's husband, Heinz, went to everyone he knew to borrow the money, but he could only get together about $ 1,000 which is half of what it cost. He told the druggist that his wife was dying and asked him to sell it cheaper or let him pay later. But the druggist said: "No, I discovered the drug and I'm going to make money from it." So Heinz got desperate and broke into the man's store to steal the drug-for his wife. Should the husband have done that? (Kohlberg, 1963, p. 19)

Kohlberg is not really interested in whether the subject says "yes" or "no" to this dilemma but in the reasoning behind the answer. The interviewer wants to know why the subject thinks Heinz should or should not have stolen the drug. The interview schedule then asks new questions which help one understand the child's reasoning. For example, children are asked if Heinz had a right to steal the drug, if he was violating the druggist's rights, and what sentence the judge should give him once he was caught. Once again, the main concern is with the reasoning behind the answers. The interview then goes on to give more dilemmas in order to get a good sampling of a subject's moral thinking.

Once Kohlberg had classified the various responses into stages, he wanted to know whether his classification was reliable. In particular, he. wanted to know if others would score the protocols in the same way. Other judges independently scored a sample of responses, and he calculated the degree to which all raters agreed. This procedure is called interrater reliability. Kohlberg found these agreements to be high, as he has in his subsequent work, but whenever investigators use Kohlberg's interview, they also should check for interrater reliability before scoring the entire sample.

KOHLBERG'S SIX STAGES

Level 1. Preconventional Morality

Stage 1. Obedience and Punishment Orientation. Kohlberg's stage 1 is similar to Piaget's first stage of moral thought. The child assumes that powerful authorities hand down a fixed set of rules which he or she must unquestioningly obey. To the Heinz dilemma, the child typically says that Heinz was wrong to steal the drug because "It's against the law," or "It's bad to steal," as if this were all there were to it. When asked to elaborate, the child usually responds in terms of the consequences involved, explaining that stealing is bad "because you'll get punished" (Kohlberg, 1958b).

Although the vast majority of children at stage 1 oppose Heinz’s theft, it is still possible for a child to support the action and still employ stage 1 reasoning. For example, a child might say, "Heinz can steal it because he asked first and it's not like he stole something big; he won't get punished" (see Rest, 1973). Even though the child agrees with Heinz’s action, the reasoning is still stage 1; the concern is with what authorities permit and punish.

Kohlberg calls stage 1 thinking "preconventional" because children do not yet speak as members of society. Instead, they see morality as something external to themselves, as that which the big people say they must do.

Stage 2. Individualism and Exchange. At this stage children recognize that there is not just one right view that is handed down by the authorities. Different individuals have different viewpoints. "Heinz," they might point out, "might think it's right to take the drug, the druggist would not." Since everything is relative, each person is free to pursue his or her individual interests. One boy said that Heinz might steal the drug if he wanted his wife to live, but that he doesn't have to if he wants to marry someone younger and better-looking (Kohlberg, 1963, p. 24). Another boy said Heinz might steal it because

maybe they had children and he might need someone at home to look after them. But maybe he shouldn't steal it because they might put him in prison for more years than he could stand. (Colby and Kauffman. 1983, p. 300)

What is right for Heinz, then, is what meets his own self-interests.

You might have noticed that children at both stages 1 and 2 talk about punishment. However, they perceive it differently. At stage 1 punishment is tied up in the child's mind with wrongness; punishment "proves" that disobedience is wrong. At stage 2, in contrast, punishment is simply a risk that one naturally wants to avoid.

Although stage 2 respondents sometimes sound amoral, they do have some sense of right action. This is a notion of fair exchange or fair deals. The philosophy is one of returning favors--"If you scratch my back, I'll scratch yours." To the Heinz story, subjects often say that Heinz was right to steal the drug because the druggist was unwilling to make a fair deal; he was "trying to rip Heinz off," Or they might say that he should steal for his wife "because she might return the favor some day" (Gibbs et al., 1983, p. 19).

Respondents at stage 2 are still said to reason at the preconventional level because they speak as isolated individuals rather than as members of society. They see individuals exchanging favors, but there is still no identification with the values of the family or community.

Level II. Conventional Morality

Stage 3. Good Interpersonal Relationships. At this stage children--who are by now usually entering their teens--see morality as more than simple deals. They believe that people should live up to the expectations of the family and community and behave in "good" ways. Good behavior means having good motives and interpersonal feelings such as love, empathy, trust, and concern for others. Heinz, they typically argue, was right to steal the drug because "He was a good man for wanting to save her," and "His intentions were good, that of saving the life of someone he loves." Even if Heinz doesn't love his wife, these subjects often say, he should steal the drug because "I don't think any husband should sit back and watch his wife die" (Gibbs et al., 1983, pp. 36-42; Kohlberg, 1958b).

If Heinz’s motives were good, the druggist's were bad. The druggist, stage 3 subjects emphasize, was "selfish," "greedy," and "only interested in himself, not another life." Sometimes the respondents become so angry with the druggist that they say that he ought to be put in jail (Gibbs et al., 1983, pp. 26-29, 40-42). A typical stage 3 response is that of Don, age 13:

It was really the druggist's fault, he was unfair, trying to overcharge and letting someone die. Heinz loved his wife and wanted to save her. I think anyone would. I don't think they would put him in jail. The judge would look at all sides, and see that the druggist was charging too much. (Kohlberg, 1963, p. 25)

We see that Don defines the issue in terms of the actors' character traits and motives. He talks about the loving husband, the unfair druggist, and the understanding judge. His answer deserves the label "conventional "morality" because it assumes that the attitude expressed would be shared by the entire community—"anyone" would be right to do what Heinz did (Kohlberg, 1963, p. 25).

As mentioned earlier, there are similarities between Kohlberg's first three stages and Piaget's two stages. In both sequences there is a shift from unquestioning obedience to a relativistic outlook and to a concern for good motives. For Kohlberg, however, these shifts occur in three stages rather than two.

Stage 4. Maintaining the Social Order. Stage 3 reasoning works best in two-person relationships with family members or close friends, where one can make a real effort to get to know the other's feelings and needs and try to help. At stage 4, in contrast, the respondent becomes more broadly concerned with society as a whole. Now the emphasis is on obeying laws, respecting authority, and performing one's duties so that the social order is maintained. In response to the Heinz story, many subjects say they understand that Heinz's motives were good, but they cannot condone the theft. What would happen if we all started breaking the laws whenever we felt we had a good reason? The result would be chaos; society couldn't function. As one subject explained,

I don't want to sound like Spiro Agnew, law and order and wave the flag, but if everybody did as he wanted to do, set up his own beliefs as to right and wrong, then I think you would have chaos. The only thing I think we have in civilization nowadays is some sort of legal structure which people are sort of bound to follow. [Society needs] a centralizing framework. (Gibbs et al., 1983, pp. 140-41)

Because stage 4, subjects make moral decisions from the perspective of society as a whole, they think from a full-fledged member-of-society perspective (Colby and Kohlberg, 1983, p. 27).

You will recall that stage 1 children also generally oppose stealing because it breaks the law. Superficially, stage 1 and stage 4 subjects are giving the same response, so we see here why Kohlberg insists that we must probe into the reasoning behind the overt response. Stage 1 children say, "It's wrong to steal" and "It's against the law," but they cannot elaborate any further, except to say that stealing can get a person jailed. Stage 4 respondents, in contrast, have a conception of the function of laws for society as a whole--a conception which far exceeds the grasp of the younger child.

Level III. Postconventional Morality

Stage 5. Social Contract and Individual Rights. At stage 4, people want to keep society functioning. However, a smoothly functioning society is not necessarily a good one. A totalitarian society might be well-organized, but it is hardly the moral ideal. At stage 5, people begin to ask, "What makes for a good society?" They begin to think about society in a very theoretical way, stepping back from their own society and considering the rights and values that a society ought to uphold. They then evaluate existing societies in terms of these prior considerations. They are said to take a "prior-to-society" perspective (Colby and Kohlberg, 1983, p. 22).

Stage 5 respondents basically believe that a good society is best conceived as a social contract into which people freely enter to work toward the benefit of all They recognize that different social groups within a society will have different values, but they believe that all rational people would agree on two points. First they would all want certain basic rights, such as liberty and life, to be protected Second, they would want some democratic procedures for changing unfair law and for improving society.

In response to the Heinz dilemma, stage 5 respondents make it clear that they do not generally favor breaking laws; laws are social contracts that we agree to uphold until we can change them by democratic means. Nevertheless, the wife’s right to live is a moral right that must be protected. Thus, stage 5 respondent sometimes defend Heinz’s theft in strong language:

It is the husband's duty to save his wife. The fact that her life is in danger transcends every other standard you might use to judge his action. Life is more important than property.

This young man went on to say that "from a moral standpoint" Heinz should save the life of even a stranger, since to be consistent, the value of a life means any life. When asked if the judge should punish Heinz, he replied:

Usually the moral and legal standpoints coincide. Here they conflict. The judge should weight the moral standpoint more heavily but preserve the legal law in punishing Heinz lightly. (Kohlberg, 1976, p. 38)

Stage 5 subjects,- then, talk about "morality" and "rights" that take some priority over particular laws. Kohlberg insists, however, that we do not judge people to be at stage 5 merely from their verbal labels. We need to look at their social perspective and mode of reasoning. At stage 4, too, subjects frequently talk about the "right to life," but for them this right is legitimized by the authority of their social or religious group (e.g., by the Bible). Presumably, if their group valued property over life, they would too. At stage 5, in contrast, people are making more of an independent effort to think out what any society ought to value. They often reason, for example, that property has little meaning without life. They are trying to determine logically what a society ought to be like (Kohlberg, 1981, pp. 21-22; Gibbs et al., 1983, p. 83).

Stage 6: Universal Principles. Stage 5 respondents are working toward a conception of the good society. They suggest that we need to (a) protect certain individual rights and (b) settle disputes through democratic processes. However, democratic processes alone do not always result in outcomes that we intuitively sense are just. A majority, for example, may vote for a law that hinders a minority. Thus, Kohlberg believes that there must be a higher stage--stage 6--which defines the principles by which we achieve justice.

Kohlberg's conception of justice follows that of the philosophers Kant and Rawls, as well as great moral leaders such as Gandhi and Martin Luther King. According to these people, the principles of justice require us to treat the claims of all parties in an impartial manner, respecting the basic dignity, of all people as individuals. The principles of justice are therefore universal; they apply to all. Thus, for example, we would not vote for a law that aids some people but hurts others. The principles of justice guide us toward decisions based on an equal respect for all.

In actual practice, Kohlberg says, we can reach just decisions by looking at a situation through one another's eyes. In the Heinz dilemma, this would mean that all parties--the druggist, Heinz, and his wife--take the roles of the others. To do this in an impartial manner, people can assume a "veil of ignorance" (Rawls, 1971), acting as if they do not know which role they will eventually occupy. If the druggist did this, even he would recognize that life must take priority over property; for he wouldn't want to risk finding himself in the wife's shoes with property valued over life. Thus, they would all agree that the wife must be saved--this would be the fair solution. Such a solution, we must note, requires not only impartiality, but the principle that everyone is given full and equal respect. If the wife were considered of less value than the others, a just solution could not be reached.

Until recently, Kohlberg had been scoring some of his subjects at stage 6, but he has temporarily stopped doing so, For one thing, he and other researchers had not been finding subjects who consistently reasoned at this stage. Also, Kohlberg has concluded that his interview dilemmas are not useful for distinguishing between stage 5 and stage 6 thinking. He believes that stage 6 has a clearer and broader conception of universal principles (which include justice as well as individual rights), but feels that his interview fails to draw out this broader understanding. Consequently, he has temporarily dropped stage 6 from his scoring manual, calling it a "theoretical stage" and scoring all postconventional responses as stage 5 (Colby and Kohlberg, 1983, p. 28).

Theoretically, one issue that distinguishes stage 5 from stage 6 is civil disobedience. Stage 5 would be more hesitant to endorse civil disobedience because of its commitment to the social contract and to changing laws through democratic agreements. Only when an individual right is clearly at stake does violating the law seem justified. At stage 6, in contrast, a commitment to justice makes the rationale for civil disobedience stronger and broader. Martin Luther King, for example, argued that laws are only valid insofar as they are grounded in justice, and that a commitment to justice carries with it an obligation to disobey unjust laws. King also recognized, of course, the general need for laws and democratic processes (stages 4 and 5), and he was therefore willing to accept the penalities for his actions. Nevertheless, he believed that the higher principle of justice required civil disobedience (Kohlberg, 198 1, p. 43).

Summary

At stage 1 children think of what is right as that which authority says is right. Doing the right thing is obeying authority and avoiding punishment. At stage 2, children are no longer so impressed by any single authority; they see that there are different sides to any issue. Since everything is relative, one is free to pursue one's own interests, although it is often useful to make deals and exchange favors with others.

At stages 3 and 4, young people think as members of the conventional society with its values, norms, and expectations. At stage 3, they emphasize being a good person, which basically means having helpful motives toward people close to one At stage 4, the concern shifts toward obeying laws to maintain society as a whole.

At stages 5 and 6 people are less concerned with maintaining society for it own sake, and more concerned with the principles and values that make for a good society. At stage 5 they emphasize basic rights and the democratic processes that give everyone a say, and at stage 6 they define the principles by which agreement will be most just.

THEORETICAL ISSUES

How Development Occurs

Kohlberg, it is important to remember, is a close follower of Piaget. Accordingly, Kohlberg's theoretical positions, including that on developmental change, reflect those of his mentor.

Kohlberg (e.g., 1968; 198 1, Ch. 3) says that his stages are not the product of maturation. That is, the stage structures and sequences do not simply unfold according to a genetic blueprint.

Neither, Kohlberg maintains, are his stages the product of socialization. That is, socializing agents (e.g., parents and teachers) do not directly teach new forms of thinking. Indeed, it is difficult to imagine them systematically teaching each new stage structure in its particular place in the sequence.

The stages emerge, instead, from our own thinking about moral problems. Social experiences do promote development, but they do so by stimulating our mental processes. As we get into discussions and debates with others, we find our views questioned and challenged and are therefore motivated to come up with new, more comprehensive positions. New stages reflect these broader viewpoints (Kohlberg et al., 1975).

We might imagine, for example, a young man and woman discussing a new law. The man says that everyone should obey it, like it or not, because laws are vital to social organization (stage 4). The woman notes, however, that some well-organized societies, such as Nazi Germany, were not particularly moral. The man therefore sees that some evidence contradicts his view. He experiences some cognitive conflict and is motivated to think about the matter more fully, perhaps moving a bit toward stage 5.

Kohlberg also sometimes speaks of change occurring through role-taking opportunities, opportunities to consider others' viewpoints (e.g., 1976). As children interact with others, they learn how viewpoints differ and how to coordinate them in cooperative activities. As they discuss their problems and work out their differences, they develop their conceptions of what is fair and just.

Whatever the interactions are specifically like, they work best, Kohlberg says, when they are open and democratic. The less children feel pressured simply to conform to authority, the freer they are to settle their own differences and formulate their own ideas. We will discuss Kohlberg's efforts to induce developmental change in the section on implications for education.

The Stage Concept

Piaget, you will recall, proposed that true mental stages meet several criteria. They (1) are qualitatively different ways of thinking, (2) are structured wholes, (3) progress in an invariant sequence, (4) can be characterized as hierarchic integrations. and (5) are cross-cultural universals. Kohlberg has taken these criteria very seriously, trying to show how his stages meet them all. Let us consider these points one at a time.

1. Qualitative differences. It seems fairly clear that Kohlberg's stages are qualitatively different from one another. For example, stage 1 responses, which focus on obedience to authority, sound very different from stage 2 responses, which argue that each person is free to behave as he or she wishes. The two stages do not seem to differ along any quantitative dimension, they seem qualitatively different.

2. Structured wholes. By "structured wholes," Kohlberg means that the stages are not just isolated responses but are general patterns of thought that will consistently show up across many different kinds of issues. One gets a sense that this is true by reading through his scoring manual; one finds the same kinds of thinking reappearing on diverse items. For example, one item asks, "Why should a promise be kept?" As on the Heinz dilemma, children at stage 1 again speak in terms of obedience to rules, whereas those at stage 2 focus on exchanging favors that are in one's self-interest (e.g., "You never know when you're going to need that person to do something for you"). Similarly, as children proceed through the stages they keep giving responses that are similar to those to the Heinz dilemma (Gibbs et al., 1983, pp. 315-82).

In addition, Kohlberg and his co-workers (Colby et al., 1983) have obtained quantitative estimates of the extent to which subjects respond in terms of one particular stage. Since some subjects might be in transition between stages, one does not expect perfect consistency. Nevertheless, Kohlberg found that subjects scored at their dominant stage across nine dilemmas about two-thirds of the time. This seems to be a fair degree of consistency, suggesting the stages may reflect general modes of thought.

3. Invariant sequence. Kohlberg believes that his stages unfold in an invariant sequence. Children always go from stage 1 to stage 2 to stage 3 and so forth. They do not skip stages or move through them in mixed-up orders. Not all children necessarily reach the highest stages; they might lack intellectual stimulation. But to the extent they do go through the stages, they proceed in order.

Most of Kohlberg's evidence on his stage sequence comes from cross-sectional data. That is, he interviewed different children at various ages to see if the younger ones were at lower stages than the older ones. Stages 1 and 2 are primarily found at the youngest age, whereas the higher stages become more prevalent as age increases. Thus, the data support the stage sequence.

Cross-sectional findings, however, are inconclusive. In a cross-sectional study, different children are interviewed at each age, so there is no guarantee that any individual child actually moves through the stages in order. For example, there is no guarantee that a boy who is coded at stage 3 at age 13 actually passed through stages 1 and 2 in order when he was younger. More conclusive evidence must come from longitudinal studies, in which the same children are followed over time.

The first two major longitudinal studies (Kohlberg and Kramer, 1969; Holstein, 1973) began with samples of teenagers and then tested them at three-year intervals. These studies produced ambiguous results. In both, most subjects either remained at the same stage or moved up one stage, but there were also some who might have skipped a stage. Furthermore, these studies indicated that some subjects had regressed, and this finding also bothered Kohlberg, because he believes that movement through his stages should always be forward.

Kohlberg's response to these troublesome findings was to revise his scoring method. He had already become uncomfortable with his first (1958b) scoring manual, believing that it relied too heavily on the content of subjects' answers rather than their underlying reasoning. and he had made some improvements on it. So, when these longitudinal findings emerged, he decided to develop a much more precise and adequate scoring system and, to some extent, to revise his definitions of the stages.

To create the latest scoring manual, Kohlberg and his co-workers (Colby et al., 1983) worked with seven boys from his original (1958) sample who had been retested every three or four years for 20 years. It was during this work that Kohlberg decided to drop stage 6.

Kohlberg then examined the hypothesis of invariant sequence for 51 other boys from his original sample, who also had been retested at least twice (every three or four years) over the 20-year period. This time, Kohlberg and his colleagues (Colby et al., 1983) found no stage-skipping, and only about 6 percent of the subjects showed signs of regressing. Four recent longitudinal studies have obtained similar results although, two have found somewhat more regression (up to 15 percent) (see Colby et al., 1983). In general, then, the new longitudinal studies seem to support the invariant-sequence hypothesis.

Kohlberg's new, longitudinal study has also changed the earlier picture of moral development in other ways. Stage 4 had become the dominant stage by age 16. In the new scoring system, however, it is more difficult to achieve the higher stages--the reasoning must be more clearly demonstrated--and Kohlberg finds that stage 4 does not become dominant until the boys are in their 20s and 30s. Stage 5, too, only appears in the mid-20s and never becomes very prevalent.

4. Hierarchic integration. When Kohlberg says that his stages are hierarchically integrated, he means that people do not lose the insights gained at earlier stages, but integrate them into new, broader frameworks. For example, people at stage 4 can still understand stage 3 arguments, but they now subordinate them to wider considerations. They understand that Heinz had good motives for stealing, but they point out that if we all stole whenever we had a good motive, the social structure would break down. Thus stage 4 subordinates a concern for motives to a wider concern for the society as a whole.

The concept of hierarchic integration is very important for Kohlberg because it enables him to explain the direction of his stage sequence. Since he is not a maturationist, he cannot simply say that the sequence is wired into the genes. So he wants to show how each new stage provides a broader framework for dealing with moral issues. Stage 4, as mentioned, transcends the limitations of stage 3 and becomes more broadly concerned with social organization. Stage 5, in turn, sees the weakness of stage 4; a well-organized society is not necessarily a moral one. Stage 5 therefore considers the rights and orderly processes that make for a moral society. Each new stage retains the insights of the prior stage, but it recasts them into a broader framework. In this sense, each new stage is more cognitively adequate than the prior stage.

If Kohlberg is right about the hierarchic nature of his stages, we would expect that people would still be able to understand earlier stages but consider them inferior, In fact, when Rest (Rest et al., 1969; Rest, 1973) presented adolescents with arguments from different stages, this is what he found. They understood lower-stage reasoning, but they disliked it. What they preferred was the highest stage they heard, whether they fully understood it or not. This finding suggests, perhaps, that they had some intuitive sense of the greater adequacy of the higher stages.

Werner, we remember from Chapter 4, described hierarchic integration as a process that occurs alongside differentiation, and Kohlberg believes his stages represent increasingly differentiated structures as well. Kohlberg points out that the stage 5 value on life, for example, has become differentiated from other considerations. Stage 5 respondents say that we ought to value life for its own sake, regardless of its value to authorities (stage 1), its usefulness to oneself (stage 2), the affection it arouses in us (stage 3), or its value within a particular social order (stage 4). Stage 5 subjects have abstracted this value from other considerations and now treat it as a purely moral ideal. Their thinking, Kohlberg says, is becoming like that of the moral philosophers in the Kantian tradition (1981, p. 171).

5. Universal sequence. Kohlberg, like all stage theorists, maintains that his stage sequence is universal; it is the same in all cultures. At first glance, this proposal might be surprising. Don't different cultures socialize their children differently, teaching them very different moral beliefs?

Kohlberg's response is that different cultures do teach different beliefs, but that his stages refer not to specific beliefs but to underlying modes of reasoning (Kohlberg and Gilligan, 197 1). For example, one culture might discourage physical fighting, while another encourages it more. As a result, children will have different beliefs about fighting, but they will still reason about it in the same way at the same stage. At stage 1, for example, one child might say that it is wrong to fight when insulted "because you will get punished for it, "while another says that "it is all right to fight; you won't get punished." The beliefs differ, but both children reason about them in the same underlying way, in terms of the physical consequences (punishment). They do so because this is what they can cognitively grasp. Later on, the first child might argue that fighting is bad "because if everyone fought all the time there would be anarchy," while the second child argues that "people must defend their honor, because if they don't everyone will be insulting everyone, and the whole society will break down." Once again, the specific beliefs differ, reflecting different cultural teachings, but the underlying reasoning is the same--in this case it is stage 4, where people can consider something as abstract as the social order. Children, regardless of their beliefs, will always move to stage 4 thinking some time after stage 1 thinking because it is cognitively so much more sophisticated.

Kohlberg, then, proposes that his stage sequence will be the same in all cultures, for each stage is conceptually more advanced than the next. He and other researchers have given his interview to children and adults in a variety of cultures, including Mexico, Taiwan, Turkey, Israel, the Yucatan, Kenya, the Bahamas, and India. Most of the studies have been cross sectional, but a few have been longitudinal. Thus far, the studies have supported Kohlberg's stage sequence. To the extent that children move through the stages, they appear to move in order (Edwards, 1980).

At the same time, people in different cultures seem to move through the sequence at different rates and to reach different end-points. In the United States most urban middle-class adults reach stage 4, with a small percentage using some stage 5 reasoning. In urban areas of other countries the picture is fairly similar. In the isolated villages and tribal communities of many countries, however, it is rare to find any adult beyond stage 3 (Edwards, 1980).

Kohlberg (Nisan and Kohlberg, 1982) suggests that one can understand these findings in terms of Piagetian theory. Cultural factors, in this theory, do not directly shape the child's moral thought, but they do stimulate thinking. Social experiences can challenge children's ideas, motivating them to come up with new ones. In traditional villages, however, there may be little to challenge a stage 3 morality; the norms of care and empathy work very well in governing the face-to-face interactions of the group. Thus, there is little to stimulate thinking beyond this stage.

When, in contrast, young people leave the village and go off to the city, they witness the breakdown of interpersonal ties. They see that group norms of care and empathy have little impact on the impersonal interactions of city life, and they see the need for a formal legal structure to ensure moral conduct. They begin to think in terms of stage 4 morality. Furthermore, as Keniston (1971) notes, if young people attend the universities, they may take classes in which the teachers deliberately question the unexamined assumptions of their childhoods and adolescences. Thus they are stimulated to think about moral matters in new ways.

Moral Thought and Moral Behavior

Kohlberg's scale has to do with moral thinking, not moral action. As everyone knows, people who can talk at a high moral level may not behave accordingly. Consequently, we would not expect perfect correlations between moral judgment and moral action. Still, Kohlberg thinks that there should be some relationship.

As a general hypothesis, he proposes that moral behavior is more consistent, predictable. and responsible at the higher stages (Kohlberg et al., 1975), because the stages themselves increasingly employ more stable and general standards. For example, whereas stage 3 bases decisions on others' feelings, which can vary, stage 4 refers to set rules and laws. Thus, we can expect that moral behavior, too, will become more consistent as people move up the sequence. Generally speaking, there is some research support for this hypothesis (e.g., with respect to cheating), but the evidence is not clear-cut (Blasi, 1980; Brown and Herrnstein, 1975).

Some research has focused on the relationships between particular stages and specific kinds of behavior. For example, one might expect that juvenile delinquents or criminals would typically reason at stages 1 or 2, viewing morality as something imposed from without (stage 1) or as a matter of self-interest (stage 2), rather than identifying with society's conventional expectations (stages 3 and 4). Again, some research supports this hypothesis, but there also are some ambiguous results (Blasi, 1980).

Several studies have examined the relationship between postconventional thinking and student protest. In a landmark study, Haan et al. (1968) examined the moral reasoning of those who participated in the Berkeley Free Speech Movement in 1964. Haan found that their thinking was more strongly postconventional than that of a matched sample of nonparticipants, but this f inding was not replicated for some other protests, apparently because moral principles were not at stake (Keniston, 1971, pp. 260-6 1).

Blasi (1980), after reviewing 75 studies, concludes that overall there is a relationship between moral thought and action, but he suggests that we need to introduce other variables to clarify this relationship. One variable may simply be the extent to which individuals themselves feel the need to maintain consistency between their moral thoughts and actions (Blasi, 1980, Kohlberg and Candee, 1981).

Moral Thought and Other Forms of Cognition

Kohlberg has also tried to relate his moral stages to other forms of cognition. He has first analyzed his stages in terms of their underlying cognitive structures and has then looked for parallels in purely logical and social thought. For this purpose, he has analyzed his own stages in terms of implicit role-taking capacities, capacities to consider others' viewpoints (Kohlberg, 1976; see also Selman, 1976, and Rest, 1983).

At first, at stage 1, children hardly seem to recognize that viewpoints differ. They assume that there is only one right view, that of authorities. At stage 2, in contrast, they recognize that people have different interests and viewpoints. They seem to be overcoming egocentrism; they see that perspectives are relative to the individual . They also begin to consider how individuals might coordinate their interests in terms of mutually beneficial deals.

At stage 3, people conceptualize role-taking as a deeper, more empathic process; one becomes concerned with the other's feelings. Stage 4, in turn, has a broader, society-wide conception of how people coordinate their roles through the legal system..

Stages 5 and 6, finally, take a more idealized look at how people might coordinate their interests. Stage 5 emphasizes democratic processes, and stage 6 considers how all parties take one another's perspectives according to the principles of justice.

The moral stages, then, reflect expanded insights into how perspectives differ and might be coordinated. As such, the moral stages might be related to stages of logical and social thought which contain similar insights. So far, the empirical evidence suggests that advances in moral thinking may rest upon prior achievements in these other realms (Kohlberg, 1976; Kuhn et al., 1977). For example, children seem to advance to stage 2, overcoming their egocentrism in the moral sphere, only after they have made equivalent progress in their logical and social thought. If this pattern is correct, we can expect to find many individuals who are logical and even socially insightful but still underdeveloped in their moral judgment.

IMPLICATIONS FOR EDUCATION

Kohlberg would like to see people advance to the highest possible stage of moral thought. The best possible society would contain individuals who not only understand the need for social order (stage 4) but can entertain visions of universal principles, such as justice and liberty (stage 6) (Kohlberg, 1970).

How, then, can one promote moral development? Turiel (1966) found that when children listened to adults' moral judgments, the resulting change was slight. This is what Kohlberg might have expected, for he believes that if children are to reorganize their thinking, they must be more active.

Accordingly, Kohlberg encouraged another student, Moshe Blatt, to lead discussion groups in which children had a chance to grapple actively with moral issues (Blatt and Kohlberg, 1975). Blatt presented moral dilemmas which engaged the classes in a good deal of heated debate. He tried to leave much of the discussion to the children themselves, stepping in only to summarize, clarify, and sometimes present a view himself (p. 133). He encouraged arguments that were one stage above those of most of the class. In essence, he tried to implement one of Kohlberg's main ideas on how children move through the stages. They do so by encountering views which challenge their thinking and stimulate them to formulate better arguments (Kohlberg et al., 1975).

Blatt began a typical discussion by telling a story about a man named Mr. Jones who had a seriously injured son and wanted to rush him to the hospital. Mr. Jones had no car, so he approached a stranger, told him about the situation, and asked to borrow his car. The stranger, however, refused, saying he had an important appointment to keep. So Mr. Jones took the car by force. Blatt then asked whether Mr. Jones should have done that.

In the discussion that followed, one child, Student B, felt that Mr. Jones had a good cause for taking the car and also believed that the stranger could be charged with murder if the son died. Student C pointed out that the stranger violated no law. Student B still felt that the stranger's behavior was somehow wrong, even though he now realized that it was not legally wrong. Thus, Student B was in a kind of conflict. He had a sense of the wrongness of the stranger's behavior, but he could not articulate this sense in terms that would meet the objection. He was challenged to think about the problem more deeply.

In the end, Blatt gave him the answer. The stranger's behavior, Blatt said, was not legally wrong, but morally wrong--wrong according to God's laws (this was a Sunday School Class). At this point, Blatt was an authority teaching the "correct" view. In so doing, he might have robbed Student B of the chance to formulate spontaneously his own position. He might have done better to ask a question or to simply clarify the student's conflict (e.g,, "So it's not legally wrong, but you still have a sense that, it's somehow wrong. . ."). In any case, it seems clear that part of this discussion was valuable for this student. Since he himself struggled to formulate a distinction that could handle the objection, he could fully appreciate and assimilate a new view that he was looking for.

The Kohlberg-Blatt method of inducing cognitive conflict exemplifies Piaget's equilibration model. The child takes one view, becomes confused by discrepant information, and then resolves the confusion by forming a more advanced and comprehensive position. The method is also the dialectic process of Socratic teaching. The students give a view, the teacher asks questions which get them to see the inadequacies of their views, and they are then motivated to formulate better positions.

In Blatt's first experiment, the students (sixth graders) participated in 12 weekly discussion groups. Blatt found that over half the students moved up one full stage after the 12 weeks. Blatt and others have tried to replicate these findings, sometimes using other age groups and lengthier series of classes. As often happens with replications, the results have not been quite so successful; upward changes have been smaller--usually a third of a stage or less, Still, it generally seems that Socratic classroom discussions held over several months can produce changes that, while small, are significantly greater than those found in control groups who do not receive these experiences (Rest, 1983).

One of Blatt's supplementary findings was that those students who reported that they were most "interested" in the discussions made the greatest amount of change. This finding is in keeping with Piagetian theory. Children develop not because they are shaped through external reinforcements but because their curiosity is aroused. They become interested in information that does not quite fit into their existing cognitive structures and are thereby motivated to revise their thinking Another Kohlberg student--M. Berkowitz (1980)--is examining actual dialogues to see if those who become most challenged and involved in the tensions of moral debate are also those who move forward.

Although Kohlberg remains committed to the cognitive-conflict model of change, he has also become interested in other strategies. One is the "just Community" approach. Here the focus is not the individuals but groups. For example, Kohlberg and some of his colleagues (Power and Reimer, 1979) set up a special democratic high school group and actively encouraged the students to think of themselves as a community. Initially, little community feeling was present. The group's dominant orientation was stage 2; it treated problems such as stealing as purely individual matters. If a boy had something stolen, it was too bad for him. After a year, however, the group norms advanced to stage 3; the students now considered stealing to be a community issue that reflected on the degree of trust and care in the group.

It will be interesting to see if the just community approach can promote further advances in moral thinking. In the meantime, this approach has aroused some uneasiness among some of Kohlberg's associates. In particular, Reimer et al. (1983) have wondered whether Kohlberg, by explicitly encouraging the students to think of themselves as a community, is not practicing a form of indoctrination. Reimer says that he has talked to Kohlberg about this, and he has come away convinced that Kohlberg is committed to democratic groups in which students are encouraged "to think critically, to discuss assumptions, and. when they feel it is necessary, to challenge the teacher's suggestions" (p. 252). Thus, moral development remains a product of the students' own thinking.

EVALUATION

Kohlberg, a follower of Piaget, has offered a new, more detailed stage sequence for moral thinking. Whereas Piaget basically found two stages of moral thinking, the second of which emerges in early adolescence, Kohlberg has uncovered additional stages which develop well into adolescence and adulthood. He has suggested that some people even reach a postconventional level of moral thinking where they no longer accept their own society as given but think reflectively and autonomously about what a good society should be.

The suggestion of a postconventional morality is unusual in the social sciences. Perhaps it took a cognitive developmentalist list to suggest such a thing. For whereas most social scientists have been impressed by the ways in which societies mold and shape children's thinking, cognitive-developmentalists are more impressed by the capacities for independent thought. If children engage in enough independent thinking, Kohlberg suggests, they will eventually begin to formulate conceptions of rights, values, and principles by which they evaluate existing social arrangements. Perhaps some will even advance to the kinds of thinking that characterize some of the great moral leaders and philosophers who have at times advocated civil disobedience in the name of universal ethical principles.

Kohlberg's theory has provoked a good deal of criticism. Not everyone, first of all, is enthusiastic about the concept of a postconventional morality. Hogan (1973, 1975), for example, feels that it is dangerous for people to place their own principles above society and the law. It may be that many psychologists react to Kohlberg in a similar way, and that this reaction underlies many of the debates over the scientific merits of his research.

Others have argued that Kohlberg's stages are culturally biased. Simpson (1974), for example, says that Kohlberg has developed a stage model based on the Western philosophical tradition and has then applied this model to non-Western cultures without considering the extent to which they have different moral outlooks.

This criticism may have merit. One wonders how well Kohlberg's stages apply to the great Eastern philosophies. One also wonders if his stages do justice to moral development in many traditional village cultures. Researchers find that villagers stop at stage 3, but perhaps they continue to develop moralities in directions that Kohlberg's stages fail to capture.

Another criticism is that Kohlberg's theory is sex-biased, a view that has been thoughtfully expressed by one of Kohlberg's associates and co-authors, Carol Gilligan (1982). Gilligan observes that Kohlberg's stages were derived exclusively from interviews with males, and she charges that the stages reflect a decidedly male orientation. For males, advanced moral thought revolves around rules, rights, and abstract principles. The ideal is formal justice, in which all parties evaluate one another's claims in an impartial manner. This conception of morality, Gilligan argues, fails to capture the distinctly female voice on moral matters.

For women, Gilligan says, morality centers not on rights and rules but on interpersonal relationships and the ethics of compassion and care. The ideal is not impersonal justice but more affiliative ways of living. Women's morality, in addition, is more contextualized, it is tied to real, ongoing relationships rather than abstract solutions to hypothetical dilemmas.

Because of these sex differences, Gilligan says, men and women frequently score at different stages on Kohlberg's scale. Women typically score at stage 3, with its focus on interpersonal feelings, whereas men more commonly score at stages 4 and 5, which reflect more abstract conceptions of social organization. Thus, women score lower than men. If, however, Kohlberg's scale were more sensitive to women's distinctly interpersonal orientations, it would show that women also continue to develop their thinking beyond stage 3.

Gilligan has made an initial effort to trace women's moral development. Since she believes that women's conceptions of care and affiliation are embedded in real-life situations, she has interviewed women facing a personal crisis--the decision to have an abortion. Through these interviews, Gilligan has tried to show that women move from a conventional to a postconventional mode of thinking. That is, they no longer consider their responsibilities in terms of what is conventionally expected, of them but in terms of their own insights into the ethics of care and responsibility.

Not everyone agrees with Gilligan's critique. Rest (1983), in particular, argues that Gilligan has exaggerated the extent of the sex differences found on Kohlberg's scale. An evaluation of this question, however, must await closer reviews of the literature.

In the meantime, Gilligan has raised an interesting theoretical possibility. Like Werner, she is suggesting that development may proceed along more than one line. One line of moral thought focuses on logic, justice, and social organization, the other on interpersonal relationships. If this is so, there is the further possibility that these two lines at some point become integrated within each sex. That is, each sex might become more responsive to the dominant orientation of the other. Perhaps, as Gilligan briefly suggests (1982, Ch. 6), this integration is a major task of the adult years. (For further thoughts in this vein, see Chapter 14 on Jung's theory of adult development.)

There are other criticisms of Kohlberg's work. Many of these have to do with empirical matters, such as the problem of invariant sequence, the prevalence of regression, and the relationships between thought and action. Since I have mentioned these earlier, I would like to conclude with a more general question. Kohlberg writes in a forceful manner and he promotes stage 6 as if it provides the decision-making tools we need for the toughest ethical dilemmas. However, there may be issues that the principles of justice frequently fail to resolve. One such issue is abortion. Stage 6 would ask us to consider the physical life of the fetus as well as all the parties' right to fulfilling lives, but does stage 6 routinely lead to decisions that we feel are right? Kohlberg's students, Reimer et al. (1983, pp. 46-47, 88-89) discuss a stage 6 approach to a hypothetical abortion decision without reaching much of a conclusion. The decision, they say, will have to vary with the situation. Stage 6. of course, is not intended to provide a set of answers--it is a mode of decision-making. Still, Kohlberg sometimes seems to skim over the incredible difficulty that some ethical problems present--a difficulty that is more directly expressed in the writing of Kant (1788).

Nevertheless, whatever criticisms and questions we might have, there is no doubt that Kohlberg's accomplishment is great. He has not just expanded on Piaget's stages of moral judgment but has done so in an inspiring way. He has studied the development of moral reasoning as it might work its way toward the thinking of the great moral philosophers. So, although few people may ever begin to think about moral issues like Socrates, Kant, or Martin Luther King, Kohlberg has nonetheless provided us with a challenging vision of what development might be.