The way in which researchers have determined whether or not people make a conceptual distinction between morality and convention has been by asking people to evaluate various actions in terms of one or more of the following criteria:
- Rule contingency: Does the wrongness of a given action depend upon the existence of a governing rule or social norm? (The reader will recognize this criterion from the interview with the 4-year-old child described above.)
- Rule alterability: Is it wrong or all right to remove or alter the existing norm or standard?
- Rule generalizability: Is it wrong or all right for members of another society or culture not to have a given rule or norm?
- Act generalizability: Is it wrong or all right for a member of another society or culture to engage in the act if that society/culture does not have a rule about the act?
- Act severity: How wrong (usually on a 5-point scale) is a given action?
In addition to being asked to make criterion judgments, people are also asked to provide justifications for the answers they give. These justifications allow researchers to determine which substantive bases people employ
to make their criterion judgments. In the example presented earlier, the young girl responded to the rule contingency question about hitting by responding that it would be wrong to hit, whether or not a governing rule were in effect. The substantive justification for judging hitting as wrong was that hitting has harmful effects on another person.
In order to gain clear-cut answers to whether or not people make distinctions between morality and convention, researchers have asked people to make judgments that would constitute prototypical examples of
moral or conventional issues. Issues have been presented in contexts in which the acts in question are generally not in conflict with other types of goals or events. In studies which have involved observations of children’s interactions, children have been asked to evaluate real situations they had just witnessed (as in the previous example). In most cases, however, issues have been presented in story or pictorial form. The types of issues used as moral stimuli have had to do with welfare and physical harm (for instance, pushing, shoving, hitting, and killing), psychological harm (such as teasing, hurting feelings, ridiculing, or name calling), fairness and rights (such things as stealing, breaking a promise, not sharing a toy, or destroying others’ property), and positive behaviors (things like helping another in need, sharing, or donating to charity).
Consistent with the assumptions of domain theory, children and adults distinguish between morality and convention on the basis of these criteria. Moral issues are viewed to be independent of the existence of social
norms and generalizable across contexts, societies, and cultures. Social conventions, on the other hand, are rule dependent, and their normative force holds only within the social system within which the rule was formed. Justifications people give for their criterion judgments are also in line with the distinctions that have been drawn between the moral and conventional domains. Judgments of moral issues are justified in terms of
the harm or unfairness that actions would cause, while judgments of conventions are justified in terms of norms and the expectations of authority.
There are, as one would expect, age and experience effects on the ability of people to make these domain distinctions. The youngest age at which children have been reported to differentiate consistently between
morality and convention is 2 1/2 years (Smetana and Braeges 1990). The toddlers in the Smetana and Braeges (1990) study were more likely to generalize moral issues across contexts (view such issues as unprovoked hitting of another child as wrong both at home and at another day-care setting) than they were to generalize conventions (putting toys away). They did not, however, make distinctions based on any of the other dimensions used in that study. By about age 3 1/2, however, children treated moral and conventional issues differently on the basis of several criteria, including seriousness and rule contingency, as well as generalizability.
The same study demonstrated that children are capable of making rudimentary distinctions between issues of morality and convention during the third year of life. This study and other work (Nucci and Turiel 1978;
Smetana 1981) have demonstrated that by age 4, children have developed fairly consistent and firm differentiations between familiar moral and conventional issues encountered in home or preschool settings. As children become older, their understandings of moral and conventional issues are extended beyond events with which the children have had direct personal experience to include the broad range of issues, familiar and unfamiliar alike, which constitute moral and conventional forms of social events (Davidson, Turiel, and Black 1983). Moreover, as children develop, they become better able to apply more abstract criteria, such as cross-cultural generalizability, to differentiate between issues within the two domains.
Studies that examine whether children differentiate between morality and convention have not been limited to the United States or Western contexts, but have been conducted across a wide range of the world’s cultures.
Such studies have been conducted with children and adolescents in northeastern Brazil; preschool children in St. Croix, the Virgin Islands; Christian and Moslem children in Indonesia; urban and kibbutz Jewish children and traditional village Arab children in Israel; children and adults in India; children and adolescents in Korea; Ijo children in Nigeria; and children in Zambia. (For a complete listing of these studies see Smetana 1995a or Turiel 1998a.) With some variations in specific findings regarding convention, the distinction between morality and convention has been reported in each of the cultures examined.
Only one study (Shweder et al. 1987) has claimed to have obtained data indicating that individuals within a non-Western culture (members of a temple village in India) make no distinction between morality and convention, and that result has been disputed by findings from a subsequent study (Madden 1992). In all cases, children and adolescents have been found to treat moral issues entailing harm and injustice in much the same way. Children across cultural groups and social classes have been found to treat moral transgressions,
such as unprovoked harm, as wrong regardless of the presence or absence of rules, and have viewed the wrongness of such moral transgressions as holding universally for children in other cultures or settings, and not just for their own group. This is a rather remarkable set of findings, and has to stand as one of the more robust phenomena to have been uncovered by psychological research.
Some cultural differences have been reported, however, in findings regarding children’s treatment of social conventions. In some respects, this has been due to differences in methods employed by researchers from different investigative groups in framing questions posed to children. For example, in a study conducted in northeastern Brazil (Haidt, Koller, and Dias 1994), children were presented with descriptions of rules from two cultures. In each case, one culture had a rule just like the child’s own culture, and the other did not. Children were asked to indicate whether one of the cultures was doing the better thing. Overwhelmingly, children indicated that the culture whose rule was like their own was doing the better thing. This form of question, however, does not differentiate between what children view as universally obligatory from what they view as preferred.
The fact that children in the study preferred their own rules to other ones may or may not mean that they saw it as wrong for other cultures to do things differently. Indeed, when children from the same area of Brazil were directly asked to evaluate whether it would be wrong or all right for people of another culture to engage in a given act if the other culture had no rule about the act, children universalized moral prescriptions (e.g., it would be wrong to hit). They did not, however, tend to universalize their own conventions (e.g., eat chicken with a fork instead of with one’s hands) (Nucci et al. 1996).
Failure to employ methods that directly and clearly assess criteria people use to evaluate moral and conventional acts accounts for studies failing to observe distinctions between morality and convention (Haidt et al. 1994; Shweder et al. 1987). When such methodological issues are taken into account, however, some cultural differences in treatment of social conventions still remain. For instance, Korean children and adolescents made much greater use of justifications pertaining to social status, social roles, appropriate behavior, and courtesy than is commonly observed in American children’s reasoning about conventions (Song et al. 1987). Ijo children and adolescents in Nigeria (Hollos et al. 1986), Arab children in Israel (Nisan 1987), and lower class children in northeastern Brazil (Nucci et al. 1996) affirmed the importance of customs and tradition to a greater degree than did American children. In each of these latter studies, however, issues of culture were confounded with social class. A general finding is that middle-class children worldwide appear to be more willing than lower-class children to view conventions as alterable and culture or context specific. Untangling what aspect of observed cross-country differences in children’s treatment of convention is due to culture, and what is due to cross-cultural effects of social-class hierarchy, is an issue for future research.
In sum, the overwhelming body of research evidence is consistent with the proposal that morality is conceptualized differently from convention. As I noted earlier, this finding has been used by some writers on moral development (viz., Wilson 1993) to sustain the view that morality is based on an innate moral sense. That is not the view being presented here. On the contrary, the observed emergence of morality and convention as distinct conceptual systems is to be accounted for in terms of the qualitatively differing forms of social interaction children experience in the context of these two forms of social regulation. Let’s turn, then, to an overview of the research that has looked at those patterns of social interaction.
