Some Critical Issues in the Study of Moral Development

We begin with an assumption about the necessity for a developmental theory of morality. In the course of our discussions we offer two research paradigms, one realized and other proposed, which might move the theory in that direction. First, let us begin with some orienting comments. Our starting assumption is that the proper place to begin with morality is with its structure. Aristotle divided causes into four types: formal, efficient, final, and material1 (Irwin & Fine, 1995; Stein, 2011). These may be translated into the following more current terms: structural, empirical, functional, and biological. What is important about this analysis of why? A clarification of the question of causation should help reveal what are real disputes and what are pseudo-disputes. That is, we distinguish between different psychological questions: a) what is structural or formal account of morality and moral development? b) what is a set of actual or possible efficient causes of (influences on) these structures? and c) what is the final or teleological end purpose(s) that these structures serve? Because we wish to provide a developmental account, we need to integrate these perspectives by inferring a mechanism that explains the successive structural changes by reference to efficient (empirical) and final (functional) causes. Thus, we believe the key point of entry lies in a structural analysis of moral decision-making. However, readers may question why we propose that a structural analysis provides the critical first step. Basically, we advance this assumption because we believe that (1) structure provides kind of explanation that transcends the specifics of place, time, and culture, which obviously any theory of morality needs to consider and explain and that might serve/govern moral judgment and behaviors (see “the first paradigm”), and (2) because it is consistent with other (traditional) accounts of development. Here, we follow Kohlberg (1981) although we do not advocate his method, which is too reliant on verbal facility. As to what kind of formal or structural analysis is needed, we also agree with Kohlberg that any psychological theory of morality needs to rest on a philosophical analysis of morality (e.g., Kohlberg, 1981). A psychological theory of morality without basing it on ethics would be like developing a psychological theory of music or musical appreciation without basing it on a structural analysis of music. The point to emphasize here is that a structural analysis of morality is necessary for understanding moral thought and judgment throughout the life span. In other words, one has to understand what is being acquired before proposing how it is acquired.

Whether starting with moral dilemmas is the correct entrance strategy is not settled, but in our view, focusing on the cognitive-structural essence of morality should be. Therefore, we follow Kohlberg’s strategy of analyzing the structures of children’s, adolescents’ and adults’ judgment and reasoning as they struggle with moral conflicts and dilemmas, but not his method, which we consider limited especially when examining early moral development and actual moral functioning in real contexts rather than in the restricted context of the formal interview about complex moral dilemmas. To this end, we briefly examine two problematic areas of study and offer two research paradigms that, we believe, may promote a better understanding of moral development.

The first concerns the early origins of morality. Kohlberg and his coworkers have been justifiably criticized for focusing on children’s judgments of inappropriately abstract, unfamiliar and complex moral dilemmas, which may underestimate the genuine moral reasoning of the child. Although many excellent studies, which have assessed children’s moral reasoning “in the wild,” e.g., in classroom or playground (Dunn, 1987; Smetana et al., 2012), have the advantage of allowing for more candid judgments of the situation by the children, they are limited by assessing the child’s judgment on-the-spot thus not providing an opportunity for the child to reflect on his/her own behavior. Thus, the methods do not allow researchers to assess children’s delayed reflections on the event, that is, children’s cognitive and affective residue of previously encountered events. Instead, what we thought was needed was a personalized real-life encounter by and for that child, but providing the opportunity for the child to reflect on the conflict after its occurrence since the child’s own and “owned” moral judgments are the focus of our concern. By “owned” we mean understanding the relevance of the knowledge to their own self.

Therefore, the first paradigm offers an opportunity for the individual child’s genuine responses to individualized, personally relevant moral and other kinds of contentious encounters with parents or others. (A published report of the study [Takagi & Saltzstein, 2015] is available for further details.) In brief, parents kept a diary for a week of problematic encounters. Two encounters for each child (3–5 year-olds) were chosen to present to that child later in the preschool laboratory as an “as if” encounter for them to judge verbally and nonverbally. Critically, rather than asking whether the act was good or bad, the child was asked to make believe he/she was the character in the story and whether the child would do it again, and why? All children, without exception, said “no,” often emphatically ! However, when asked to explain why they would not repeat the act, a majority of the children, no matter of their age, could not explain why they would not commit a moral transgression whereas most of the older children (4 and 5-year-olds) could explain why they would not commit a prudential one, such as running on a subway platform. That is, reasoning was congruent with the transgressive nature of prudential but not with moral transgressions. These results suggest to us that young children may not “own,” i.e., understanding should govern their own behavior. (See Takagi et al. (2018), for a preliminary report of the follow-up study). Thus, in our view, knowing something is “wrong” in the general sense, as in past children being assessed with evaluative questions and hypothetical stories, is an important step towards developing a moral sensibility, but is far from achieving one. That is, we do not equate “knowing right from wrong in the abstract” and “owning” that knowledge, in the sense that he/she understands its relevance to their own behavior. We believe that the research paradigm we have offered is one way of differentiating between the two. One might comment that this finding is consistent with Haidt’s (2001) report that people frequently express strong moral outrage, but cannot justify their outrage with reasons, what Plato called aporia. However, we place this in a developmental framework, in that this inability to justify precedes rather than substitutes, for moral reasoning2.

The second research paradigm offered is proposed. It concerns the relationship between moral reasoning and construals or interpretations of the problematic situation and by extension to the tricky issue of cultural variations in moral thinking and reasoning. Turiel (2002) and Shweder (2012), using different cultural examples and arguments, have offered basically similar explanations of cultural variations in moral thought and practice. For example, they have explained what to Westerners seem strange beliefs such as those in traditional high caste Hindu culture: The first is the belief among some Hindus that widows should refrain from eating red meat (Turiel, 2002), and the second is that a son is enjoined from having a haircut for some period after his father’s death (Shweder, 2012).

The first example is purportedly explained by reference to the belief that eating red meat excites sexual appetites, which might induce a widow to engage in sexual activity, which would interfere with the deceased husband’s well-being in the afterlife, and the second example of not having a haircut for some period after the death of a father as desecrating the memory of the deceased father. So, the argument (e.g., Turiel, 2002) is that if you believed these facts about the situation (construals), you would also see that refraining from eating meat is understandable and rests on universal moral concerns.

Although these explanations seem plausible and indeed may, in part, be true; however, theoretically they assume that the process of moral decision-making is a linear one, always or at least normally involving the following sequence: first, encountering the situation, then interpreting it to produce a construal or understanding of it, and then selecting which principle or domain applies to that construal. However, it is possible, indeed in our view likely, that the process may not always be linear but sometimes recursive. We would argue that the construal of the situation may be, indeed probably is, dependent of their moral code. That is, it is possible that the construal process sometimes occurs after, but then influences, the initial moral judgment. For example, evidence has been obtained, which suggests that moral judgments elicited by severe outcomes may induce inferences of intentionality more than less severe outcomes (Knobe, 2003, 2004).

With this as background, we now propose a second paradigm to study the phenomenon to question the implicit assumption that the construal-evaluation sequence is always or at least regularly linear and suggest that structure of reasoning (the developmental state of the child reasoner) may, indeed probably does, influence the aspects of construal or interpretation of the situation, i.e., what information is relevant for making a moral judgment. What we propose, instead of providing features of the moral situation that the researcher thinks might be relevant, is to ask children of different ages and cultural backgrounds what information they would want or need to judge a particular act as right or wrong, or perhaps (ask them) to impose a penalty about an act simply described (i.e., without context or relevant information, such as intentions). Presumably, if we are correct, the kind of information requested would systematically change with age. For example, (a) the youngest might not see the need to request any further information since rule-breaking is always and automatically wrong, but this would change with age to a different question: In contrast, (b) older children might want to know whether the act was seen by authorities (parents, teachers); (c) still older children might want to know whether the outcome was accidental or intentional, etc. Other questions might be, e.g., whether the victim was a friend or a stranger, etc. Such a method would have the advantage of eliciting what the respondent (child, adolescent, adult, etc.) thinks is relevant information rather than focus what the researchers think is relevant. Our hypothesis is that the construal or interpretive process is not independent of the kind (structure) of morality in operation. This seems to us self-evident, but has not, to our knowledge, been studied systematically.

What we have attempted to do in this brief viewpoint paper is present what we believe a theory of moral development should accomplish, specifically focusing on structural analysis. Thus, we do not offer a new “theory” of morality, but rather an outline of what such a theory should consist of and two research paradigms, one realized and the other proposed, that would help test and further develop such a theory.