UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Kohlberg and ethical universalism Yeung, Kwok Wing Anthony 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-34648X.pdf [ 12.36MB ]
Metadata
JSON: 831-1.0088492.json
JSON-LD: 831-1.0088492-ld.json
RDF/XML (Pretty): 831-1.0088492-rdf.xml
RDF/JSON: 831-1.0088492-rdf.json
Turtle: 831-1.0088492-turtle.txt
N-Triples: 831-1.0088492-rdf-ntriples.txt
Original Record: 831-1.0088492-source.json
Full Text
831-1.0088492-fulltext.txt
Citation
831-1.0088492.ris

Full Text

Kohlberg and Ethical Universalism by Kwok Wing Anthony Yeung B.A. (Hons.), The Chinese University of Hong Kong, 1990 M . Phil., the Chinese University of Hong Kong, 1992 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Philosophy We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A September 30, 1998 © Kwok Wing Anthony Yeung 1998  In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or by his or her representatives.  It is understood  that copying or  publication of this thesis for financial gain shall not be allowed without my written permission.  Department The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  C?d.  I . /ff<2-  ii  Abstract This dissertation is a study of Kohlberg's moral psychology, which is a six-stage model of moral development. Kohlberg claims that his stages form a universal invariant sequence and that they are hierarchical, i.e., higher stages are better than lower stages. Accordingly, he claims that Stage 6 morality, which centers on justice, is universally valid. This ethic of justice is embodied mainly in respect for persons, fairness, and the procedural principle of ideal role taking. Kohlberg claims not only that Stage 6 values and principles are universally valid, but also that they are determinate. In other words, reasoning in terms of these values and principles guarantees that, for each particular moral problem, there will be a distinct solution on which all morally mature people could agree. By making these claims Kohlberg is advocating a strong and traditional version of universalism, which I call 'paradigm universalism.' The dissertation is divided into five chapters. In the first two chapters I outline Kohlberg's theory and explore its philosophical implications. In Chapter 3 I discuss Kohlberg's debates with two important critics, Gilligan and Flanagan. Gilligan claims that Kohlberg's emphasis on justice rather than care indicates a gender bias in his model. Flanagan, on the other hand, argues that since morality is multifarious it is wrong to i equate morality either with justice or care of a combination of both. While these criticism do point out certain shortcomings of Kohlberg's theory, I argue that they do not seriously threaten the universal validity of Stage 6 moral values and principles in general. Chapter 4 introduces the main philosophical arguments of this dissertation. In this chapter I argue that (1) moral psychology is relevant to moral philosophy; (2) that the claim of hierarchy for the Kohlbergian stages does receive significant support from his research; and  Ill  therefore (3) Stage 6 does plausibly reflect certain universal moral ideals. At the same time I allow (4) that there is clearly certain cultural bias in Kohlberg's theory and (5) that he is excessively optimistic about the determinacy of Stage 6 moral reasoning. In the final Chapter, I reflect on the universalism-relativism debate in light of Kohlberg's theory. I argue that paradigm universalism is too strong for Kohlberg to support, and that universalism is acceptable only in a weakened form which I call 'minimal universalism.' Contrary to the hope of paradigm universalists, this minimal universalism cannot serve as a comprehensive theory for solving moral problems. Neither does it exclude all forms of ethical relativism, but it does set important limits to any acceptable relativist theory.  TABLE OF CONTENTS  Abstract Table of contents Acknowledgment Introduction Chapter 1: Kohlberg's theory — the early model Section 1. The stage model Section 2: Empirical tests Section 3: Philosophical implications Section 4: Explanatory difficulties Section 5: Summary Chapter 2: Kohlberg's theory — the late model Section 1: Modifications made in the late model Section 2: Empirical support Section 3: Postconventional moral conception Section 4: Philosophical implications updated Chapter 3: Criticisms of Kohlberg Section 1: Gilligan and the ethic of care Section 2: Reply from Kohlberg and his school Section 3: Reflections on the Kohlberg-Gilligan debate Section 4: Flanagan and pluralism Section 5: Kohlberg's reply Section 6: Flanagan's counter-arguments Section 7: Reflections on the Kohlberg-Flanagan debate Section 8: Summary Chapter 4: From moral psychology to moral philosophy Section 1: The relevance of psychology to moral philosophy Section 2: Supremacy and universality of Stage 6 Section 3: Summary Chapter 5:A reflection on the universalism-relativism debate Section 1: A reflection on Kohlberg's universalist claims Section 2: Minimal universalism Section 3: From minimal universalism to limited relativism Section 4: The study of human good Bibliography Appendices  Acknowledgment There are many people I wish to thank for their assistance and support in the preparation of this dissertation. I would like to thank my supervisor, Professor Earl Winkler, for his guidance and encouragement. He patiently went through my rough, disorganized drafts, gave advice on how to refine my arguments, and helped to shape this dissertation as it is today. I owe much to Professor Lawrence Walker, with whom I have studied moral psychology. His instruction provided me with a good foundation for understanding Kohlberg's theory and moral developmental psychology in general. I am also indebted to Professor Jerrold Coombs, who helped me clarify important ideas for understanding Kohlberg's theory and method on the one hand, and its philosophical implications on the other. I am also grateful to my proof-readers, Angela Yee, Emma Poon and Jeaneette Lim. Finally, I have to thank my wife, Carole Hang Fung Hoyan, who helped me edit my final drafts, and advised me on the writing of the dissertation. I am most grateful for her love and support.  1  Introduction The project This dissertation is a study of Kohlberg's moral theory. Kohlberg is probably the most important moral developmental psychologist of the century. What is especially interesting about his theory is that it concerns not only the phenomena of human moral development, but also the possibility of developing a certain universalist moral philosophy based on these phenomena. Kohlberg's theory and arguments suggest a strong version of universalism, although he may not have advocated it explicitly. According to this universalism, there are some ultimate and all-embracing moral principles and decision-making procedures that determine the solutions of moral dilemmas in a definite way. Two questions then follow. First, is it possible to develop any ethical theory based on empirical knowledge at all? Second, does Kohlberg's research successfully defend the form of universalism he wants to defend? If not, precisely what theory does his research support? This dissertation is an attempt to reply to these questions. I will argue that empirical studies are relevant to moral philosophy in a way comparable (though not identical) to the way in which observations of particular events support general scientific theories. But I will also argue that Kohlberg's research does not support the kind of universalism he wants to defend. The only way we can defend universalism is to weaken our universalistic claim by abandoning the idea that universal moral principles or decision-making procedures can guarantee unique solutions for all moral problems. But a kind of universalism, weakened this way, does receive significant support from  Kohlberg's research. In the course of developing this view I will explain what remains universalitc about this "minimal" universalism.  Why moral psychology That Kohlberg's research supports a weak version of universalism has great significance. We live with a mixture of many moral intuitions. When philosophers discuss moral problems, they cannot help starting from premises which they find undeniable. Different philosophers begin with different fundamental intuitions, all of which are believed by their proponents to be obvious, but people nevertheless disagree. Bentham intuits that all pleasures are good and that they differ only in quantity. Mill agrees that pleasure alone is good but insists that pleasures differ in quality as well as in quantity. Moore rejects the idea that all good is ultimately grounded in pleasure. And so on. Who is to be the final judge? If empirical data about moral development do support a certain kind of ethical theory, we can say, as physicists may do, that there is no final judge, or rather that the only judges we have are logic and the empirical data, at least ideally speaking. If I believe that time and space are absolute, I am wrong in spite of my strong intuition that they are so. In face of all the empirical evidence, it is no use for me to complain that it is unimaginable that time and space are relative. Likewise, if empirical data about moral development do support a certain kind of ethical theory, then there will be real hope that we can settle some important controversies in moral philosophy. Since Kohlberg's (1958) study, the cognitive developmental approach in moral developmental psychology has become a distinctive field of study. Kurtines & Gewirtz  say that Kohlberg was "almost single-handedly responsible for a 'cognitive' revolution in the moral development literature that paralleled the cognitive revolution that was taking place in developmental psychology during the decades of the 60s and 70s."1 Kohlberg's developmental theory is a six-stage model of moral development.2  ,  Every stage represents a "form" of moral reasoning. Two claims of Kohlberg's are especially important and philosophically interesting. First, he claims that development through these stages is universally invariant. Everybody in different cultures develops morally through the same sequence of stages. Second, when one moves from a lower stage to a higher one, the moral adequacy of one's mode of reasoning increases. From these two claims it follows that Stage 6 represents the universal destiny of moral development, and hence represents the universal moral ideal. In this sense the moral standard at Stage 6 is ultimate and supreme, and the standard constitutes a form of ethical universalism. > There are also other interesting philosophical claims in Kohlberg's theory. These include his formalism, rationalism, objectivism, and constructivism. Kohlberg regards his theory as a formalism in that moral adequacy is determined by the "form" of reasoning behind particular moral judgments. It is rationalistic in that, firstly, moral correctness is largely determined in terms of moral reasoning,  and secondly, the correct way of moral  William M. Kurtines & Jacob L. Gewirtz, Prologue to "Part I: Cognitive Developmental Perspectives'^In Moral Development: An Introduction, ed. William M. Kurtines & Jacob L. Gewirtz (Boston: Allyn and Bacon, 1995), 19. 2 Kohlberg also assumes the existence of a seventh stage. Stage 6 reasoning, being the most adequate form of reasoning, deals only with the question of what is the morally right thing to do. Stage 7 goes beyond Stage 6 in that it proceeds to address the quesiton of why one should be moral at all. It does not introduce any new moral principles or values, but accomodates morality by constructing a view about the relation between the individual and the universe: Generally speaking, Stage 7 involves an identity or unity with being, life, or with God, and moral life is an expression.such identity (Kohlberg 1984, p. 41). Since Stage 7 1  4 reasoning is nearly all one needs to be a moral person. It is objectivist in that there is a method by which we can judge the validity of a moral judgment, which is independent of personal feelings or opinions. But his objectivism takes a naturalistic form. It does not, like Plato's theory of idea, regard moral values as something that exists totally independent of the human mind or human activities. Rather, it manifests itself as a claim that morality is a human construct. He calls this view "constructivism." Though the validity of these claims is not the main concern of this dissertation, I will also discuss these topics briefly.  Gilligan and Flanagan Although Kohlberg's theory has become a powerful theory in the field of moral psychology, it is not immune from criticisms. Many argue that Kohlberg's theoretical model fails to provide a complete account of the kinds of moral reasoning people actually employ. Some challenge his methodology, and some challenge his derivation of normative claims from empirical phenomena. Some even deny that Kohlberg is studying moral reasoning at all. The MJI method (Moral Judgment Interview, the method Kohlberg uses to elicit the moral reasoning of his subjects), it is maintained, elicits people's moral justification rather than their reasoning.4 The difference between moral reasoning and moral justification is that, while the former is what the subject employs in order to reach his decision, the latter is what the subject uses to rationalize his choice after the choice is made. Therefore it can hardly be the case that Kohlberg's model is a model about the reasoning that determines moral decisions.  has nothing to do with adequacy of moral reasoning as such, I will not discuss its status in detail in this dissertation. 3 As we shall see, the second idea has been weakened with the later development of Kohlberg's theory.  5 I will deliberately avoid discussing the last of the above objections. M y project in this dissertation is to explore and critically examine the claim of universality in Kohlberg's theory. Since universalism only deals with the problem of universal moral standards,  we can avoid the intriguing problem about whether Kohlberg is measuring  reasoning  or justification.  Even if what Kohlberg measures is justification but not  reasoning, it still reveals something about the moral standards of people, for both moral reasoning and moral justification presuppose certain moral standards.  Whether one is to  reason about a moral problem or to justify one's moral decision, he has to rely on some moral standards. He can be satisfied or dissatisfied with his moral justification as much as he can with his reasoning. As far as he is dissatisfied with his moral justification, he will either throw doubts on the standards behind his justification or admit that his moral decision is not justifiable. Therefore, even if what the MJI method elicits is moral justification, it still helps to reveal the moral standards of the subjects. The first two questions, however, cannot be so avoided. If Kohlberg's account of people's approaches to moral reasoning is incomplete, then one can reasonably doubt whether his theory is universally applicable. If there is any problem in his research method or in his basic ideas about the relation between psychology and philosophy, then his arguments for his philosophical claims will not be sound. I will concentrate on two principal critics of Kohlberg, Gilligan and Flanagan. Gilligan is doubtlessly the most famous opponent of Kohlberg. She is a moral psychologist who used to work with Kohlberg but became convinced that Kohlberg's model is flawed. Flanagan, on the other hand, criticizes Kohlberg from a more philosophical perspective. They are not the only  4  This point is mentioned by Professor Coombs to me in a personal discussion.  6 opponents of Kohlberg's theory, but their criticisms represent principal objections from the views of both disciplines. By considering their objections to Kohlberg we can have a fairly complete understanding of both the strengths and weaknesses of Kohlberg's theory.  Structure of the dissertation This dissertation is composed of five chapters. In the first two chapters, I will explore Kohlberg's empirical studies, the various formulations of his theory, and their philosophical implications. In Chapter 3 I will review Gilligan's and Flanagan's objections to his theory. I will assess how well Kohlberg's views can survive these objections. Then, in Chapter 4,1 will argue in support of Kohlberg's claim for the supreme adequacy of Stage 6 moral reasoning, partially on the basis of relevant empirical research. The final chapter will be a reflection on the universalism-relativism debate in light of Kohlberg's theory and research. Based on what we have learned from the debate between Kohlberg and his critics, I will propose a minimal universalism. In doing so, I will also briefly discuss how universalism and relativism can be reconciled.  7  Chapter One: Kohlberg's theory — the early model Kohlberg's developmental theory was changing almost all through his academic life. His theoretical claims and study methods at different times are intricately related to each other, and are impossible to be summarized in a few paragraphs. In order to provide the reader with a relatively clear picture of Kohlberg's theory, I will describe his theory in terms of an early model and a late model, using Kohlberg's two anthologies as representatives of these models. The first anthology is The Philosophy of Moral  Development, published in 1981 (abbreviated as Kohlberg 1981 below). This anthology 1  is a collection of a number of Kohlberg's early papers, which were written between the years 1967 and 1979. These include most of his seminal research in moral psychology, as well as his early views on moral philosophy and moral education. The second anthology is The Psychology of Moral Development, published in 1984 (abbreviated as Kohlberg 1984). This second volume also includes a few of Kohlberg's early papers, but starting with Chapter 3 it represents the a substantially revised version of his theory, and reports studies using revised research methods. Since the publication of the (1984) anthology, the revision of the theory is largely complete, though there has still been some minor change in the description of the top developmental stage (i.e., Stage 6). For the sake of convenience, I will refer to the theoretical model as established up to the publication of the (1981) anthology as the 'early model,' while referring to the model developed since the (1984) anthology as the 'late model.' In this first chapter we shall review Kohlberg's early model of moral development. Reference will be made mainly to the (1981) anthology, but when necessary to other essays of his as well. The  Lawrence Kohlberg, The Philosophy of Moral Development (Cambridge: Harper & Row, 1981). Lawrence Kohlberg, The Psychology of Moral Development (Cambridge: Harper & Row, 1984). Part 2 of the (1984) anthology, which is co-authored with Charls Levine and Alexandra Hewer, has been published 1  2  as a monograph in 1983, namely, Moral Stages: a Current Formulation  and a Response to Critics (Basel:  S. Sarger, 1983). I will refer to this monograph by Kohlberg et al. 1983 in this dissertation.  8  first half of this chapter is a review of his psychological theory; after this we shall discuss its philosophical implications. By doing so, I wish to discuss the initial formulation of Kohlberg's theory, which sets out the basic issues and provides the historical context for the understanding of the more recent development of his theory.  Section 1: The stage model Kohlberg's research on moral development can be traced back to the late 1950's, when he conducted research on a group of American school boys for his Ph.D. program. In this research Kohlberg interviewed his subjects and asked them to solve hypothetical dilemmas in order to see how they reasoned morally.3 The project ended up with six typologies of moral conceptions. These typologies provided him with the foundation for developing his stage model of moral development.4 Kohlberg's theory then develops into a six-stage cognitive model, in which every stage represents a distinct form of moral reasoning. Each "form of moral reasoning" involves a way of resolving "moral conflicts," by which he means conflicts of claims. He then calls a principle that serves the purpose of resolving moral conflicts a "principle of justice" (Kohlberg 1981, p. 143). For Kohlberg, the relation between justice and morality is so close that they are hardly distinguishable from each other. This strong correlation between justice and morality is an important theme of his early works, and his model is often understood as a "justice" model of morality. 1.1 Stage descriptions In the (1981) anthology, each stage is defined in terms of (1) a set of criteria for judging the morally right from the morally wrong, and (2) a set of motives for moral  See Appendix 1 for a list of the hypothetical dilemmas employed by Kohlberg in his research. For details about the research see Lawrence Kohlberg, "The Development of Modes of Moral Thinking and Choice in the Years 10-16" (Dissertation, University of Chicago, 1958). 3  4  9  behavior. The stages fall into three different levels, the preconventional, the conventional and the postconventional. Below is a brief review of the definitions of the stages.5 Preconventional  Level  The morality of this level is marked chiefly by an apparent egocentrism. The Tightness and wrongness of actions are judged on the basis of their physical consequences in relation to the subject himself/herself. It is not perfectly accurate to say that people at this level are selfish, for an alternative to a self-oriented morality is still not available to them. Rightness and wrongness are interpreted "in terms of either the physical or the hedonistic consequences of action (punishment, reward, exchange of favors) or in terms of the physical power of those who enunciate the rules and labels." The preconventional level is composed of two stages. Stage 1 morality is called 'Punishment  and Obedience  Orientation.''  At this stage the subject interprets rightness in  terms of obedience and wrongness in terms of disobedience and punishment. The motive for moral behavior is simply to avoid punishment. "The avoidance of punishment and unquestioning deference to power are valued in their own right." At Stage 2, the Instrumental  Relativistic  Orientation,  rightness of actions is still  understood in terms of self-interest. However, unlike Stage 1, in which people understand rightness in terms of the avoidance of punishment alone, reward comes into play at this stage. Whatever brings about good results for the self is right. Cooperation with others in one's own interest starts to make sense to the subject. Therefore, "elements of fairness, reciprocity, and equal sharing are present, but they are interpreted in a physical and pragmatic way." Reciprocity is a matter of "you scratch my back and I'll scratch yours." The stage descriptions reported here are first published in 1970, in his paper "Education for Justice: A Modern Statement of the Socratic View," which has become a chapter of his (1981) book (see Kohlberg 1981, pp. 29-48, esp. 17-19). In reporting his stage descriptions, I have tried to retain many of Kohlberg's own words, which I put inside quotation marks. 5  10  The motivation at this stage for right behavior is "to obtain reward, have favors returned, and so on." Conventional  Level  At the second level, which Kohlberg calls the 'conventional level,' the subject's moral concern is extended from the self to the group one belongs to. Such a group can be one's family, one's nation, or a social group of any other kind. "Conformity" with the expectations of one's group, "loyalty to it," and "actively maintaining and supporting its order" are considered the right things to do. The conventional level is also marked by a recognition of the difference between intentions and immediate consequences, and the judgment of rightness and wrongness proceeds in light of intentions (see the definition of Stage 3). The conventional level includes Stage 3 and Stage 4. At Stage 3, The Interpersonal Concordance or "GoodBoy — Nice Girl"  Orientation, the central concept of moral rightness is that of social approval. What makes an action right is that it "pleases or helps others and is approved by them." There is much conformity to stereotypical images of what constitutes majority or "natural" behavior. "Behavior is frequently judged by intention — the judgment "he means well" becomes important for the first time. One earns approval by being 'nice.'" The motive for moral behavior is the avoidance of "disapproval and dislike by others." At Stage 4, the Society Maintaining Orientation, the orientation becomes one of "maintenance of the social order." "Right behavior consists of doing one's duty, showing respect for authority, and maintaining the given social order for its own sake." The motive for moral behavior becomes the desire to "avoid censure by legitimate authorities and resultant guilt." Postconventional,  Autonomous,  or Principled  Level  11  "At this level, there is a clear effort to define moral values and principles which have validity and application apart from the authority of the groups or persons holding these principles, and apart from the individual's own identification with these groups." This level has two stages, namely Stage 5 and Stage 6. At Stage 5, The Social-Contract Legalistic Orientation, right actions are defined in terms of "general individual rights" and consensus among individuals (or, in Kohlberg's own words, "standards that have been critically examined and agreed on by the whole society"). The "legal point of view" is acknowledged, but with "an emphasis on the possibility of changing law in terms of the rational consideration of social utility (rather than freezing it in terms of Stage 4 'law and order')." "There is a clear awareness of the relativism of personal values and opinions and a corresponding emphasis on the procedural rules for reaching consensus. Aside from what is constitutionally and democratically agreed on, the right is a matter of personal 'values' and 'opinion,'" and "free agreement and contract" become "the binding elements of obligation." The motive for moral behavior is "to maintain the respect of the impartial spectator judging in terms of community welfare." In short, the Stage 5 moral conception is a mixture of social contract, human rights, and utilitarianism. Stage 6 is The Universal Ethical Principle Orientation. Moral Tightness is defined by "the decision of conscience in accord with self-chosen ethical principles appealing to logical comprehensiveness, universality, and consistency. These principles are abstract and ethical (the Golden Rule, the Categorical Imperative); they are not concrete moral rules like the Ten Commandments. At heart, these are universal principles of justice, of the reciprocity and equality of human rights, and of respect for the dignity of human beings as individuals." The motive for moral behavior is "to avoid self-condemnation."  12  1.2 Features of the stages  Three features are essential for understanding Kohlberg's concept of moral stage. First, a moral stage is a structured whole, a total way of thinking. The moral stage of a subject is not determined by any particular moral judgment he/she makes, but by the salient features of his/her process of moral reasoning as a whole. This structural wholeness can be further explained by the introduction of Kohlberg's form-content distinction. He defines moral content as the judgment of what is morally right in a particular situation, and moral form is defined as the way in which a subject reasons in order to arrive at that judgment. This usage of the terms .'form' and 'content' is somewhat atypical in philosophical writings, but is crucial to the understanding of Kohlberg's so called "formalism." Kohlberg observes that individuals tend to employ the same "forms" of moral reasoning when faced with different moral problems, and hence for one particular subject there is a particular "form" of reasoning which represents his "total way of thinking." It is this total way of moral reasoning instead of the "content choices" that defines the moral stage of that subject (Kohlberg 1981, pp. 120-2). A second feature of the moral stages defined by Kohlberg is that of a universally invariable sequence. The order of the stages (from Stage 1 through Stage 6) represents the chronological order of their appearance in the moral development of human beings. Except in situations where keeping up with one's moral standards is excessively threatening to the self, people never (1) skip stages or (2) reverse stages in their development. This sequence is expected by Kohlberg to be constant in all (or almost all) individuals regardless of their society or culture. However, it is not expected that every single person goes through the whole sequence from Stage 1 to Stage 6. It is possible for an individual to stop at a particular stage in his development, failing to proceed any further. It is also not expected that every stage appears in every society or cultural group. It is possible that there exists a society in which no member is capable of exceeding a  13  certain stage of moral development. Indeed Kohlberg embraces a mild version of "social evolutionism." Some societies can be less developed (morally) than others in that their members stop their development at a lower level, and thus in them the highest stages entirely fail to exist (Kohlberg 1981, p. 128). However, whatever stage an individual reaches, he/she has to develop through the sequence of earlier stages to arrive there (Kohlberg 1981, pp. 120 & 122). A third essential feature is the hierarchical nature of moral stages. There is a normative side and an empirical side of the hierarchical claim. On the normative side, it means that later stages are better than earlier stages. Proving that the later stages are more adequate than the early ones is an important task for Kohlberg. He claims that there are a number of criteria by which the later stages can be judged to be better. He supposes that these criteria are widely accepted and need no justification. We shall discuss these criteria in detail shortly. On the empirical side, the hierarchical claim means that the sequence reflects a certain logical order. Higher stages "include lower stages as components," that is, lower stages are "reintegrated" in higher ones. People at higher stages have no difficulty comprehending moral concepts of lower stages and employing them in their reasoning, while those of lower stages do have difficulty in understanding and employing the concepts of higher stages. Very few people can understand two stages above their own modal stages, and many do not understand any stage above their own at all. Also, people tend to prefer higher stages to lower stages (or, more precisely, prefer statements that instantiate higher-stage moral reasoning to those instantiating lower-stage one) as far as they can comprehend them (Kohlberg 1981, p. 137).  14 1.3 The claim of hierarchy of the moral stages Among all the features of the Kohlbergian moral stages, their hierarchical nature is most important to our discussion. A n ultimate concern of this paper is to examine the question of whether any universal moral standards exist. The claims of structural wholeness and of universal invariable sequence are purely empirical. Although I believe that these empirical claims are not totally irrelevant to normative ethics, their relation to ethics is less direct than the claim of hierarchy. The hierarchical claim is itself strongly normative, and hence it deserves close examination. How do we judge that the stages are hierarchical? Kohlberg uses two criteria to explain this, one cognitive and the other normative. The cognitive criteria: integration  and  differentiation  Kohlberg's theory of moral development is a cognitive-developmental model. In this model, moral maturity is closely related to cognitive maturity. Higher stages are cognitively more sophisticated. Kohlberg describes these cognitive sophistications by saying that higher stages are more integrated and differentiated than the lower stages. The increasing integration and differentiation, Kohlberg claims, explain why the stages form an invariant sequence. To him, there is an inner logic in the sequence of the moral stages; the new integration and the new differentiation made at each stage logically depend on those made in its preceding stage (Kohlberg 1981, p. 137). The idea of integration is quite clearly presented in Kohlberg's writings. A higher integration means a wider scope of applicability, better equilibrium, 6 greater stability, and greater consistency (Kohlberg 1981, pp. 135-136, & 147). To explain this, we can start with the comparison of the conventional level and the postconventional level. The conventional level has its values defined within the context of a particular community, Equilibrium seems to mean the balance between the perspectives of different parties involved in a moral problem. See Kohlberg 1981, pp. 211-214. 6  15  while postconventional morality strives to define values independent of any particular culture or subculture. There is clearly a difference in the scope of applicability between the two levels. Kohlberg states that conventional morality defines "good behavior for a Democrat but not a Republican, for an American but not for a Vietnamese, for a father but not a son" (Kohlberg 1981, p. 135). This seems to be an overstatement, as the Stage 4 society-maintenance conception of morality surely can assign the rights for people of different roles in the society, and thus prescribe right behaviors for both fathers and sons. The point he actually wants to make is probably this: when there are real conflicts between different parties, there exists no higher principle available within the conventional moral conception to resolve the conflicts, especially when these parties come from different societies. This point can be made clear by considering Kohlberg's explanation of consistency. In Kohlberg's terminology, "the different definitions of right for Republicans and Democrats, for Americans and Vietnamese, for fathers and sons" are enough to constitute "self-contradiction" (Kohlberg 1981, p. 136). This makes conventional morality less consistent than postconventional morality. At first glance, the use of the term 'self-contradiction' is illegitimate. How can different definitions of the rights of different parties cause se/^contradiction? They might contradict each other, but surely nobody is contradicting himself. I believe I should follow the policy of the American government, therefore I should fight against the Vietnamese communists. Where is the inconsistency in my belief? Kohlberg does not state this clearly, but a plausible answer might go like this: Suppose I am the American soldier. My society tells me that the North Vietnamese are evil, or that they have evil intentions, and I must fight against them. Following the values of my society I see them as evil. But so far as I accept that I am right because I am following the values of my society, I have to admit that the Vietnamese I fight is also  16  right because he, too, is only doing what his society requires. But this appears to contradict the belief that the North Vietnamese, or their intentions, are evil. It is in this sense that conventional morality is inconsistent. Similarly, conventional morality is more consistent than preconventional morality. For instance, a preconventional thinker judges that it is good to do something if it benefits herself. When another person, out of self-interest, does something which conflicts with her interests, she will naturally say the other is wrong. What she may not be fully aware of is that her judgment is only based on her own interest. If she recognizes this, she will have to say that the other is also doing the right thing, because the other is also acting according to his own interest. Again there is no way for resolution unless our protagonist can draw upon a value system which takes the needs or claims of both parties into consideration. For a case like this, a morality of the conventional level could serve as such a mediating value system. From the above interpretation one can infer that the scope of applicability, stability, and consistency are very much one and the same thing. Kohlberg himself does say that universality and consistency are closely linked (Kohlberg 1981, p. 135). The reason for their close relation is apparent in the above examples. With a wider scope of applicability, a set of moral standards can handle more moral conflicts with more consistency, hence it is less vulnerable to challenge from different value systems and less subject to change, i.e., more stable. A l l these concepts are unified under the name of integration. From this point of view, the hierarchy of the Kohlbergian stages can be understood in a simple way. Preconventional morality defines morally right behavior in terms of personal interests, and has the narrowest scope of applicability. When conflicts emerge between individuals making compromise impossible to achieve, conventional  17  morality provides solutions. Conventional morality presupposes a community of some kind, e.g., a family, a society, or a social organization, where there are conventional norms that are designed to stipulate the right thing to do in different situations of conflict. People can therefore determine the right solution when conflicts arise between them. But conventional morality has its limits too. The most obvious one is that it presupposes the absolute value of a particular set of conventions and social institutions. Therefore when conflicts arise between communities, or between individuals from different communities, there is no higher authority to whom to appeal in resolving these conflicts. Thus, it is essential for moral progress that they find some universalizable standard to determine a mutually acceptable method of resolution. This is exactly what leads to postconventional morality. We can now see how higher stages are more integrated than lower stages. They are more integrated because their scopes of applicability for conflict resolution are wider. Preconventional morality is applicable to the individual only. Conventional morality is applicable only to people within a community. Postconventional morality is applicable to the entire human species. By virtue of wider scope of applicability for conflict resolution, each stage is also more consistent and stable. Moral stages are also hierarchical in that they are cumulative, i.e., the later stages include the earlier stages as their elements (Kohlberg 1981, pp. 180, 190). Sometimes Kohlberg even says that there is a "logical order" among the stages, and this logical order requires that every stage must "imply" its lower stages (Kohlberg 1981, p. 137). Such wording is highly misleading and cannot be understood literally.7 A better understanding The use of the word "imply" is misleading because it leads us to think that the definition of a morally right behavior at a lower stage can be logically derivedfromthat at a higher level. If this is so, what is regarded as morally right at a lower stage would be necessarily regarded as right at a higher level too. But this is obviously not the case in Kohlberg's theory. What is admitted as morally right at a lower stage can be, and is often, rejected at a higher stage due to the change in the moral conception held by the subject. 7  18  of the cumulative nature of stage progression is made possible by Kohlberg's statement that every stage is a structure which "includes elements of earlier structures but transforms them in such a way as to represent a more stable and extensive equilibrium." Seen in this light Kohlberg's talk of implication or logical order has at least two meanings. The first is that the moral standards represented by the lower stages are "available to, or comprehended by" the higher-stage thinker, though the thinker shows a preference for the higher stages over the lower stages (Kohlberg 1981, p. 137). The second meaning is that the values of the lower stages are in one way or another adopted by the higher stages, though the values then function quite differently from the way they did at the lower stages. Kohlberg illustrates this by comparing Stages 4 and 5. I said that the superiority of Stage 5 is partly a cognitive superiority, that the judgments of Stage 5 are more cognitively complex (differentiated) and more cognitively inclusive than Stage 4 judgments (inclusive meaning that Stage 5 ideas include Stage 4 ideas as elements or parts). I cited, as an example, the cognitive perspective of Stage 5 as compared to Stage 4.1 claimed that Stage 5 ideas arose from a social contract, utilitarian, "prior-to-society" law-making perspective, while Stage 4 judgments arose from a "member-of-society" law-maintaining perspective. As contrasted to Stage 5, at Stage 4 the authority of laws does not rest on free contract but rests directly on divine, natural, or societal authority. Laws are not judged functionally as revisable in the light of maximizing utility or public welfare, but maintaining laws is necessarily utilitarian in preventing disorder. In including Stage 4 considerations of authority and functions of law, the Stage 5 perspective is cognitively better (Kohlberg 1981, p. 190). He says here that Stage 4 morality is a law-maintaining.perspective, while Stage 5 morality represents a law-making perspective. The recognition of the law-making perspective implies the recognition of the law-maintaining perspective, for it is useless to make law without having the law maintained or followed. What distinguishes the lawmaking perspective from the law-maintaining perspective is that, the law-making perspective recognizes that law is made by human beings to serve human goals. As it is  Lawrence Kohlberg, "The Claim to Moral Adequacy of a Highest Stage," 632.  19  made to serve human goals, it can also be violated or changed when doing so serves human goals better. It seems that the Stage 4 subject may possibly be aware that laws are made for human goals as well. However, as he/she treats the laws as societal authority, he/she seems to be unable to take into consideration the point that the legal system should be adjustable to human goals. In this sense Stage 5 is cognitively better than Stage 4. The recognition that law is made to serve human goals deprives law-maintenance of its ultimate value status. Nevertheless, it allows law-maintenance to retain some of its value as far as it carries out its proper function. The social order, for the Stage 5 reasoner, is what people have agreed upon, or would agree upon to the extent that they are rational, in forming social rules of cooperation so as to enable everyone equally to pursue his/her own good. If a law clearly fails in this function, there is no reason to maintain the law. Kohlberg does not illustrate how each stage includes the elements of its predecessors. However, it is not hard to understand how this works. Stage 2 is an orientation of exchange for one's own good. This orientation still acknowledges the value of avoidance of punishment, since avoidance of punishment is certainly a kind of good. Stages 3.and 4 are oriented to social norms and social order, but they do not negate the value of self interest. Conventional morality may sometimes require a person to sacrifice his/her personal interests, but it surely does not define the pursuit of personal interest as a moral evil. The idea of increasing differentiation, on the other hand, refers to the increasing separation of "ought" from "is," and the separation of non-moral value judgments from moral ones (Kohlberg 1981, p. 135). Consider the conception of right at different stages. At Stage 1, having a right is confused with having power or authority, and obligation (i.e., what one should do) is defined as demand from external authorities. At Stage 2, the  20  concept of right and the concept of obligation are differentiated from external authority. Having a right is now defined as having the ability to control one's own behavior and one's possessions. On the other hand, an obligation is understood as what is required in order to achieve one's goal. By defining right and obligation in this way, Stage 2 is still confusing right and the ability to control, and obligation is confused with or limited to prudence. At Stage 3, people are able to separate right from ability to control on the one hand, and obligation from means to desired ends on the other. They define rights in terms of group expectations. According to their understanding, one has to earn his/her rights by doing things that his/her society expects him to do. For instance, one has the right of property because one earns it by working hard. Obligation is defined in terms of the social role one plays. Despite these advances, the Stage 3 conception of right is still confused with expectations and demands assigned to the individual according to his/her role. Stage 4 transcends Stage 3 by separating rights and obligations from particular role expectations. There are rights which are awarded by society to particular members according to the roles they play, but there also exist rights which represent freedoms that all members of society are entitled to, and the latter have priority over the former. Obligation is likewise defined in terms of (1) one's membership in society, and (2) the responsibility of the role one voluntarily enters. By introducing rights and obligations that one has due to his/her membership in society, Stage 4 differentiates right and obligation from any particular role expectations. However, they are still confused with societal expectations. At Stage 5, the subject proceeds to separate these concepts. Rights become something prior to society. Freedom is valuable for its own sake, and should not be limited by society unless it is incompatible with the like freedom of others. Obligation is defined in terms of a rational concern for the welfare of others. At Stage 6, there are universal rights of just treatment in addition to liberty rights. Rights represent universalizable claims that each individual can make on  21  others. Obligation is defined in terms of respect for others' rights. By defining obligation as concern for the welfare of others, Stage 5 reasoning is somewhat incapable of differentiating supererogation from duty, whereas Stage 6 reasoning sharpens the demarcation (Kohlberg 1981, pp. 215-219). The claim that higher stages are better than lower ones because they are more differentiated is itself not particularly interesting. Since 'more differentiated' means greater ability to distinguish moral values from non-moral ones, saying that the more differentiated is better amounts to saying that it is better to distinguish moral values from non-moral ones than to mix up the two. What is more interesting, however, is his particular view about what constitutes non-moral values: self-interest at level 1, and social norms at level 2. The presupposition behind his claim that greater differentiation makes a stage better is that personal interest and social convention themselves do not provide moral justification to a behavior. Of course such a presupposition is a normative one, and therefore the criterion of differentiation is not purely cognitive. But now let us go to another topic, namely the normative criteria. The normative criteria: prescriptivity,  universalizability  Kohlberg emphasizes that higher stages are not merely cognitively superior, but also normatively so. To argue for the normative superiority of higher stages, he begins with the formalist's convention that an adequate moral judgment must have certain formal properties. These are, essentially: prescriptivity, primacy, and universality (Kohlberg 1981, pp. 135, 171, 191).9  In many places Kohlberg also mentions "reversibility" and "equilibrium" as formal properties which determine the adequacy of certain moral judgments. However, it seems reasonable to suppose that these properties are not independent of universality. To the extent a moral judgment is universalizable, it should be equilibrated and reversible as well.  9  22  Though Kohlberg does not explicitly say so, prescriptivity and primacy of moral judgment come together to form a parallel with differentiation. What prescriptivity means is that moral judgment represents "a distinct concept of an internal duty" (Kohlberg 1981, p. 191). Thus understood, "prescriptivity" seems to mean something close to what philosophers mean by 'categoricalness.'10 Primacy, on the other hand, means the "superiority of moral considerations over non-moral ones" (Kohlberg 1981, p. 191). The two combine to form a distinct deontological conception of morality. This is what the criterion of differentiation is all about. Differentiation is the subject's ability to separate moral from non-moral values, and make moral judgments only with reference to moral values. Universality, on the other hand, parallels the criterion of integration. Given that integration has much to do with the scope of applicability and equilibrium, as we have already seen, the parallelism between universalizability and integration is clear, and needs no further elaboration here. Universalizability means that a moral judgment has to be one which "all people can act on" (Kohlberg 1981, p. 191). This obviously echoes the Kantian conception of morality presented by his Categorical Imperative. Kohlberg is confident that these formal requirements of moral judgments are fairly noncontroversial (Kohlberg 1981, p. 173). Granted these requirements, Kohlberg infers that higher stages are more adequate as they yield moral judgments which conform more completely to these formal requirements. In Kohlberg's view, prescriptivity and universality are closely related. A highly prescriptive principle implies valuing something independently of other factors. This independence in turn implies that the value of the particular thing at stake is recognized under all circumstances, which entails universality. Consider Dilemma I A (I call it This use of the word 'prescriptivity' sounds, of course, a little strange for philosophers. The word 'prescriptive' is usually contrasted with 'descriptive', and has no necessary connection with "internal duty". What we usually regard as related to internal duty are rather concepts like "unconditionality" or "obligatoriness". To be faithful to Kohlberg's own wording, 1 will still use the word 'prescriptivity' to refer to the formal character of moral judgments that they are expressions of internal duties. But the reader should bear in mind that it means something like unconditionality or obligatoriness. 10  23  'Bob's dilemma,' see Appendix 1 for details of the dilemma). When considering whether Joe (=Bob in the dilemma quoted in our appendix) should tell on his brother, a Stage 1 subject deliberated in terms of the probabilities of Joe getting beaten up by his father or by his brother. His answer was a typical preconventional response because it had its basis in personal interest, which showed little or no awareness that an adequate moral judgment should be universalizable. He does not answer with a moral judgment that is universal or that has any impersonal or ideal grounds. In contrast, a Stage 6 subject said that one should keep one's promise "regardless of who it (sic) was." By saying this he implies that a moral judgment should be universally applicable. Stage 6 subjects would also say things like "Morally I would do it in spite of fear of punishment," implying that a moral obligation should be an ideal (more or less) independent of personal interests. When asking "Is it morally right?", Stage 6 subjects are asking something different from the question about avoiding punishment (Stage 1), prudence (Stage 2), conforming to authority (Stages 3 and 4), and so on (Kohlberg 1981, p. 170). The same is true for people's reasoning about the moral value of human life. Kohlberg observes that the value of human life becomes increasingly independent of the factual properties of the life in question, which means for him that the "prescriptivity" of the value of human life increases with the stages (see Appendix 2). Similarly, and closely related to prescriptivity, the order of the stages also shows an increase in the universality of the moral value of human life. At Stage 1, only the lives of important people are valued. At Stage 3, only the lives of those who are close to the subject are important. But at Stage 6, everyone's life is taken seriously (Kohlberg 1981, p. 135). "With each stage," writes Kohlberg, "the obligation to preserve human life becomes more categorical, more independent of the aims of the actor, of the commands or opinions of others, and so forth" (Kohlberg 1981, p. 171).  24  To the extent that higher stages reflect a higher awareness of the notion that a moral judgment should be universalizable, and that higher stages better disentangle moral values from non-moral ones, they better instantiate the properties of universalizability and prescriptivity. When he says that higher stages better disentangle moral values from nonmoral ones, Kohlberg has presupposed some particular view about what moral values are and what they are not. The above discussion indicates that for Kohlberg neither prudence nor conformance to authority is a moral value. For him, moral values have much to do with equality, fairness and respect, which we shall see more clearly later. 1.4 Explaining stage progressions  How do stage progressions happen? Kohlberg says that the key factor determining stage progressions is disequilibrium. As we have seen in section 1.3, Kohlberg claims that the reason the stages form an invariant sequence is because the higher stages are more consistent and equilibrated. What is implicit is that, since higher stages are more equilibrated, a person will develop into his/her +1 stage (the stage next to his/her own) i f (1) he/she is aware that his own stage is not equilibrated enough, and (2) the +1 stage is cognitively available to him/her. The result is, of course, that this subject will eventually move into his/her +1 stage. Two things result from this hypothesis. First, the more a subject is aware of the limits of his/her own moral stage, the more likely he/she will move to a higher stage. Consequently, the opportunity for role-taking becomes an important factor affecting a person's rate of moral development. The reason is that opportunities for role-taking provide chances for a subject to discover the limitations of his stage (Kohlberg 1981, pp. 140-141). Second, the availability of higher stages to a subject is also an important determinant of his/her development. The factor of availability is in turn determined by  25  two factors: the cognitive ability of the subject and his exposure to moral reasoning of higher stages. It can easily be imagined, then, that moral development will be affected by various factors. Related to cognitive ability we have the educational level of a person; related to the involvement in role-taking we have his/her peer participation (Kohlberg 1981, p. 142), his/her involvement in free discussions about moral problems with others (as in larger social institutions or just in the family) (Kohlberg 1981, pp. 141-142), and experience in leadership." Some later studies also find that group discussions will more effectively prompt stage development if advice from a higher-stage point of view is given to the participants. Even social class serves as a factor of the rate of development (Kohlberg 1981, pp. 141-142). Social class and the justice structure of a society influence the opportunity for role-taking of its members, and hence influence their rates of development (Kohlberg 1981, pp. 141-144).12  Section 2: Empirical tests As a psychologist, Kohlberg's mission is not just to compare a set of forms of moral reasoning at the conceptual level, but to have his theory empirically confirmed. Attempts have been made to test the core assumptions of Kohlberg's theory, i.e., the claim of universal invariant sequence, and the claim of hierarchy.  2.1. Universal invariant sequence To be sequential, the order of the stages must represent the chronological order of the emergence of the stages in the moral development of individual human beings. A sequence is more than an age trend. It is not enough to show that older people tend to be " It is supposed by some supporters of Kohlberg that leadership requires one to solve conflicts among subordinates in a way that yields more opportunity for role-taking than non-leaders typically have. 12 One may wonder why social class should influence opportunities for role-taking. A plausible explanation is: middle class parents are in general more educated than lower class parents, which in turn influences their ways of educating their children.1  26  at higher stages. People of higher stages must have progressed through all of the lower stages. There should be, in normal situations, no stage skippers and no regressors. Also, let us recall that the invariant sequence does not require that everybody develops at the same rate, nor does it require that everybody eventually reaches Stage 6. Since the sequence is claimed to be universal, it should be applicable to all cultures. In his "Stage and Sequence: the Cognitive-Developmental Approach to Socialization," Kohlberg reported a study in which 50 American boys originally aged 1016 were studied over a 12-year period, which is a followup on his dissertation. Though the data analysis was not yet complete when the paper was published, the result of the research suggested that the subject's patterns of moral reasoning form a general sequence that Kohlberg's model predicts.13 But the support of Kohlberg's theory from the research was far from perfect. Even with the incomplete analysis of the data, exceptions to the sequence were found. Some subjects showed stage regressions at the end of high school. Since virtually all these subjects later on continued advancing to the higher levels, Kohlberg believes that this anomaly can be explained by adding a transitional stage to the sequence between Stages 4 and 5. 1 4 In the same paper, Kohlberg reports a series of cross-cultural studies conducted in Taiwan, Great Britain, the United States, and two villages in Mexico and Turkey. 1 5 In the USA, Taiwan, and Mexico, use of both Stage 1 and Stage 2 reasoning dropped strikingly with age (from 10-16). Uses of Stage 4 to Stage 6 reasoning, on the contrary, increased with age. In the isolated villages, the cases were similar in that the use of higher-stage reasoning increased with age, while that of lower stages decreased with age. The difference between these isolated villages and the urban areas was that Stages 5 and 6 13 Lawrence Kohlberg, "Stage and Sequence: the Cognitive-Developmental Approach to Socialization." In Handbook of Socialization on Theory and Research, ed., David A. Goslin (New York: Rand McNally, 1969), 383-431. 14 Ibid, 388. 15 Strangely enough, the research data for Great Britain are not reported in his paper.  27  were virtually absent in these villages. 1 6 Since these studies are cross-sectional instead of longitudinal, they only show that the stages represented age trends, but not that they are in sequence. Nevertheless, age trends can still be regarded as a weak support of the sequential claim. Since these studies were carried out across different cultures, they supported the claim that the sequence was universal (if the age trend is really a result of the sequentiality of the stages). One may question how Kohlberg can make any universal claim of moral developmental sequence if two of the stages in the sequence were absent in some cultures.17 Kohlberg has an easy reply: the universality claim for the sequence does not require every single stage to be present in every culture. What is required is only that (1) the six stages provide a comprehensive account for the forms of moral reasoning in individuals in all cultures; and (2) in all cultures there are no stage skippers or regressors. Kohlberg's universality claim does not exclude the possibility of having an individual or a group stopping at any particular stage in the course of his or their moral development. Since the two isolated villages did show an increase of use of higher stages, and a decrease in the use of the lower stages, with the increase of age, the findings are consistent with confirm Kohlberg's theory (see Kohlberg 1981 pp. 126-127). Not only is the sequence of development independent of the society in which it occurs, it is also independent of social class and religion. Kohlberg's studies showed that middle-class children and working-class children move through the same sequences (though middle-class children move faster), and that there is no important difference in  Ibid., 384-385, also see Kohlberg 1981, pp. 24-25. For example see Don Locke, "A Psychologist among the Philosophers: Philosophical Aspects of Kohlberg's Theories." In Lawrence Kohlberg: Consensus and Controversy, ed. Sohan Modgil and Celia Modgil (Philadelphia, The Falmer Press, 1986), 26. 16  17  28  the development of moral thinking among Catholics, Protestants, Jews, Buddhists, Moslems, and atheists (Kohlberg 1981, p. 25). 1 8 2.2 The hierarchy claim  The empirical support for the hierarchical claim is also rich. For Kohlberg and his supporters, the hierarchical claim implies that (1) people comprehend all the stages below their own stages but not those higher then theirs (except the +1 stage), and (2) they prefer judgments of higher stages than those of lower stages as far as they can comprehend them. These predictions were confirmed by various studies.19 Rest (1973), for instance, conducted an experiment on people's preferences for the stages.20 He found that people, when given various suggestions for solving moral dilemmas, preferred suggestions from stages above their own. The higher stages are judged better than their own stages.21 The experiment also tested the subjects on their ability to understand the principles at different stages. The result showed that the majority of people experienced no difficulty in understanding stages lower than their own, but had difficulty understanding those above them. The greater the discrepancy between the tested stage and the subject's own stage, the more difficult it was for the subject to understand it. This  Since Kohlberg does not provide us a systematic analysis of the research data, the systematicity of this research remains a question. 19 At first glance, (1) and (2) together imply that no one will stay in their own stage very long if they understand their +1 stage. In fact, a subject's comprehension of a higher stage does always prompt him to progress into that stage. This progression appears as a process in which the subject increases the usage of his/her +1 stage and reduces using his modal stage in his moral judgments, and finally gets rid of his/her original stage entirely. But such a process is usually quite slow. 20 James R. Rest, "The Hierarchical Nature of Moral Judgment: A Study of Patterns of Comprehension and Preference of Moral Stages," Journal of Personality 41 (1973): 86-109. 21 Ibid., 102-103. In Rest's study there is no marked difference between people's preferences for their +1 stages and +2 stages. But as we shall see in Chapter 2, with later modification of Kohlberg's model and the scoring manual, people typically prefer their +1 stages but not their +2 stages. 18  29  confirmed the hypothesis that the higher stages are cognitively more complex than the 22  lower stages.  Since the hierarchical claim means that the higher stages are better than the lower stages, and this superiority is connected with the higher stability and better equilibrium of the higher stages, the following studies that tested the stability of judgments from different moral stages can also be regarded as confirmations of the hierarchical claim. The first is Blatt and Kohlberg's (1975) test.23 The study was carried out in a Reform Jewish Sunday school class. The experimenter taught a group of 30 students aged 11-12. Thirteen of these students were randomly chosen for a pre-test to determine their moral stage. Among these 13 children, 11 were available for a post-test offered by the experimenter right after the program, and 10 of them were available for a follow-up test a year later. Among the 11 children in the post-test, six were boys and five were girls. In the experiment, the experimenter organized a teaching program for the children, which lasted for 12 weeks, involving a total of 12 hours of discussions. In these sessions, hypothetical moral dilemmas were provided for class discussion. Free expression of opinions was encouraged, and the experimenter helped to point out the psychological and social dimensions of the experiences of different parties involved in the dilemma. The result was that 63% of the experimental children moved up one stage or slightly more, and 9% moved up half a stage. In the follow-up test, the results remained essentially the same.24 In part 2 of the same study, which was carried out on a larger scale to assure the reliability of the first part, 132 subjects were chosen from a suburban Chicago school Ibid, 233. Moshe M. Blatt & Lawrence Kohlberg, "The Effect of Classroom Moral Discussion upon Children's Level of Moral Judgment," Journal of Moral Education 4, no. 2 (1975): 129-161. 24 Ibid, 137.  22 23  30  system and tested in essentially the same way as part one. In this part of the study, only 7 out of 46 (15%) of the subjects in the main experimental group had moved upward by one stage or more after the class sessions. In a second experimental group, in which discussion was mostly leaderless, 6 out of 47 (13%) of the subjects moved upward for one stage or more. The percentage of subjects showing upward movement was far less than in part one. Though the upward stage movement in part 2 was less prominent than part 1, Blatt and Kohlberg still claim that the result was all in all a replica of the first part. They thus conclude that rational discussion is an important factor that stimulates moral development.25 Another experiment relevant to the claim of increasing equilibrium was Erdynast's experiment conducted in 1973. 26 In the experiment, Erdynast tested 20 adult business executives and 10 graduate students. The subjects were asked to solve two moral dilemmas. One dilemma involved distribution of income in a new cooperative business. Two solutions were open to the subject: a "utility" solution and an "equity" solution (as Erdynast calls them). The other dilemma was the suicide mission dilemma (Appendix 1). Similar to the first dilemma, the subject had to choose between a "utility" solution (to order the demolition man to stay behind) and an "equity" solution (to go for a lottery). After a pre-test, the subjects were asked to take the viewpoint of each party in the situation, and were also asked to decide on a solution imagining that they did not know which of these parties they would become in that situation. A few days later they were tested again. The result was that all the Stage 6 subjects in the pre-test agreed in their choice on the equity solution in each of the dilemmas, and did not change their choice in the retest. Ibid., 153. A. Erdynast, "Improving the Adequacy of Moral Reasoning: An Exploratory Study with Executives and with Philosophy Students" (Ph.D. diss. Harvard University, 1973). My report of the experiment is based on Kohlberg's review in Kohlberg 1981, pp. 211-4. 25  26  31  In contrast, subjects at Stage 4 and 5 disagreed in their solutions. Some chose the utility solution while others chose the equity solution. In the retest, after the role-taking process, many of the subjects who initially chose the utilitarian solution changed their minds and preferred the equity solution. By shifting to a view with a greater emphasis on equity, the type of moral reasoning they used in solving the dilemmas had shifted to that of a higher stage. This, Kohlberg says, demonstrated that the lower stages were less equilibrated than the higher stages. The experiment thus supported the idea that (1) higher moral stages are superior since they are in better equilibrium than the lower stages; (2) such differences in equilibrium can be defined in terms of the different degrees of reversibility — which means "the capacity of moral principles to generate concrete judgments that do not vary across the perspectives of the actors in the dilemma" (Kohlberg 1981, p. 279), or the acceptability of one's own judgment when one trades off the position with the other party in a moral situation; and (3) the awareness of the lack of reversibility of one's own moral stage motivates stage progression because of the higher reversibility of the higher stages. There is still another type of studies that is related to the hierarchical claim, though perhaps less directly. Since the superiority of the higher stages is in part a  cognitive superiority, one can thereby infer that the higher cognitive dimensions of higher moral stages is one aspect of the superiority of the higher stages. This is confirmed by the deVries and Kohlberg (1977) test, which found that children aged five to seven who passed a moral reasoning task at Stage 2 passed a corresponding task of logical reciprocity or reversibility. However, many (52%) children who passed the logical task 27  did not pass the moral task.  27  R. deVries & Lawrence Kohlberg, "Relations Between Piaget and Psychometric Assessments of  Intelligence." In Current Topics in Early Childhood Education vol. 1, ed., L. Katz (Norwood N.J.: Ablex,  1977), 119-137. Also see Kohlberg 1981, p. 138.  32  Another study supporting Kohlberg's claim about the relation between moral and cognitive stages is the Kuhn et al. (1977) study. The study showed that all adolescents and adults using Stage 5 or 6 reasoning were capable of formal reasoning on the Inhelder and Piaget pendulum and correlation problems. Many adolescents and adults capable of the latter showed no Stage 5 or 6 moral reasoning.28 Another group of studies related to the hierarchical claim concerns the relation between moral stages and moral behavior. Kohlberg's stage model is, after all, a device to measure moral maturity. To see how morally mature a person is, the best thing to do is just to watch his/her behavior. I may forcefully argue for the moral maturity of higherstage subjects, but if they often behave poorly or more poorly than lower-stage subjects, then one can hardly believe that higher stages indicate moral maturity at all. On the other hand, if we see that higher-stage subjects usually behave better than lower stage ones, then we might feel that higher-stage subjects are generally more mature, and can therefore reasonably conjecture that there is some linkage between higher-stage moral reasoning and moral maturity. But, of course, if we want to see whether one's behavior is moral, we at least have to find some situations in which the morally right action can be identified with certainty. Below are a few examples. A study has been conducted in the so called "Milgram Situation." In 1963, Stanley 29  Milgram conducted an experiment to investigate people's obedience behavior.  40 male  subjects were recruited by newspaper advertisements and were paid to participate in a "learning experiment." In the experiment, the subject worked in a lab with an experimenter where they were training a learner to memorize words. The learner was actually an accomplice. The subject was told to read words to the learner and see how D. Kuhn, J. Langer, L. Kohlberg, & N. Haan, "The Development of Formal Operations in Logical and Moral Judgment," Genetic Psychology Monographs 95 (1977): 97-188. 29 Stanley Milgram, "Behavioral Study of Obedience," Journal ofAbnormal and Social Psychology 67, no. 4 (1963): 371-378. 28  33  well he could remember them. When the learner made mistakes, the subject was required to administer an electrical shock (which was fake), and to increase the intensity of the shock when mistakes repeated. In Milgram's study, no subject stopped before the electric shock reached 300 volts, at which point the victim "kick[ed] on the wall and no longer provided answers to the teacher's multiple-choice questions."30 Although shocks between 315 volts and 420 volts were labeled "Danger: Severe Shock," 26 of the subjects (65%) continued on up to 450 volts, i.e., 30 volts beyond the maximum of the "danger-severe shock" range. At this point the experimenter called a halt to the session.31 Most subjects showed nervousness when they went on to the high voltage shocks, but they failed to quit. Kohlberg supposed that the failure to quit represents a weakness in one's moral character, and predicted that subjects of higher stages should be more ready to quit than subjects of lower stages. He tested the subjects' moral stages before they participated in Milgram's study in order to see whether there was correlation between moral stage and quitting. And there was: 75% of Stage 6 subjects quit (Kohlberg does not tell when they did so), as compared to only 13% of all the subjects at lower levels who did so (Kohlberg 1981, pp. 44-45). Another study conducted by Krebs in 1967 showed a similar relation between moral behavior and moral stages. 100 sixth-grade children were given experimental cheating tests.32 The result was that 75% of the preconventional and conventional subjects cheated. In contrast, only 20% of the Stage 5 and Stage 6 subjects did so. A similar study was conducted at the college level. Only 11% of the Stage 5 and Stage 6  Ibid, 375. ' Ibid, 376. . 32 For details of the experiment please see R. L. Krebs, "Some Relationships Between Moral Judgment, Attention, and Resistance to Temptation" (Ph.D. diss. University of Chicago, 1967). 30 3  34  subjects cheated, while up to 42% of the students at the lower levels did so (Kohlberg 1981, p. 44). Honest behaviors are, therefore, predicted by stages of moral judgments. Haan, Smith and Block conducted a survey in 1968, studying a group of students at the University of California at Berkeley on their participation in a free speech movement.  The situation was that the Board of Regents of the university were  enforcing a rule banning the use of university grounds for political recruitment and the distribution of political material. The students organized a sit-in to contest the rule as a violation of civil rights. Haan, Smith, and Block administered moral judgment interviews to over 200 of these students. Only 10 % of the Stages 3-4 subjects sat in, whereas 50 % of the Stage 5 subjects did so. Among Stage 6 students, the issue was apparently clearcut: 80 % of them sat in. To summarize, empirical studies suggest that higher stages are preferred by people over their own stages, represent higher equilibrium and reciprocity, and to a considerable extent predict moral behavior (assuming, of course, a common consensus concerning morally correct behavior in the experimental situations). Therefore, there seems to be significant empirical support for the claim that the moral stages constitute a hierarchy.  Section 3: Philosophical implications Certain ideas of Kohlberg's are of special philosophical interest. Firstly, he claims that what largely distinguishes the morally right from the morally wrong choice is the "form" of moral reasoning, and not the actual content of the choice. Because of this Kohlberg later writes that "formalism" is an assumption of his theory (Kohlberg et al. 1983, p.81). We pointed out in section 1.2 that Kohlberg's use of "formalism" is not typical among philosophers. Of course Kohlberg is not supposed to use the word in the  33 N. Haan, Smith, M. B., and Block, J., "Political, Family and Personality Correlates of Adolescent Moral Judgment," Journal of Personality and Social Psychology 10(\ 968): 183-201.  35  same way as philosophers do, but since we are discussing his theory from a philosophical point of view, it is important that we distinguish the different uses of word 'form' by Kohlberg and by philosophers. Because Stage 6 subjects typically agree on right action, and because Stage 6 is characterized by a particular "form" of reasoning, it is reasonable to suggest that they agree on the solution because they share the same "form" of reasoning. If this is true, then the correct "form" of moral reasoning does determine the correct solution. By investigating what the Stage 6 moral "form" includes we will see what Kohlberg's normative ethical theory is like. Secondly, his theory is a strong version of universalism. To Kohlberg, morality is universal in that there exists a general form of reasoning which represents the highest degree of moral maturity, without being relative to any particular cultural setting or particular value system. This form of moral reasoning is applicable to all moral situations and capable of determining uniquely correct solutions in all or most of them. Also, to the extent that this form of moral reasoning does determine content, there are universal and substantive moral values. Thirdly, his theory is a rationalism with two manifestations. On one hand, it is manifested in the claim that moral maturity is a matter of reasoning correctly. In this claim he believes that his studies support a Socratic view of virtue: virtue is knowledge of the good, and therefore "he who knows the good chooses the good" (Kohlberg, 1981, p. 189). On the other hand, his rationalism is presented as if it is a kind of principlism, which claims that moral reasoning is basically a matter of application of principles to cases. The principles he attempts to support by his studies are the Golden Rule, the Categorical Imperative, and the principle of equal respect for persons. However, as I will argue, although Kohlberg always emphasizes the importance of principles in adequate moral reasoning, his theory is not principlism in a proper sense (Chapter 5, section 2).  36  Lastly, Kohlberg's theory is also relevant to the objectivism-subjectivism debate. His theory is a version of constructivism, in which morality is viewed as a human construction devised to resolve conflicts of claims. Therefore morality to him is subjective (or better intersubjective) in the sense that it is a kind of human construct. It is, on the other hand, objective in the sense that there exists an adequate or ideal form of moral reasoning and, for particular cases, there exist right courses of action, which are determined by this form of reasoning and which are to this extent independent of any particular  subject's personal values.  3.1 Universalism  Kohlberg's objection to relativism No defense of universalism is complete if not accompanied by a reply to relativism. In Chapter 4 of his (1981) anthology, Kohlberg spends much effort undermining relativistic arguments. These arguments are not attempts to falsify relativism conclusively, but are attempts to show that relativism is less attractive as an account of the diversity of moral beliefs in different cultures than it seems. Kohlberg stresses that the concept of morality is not a behavioral one, but a philosophical and ethical one. Therefore, a psychological study of morality must orient itself towards philosophical concepts of morality. However, he also sees that empirical study of moral development can contribute to the establishment of an ethical theory (Kohlberg 1981, p. 102). Thus his method of assessing a moral theory, whether psychological or philosophical, is a mixed one. For him, neither isolated empirical study nor philosophical contemplation is a satisfactory way of providing solid ground for a moral theory. Before fruitful empirical psychological work on moral development can be undertaken, there must be a moral philosophic clarification and justification of the terms moral and development. Philosophy, then, must be present before starting  37 adequate e m p i r i c a l w o r k i n p s y c h o l o g y . I also c l a i m , h o w e v e r , that m o r a l  p h i l o s o p h y needs the w o r k o f the e m p i r i c a l m o r a l p s y c h o l o g i s t . F i n d i n g s o n m o r a l d e v e l o p m e n t c a n support o r refute i n i t i a l p h i l o s o p h i c assumptions about m o r a l  d e v e l o p m e n t a n d so c a n help revise a n d correct t h e m w h e n they f l y i n the face o f e m p i r i c a l r e s u l t s . . . . P s y c h o l o g i s t s c a n , a n d s h o u l d . . . m o v e b a c k a n d forth  between p h i l o s o p h i c a n d e m p i r i c a l f i n d i n g s . ( K o h l b e r g 1.981, x x x i )  H i s v i e w about the r e l a t i o n s h i p between m o r a l p s y c h o l o g y and m o r a l p h i l o s o p h y is then comparable to R a w l s ' s m e t h o d o l o g y . R a w l s calls h i s m e t h o d o l o g y i n m o r a l p h i l o s o p h y " r e f l e c t i v e e q u i l i b r i u m , " w h i c h is a c h i e v e d b y means o f m u t u a l adjustment between o u r m o r a l p r i n c i p l e s a n d considered j u d g m e n t s o n particular m o r a l p r o b l e m s .  34  With  K o h l b e r g , ' t h e m u t u a l adjustments between p r i n c i p l e s and j u d g m e n t s b e c o m e s the m u t u a l adjustment between m o r a l theory and the w a y w e c l a s s i f y a n d assess p e o p l e ' s m o r a l reasoning. E t h i c a l theory p r o v i d e s us w i t h a conceptual f r a m e w o r k to c l a s s i f y m o r a l reasoning, but h o w p e o p l e ' s reasoning actually develops also helps us to reflect o n and revise o u r ethical theory. K o h l b e r g talks as i f there is a b i g difference between the study o f m o r a l development a n d other s c i e n t i f i c research. H o w e v e r , I b e l i e v e that h i s research project is not m u c h different f r o m that o f other sciences, such as p h y s i c a l science. B e f o r e f r u i t f u l e m p i r i c a l research o n p h y s i c a l science c a n be d o n e , the p h y s i c i s t also needs some theoretical assumptions w h i c h l i e b e y o n d and i n f o r m the issue he/she is w o r k i n g o n . W h e n a p h y s i c i s t generalizes f r o m his/her e x p e r i m e n t a l results that the relative speed o f light is the same f o r a l l observers, he/she has supposed, f o r e x a m p l e , that the universe i s a u n i f o r m one, i.e., that things h a p p e n i n g i n different places at different times f o l l o w the same set o f l a w s . T h i s a s s u m p t i o n is p r i o r to the experiment c o n d u c t e d b y the p h y s i c i s t , as is the general c o n c e p t i o n o f m o r a l i t y that K o h l b e r g has i n m i n d p r i o r to h i s research o n moral psychology.  35  B y s a y i n g that p s y c h o l o g i s t s c a n a n d s h o u l d " m o v e b a c k and forth  between p h i l o s o p h i c and e m p i r i c a l f i n d i n g s , " K o h l b e r g e v i d e n t l y accepts that ethical 3 4  John Rawls, A Theory of Justice (Oxford University Press, 1972), 20f.  35  The parallelism between the relation ethical theory and natural science w i l l be further explored in Chapter  4.  38  theory can somehow be confirmed by empirical observations. For Kohlberg, then, an adequate moral theory must pass two sets of tests, one philosophical and one empirical. The Four Confusions of the Relativist  Kohlberg believes that relativism fails both kinds of tests. On the empirical side, he asserts, relativism is incorrect in claiming that there are no universal values. On the philosophical side, the relativist has made a fallacious inference from the factual claim that there are no universally accepted values (which, indeed, is not even true) to the normative statement that there are no universally valid values. Kohlberg claims that there are, first of all, some universal categories of moral judgments (Appendix 3; Kohlberg 1981, p. 117). He also claims that there are 25 aspects of moral judgments which are common to all cultures.36 Kohlberg does not tell us how he arrives at these claims, nor is the truth of these claims obvious. Nevertheless, Kohlberg makes use of these universal aspects of moral judgments to design his research program. From the 1950's to the 1970's, he conducted various research, some longitudinal and some cross-sectional, in different cultures, using his MJI method. People's answers to his hypothetical moral dilemmas were grouped into six types, with respect to these aspects of morality. 37 He finds (1) that these types form a universal sequences; and (2) that they are structural wholes. (A person at stage 6 on the value-of-human-life aspect is also likely to be at stage 6 on the moral-motivation aspect.) These findings made him believe that he is entitled to call the types "stages" (Kohlberg 1981, p. 120). Kohlberg's argument against relativism can be summarized as follows:  In The Philosophy of Moral Development, only two examples of these aspects are given, i.e., the motivation for moral action and the value of human life (Kohlberg 1981, pp. 19-20). For a complete list of the 25 basic moral aspects, please see Kohlberg's "Stage and Sequence," 378-379. The list is also presented as the table of moral aspects in Appendix 4 of this paper. 37 Lawrence Kohlberg, "The Development of Modes of Moral Thinking and Choice in the Years 10-16," 384-428. 36  39  (1) There are universal moral concepts, values and principles. There also exists a universal sequence of moral development. (2) The sequence is composed of hierarchical moral stages. Individual and cultural differences in forms of moral reasoning are differences of stage or developmental status. (3) Since these individual and cultural differences are differences in developmental status, they are not morally neutral or arbitrary. Certain forms of moral reasoning are more adequate or more mature compared to others, and there exists a form of moral reasoning which is the most adequate of all. In this way, Kohlberg can admit that people have different moral beliefs but still maintain that there exists some universal scheme for comparing the adequacy of the moral principles they hold (Kohlberg 1981, pp. 126-127). Below is a review of his reply to the relativist arguments, followed by a consideration of his positive claims about universalism. Following Brandt, Kohlberg distinguishes between cultural relativism and ethical relativism. According to Brandt, ethical relativism consists of three beliefs: (1) moral principles are culturally variable in a fundamental way; (2) such divergence is logically unavoidable since there are no rational principles and methods that can reconcile observed divergence of moral beliefs; and (3) people ought to live according to the moral principles they themselves hold. Brandt calls "a person who accepts the first principle a cultural relativist," and "reserves the term ethicafrelativism for the view that both the first and second principles are true" (Kohlberg 1981, pp. 106-107).  38  Also see R. B, Brandt, "Toward a Credible Form of Utilitarianism," Contemporary Utilitarianism, ed., M. D. Bayles (Garden City, N.Y.: Anchor Books, 1968), 433. People may question why Brandt should reserve the term 'ethical relativism' only for the view that accepts the first two principles but not all three. The answer is, I suggest, that the third principle is not essential to ethical relativism. Once we accept that disagreements in moral principles are incommensurable, we are ethical relativists. It does not matter whether we believe that people should thus follow their own principles. Indeed, If one does not believe people should live up to their own principles, one can be a relativist in an even more radical sense. 38  40  Among the above three theses of relativism, the first thesis, i.e., the thesis of cultural relativity, is an empirical fact and there is nothing controversial about it (though there can be disagreement over how fundamental cultural variability is). Kohlberg's objection against relativism, therefore, is mainly concerned with the second and third theses. Kohlberg argues that the belief in ethical relativity is based on various confusions. The first confusion is what Kohlberg calls the "naturalistic fallacy." The term was first introduced by G. E. Moore in his Principia Ethica to refer to the fallacy of defining ethical concepts in terms of any complex of natural properties. The reason why this is a fallacy is demonstrated by the so called "open question argument." "[Wjhatever definition be offered, it may be always asked, with significance, of the complex so defined, whether it is itself good." 3 9 Suppose one wants to define "good" in terms of what is "pleasant." Moore points out that it is always possible to ask, "Is pleasure good?", and we are not asking whether pleasure is pleasant. We surely do not regard the question "Is pleasure good?" and the "Is pleasure pleasant?" as one and the same question. However, if "pleasant" defines "good," or if "good" just means "pleasant," then the two questions should be identical. 40 By the same token, not only is it impossible to define "good" as "pleasant," but it is also impossible to define "good" in terms of any natural property at all. The predicate "good," says Moore, is "simple, unanalysable, indefinable." 41 Whatever property P a thing has, we can always ask whether it is good to have the property P, and the question is always meaningful. Likewise, Frankena applies the concept of the naturalistic fallacy to reject any attempt at reducing ethical terms to non-ethical terms.42 G. E. Moore, Principia Ethica (New York: Prometheus Books, 1988), 15. Note, however, that Moore does not deny the possibility that the question "What is good?" can be answered, as far as the question is not one of definition, but of co-extensiveness. In other words, it is possible that we can find a property P such that all good things have P and everything having P is good. Moore's idea is only that, even if we can find such a P, "good" still does not mean P. 41 Ibid., 36. 42 William K. Frankena, Ethics, (Englewood Cliffs, N.J.: Prentice-Hall, 1963), 81-82. 39  40  41  In general, the impossibility of exhaustively defining value terms by reference to natural properties yields the fact-value dichotomy. According to this dictum no set of purely descriptive statements can logically entail a normative statement. Given that different cultures have different values, one can always ask whether it is a good thing that a culture has the values it has. In this sense, the relativists' inference from the factual statement "Every person/society has their own values" to the normative statement "Every person/society ought to have their own values" is not warranted (Kohlberg 1981, p. 107). The second confusion of the ethical relativists is the tendency to conflate the liberal value of tolerance with ethical relativity. Kohlberg observes that many social scientists believe in ethical relativism because they think that such relativism is necessary if one is to be fair to other cultures and minority groups (Kohlberg 1981, pp. 107-108). However, Kohlberg maintains, this is not true. To be fair to another culture, one need only be tolerant of their values and deeds.43 To say that a certain value system is more adequate than another is not to say that people living in accordance to the inferior value system are more blameworthy, much less is it to say that one should or has the right to punish them or treat them badly. Indeed by saying that we should be fair or tolerant to other cultures we are advocating the non-relative value of tolerance. Therefore we cannot jump from tolerance or fairness to ethical relativism (Kohlberg 1981, p. 110). The third confusion is the confusion between ethical relativism and value neutrality or scientific impartiality (Kohlberg 1981, p. 112). Kohlberg's point is that some social scientists mistakenly think that to be value neutral one has to adopt ethical relativism. Ethical relativism itself is not value neutral, for it states that any value system actually held by any society is correct or cannot be rationally criticized. This is itself a Whether we should be tolerant of every value and deed of them remains a question. Though Kohlberg does not specify, it seems possible that there is, or should be, a limit to the extent of this tolerance. In other words, there may be some practices of other cultures which we need not or should not tolerate. Cannibalism, among others, is a highly plausible example of such practices.  43  42  value judgment. It "prejudges" that there is no "culturally universal criteria that might aid in defining the field of the moral" and that "variations in cultural evaluations may not themselves be assessed as more or less adequate or moral in terms of some universal criterion" (Kohlberg 1981, p. 112). The fourth and last confusion is the confusion between being scientific and being value neutral. To be scientific one must be responsive to facts and to the connections between them. One should be careful, during research and observation, not to read one's own values or value laden expectations into the facts. Yet this does not mean that there are no evaluative criteria which can be used to make any judgment at all about the observed facts. By conflating a scientific attitude with complete value neutrality, social scientists are apt to avoid any value judgment at all. Such an attitude, together with the confusion between value neutrality and ethical relativity, forces them to adopt ethical relativism (Kohlberg 1981, p. 113-1.14). In summary, according to Kohlberg, all or most arguments for relativism include one or more of the following four confusions: (1) the confusion between the alleged fact that there are no universally accepted values and the norm that everybody should have their own values; (2) the confusion between ethical relativity and the liberal value of tolerance of people of different cultures and different moral beliefs; (3) the confusion between ethical relativism and "value neutrality or scientific impartially"; and (4) the confusion between being scientific and being value neutral.  The Kohlbergian Claim of Universalism The above objections serve the goal of undermining certain arguments for ethical relativism. They are attempts to show that, even if the cultural relativists' premises are accepted, there is no logical necessity for us to accept ethical relativism. After this, Kohlberg pushes further by presenting his reason for believing in universalism.  43  He starts with the claim that, despite the beliefs of many sociologists, there are culturally universal values. He makes up a list of twelve such values: (1) life, (2) property, (3) truth, (4) affiliation, (5) erotic love and sex, (6) authority, (7) law, (8). contract, (9) civil rights, (10) religion, (11) conscience, and (12) punishment (Kohlberg 1981, p. 117, also see Appendix 3 of this dissertation). Unfortunately, Kohlberg explains neither how he reaches the conclusion that all these values are universal, nor in exactly what sense they are universal. In addition, clear definitions of these values are not provided either. Taking for granted that there are certain very general universal values, Kohlberg argues that the only possible way of defending relativism is "to say that the ordering or hierarchy of these values is relative" (Kohlberg 1981, p. 123). However, Kohlberg rejects even this, believing in some universal ordering of these values. It might be true that people rank these values differently, but such differences are "primarily reflections of developmental stages in moral thought" (Kohlberg 1981, p. 123). Because different stages of moral development are not equally adequate, the different rankings are not equally adequate either. As mentioned before, Kohlberg claims that there are two universal aspects of moral phenomena. First, people in all cultures go through the same order or sequence of stages in their moral development (Kohlberg 1981, p. 126). Second, people at the highest level of moral development share a set of moral principles independently of their culture (Kohlberg 1981, pp. 127-128). The philosophical implication of these two points taken together is that there exists a form of moral reasoning (defined in terms of a set of moral principles, according to Kohlberg) which is the most adequate without being relative to any particular culture or value system. On the basis of this, Kohlberg claims that there exists a single highest virtue, which he calls 'justice.' To be just is, Kohlberg says, to respect and maintain the rights of individuals. This conception of morality as justice "[t]o  44  a large extent... coincides with a culturally universal definition of morality." In his view, "a culturally universal definition of morality can be arrived at if morality is thought of as the form of moral judgments instead of the content of specific moral beliefs" (Kohlberg 1981, p. 300). 3.2 Kohlberg's moral " f o r m "  What then is this ideal form of moral reasoning? To answer the question, we have to first find out what Kohlberg means by the word 'form.' Philosophically, moral form means the set of characteristics that distinguishes moral from non-moral judgments. A typical list of these characteristics include consistency (does not entail P and not P); prescriptivity (judgments are made in a mode that is expressed by words like 'should' or 'ought'); and universality (if a moral rule is valid it is valid for everybody). Some of Kohlberg's moral forms, such as universalizability, consistency, and prescriptivity (note that Kohlberg's meaning of'prescriptivity' is very similar to 'unconditionality'), are very close to these forms defined by philosophers. 44 In this sense a "universal moral form" is contrasted with the basic universal "content principles" (Kohlberg 1981, p. 126).  But more often Kohlberg's "moral form" means something very different from the above sense. According to this second meaning, moral form is contrasted with the actual judgment one makes about the correct solution to a moral problem. Moral form, thus understood, refers to the reason behind one's solution to a moral problem (Kohlberg  See Kohlberg 1981, p. 191, where he introduces prescriptivity, universalizability, and primacy as the formal conditions; and Kohlberg 1981, p. 279, where he introduces universalizability, consistency and reversibility as the formal conditions. Because of Kohlberg's emphasis on disequilibrium as a factor leading to moral development, it may be desirable that we add to the list the condition of equilibrium or reversibility as weil. But whether we actually add it to the list is not important, for, as we shall see shortly, it is already included in the requirement of universality. 44  45  1981, p. 120). Moral form therefore includes one's principles, values, and the procedural factors in one's moral reasoning. A l l these factors can be quite substantive (though usually highly abstract) in the eyes of a philosopher. When Kohlberg says that moral maturity is determined by the moral form but not content, he is talking about moral form in this sense. Thus Kohlberg's claim that moral form determines moral adequacy should not be confused with ethical formalism. Formalism is the view that the correct form of moral judgment (universalizability, consistency, and unconditionality) guarantees the correctness of the judgment itself. Kohlberg's claim is much weaker. His claim is merely that what is important for moral adequacy is not the action choice itself but the reason behind it. This does not amount to ethical formalism in the typical philosophical sense. What we can say about Kohlberg's moral principles is, then, not that they are formal, but that they are substantive and yet highly abstract. Stage 6 moral principles do not directly prescribe any particular actions, as the Ten Commandments do. Three principles are central at Stage 6 moral reasoning: (1) the principle of justice, (2) the principle of role-taking, and (3) the principle of respect for persons (Kohlberg 1981, p. 162). The principle of justice requires that we recognize "the right of every person to an equal consideration of his claims in every situation," regardless of any particular social convention or legal code. The principle of role-taking requires that we put ourselves in others' situations in order to reach a solution that is acceptable to all parties provided that they are willing to do the same. The principle of respect for persons can be represented by  46 Kant's formula: Act so as to treat each person as an end, not merely as a means (Kohlberg 1981, p. 164). Although Kohlberg's theory is, strictly speaking, not an ethical formalism, it resembles ethical formalism at least in one way. Like ethical formalism, it emphasizes the importance of abstract criteria or standards in the assessment of moral adequacy, and the possibility of deriving distinct solutions of moral problems from these criteria. For Kohlberg, Stage 6 moral standard is sufficient for determining a correct moral decision in solving moral dilemmas: Not only are Stage 6 principles designed to be acceptable to all rational people, but all those who were using Stage 6 methods and principles will eventually agree on the "right" solution in concrete situations, our empirical data suggest. The people we have studied whose reasoning is at Stage 6 have agreed on the dilemmas we have presented them. In contrast, Stage 5 people all agree on certain dilemmas but not on others (Kohlberg 1981, p. 193). To see how particular moral judgments are, with Kohlberg's theory, derived from abstract moral standards, let us take the principle of ideal role-taking as an example. Kohlberg describes ideal role-taking as a second-order use of the Golden Rule: In essence, Stage 6 is a second-order conception of Golden Rule role-taking. Judgments are formed by role-taking the claim of each actor under the assumption that all the other actor's claims are also governed by the Golden Rule and accommodated accordingly. The steps involved in making a decision based on Stage 6 role-taking are (1) to imagine oneself in the position of each person in the situation and to consider all the claims that a rationally self-interested person could make in each position; (2) then to ask which claims one would uphold and which claims one would relinquish, imagining that one does not know which person in the situation one is to be; and (3) then to formulate a moral judgment in accordance with the fully reversible claims (those that one would uphold not knowing who one was to be) (Kohlberg 1981, p. 281). If all the actors involved in a dilemma were to follow this hypothetical three-step procedure, they would necessarily give equal consideration to the claims of all before reaching a decision. Moreover, deliberating under these constraints, they  47 would almost inevitably reach agreement with one another as to what the fair resolution of the situation would be. Thus, at Stage 6, the criteria of consistency and reversibility are fully met, because one and only one resolution would be agreed on and this resolution would be accepted as fair by all concerned. At the same time, the criterion of universalizability is fully satisfied, because a roletaking procedure based on the idea that equal consideration should be given to the claim of all whose interests are at stake commends itself as a principle we would want everyone to follow in resolving moral dilemmas. This is what it means to say that Stage 6 is fully adequate moral principle (Kohlberg 1981, p. 281) When applying the Golden Rule in this way, one asks whether one can still acknowledge one's claim if one trades places with the other party (Kohlberg 1981, p. 203-204). Consider Heinz's dilemma as an example (see Appendix 1). The druggist's claim to withhold property at the expense of the other's life is irreversible in that he cannot recognize such a claim in the wife's role. In contrast to this, the wife's claim to life can be recognized even when she switches roles with the husband or the druggist. Therefore a Stage 6 reasoner judges that it is right for Heinz to steal the drug for his wife (Kohlberg 1981, p. 167). The application of the Golden Rule aims at a solution to a moral dilemma which is acceptable to all parties, and this acceptability to all parties determines what a just solution is. Stage 6 morality is a method of moral reasoning aiming at the resolution of conflicting claims that is acceptable to all parties involved if they are rational. As we have seen from the example of Heinz's dilemma, by the mere requirement that a morally right action must be one which is acceptable to all, we are able to decide what the right course of action is when faced with particular cases. Another example is Kohlberg's treatment of capital punishment. Kohlberg maintains that capital punishment is, whatever the situation, unjust or morally wrong. His argument runs like this: the ideal role-taking at Stage 6 involves a decision on what social  48 justice principle would be chosen if the decision is made by rational, self-interested people behind Rawls's famous "veil of ignorance" (Kohlberg 1981, p. 281). Capital punishment is irreversible in the sense that i f one turns out to be the capital offender, he/she would be executed, an outcome which he/she surely cannot accept. One might argue that if one turns out to be an ordinary citizen, then he/she will benefit from the deterrent effect of capital punishment. Then he/she surely has a reason to vote for capital punishment. Kohlberg's reply is that such a consideration still cannot justify capital punishment for two reasons. First, imagining ourselves behind the veil, we can see that the risk of being executed is very high if we turned out to be the capital offender. However, "our risk of being murdered i f we turned out to be an ordinary citizen would be much lower" (Kohlberg 1981, p. 287). Under this consideration, "it would never be rational to prefer one's prospects under capital punishment to one's prospects under an alternative system, no matter how great a deterrent effect the death penalty might have" (Kohlberg 1981, p. 286).  The second reason is that "from an original position point of view, the penalty of death would probably seem more severe than necessary to maximize deterrence." "[W]e must consider briefly the presuppositions about the motivation of prospective criminals that are implicit in the concept of deterrence i t s e l f . . . . At least in theory, a potential criminal acting rationally would be unlikely to decide that a particular crime was not worth risking his/her life to commit but was worth the risk of a lesser but extremely severe penalty such as mandatory life imprisonment without parole. Even if some people contemplating the commission of a crime would refrain from doing so only if they feared  49 execution, the overwhelming majority of rational individuals would probably be deterred just as effectively by a punishment less harsh than death" (Kohlberg 1981, pp. 287-288). We have seen that Kohlberg's theory implies that there is a set of supreme moral principles that are universally adequate. Following Paul Taylor's terminology, we may say that Kohlberg's theory is a universalism but not an absolutism. Absolutism, according to Taylor, comes in two forms. The first form of absolutism views that "at least some moral norms are justifiable on the grounds that can be established by a cross-cultural method of reasoning and that, consequently, these norms correctly apply to the conduct of all human beings." This kind of absolutism amounts to what he calls "universalism." In the second sense, absolutism means that there are moral norms that "have no exceptions."45 Clearly absolutism in the first sense, i.e., universalism, makes a much weaker claim than absolutism in the second sense. One can be an absolutist in the first sense without being an absolutist in the second sense. For clarity and convenience, I will use the word 'absolutism' to refer only to Taylor's absolutism in the second sense, while referring to absolutism in his first sense as 'universalism.'  Using 'universalism' and 'absolutism' in this way, we can say that Kohlberg's ethical theory is universalistic but not (generally) absolutistic. Kohlberg's universal principles do not prescribe certain types of action unconditionally. Instead they provide certain method of moral deliberation, based on some basic moral values such as equality and fairness. Kohlberg believes that one will be able to find out the best solution in a moral situation if one upholds the Stage 6 principles and carefully carries out the  50 procedure of ideal role-taking. We do not need any absolute rules of the type of the Ten Commandments. Regarding substantive values, the most notable feature of Kohlberg's theory is that the endorsement of human rights is essential for moral maturity. Richard Shweder presents Kohlberg's value system as one which holds out liberalism and the American Bill of Rights as the ideal of morality and social justice (quoted from Kohlberg's report in Kohlberg et al. 1983, p. 104). Whether Shweder's reading of Kohlberg is accurate remains a question (Kohlberg et al. 1983, pp. 104 f), but he is quite right in that Kohlberg is sympathetic to liberalism, at least in the Rawlsian form of it, and to the Bill of Rights.  Kohlberg says that a Stage 5 conception of rights is that every person has a right to liberty prior to society, and society has an obligation to protect it. Stage 5 subjects believe that freedoms should be limited by society and law only when they are incompatible with the like freedoms of others. At Stage 6, having rights means there are universal rights of just treatment that go beyond liberties and that represent universalizable claims of one individual on another (Kohlberg 1981, pp. 215-216). In this sense, "[t]he procedural arrangements called constitutional democracy can make law and society attractive to rational members because they rest on their consent, provide equal representation for their self-interest, and include a Bill of Rights protecting their individual liberties (natural rights prior to law and society)" (Kohlberg 1981, p. 154). Looking at Kohlberg's value system from this point of view, Kohlberg's sympathy for the Bill of Rights is clear. Paul Taylor, "Ethical Relativism." In Moral Philosophy: Selected Readings, ed, George Sher (Fort  51 By reviewing Kohlberg's solutions to the hypothetical dilemmas, we can see more about how Kohlberg ranks human values. In Heinz's dilemma we see that the value of human life is more fundamental than property, and the value of human life is equal for all persons (Kohlberg 1981, p. 200). In the doctor's dilemma, however, we see that the value of autonomy is even more fundamental than the value of life, and thus the doctor ought to let the patient die as she asks (Kohlberg 1981, p. 204). In the life-boat dilemma, we can see how Kohlberg understands fairness and equality. What the captain should do, according to Kohlberg, is to go for a lottery (Kohlberg 1981, p. 208). The solution shows that, with Kohlberg, to be fair is to consider everybody equally. Nobody can be considered  more important  than others. Morally  speaking, everybody weighs the same. The only qualification for being treated equally is being a person. Therefore in the life-boat dilemma, the life of the old man should not be valued less only because he is old and is less able to contribute to the chance of team's survival. Therefore, a lottery should be drawn to decide who is to go overboard. In this sense, a fair system of punishment should not be understood as giving someone his desert. To understand punishment this way is to exercise only conventionallevel reasoning (Kohlberg 1981, pp. 257-259, 261-262). As we have seen in Kohlberg's argument against capital punishment, the murderer and the possible victims should be considered on a par, i.e., their claims are weighed equally.  Worth: Harcourt Brace Jovanich College Publishers, 1987), 157-158.  52 3.3 Constructivism and Rationalism Kohlberg's theory represents the "cognitive-developmental approach" of moral developmental psychology. Under this approach, moral development is viewed as a process in which an active subject struggles for a better equilibrium in his interaction with his environment. What is changed when the subject advances through the moral stages is the cognitive structure of their conception of the self in relation to society (Kohlberg 1981, p. 134).  There are two important philosophical implications of a model like this. The first is what we might call 'constructivism.' In this view, "moral judgments and norms are ultimately understood as universal constructions of human actors that regulate their social interaction, rather than as passive reflections of either external facts (including psychological states of other humans), or of internal emotions" (Kohlberg 1981, p. 134). Since morality is a matter of equilibrium of the interaction between an active agent and his/her environment, the environment provides for the most part opportunities for the agent to explore his/her own moral nature, not a context for implanting values. By saying this, Kohlberg probably has underestimated the role of internalization in human moral development. It will be more plausible to claim, following Kohlberg's main idea, that the moral agent is an active agent, and moral development is not totally a passive process of adopting existing social values. A moral problem emerges when there are conflicts between people's claims. A moral principle is, therefore, a principle which prescribes resolutions of conflicting claims. This is the reason why Kohlberg believes that there is  53, one and only one virtue, namely justice, for justice is by definition giving each her due in situations where interests or claims compete. The second implication is rationalism, which is manifested in two dimensions. One dimension is what we might call 'principlism,' while the other concerns the relation between knowledge and virtue. His principlism claims that there are some principles that are central to moral reasoning. As I have mentioned, three principles are central to his theory. They are the principle of justice, the principle of role-taking, and the principle of respect for persons. He also refers frequently to Kant's Categorical Imperative and the Golden Rule. By applying these principles correctly, he believes, one can derive the correct courses of action concerning concrete moral issues. Kohlberg's rationalism is further seen in his identification of moral virtue with moral knowledge: "[Vjirtue is the knowledge of the good. He who knows the good chooses the good" (Kohlberg 1981, p. 189). This is certainly an overstatement, for we see even in Kohlberg's own tests that hot all Stage 6 reasoners quit in the Milgram study. But a mild rationalist claim still gains support from two empirical phenomena. Firstly, it is found that subjects have to be cognitively mature to reason morally, though cognitive maturity does not guarantee moral maturity (Kohlberg 1981, p. 138). Secondly, it is also found that the "amount of opportunities for role-taking" has a great effect on the possibility of the subjects promotion in moral stages (Kohlberg 1981, pp. 141-142).  54  Section 4: Explanatory Difficulties Though there is positive empirical support for Kohlberg's theory, counter evidence has also been found. The most threatening finding to this early model is that there are considerable numbers of stage skippers and regressors. Most notably, about 20% of college youths experienced an "apparent" regression from Stage 4 to Stage 2 (Kohlberg 1981, p. 130).46 In the Blatt and Kohlberg classroom discussion experiment (1975), study two, both stage regressors and skippers were found. Taking the two experimental groups together, 5 of the 12 stage progressors (42%) were stage skippers, while 5 out the 97 subjects (5%) in the two experimental groups were regressors. While the percentage of regressors was small enough to be attributed to scoring errors, the percentage of skippers are surely too high to be neglected. Besides these, Holstein's (1974) study of the influence of parenting styles on the moral development of children also showed that there was a significant number of stage skippers and regressors.47 A l l these findings powerfully challenges Kohlberg's sequential claim. Unless Kohlberg can find ways to explain away these anomalies, his theory will be falsified by such counter evidence.  48  Support for the claim of structured-wholeness is not very strong either. As Lawrence Walker points out, Kohlberg's (1969) study showed that "only 45% of reasoning was scored at the modal stage and with considerable dispersion over the other  Also see Lawrence Kohlberg and R. Krammer, "Continuities and Discontinuities in Childhood and Adult Moral Development." Human Development 12 (1969): 109. 47 Constance B. Holstein, "Moral Judgment Change in Early Adolescence and Middle Age: A Longitudinal Study," paper presented at the biannual meeting of the Society for Research in Child Development, Philadelphia, March 1973. Also see William Kurtines and Esther B. Greif, "The Development of Moral Thought: Review and Evaluation of Kohlberg's approach," Psychological Bulletin 81, (1974): 467. 48 For still more counter evidence of the sequential claim, please see C. B. Holstein, "Irreversible, Stepwise Sequence in the Development of Moral Judgment: A Longitudinal Study of Males and Females," Child Development 47 (1976): 51-61; D. Kuhn, "Short-term Longitudinal Evidence for the Sequentiality of Kohlberg's Early Stages of Moral Judgment," Developmental Psychology 12 (1976): 162-166; and Kuhn et al., "The Development of Formal Operations in Logical and Moral Judgment," Genetic Psychology Monographs 95 (1977): 97-188.  46  55  stages (e.g., 21% of reasoning was scored at two or more stages from mode)." 4 y The support of the claim of structured-wholeness from empirical data was far from satisfactory. Difficulties also appeared in studies of the relation between moral reasoning and behavior. In Haan, Smith & Block's (1969) study of participation in the Free Speech Movement (this chapter, section 2.2), as many as 60% of the Stage 2 subjects sat in, a percentage much higher than that of conventional reasoners (10%), and indeed even higher than that of Stage 5 reasoners (50%). The first attempt by Kohlberg to explain these unexpected findings is to suggest that there exists a transitional stage, between Stage 4 and Stage 5. People who apparently regressed from Stage 4 to Stage 2 were not real regressors, because these subjects eventually moved to Stage 5. Kohlberg thus suggests that there exist a transitional stage between Stage 4 and Stage 5, which he calls Stage 4!/2 (Kohlberg 1981, p. 130). People at this stage are in a stage of confusion. They have abandoned their conventional moral ideas, but have not yet established the postconventional values, and because of this seem to regress back to Stage 2 instrumental hedonism. After they overcome their confusion, they will move into Stage 5 (Kohlberg 1981, pp. 45 & 130).50 Unfortunately the attempt to save the theory fails. As Kohlberg later admits, Stage  4Vi can only account for a few cases of stage regression. Furthermore, the existence of 51  Lawrence Walker, "The Development of Moral Reasoning," Annals of Child Development vol. 5 (1988): 39. 50 The transitional Stage 4'/2 is defined as follows (Kohlberg 1981, p. 411): This level is postconventional but not yet principled. At Stage 4 A, choice is personal and subjective. It is based on emotions, conscience is seen as arbitrary and relative, as are ideas such as "duty" and "morally right." ... At this stage, the [social] perspective is that of an individual standing outside of his own society and considering himself as an individual making decisions without a generalized commitment or contract with society. One can pick and choose obligations, which are defined by particular societies, but one has no principles for such choice. 51 Anne Colby, " A Longitudinal Study of Moral Development," Monograph of the Society for Research in Child Development 48, no. 4 (1983): 5.  49  l  56  Stage 4/4 does not help to explain the existence of stage skippers at all. The upshot is that there must be problems in Kohlberg's project, either in the stage model itself or in the research method (particularly the scoring manual). However, it surely does not mean that Kohlberg's theory is completely wrong. If it is completely wrong, there should be no correspondence whatsoever between the Kohlbergian moral stages and moral behavior. This explains why, on the one hand, Kohlberg's theory has not been totally abandoned by developmental psychologists, and, on the other hand, why it needs revision to survive.  Section 5: Summary To summarize, Kohlberg's theory is a mixture of moral psychology and philosophy. On the psychological side, it is a six-stage model of moral development. At Stage 1, people reason about moral problems in terms of avoidance of punishment and obedience to authority. At Stage 2, they reason in terms of self-interest and exchange. Stage 3 subjects extend their concern to others on an interpersonal base. The moral standard is largely that of winning the approval of others, and therefore the fulfillment of role expectations is important. At Stage 4, the moral conception is that of social maintenance. Stage 5 moral standard transcends the societal perspective and views individuals as prior to society. It thus relies on human rights, social contract,-and social utility. Stage 6, the highest stage in moral development, upholds the principle of justice, the principle of role-taking, and the principle of respect as the ultimate standard. Three claims are especially important for these stages. First, every stage is a structural whole. Second, they form a universally invariant sequence. Third, the stages are hierarchical. On the philosophical side, Kohlberg's theory is a non-absolutistic version of universalism. It proposed that the most adequate "form" of moral reasoning involves the employment of three principles, namely the principle of justice, the principle of roletaking, and the principle of equal respect for persons. The substantive value embraced by  57  the theory is chiefly that of human rights, in which the right to liberty or autonomy is most important. He believes in fairness, but does not believe in desert. As a universalist, Kohlberg argues that (1) there exists a universal pattern of development, and (2) all subjects at the highest stage share the same basic set of ethical principles. It is important to note that, for Kohlberg, these principles determine right actions when applied to particular cases. He also argues that the acceptance of ethical relativism is the result of confusion. The confusion is (1) the confusion between the fact that people do not agree in their moral judgments and the idea that no basis of argument can be formed; (2) the confusion between the universal value of tolerance and the relativist view that there is no universal value; (3) the confusion between ethical relativism and scientific impartiality; or (4) the confusion between being scientific and being value neutral. Kohlberg claims that higher stages are better than lower stages in a cognitive and a normative sense. They are cognitively better since they are more integrated and more differentiated; they are normatively better since they are more prescriptive (in his sense) and more universalizable. Four empirical findings are relevant to the normative claim that higher stages are more adequate than lower stages. First, higher-stage reasoners comprehend the lower stages, but lower-stage reasoners cannot comprehend the higher stages (except that some subjects can comprehend their +1 stages). Second, people prefer moral judgments of the highest stages which they comprehend, and this remains true even when their highest comprehended stages are higher than their modal stages. Third, reasoning of higher stages requires higher cognitive ability. Fourth, higher-stage reasoning predicts better moral behavior. The first two empirical findings are regarded as the direct consequence of Kohlberg's hierarchical claim. The last two findings, I argue, are related to the claim that higher stages are more adequate than lower stages, although they may not be directly related to the hierarchical claim as understood by psychologists..  58  Empirical tests of the claim of invariant sequence show mild support for the claim. However, there is also considerable counter-evidence to it. As a result, modifications have been made since the mid 1970's to improve the predictive power of the theory.  59  Chapter Two: Kohlberg's theory — the late model  Section 1: Modifications made in the late model In the first chapter we reviewed Kohlberg's theory in its early form, and reported some phenomena that the early model fails to explain. In this chapter we shall review the recent development of the theory and see how this late model better fits empirical findings. I will refer mainly to five of his recent publications: Moral Stages: A Current Formulation and a Response to Critics (abbreviated as Kohlberg et al. 1983),' The  Psychology of Moral Development (abbreviated as Kohlberg 1984), " A Current 2  Statement on Some Theoretical Issues" (abbreviated as Kohlberg 1986),3 The Measurement of Moral Judgment vols. 1&2 (co-authored with Colby and others,  abbreviated as Colby et al. 1987 vols. 1&2), 4 and "The Return of Stage 6: Its Principle and Moral Point of View" (abbreviated as Kohlberg et al. 1990).5 Although I use the term "late model" to refer to the body of the newly introduced elements of Kohlberg's theory, this "late model" of Kohlberg's theory is not presented to us once and for all. It has emerged bit by bit since the late 70's (i.e., before the publication of the [1981] anthology) as Kohlberg strove to make his theory more compatible with empirical findings. The reason for choosing the above works to represent  Lawrence Kohlberg, C. Levine, and A. Hewer: Moral Stages: A Current Formulation and a Response to Critcs (Basel: Karger, 1983). 2 Lawrence Kohlberg. The Psychology of Moral Development (Cambridge: Harper and Row, 1984). 1  3 Lawrence Kohlberg. " A Current Statement on Some Theoretical Issues." In Lawrence Kohlberg: Consensus and Controversy, ed. Sohan Modgil and Celia Modgil (Philadelphia & London: Falmer, 1986), 485-546. 4 Anne Colby and Lawrence Kohlberg. The Measurement of Moral Judgment, 2 vols. (London and New York: Cambridge University Press, 1987).  60  his late model is largely a technical one: they present us with a relatively comprehensive picture of the recent development of the theory. The process of revising the theory is largely settled in 1987, with the publication of The Measurement  of Moral  Judgment  vols.  1&2. But even then he was not totally satisfied with his theory, mainly because the nature and status of Stage 6 is not fully clarified. His (1990) paper is an attempt, perhaps a preliminary one, to solve the problem. With the death of Kohlberg in 1987, the (1990) paper becomes his last words on the nature of Stage 6, and his project remains unfinished. In the late model, refinements have been made both in terms of the research method and the theory itself. Earlier versions of the scoring manual are considered not well-formulated, and the scorings are regarded as too subjective. These problems were rectified by the publication of the (1987) scoring manual (Colby et al. 1987 vol. 2). Since our main concern is the philosophical implications of his theory but not the research method, we shall go directly to the modifications of the theoretical structure.6 Three changes in the late model seem to be most significant: (1) modification of the "form" and "content" distinction, (2) redefinitions of the stages, and (3) introduction of moral orientations and the A and B Substages (or Types).  1.1 Form and content distinction  In the (1958) project, Kohlberg assessed moral stages partly in terms of decisional content (Kohlberg et al. 1983, p. 42). In the 1970's his definitions of the moral stages  5 Lawrence Kohlberg, Dwight R. Boyd, & Charles Levine, "The Return of Stage 6." In The Moral Domain, ed. Thomas E. Wren (Cambridge MA, The MIT Press, 1990), 151-181. 6 Readers who are interested in the problems of research method can refer to William Kurtines and Esther B. Greif, "The Development of Moral Thought," Psychological Bulletin 81, no. 8 (1974): 453-470; Lawrence J. Walker, "The Development of Moral Reasoning," Annals of Child Development 5, 37-78, esp.  61  moved in the direction of a higher degree of formality, and the stages become less dependent on the content choices. As we have seen, in the early model, Kohlberg defined moral content as the actual judgment on what is the right thing to do, and form as the reason behind the choice. In this way the form refers to quite a wide range of things, but principally includes basic norms and values upheld by the subject (see Appendix 3 for various universal values and norms people use to justify their moral judgments). In the late model the meaning of moral form is further narrowed down, coming in the end to contain only two elements, the social perspective and the justice structure.7 The social perspective refers to the way one views one's relation with others on the one hand, and with society as a whole on the other. We shall review the relation between social perspectives and moral stages in the next section. The second element, i.e., the justice structure, refers to the subject's understanding of equality, equity and reciprocity (Kohlberg etal. 1983, p. 42). 8 Accordingly, the values or norms of Appendix 3 no longer have direct bearing on the moral stage of a person. For example, according to the early model, Stage 6 moral motivation includes that of avoiding self-condemnation. Hence, if a subject judges that Heinz should not steal the drug because otherwise he will condemn himself and will not  36-37; and Anne Colby and Lawrence Kohlberg, The Measurement of Moral Judgment, vol. 1, (Cambridge: Cambridge University Press, 1987), 35-61. 7 The sharpening of the distinction between moral form and moral content is partly accomplished by the refinement of the scoring method. This involves further breaking down the moral judgment of a subject into each value and norm involved, and assessing how the subject reasons when these norms and. values are at stake. For details of the relation between the form-content distinction and the scoring method please see Colby etal. 1987 vol. 1, pp. 41-61, esp. 41-43. 8 See Appendix 5 for a summary of the concept of equality (including equity) and reciprocity in each of the moral stages. The reader may also find Appendix 5 useful for the understanding of the moral stages when combined with the section on the stage descriptions.  62  have l i v e d up to his own conscience (norm 11, see Appendix 3), this would have been assessed as Stage 6 reasoning. However, in the late model, such a judgment does not necessarily represent postconventional reasoning.9 Postconventional moral motivation in the late model is either a commitment to a social contract or the promotion of the welfare of the greatest number (Stage 5), or a commitment to some universal ethical principles (Stage 6). Similarly, in the early model, if a subject claims that Heinz should steal the drug because of the importance of a loving relationship, it would be ranked as a Stage 3 response because of its employment of an affiliation norm (Appendix 3, norm 4). But in the late model it is still an open question whether this subject is applying the affiliation norm in a manner that reflects a moral conception higher than Stage 3 (Kohlberg et al. 1983, p. 43).  1.2 T h e re-formulation of stage descriptions  Like the early model, the late model defines moral stages in terms of their criteria of Tightness and their motives for doing right. But they are now complemented by a set of social perspectives. Also, the definitions of the stages now reflect a sharper form-content distinction, the values defining a stage are more "formal," i.e., more abstract and less substantive than before. Below is a table of redefinitions of the stages in terms of what is right and the motive for doing right (Kohlberg 1984, pp. 174-176).10  See Kohlberg 1981, p. 122. There Kohlberg lists the motive for doing the right things at different stages, followed by a note that states that the definitions of Stages 5&6 are no longer valid. Note that the book is published in 1981 while the papers anthologized in it are published in the seventies, and there is a gap between the model presented in the book and the up-to-date theory when the book is published (some of them are indeed revisions of earlier papers). 1 0 The stage definitions used in this section are quoted from the second chapter of Kohlberg's (1984) book, "Moral Stages and Moralization," which is first published in 1976. The stage definitions used by Kohlberg have virtually been kept unchanged since then. The table of stage definitions is still used in the Colby et al. (1987) book.  9  63  Level and Stage  What is Right  Level I: Preconventional  To avoid breaking rules backed by punishment, obedience for its own sake, and avoiding physical damage to persons and property.  Stage 1 — Heteronomous Morality  Stage 2 — Individualism, Instrumental Purpose, and Exchange  Reason for doing right: Avoidance of punishment, and the superior power of authorities.  Following rules only when it is to someone's immediate interest; acting to meet one's own interests and needs and letting others do the same. Right is also what's fair, what's an equal exchange, a deal, an agreement. Reason for doing right: To serve one's own needs or interests in a world where you haVe to recognize that other people have their interests too.  Level II: Conventional Stage 3 — Mutual Interpersonal Expectations, Relationships, and Interpersonal Conformity  Living up to what is expected by people close to you or what people generally expect of people in your role as son, brother, friend, etc. "Being good" is important and means having good motives, showing concern about others. It also means keeping mutual relationships, such as trust, loyalty, respect, and gratitude. Reason for doing right: The need to be a good person in your own eyes and those of others. Your caring for others. Belief in the Golden Rule. Desire to maintain rules and authority which support stereotypical good behavior.  64  Stage 4 — Social System and Conscience  Fulfilling the actual duties to which you have agreed. Laws are to be upheld except in extreme cases where they conflict with other fixed social duties. Right is also contributing to society, the group, or institution. Reason for doing right: To keep the institution going as a whole, to avoid the breakdown in the system " i f everyone did it," or the imperative of conscience to meet one's defined obligations. (Easily confused with Stage 3 belief in rules and authority; see text.)  Level III: Postconventional, or Principled Stage 5 Social Contract or Utility and Individual Rights.  Stage 6 — Universal Ethical Principles  Being aware that people hold a variety of values and rules are relative to your group. These relative rules should usually be upheld, however, in the interest of impartiality and because they are the social contract. Some nonrelative values and rights like life and liberty, however, must be upheld in any society and regardless of majority opinion. Reason for doing right: A sense of obligation to law because of one's social contract to make and abide by laws for the welfare of all and for the protection of all people's rights. A feeling of contractual commitment, freely entered upon, to family, friendship, trust, and work obligations. Concern that laws and duties be based on rational calculation of overall utility, "the greatest good for the greatest number."  Following self-chosen ethical principles. Particular laws or social agreements are usually valid because they rest on such principles. When laws violate these principles, one acts in accordance with the principle. Principles are universal principles of justice: the equality of human rights and respect for the dignity of human beings as individual persons. Reason for doing right: The belief as a rational person in the validity of universal moral principle, and a sense of personal commitment to them.  The moral standards at the moral stages here are not significantly different from the early model, they are just described in greater detail. Stage 1 is obedience to authority and avoidance of punishment; Stage 2 is concrete exchange; Stage 3 is interpersonal  relationships; Stage 4 is social-order maintenance; Stage 5 is a mixture of utility, rights and contract; and Stage 6 is equal respect for individuals. All these are the same as the early model.  But if we examine the moral incentives, something more interesting is found. Consider the Stage 3 moral incentive. According to the early model, Stage 3 subjects do the right thing in order to please others. But in the late model, Stage 3 incentive is extended to include the desire to be good in one's own eyes, and the desire to maintain stereotypical values (presumably ones that one sincerely believes in). This allows a larger degree of autonomy in Stage 3 moral character, for its incentive, thus defined, acknowledges that the Stage 3 subject may do the right thing because he/she treasures these values, not solely because he/she wants to please others. The same is true for the revised Stage 4 moral incentive, which is a sense of being obliged to do one's duty as defined by the society, and a concern for the maintenance of the social order. In contrast to this, the early model describes the Stage 4 moral incentive as to avoid censure by legitimate authorities, which is obviously more heteronomous than the incentive in the late model.  The changes made to Stages 5 and 6 are also significant. In the early model, Stage 5 moral incentive is to maintain the respect of the impartial spectator, while Stage 6 moral incentive is to avoid self condemnation. In the late model, Stage 5 and 6 incentives no longer contain any idea of avoiding condemnation or gaining respect, whether from others (impartial spectator) or from the self. Postconventional moral incentive is now defined purely as the recognition that if there are any moral obligations at all one should always  66  be bound by them. Stage 5 moral incentive is thus a sense of obligation, a feeling of commitment, and a concern for the Stage 5 values themselves. Stage 6 mOral incentive is a rational belief in and a commitment to universal moral principles. So defined, the use of conscience no longer guarantees a postconventional moral stage. These changes also help to clarify the distinction between stages. In the early model, Stage 6 moral incentive differs from Stage 3 in that the former is the avoidance of self-condemnation, while the latter is the avoidance of others' dislike. Defining Stage 6 morality in terms of avoiding self condemnation inevitably blurs the line between Stage 6 and the lower stages. We can easily notice that self-condemnation appears quite early in our childhood, surely earlier than the ability to comprehend the abstract postconventional moral principles. Now, in the late model, Stage 3 incentive has included the need to be good in one's own eyes (this is quite close to Stage 6 moral incentive in the early model), whereas in the late model Stage 6 moral incentive becomes a sense of commitment to the universal moral principles. This change posts a stricter standard for Stage 6, and conforms to our experience better. Likewise, the distinction between Stage 4 moral incentive and that of preconventional level is sharpened. In the early model, Stage 4 moral incentive is to avoid censure by legitimate authorities and resulting guilt. Since censure is also a kind of punishment, the Stage 4 moral incentive thus defined is confused with Stage 1 moral incentive, i.e., the avoidance of punishment. In the late model, Stage 4 moral incentive becomes that of keeping the institution or society going as a whole. This change in the definition of Stage 4 helps to separate conventional morality, which is marked by a  67  concern for one's social group, from preconventional morality, whose concern is mainly egocentric. Kohlberg also retreats from some of his early claims about Stage 6. In the early model, he claims that the existence of Stage 6 is empirically verified. But in his (1983) book he withdraws from this claim, and admits that he has failed to "construct a detailed scoring manual description which would allow reliable identification of a sixth stage" (Kohlberg et al. 1983, p. 60). And the stage does not even appear in the (1987) standard scoring manual (i.e., Colby et al. 1987 vol. 2). Nevertheless, Stage 6 still exists in the late model, serving as a postulate and a device to define the nature and endpoint of moral development (Kohlberg et al. 1983, p. 61). In the late model, Kohlberg attaches to every moral stage a social perspective. By 'social perspective' Kohlberg means the way "[a] person sees other people, interprets their thoughts and feelings, and sees their role or place in society." These social perspectives are, strictly speaking, not part of the definitions of the stages. They are "very closely related to moral stages, but are more general," and constitute a way of viewing human interactions (Kohlberg 1984, p. 171). Since it is more general, it is easier for a person to take up a certain social perspective than to employ its corresponding moral stage. In this way there is very often a "horizontal development" from a certain social perspective stage to its corresponding stage. In the "horizontal development" one attains the social perspective that is connected to a certain moral stage first, and then solidifies the moral conception of that stage later (Kohlberg 1984, pp. 171-172).  68  The social perspective at the preconventional level is a concrete individual perspective. The subject at this level is not yet able to comprehend the concept of society as a collective entity, but can only comprehend individual persons. At the conventional level, the subject starts to understand the concept of a collective entity such as a family or a society. But with the understanding of the concept of the group the subject virtually loses the concept of the individual in the proper sense, and thereby treats the individual merely as a member of a group or society. At the postconventional level, the subject regains the concept of the individual, and combines it with the idea that individuals are members of society. The result is that persons are viewed as members of society who are prior to society. This is why postconventional morality differs from conventional morality by its law-making perspective, in contrast to the law-maintaining perspective at the conventional level (Kohlberg 1984, pp. 173-180). Below is a table of the relationship between the social perspective stages (with definitions) and moral stages (Kohlberg 1984, pp.174-176).11 Stage  Social Perspective Stage  Stage 1  Egocentric point of view. Doesn't consider the interests of others or recognize that they differ from the actor's; doesn't relate two points of view. Actions are considered physically rather than in terms of psychological interests of others. Confusion of authority's perspective with one's own.  Stage 2  Concrete individualistic perspective. Aware that everybody has his own interest to pursue and these conflict, so that right is relative (in the concrete individualistic sense).  The list was first constructed by R. L. Selman in his "The Development of Social-Cognitive Understanding: A Guide to Education and Clinical Practice." In Moral Development and Behavior: Theory, Research, and Social Issues, ed. T. Lickona (New York: Holt, Rinehart and Winston, 1976). Also see Kohlberg, "Moral Stage and Moralization," 33-35. 11  69  Stage 3  Perspective of the individual in relationships with other individuals. Aware of shared feelings, agreements, and expectations which take primacy over individual interests. Relates points of view through the concrete Golden Rule, putting yourself in the other person's shoes. Does not yet consider generalized system perspective.  Stage 4  . Differentiates societal point of view from interpersonal agreement or motives. Takes the point of view of the system that defines roles and rules. Considers individual relations in terms of place in the system.  Stage 5  Prior-to-society perspective. Perspective of a rational individual aware of values and rights prior to social attachments and contracts. Integrates perspectives by formal mechanisms of agreement, contract, objective impartiality, and due process. Considers moral and legal points of view; recognizes that they sometimes conflict and finds it difficult to integrate them.  Stage 6  Perspective of a moral point of view from which social arrangements derive. Perspective is that of any rational individual recognizing the nature of morality or the fact that persons are ends in themselves and must be treated as such. Although a social perspective stage is not identical to its corresponding moral  stage, Kohlberg says that it helps to understand the moral stage. He does not explain how, but the following example may be useful. Consider the definition of Stage 4. We know that the moral standards of the stage include fulfilling actual duties one has accepted, obeying laws except when the laws are in conflict with some explicit social duties, and contributing to one's society. But there remains the question why these standards should be put together, and to answer it some kind of underlying spirit that unifies them is needed. This underlying spirit is the social perspective. Since the Stage 4 subject views the self as a member of society which is a huge system, he/she judges an action in terms of how well it facilitates the operation of the society as a whole. And this gives rise to all  70  the Stage 4 moral standards. Therefore, the introduction o f the corresponding social 12  perspective is a step towards clarifying and integrating the meaning of the moral stages.  1.3 The introduction of Type A and Type B moral reasoning In section 1.3 we saw that the use o f conscience no longer guarantees postconventional reasoning in the late model. But does it not seem obvious that the use o f conscience should have some importance in determining one's moral maturity? D o we not think that it is better for someone to do something because o f a requirement of his/her conscience rather than, say, the desire to please others? If this is right, then we can 13  reasonably expect that a theory o f moral development should reflect this sort o f belief about the relation between conscience and moral maturity. It is the same with content choice. If, as Kohlberg believes, some solutions are objectively better than others (as it is better for Heinz to steal the drug than to let his wife die), then this should be reflected in the developmental model as well. The introduction o f Substages A and B (later called Types A and B) into the stage model is, at least in part, an attempt to cater to these considerations.  14  12 There is a puzzle concerning how Stage 4 moral standards instantiate the social perspective at the corresponding stage. The correlation of the last two standards with Stage 4 social perspective is clear and needs no explanation. But it seems problematic when we try to explain in what sense the standard of fulfilling actual duties one has agreed to can be regarded as an instantiation of the "full-fledged member-ofsociety perspective" (in Kohlberg's own word, see Kohlberg 1984, p. 181). To say that one should fulfill duties one agreed to seems to suggest that one does not have to fulfill those which one has not agreed to, and the latter surely reflects a "prior-to-society" individualistic perspective, which is the characteristic of postconventional reasoning. One possible way to explain this is to claim that the fulfillment of duties to which one has agreed should be understood as something similar to promise keeping, while promise keeping is regarded by the Stage 4 subject as something necessary for maintaining a society. The moral obligation to fulfill the agreed upon duty does not come directly from the fact that one has agreed to it, but from the need for promise keeping to maintain social order. 13 One might ask: What if our conscience requires us to please others? It seems that even in this case it is better if one is ready to do something against others' expectations when our conscience requires us to do so. 14 Kohlberg presents the substages as part of his attempts to clarify the form and content distinction. See Kohlberg etal. 1983,6-7.  71  For Kohlberg, the preconventional moral conception is largely heteronomous, while the postconventional conception is largely autonomous. The conventional level is midway between the two extremes. Therefore, it seems reasonable to subdivide conventional moralities into two types, each leaning toward one end of this autoriomyheteronomy spectrum. The more autonomous type is, not surprisingly, more adequate than the more heteronomous type. Kohlberg calls the more heteronomous type Substage A , and the more autonomous type Substage B. The substages are sometimes regarded also as midway between stage structure and content choice (Kohlberg, 1983, p. 44). They are first introduced in Kohlberg's paper "Moral Stages and Moralization," published in 1976. In the paper Kohlberg classifies four orientations of moral reasoning: normative order, utility consequences, justice or fairness (I will just call it the orientation of justice below), and ideal self. Substage A focuses more on the orientations of normative order and utility consequences. The orientation of normative order centers on moral rules, whereas the orientation of utility consequences centers on welfare of others and/or the self. Substage B, on the other hand, focuses on the orientations of justice and the ideal self. The orientation of justice centers on the relations of liberty, equality, reciprocity, and contract between persons, whereas the orientation of the ideal self centers on the consideration of what a good person should be like, in terms of conscience, motives and virtue. Substage A makes judgments more descriptively, in terms of something given "out there." Substage B makes judgments more prescriptively, in terms of what ought to be, and what is internally accepted by the self.  72  When compared to Substage A , Substage B is more ready to make the recognized values universal ones (Kohlberg 1976, pp. 40-41). In 1983, Kohlberg further clarifies the conceptions of the substages by making up a list of criteria of Substage B in contrast to Substage A . These are composed of a set of "Kantian" criteria and a set of "Piagetian" criteria (Kohlberg 1983, pp. 46-47). Here is a summary of these criteria. Five Kantian criteria are drawn up for Substage B. The first is the criterion of choice, which requires the subject to choose the more "just" course of action or solution  to the dilemma (i.e., the choice which is empirically agreed upon by most subjects at Stage 5). In Heinz's dilemma, the right choice is to steal the drug; in the judge's dilemma, the right choice is for the judge to set Heinz free, or to put him on probation; and in Joe's dilemma, the right choice is to refuse to give his father the money. The second Kantian criterion requires that the subject's reason for action must reflect the second formulation of the Categorical Imperative: "Treat persons never simply as means, but always at the same time as ends." Accordingly, the subject places the right to life over the right to property, conscience over fixed legal laws, and free agreement over authority. This is the criterion of hierarchy. The third Kantian criterion is that the subject has to uphold the  intrinsic moral worth of persons, and show an intrinsic respect for them. Kohlberg calls this the criterion of intrinsicalness. The fourth Kantian criterion is the criterion of prescriptivity, the recognition of a categorical moral "ought" regardless of the inclinations  of the actor, or any other pragmatic considerations. The final Kantian criterion is the criterion of universality or universalizability, which requires that the choice must be  .73  made with an acceptance that all moral agents in the same position should behave in the same way and, in particular, an acceptance that he/she be treated in the same way by others in comparable circumstances. There are also four Piagetian criteria for Substage B. The first Piagetian criterion is the criterion  of autonomy. The response of the subject must reflect an understanding  that the actor in question is an autonomous moral agent. The second criterion, the criterion  of moral respect, requires the subject to consider all the other parties in the  dilemma as rational and autonomous moral agents. This seems quite similar to the Kantian criterion of intrinsicalness. The third Piagetian criterion, the criterion reversibility,  of  is closely related to the second one and to that of universalizability. It  requires that the subject must be able to consider the problem not only from the view point of the protagonist of the story, but also from the view points of all the other parties. The fourth criterion is the criterion  of constructivism.  The subject must view moral rules,  laws, and principles as a product of human communication and cooperation, and something actively constructed by the human mind. These criteria help clarify the sense in which the substages are intermediate between form and content. Substage B is characterized by formal properties of moral reasoning such as intrinsicalness, prescriptivity, universalizability, and reversibility. On the other hand, it is also characterized by an unconditional respect for persons. Connected to respect for persons, there are the values of human rights, among which the rights to autonomy and to life are most basic. This constitutes the substantive side of Type B moral reasoning.  74  In the (1983) monograph, Kohlberg claims that whenever there is any change in substage (within a single stage), it is always from A to B but not vice versa. For example, whereas a 3 A can move either to 3B or 4A, a 3B can only move to 4A or directly into 4B, but never to 3A. Although it is possible for a 3B to move to 4A, it happens more often that a B subject retains his Type B thinking when he moves into a higher stage. Moreover, it is observed that B subjects are more likely than A subjects to actually carry out what they judge to be the "just" choice. In other words, behaviors of B subjects are more consistent with their moral beliefs than A subjects (Kohlberg et al. 1983, pp. 45 & 61). 1 5 These, Kohlberg believes, indicate that B subjects are morally more mature than A subjects. Kohlberg soon abandons the use of 'substage' to refer to these orientations of moral reasoning because he finds that people do sometimes move from Type B moral reasoning to Type A moral reasoning (e.g., from 3B to 3A). This violates the definition of a substage. Instead of "substages," he finds it more appropriate to call them "types." Nevertheless, it remains evident that (1) there is an age-development trend towards increase usage of Type B reasoning; (2) when people change from one type to the other, it is more likely to be a change from A to B than from B to A ; and (3) when a person terminates stage development, it is more likely that he/she will be stabilized as a Type B reasoner. Therefore, although Types A and B moral reasoning are no longer viewed as two substages, they remain a useful tool to classify people's orientations of moral  15  Also see Lawrence Kohlberg & D. Candee, "The Relationship of Moral Judgment to Moral Action." In  Morality, Moral Behavior, and Moral Development, ed. William M. Kurtines & Jacob L. Gewirtz (New  York: John Wiley & Sons, 1984), 41-73.  75  reasoning, and Type B moral reasoning is still claimed to be more adequate than Type A (Colby et al. 1987 vol. 1, pp. 324-325). Types A and B moral reasoning have shared some of the job of the moral stages. Because Type B is better than Type A , difference in moral maturity is no longer necessarily manifested as a difference in moral stages, but can be manifested as a difference in moral reasoning types. Some of the judgments previously regarded as postconventional are now assessed as Stage 3B or Stage 4B (Kohlberg et al. 1983, p. 61). This helps to remove the anomalies in observational data such as stage regressions. 1.4 The moral domain ' Kohlberg's definition of morality in terms of justice has invited many criticisms. Carol Gilligan, for example, argues that the definition of morality in terms of justice involves a bias towards the male perspective. She suggests that instead of justice and fairness, females tend to view moral problems in terms of care and responsibility, which constitutes an equally important moral orientation. Kohlberg's model fails to include the orientation of care and responsibility, and the moral domain thus defined is too narrow. Kohlberg has recognized the power of Gilligan's arguments, and draws back from the claim that morality is identical to justice (Kohlberg et al. 1983, p. 19).16 Accordingly, he acknowledges that his moral domain needs enlargement. As a result a number of new  Gilligan claims that she has discovered the gender difference in moral reasoning through empirical studies. Kohlberg is not convinced that the claim of gender difference is proved by Gilligan's studies. Nevertheless, in the so called ethic of care (which Gilligan thinks is characterized by females' moral reasoning), Kohlberg does find some elements that his early model fails to fully recognize. 16  76  principles are regarded as Stage 6 in the late model, such as the principle of Agape and the principle of benevolence. We shall return to their debate in Chapter 3.  Section 2: Empirical support With the modified research method and amended theoretical claims, Kohlberg's model has received much wider support from empirical research. To see how this is so, Walker's (1988) and Snarey's (1985, republished 1994) reviews of relevant studies are especially useful. 17 Walker's paper examines the validity of Kohlberg's stage model by reviewing a large number of empirical studies, and estimates how well they support the model. He wants to check the validity of Kohlberg's three major claims, the claim of invariant sequence, the claim of structural wholeness, and the claim of hierarchical order. Regarding the claim of invariant sequence, Walker reports the results of six longitudinal studies and one experimental study in North America, concluding that the studies are "strongly supportive" of the sequence assumption.18 There are "few instances of either regression (0-17%) or stage-skipping (0-4%) in the modal stage scores."19 One  17 Lawrence J. Walker, "The Development of Moral Reasoning," Annals of Child Development 5 (1988), 33-78; John R. Snarey, "Cross-cultural Universality of Social-Moral Development: A Critical Review of Kohlbergian Research," Psychological Bulletin 97, no. 2 (1985), 202-232, reprinted in New Research in Moral Development, ed. Bill Puka (New York & London: Garland Publishing, 1994) 268-298. 1 8 The studies include A. Colby et al, " A Longitudinal Study of Moral Judgment," Monograph of the Society in Child Development 48 (1-2, Serial no. 200); V. L. Erickson, "The Case Study Method in the Evaluation of Developmental Programs." In Evaluating Moral Development, ed. L. Kuhmerker et al. (Schenectady, NY: Character Research Press, 1980), 151-176; L. Walker, "The Sequentiality of Kohlberg's Stages of Moral Development," Child Development 51 (1980): 131-139; and J. M. Murphy & C. Gilligan, "Moral Development in late Adolescence and Adulthood: A Critique and Reconstruction of Kohlberg's Theory," Human Development 23 (1980): 77-104. 1 9 Walker, "The Development of Moral Reasoning," 43. For a more detailed review of the studies, refer to Lawrence Walker, "Cognitive Processes in Moral Development." In Handbook of Moral Development:  77  problem with Walker's report is that all the studies have been conducted in North America, which, of course, is not sufficient to prove that Kohlberg's stages form a universal invariant sequence. Yet when Snarey reviews seven cross-cultural studies, he finds a similar result. After reviewing these longitudinal studies from the Bahamas, Canada, Indonesia, Israel, United States, Turkey, and India, Snarey concludes that in all these studies stage regressions are within the range of measurement error. 20 The claim of structured wholeness has also received wide empirical support. According to a paper co-authored by Colby and Kohlberg, in Kohlberg's longitudinal study, 68-82 % of reasoning of the subjects was scored at the modal stage in late model scoring, and 97-99% of reasoning was at two most frequent (and always adjacent) stages.21 Walker, after reviewing the data of four studies conducted by him and his colleagues, also finds that 66% of the reasoning of the subjects was scored at their modal stages, while 94% of it falls in one of the two most frequent (and always adjacent) stages.22 It therefore seems that the claim of structured-wholeness is also well supported by empirical data.  Models, Processes, Techniques, and Research, ed. G. L. Sapp (Birmingham, AL: Religious Education Press), 109-145. 2 0 John R. Snarey, "Cross-cultural Universality of Socio-Moral Development: A critical Review of Kohlbergian Research." In New Research in Moral Development, ed. B. Puka (New York & London: Garland Publishing Inc., 1994), 215-216. 2 1 Colby et al., " A Longitudinal Study of Moral Judgment." In New Research in Moral Development, ed. Bill Puka, 1-76. 2 2 The studies he reviews include L. Walker, "Sources of Cognitive Conflict for Stage Transition in Moral Development," Developmental Psychology 19 (1983): 103-110; L. Walker, "Sex Differences in Moral Reasoning: A Critical Review," Child Development 55 (1986): 677-691; L. Walker et al., "The Hierarchical Nature of Stages of Moral Development," Developmental Psychology 20 (1984): 960-966; and Walker et al., "Moral Stages and Moral Orientations in Real-life and Hypothetical Dilemmas," Child Development 58 (1987): 842-858. See Lawrence J. Walker, "The Development of Moral Reasoning," 39.  78  The third assumption, the claim of hierarchy, is a little more complicated. According to Walker, the claim has two main empirical implications. First, it implies that subjects have difficulty comprehending stages higher than their modal stages. They can usually comprehend no higher than their +1 stages, i.e., the stages that are one stage higher than their predominant stages; Second, it implies that people prefer, among the stages they can comprehend, the higher to the lower stages. And this is true even for those whose highest comprehended stages are higher than their predominant stages. After reviewing a large number of studies, Walker concludes that all of them confirm these implications. 23 However, the validity of the method researchers used to test the assumption of hierarchy is questioned by Moran & Joniak. They suggest that higher stages were usually formulated in more sophisticated language. It was therefore the language, instead of the form of reasoning itself, which caused the apparent preference for the higher stages.24 As a response to Moran & Joniak's suggestion, Walker et al. conducted a test of Kohlberg's  The studies Walker uses to support the claim can be divided into two groups. Thefirstgroup are tests of the claim that subjects tend to prefer the highest stage they comprehend. They include: J. L. Carrol & J. R. Rest, "Conflict, learning, and Piaget: Comments on the Development of Moral Judgment," Dissertation Abstract International 34 (1981): 2331B (University of Microfilms No. 73-28193); M. L. Divison et al., "Stage Structure in Objective Moral Judgment," Developmental Psychology 14 (1978): 137-146; C. B. Keasey, "The Influence of Opinion Agreement and Quality of Supportive Reasoning in the Evaluation of Moral Judgments," Journal ofpersonality and Social Psychology 30 (1974): 477-480; J. R. Rest, "The Hierarchical Nature of Moral Judgment: A Study of Patterns of Comprehension and Preference of Moral Stages", Journal of Personality 41 (1973): 86-109; and J. R. Rest et al., "Level of Moral Development as a Determinant of Preference and Comprehension of Moral Judgments made by Others," Journal of Personality 37 (1969): 225-252. The second group are tests of the claim that subjects have difficulty comprehending stages higher than their modal stage. They include: Kuhn et al., "The Development of Formal Operations in Logical and Moral Judgment," Genetic Psychology Monographs 95 (1977): 97-188; R. N. Tsujimoto, "Guttman Scaling of Moral Comprehension Stages," Psychological Reports 51 (1982): 550; and the two papers by J. R. Rest just mentioned above. See Walker, "The Development of Moral Reasoning," 45. 2 3  J. J. Moran & A. J. Joniak," Effect of Language on Preference for Responses to a Moral Dilemma," Developmental Psychology 15 (1979): 337-338. 2 4  79  hierarchical claim with the language controlled by (a) writing all statements representing different moral stages as single sentences each containing the same number of words and (b) equalizing the difficulty of the vocabulary used. The result shows that the evidence for the hierarchical claim is still clear. Understanding of the statements at and below the modal stage is nearly perfect (97% and 96% respectively). (The percentage drops to 67% at the +1 stage, to 21% at the +2 stage, and 0% at the +3 stage). With regard to preference when comparing two statements of different stages, subjects in most cases (73%) prefer the one at the higher stage, given that they understand both. 2 5 As I have suggested in Chapter 1, since the hierarchical claim holds that higher stages are more adequate, showing that higher stages predict better moral behavior will provide important confirmation of the model. With the early model and its scoring method, a certain rough correlation was found between moral judgment stage and moral behavior, but considerable unexpected phenomena were also observed. Under the late model, the correlation between judgmental stage and behavior becomes much more consistent. In 1978, McNamee conducted an experiment to test how moral judgmental stages were related to helpfulness. The situation of the test is described as follows: Undergraduates who agreed to be interviewed on the standard moral dilemmas were led to a testing room. As they were entering the room they were intercepted by a student presenting himself as the next subject for the experiment. The student stated that he had just taken drugs and was having a bad time. He had come to the experiment because he thought that the experimenter, being a psychologist, could help him. The response of the experimenter was that she was a research psychologist, not a therapist. The drug-user persisted in soliciting aid, hoping that Lawrence Walker, "The Development of Moral Reasoning," 44-47. Also see Walker et al, "The Hierarchical Nature of Stages of Moral Development."  2 5  80  the experimenter could refer him to help. The experimenter replied that she had no experience with drugs and did not know what facility could help him. She told him to call to reschedule his testing session. The drug-user slowly left the room. The subject was faced with the choice of whether to remain an uninvolved bystander or whether to intervene. The result was that helpful behavior increased with moral stages. Only 9% of the Stage 2 subjects helped the victim by offering information about other sources of assistance. The percentage of Stage 3 subjects who intervened increases to 27%; Stage 4 subjects to 38%; and Stage 5 subjects to 73%. Nobody at and below Stage 4 offered any personal help (such as to take him'home or to a source of help), while 20% of the Stage 5 subjects did so. 2 6 As we saw in Chapter 1, according to Haan, Smith and Block, Stage 2 subjects involved in their (1968) study generally performed even better than Stage 5 subjects. Using the original interview records, the subjects in this study were later rescored using the revised scoring manual. With rescoring there were no more Stage 2 subjects, nor any Stage 5 or Stage 6 subjects (Stage 6 is not even present in the new scoring manual). The result was that 11% of Stage 3 subjects, 31 % of Stage 3/4 subjects, 42% of Stage 4 subjects, and 71 % of Stage 4/5 subjects sat i n . 2 7 Interviews of the subjects participating in the Milgram study have also been rescored using the new scoring manual. The rescoring showed that all subjects were conventional reasoners. 5 out of the 6 Stage 4 subjects (87%) quit the test, whereas only 1 of the 17 Stage 3/4 subjects(6%), and 2 out of the 4 Stage 3 subjects (50%) quit. The fact See S. McNamee, "Moral Behavior, Moral Development and Motivation," Journal of Moral Education 7 (1978): 179-188; also see L. Kohlberg & D. Candee, "The Relationship of Moral Judgment to Moral  2 6  81  that the percentage of Stage 3. subjects who quit (50%) is higher than that of Stage 3/4 subjects (6%) makes the result a little ambiguous. But the ambiguity can be explained by the small number of subjects at Stage 3. Overall, there is still a mild but clear correlation between moral judgmental stage and moral behavior.28 Section  3: Postconventional  moral  conception  3.1 The postconventional level  What kind of normative ethics does this new model support? To answer the question, we have to understand the postconventional morality present in the model. Roughly speaking, postconventional moral reasoning is marked by an awareness that personal interests and conventions are often arbitrary, and are by themselves insufficient for justifying moral values or norms. Thus, postconventional level reasoners look for fundamental universal values, and try to make moral judgments based on them. At Stage 5, these universal values are defined in terms of utility, social contracts and human rights, whereas Stage 6 further abstracts fundamental values into formal properties such as fairness, impartiality, and reversibility (Kohlberg 1987, pp. 29-30). These are fairly similar to the early Kohlbergian model. By comparing the three levels of moral development without referring to the details of the stages, something significant about ethical theory can be seen. The  Action." In Morality, Moral Behavior, and Moral Development, ed. William M Kurtines & Jacob L. Gewirtz (New York: John Wiley & Sons, 1984), 41-73, esp. 58-59. 2 7 Kohlberg & Candee, "The Relationship of Moral Judgment to Moral Action," 66-67. 2 8 Ibid., 69. For more reviews of the validity of Kohlberg's theory, see L. Kohlberg and D. Candee, "Invariant Sequence and Internal Consistency in Moral Judgment Stages." In Morality, Moral Behavior, and Moral Development, 41-51; and John R. Snarey, "Cross-cultural Universality of Socio-Moral Development: A critical Review of Kohlbergian Research." in New Research in Moral Development, ed. B. Puka (New York & London: Garland Publishing Inc., 1994), 268-298.  82  preconventional level represents a largely egoistic morality. The self is the only real concern of the subject. The conventional level is a morality of the social group. The primary concern of the subject is the group to which he/she belongs. At the postconventional level, moral concern is directed to the human species as a whole. The goal of moral practice is viewed as the resolution of any possible moral conflicts between persons, regardless of the social groups of the involved parties. We can see, then, that the Kohlbergian model of moral development depicts a moral agent's effort to extend his/her moral concern from the self to a particular social group and then to the human species as a whole. Higher degrees of moral maturity require a subject to bring a wider scope of people into his realm of moral concern. To treat others with moral concern, one has to endeavor to act toward them in a way that is acceptable from their points of view. This already constitutes equity or equality as a universal value, though this value is highly abstract and formal. The second point is that, the lower the moral developmental level of the subject, the more the subject sticks to the values of a particular person or group. At the lowest level, the subject simply obeys authority. At level two the subject obeys social norms of his group, and by doing this he embraces the relativity of goodness among different people in his own group. However, at this level, the subject still does not possess the ability to comprehend the relativity of values among different cultures, and tends to take the values of his own culture as absolute. At the postconventional level, the subject embraces the contingency of social norms, and therefore bases his moral considerations on values that can been seen as universal, such as human rights at Stage 5, and the Golden  83  Rule or the principle o f Agape at Stage 6. The upshot is that the more mature the subject is, the more abstract is the moral standard the subject would likely hold.  Here we can see a seemingly paradoxical phenomenon in moral development. T o be morally mature, one has to strive for a universalistic moral view. But to develop such a view, one has to give up concrete and absolute moral rules, and embrace more relativity in one's moral concepts. When a Stage 6 reasoner admits that his/her cultural code is not absolute, a natural consequence w i l l be that there can be different cultural codes that are equally valid than his/her own, and this gives rise to certain weak form o f cultural relativism. 2 9 To view the universalism-relativism debate in light o f this would inevitably lead us to the conclusion that universalism and relativism are not totally incompatible, though they may be opposites of each other in some sense. It seems that, i f Kohlberg's theory is correct, the two doctrines can be reconciled in some way. Based on what we have seen from Kohlberg's study, one may even say that a universalistic moral theory can neither be complete nor credible without introducing some relativistic elements into the theory. Such a universalism w i l l embrace, for instance, certain relativity in the realm of personal good, and can perhaps admit that since no cultural code is absolute therefore different cultural codes can sometimes be equally adequate.  3.2. Comparing Stage 5 and Stage 6 A s we have mentioned, Kohlberg's early model claims that there are two universal aspects o f moral phenomena, namely a universal developmental pattern of moral development, and a universal set of moral principles appearing at the highest stage  2 9  We shall discuss a few theories of this kind in Chapter 5.  84  of moral development. In other words, universalism is true for Kohlberg in that there exists an ideal "form" of moral reasoning which serves as the universal destination Of moral development. Since Stage 6, qua the universal destiny of moral development, represents a moral ideal, examination of the stage will help understanding the implicit ethical theory in Kohlberg's model. A straight-forward way of understanding the Stage 6 moral conception is to look at the principles used at Stage 6. Possible Stage 6 principles include the principles of justice, respect for human personality or dignity, utility or benevolence, universal human care or Agape, maximum liberty compatible with the like liberty of others, maximization of equality of life for each, and the principle of equity and fairness (Colby etal. 1987 vol. 1, p. 31). These principles are, however, all so abstract that they are not very useful in guiding actions. Kohlberg repeatedly emphasizes the importance of the principle of justice in Stage 6 moral reasoning (in the earlier model he even equates justice with the Stage 6 moral conception). But what is justice? What does justice require us to do? Kohlberg says that justice requires respect for human rights, but he also says that at Stage 6 human rights are no longer absolute. So it needs further elaboration to grasp the exact meaning of the principle. Likewise, the principle of respect also needs clarification. What must we do regarding others in the name of respect for them? There are easy cases. If I say to a person he/she is worthy of nothing because he/she is stupid, I am surely not expressing proper respect. But there are also many difficult cases. Consider capital punishment. Kant claims that capital punishment is morally justifiable, because in doing  85  so we recognize the offender as a moral agent who is responsible for what he has done. What is implicit is that capital punishment is precisely what our respect for the offender requires. Obviously Kohlberg does not agree with Kant on this point (for Kohlberg's view see Chapter 1, section 3.3). The formal properties of Stage 6 moral reasoning are fairness, impartiality, and reversibility. Kohlberg suggests four ways of realizing the spirit of Stage 6 through a process of role-taking. The first method is Rawls's original position thought experiment, where people behind "veils of ignorance" seek agreement on a set of basic principles guiding their behavior. The second is Kohlberg's own method, namely a second-order application of the Golden Rule, or moral musical chairs (Kohlberg 1981, pp. 203 ff, Kohlberg et al. 1983, pp. 101-3).30 In this second-order application of the Golden Rule, the moral agent has to take the roles of other parties, imagining that they do the same. If the subject can reasonably claim that all parties will then agree upon a certain solution, then the solution is a just one. The third method is Habermas's ideal communication, "a dialogue among free and equal persons considering and modifying their claims in light of one another." The fourth method is a utilitarian approach (rational-preference utilitarianism of Harsanyi), which requires one to consider people's preferences and balance these view points. In order to do this, one has to imagine that he has an equal probability of being any of those involved in the situation where the moral problem arises, and thus decide what he will prefer (Colby and Kohlberg 1987 vol. 1, pp. 30-31).  Also see Lawrence Kohlberg, " A Current Statement on Some Theoretical Issues." In Lawrence Kohlberg: Consensus and Controversy, ed, Sohan Modgil and Celia Modgil (Philadelphia & London, The Falmer Press, 1986), 526.  3 0  86  Viewing the Stage 6 moral conception in light of these decision-making strategies, one may argue that Kohlberg's moral ideal is more a method or a set of methods for decision-making than a set of principles that determines all morally correct actions. Kohlberg obviously does not use the word 'principle' in the same way as philosophers usually do. Philosophers seldom call something a 'principle' if it cannot be put into a well-defined formula. But for Kohlberg the name 'principle' can be attached to virtually any value or procedure for the determination of values. It seems, then, the correct moral actions that follow from Stage 6 moral reasoning are not clearly determined by the principles themselves, such as those of Agape or justice. Rather it is the procedures of role-taking and the like which, in particular cases, come closer to being determinative.  Stage 5 morality is sometimes understood as a utilitarian conception of morality, whose main concern is total utility (see the stage descriptions). This is misleading, for the concepts of human rights and social contract are equally essential for Stage 5 morality. Kohlberg writes that, at Stage 5, moral rightness is determined by contract or agreement made among members of a group. The legitimacy of agreements is not questioned. By contrast, a Stage 6 reasoner proceeds to question the legitimacy of such an agreement by asking whether it is a fair one (Colby et al. 1987 vol. 1, pp. 31-32). But even this demarcation of Stages 5 and 6 does not work. The contractarian moral values do not exhaust Stage 5 moral reasoning. There are also the values of human rights and social utility. Perhaps a reasonable reading of Kohlberg's words is that there is no principle underlying the three equally legitimate values (contract, utility, and human rights) at  87  Stage 5. B u t this renders Stage 5 subjects unable to decide i n any systematic rational manner w h a t is right w h e n these values are i n c o n f l i c t .  T h i s helps e x p l a i n the different roles h u m a n rights p l a y at Stage 5 and Stage 6. A t Stage 5, says K o h l b e r g , the subject resorts to u n i v e r s a l h u m a n rights i n s o l v i n g m o r a l p r o b l e m s . H o w e v e r , h u m a n rights are not as fundamental as the p r i n c i p l e s u p h e l d at Stage 6. T o be c o n s i d e r e d a fundamental p r i n c i p l e , a p r i n c i p l e has to be (1) generally relevant to a l l m o r a l situations, a n d (2) u n i v e r s a l i z a b l e to a l l m o r a l agents. H u m a n rights are u n i v e r s a l i z a b l e to a l l m o r a l agents but are not generally relevant to a l l situations. T h e right to l i f e , for e x a m p l e , is irrelevant to the m o r a l issue i n the J o e ' s d i l e m m a . B u t at Stage 6, the subject resorts to general u n i v e r s a l p r i n c i p l e s s u c h as the G o l d e n R u l e a n d the P r i n c i p l e o f A g a p e . These p r i n c i p l e s are b o t h general and u n i v e r s a l i z a b l e i n the sense e x p l a i n e d above ( C o l b y et a l . 1987 v o l . 1, p. 33). In this sense, t h o u g h Stage 5 m o r a l i t y is already " g r o u n d e d o n the operations o f equality, equity, and the l i k e , " o n l y at Stage 6 do these operations b e c o m e self-conscious p r i n c i p l e s ( C o l b y et a l . 1987 v o l . 1, p. 31).  It is therefore fair to say that Stage 5 and Stage 6 are essentially s i m i l a r to each other i n spirit. T h e standards used at Stage 5 are still important at Stage 6, but they cease to be ultimate. W h a t is ultimate is the f o r m a l operations o f m o r a l d e c i s i o n m a k i n g . T h e c o n c e r n f o r total u t i l i t y , h u m a n rights, a n d s o c i a l contracts, and their p r o p e r b a l a n c i n g , is the product o f these operation. Stage 6 m o r a l i t y transcends Stage 5 m o r a l i t y i n that ultimate m o r a l p r i n c i p l e s , l i k e that o f respect f o r persons, are m o r e p o w e r f u l l y i m p l e m e n t e d and integrated b y abstract p r o c e d u r a l operations. P r i n c i p l e s are e m b o d i e d by  88  cognitive operations of ideal dialogues and reciprocal role-taking leading to universalizability (Kohlberg et al. 1990, p. 153). Having seen how Stage 6 seeks to overcome the difficulty of Stage 5 moral reasoning, we can understand why Kohlberg has to postulate a Stage 6 above Stage 5. There are potential conflicts among fundamental values at Stage 5 with no hope of systematic resolution. Kohlberg believes an end stage is needed to resolve these conflicts. 31 Although Kohlberg and his colleagues fail to verify the existence of Stage 6 as the universal destiny of moral development with the new model and scoring method, they keep it in the model as a postulate. They also retain universalism as part of the theory, where universalism is defined as the view that "there is a universalistically valid form of rational moral thought process which all persons could articulate, assuming social and cultural conditions suitable to cognitive-moral stage development" (Kohlberg, 1983, p. 75).  3.3 The return of Stage 6 Formulating the characteristics of Stage 6 moral reasoning has been an ongoing process since Kohlberg first started the project of establishing the cognitivedevelopmental theory of moralization. He has not given up this goal despite his admission that Stage 6 is not verified. In 1990, a paper titled "The Return of Stage 6" was published, in which he and his colleagues attempt to reestablish the status of Stage 6 as an actually  As we have seen in Chapter 1, section 2.2, studies showed that Stage 6 reasoners typically agree on the right solution to a moral dilemma. Though Stage 6 does not appear in the (1987) scoring manual, he obviously retains his assumption that Stage 6 reasoners do agree on the solution to a moral dilemma when he says that Stage 6 resolves conflicts that Stage 5 cannot resolve. This assumption might have been given up in Kohlberg's (1990) paper, in which two subjects giving different solutions to the same moral dilemma are both ranked as Stage 6. See section 3.3 in this chapter.  3 1  89  existing stage. At the same time they present their latest thoughts about the characteristics of Stage 6 morality. Although there is nothing substantially different from the (1983) and (1984) books, this account provides us with a more refined formulation of the formal characteristics of Stage 6 moral reasoning. Owing to his death in 1987, this paper also gives us Kohlberg's last words on the issue of Stage 6's existence and nature. In "The Return of Stage 6," Kohlberg et al. report the interviews of two Stage 6 subjects, whom they call Joan and Judge D. They emphasize that the core idea of Stage 6 moral reasoning is respect for persons. In addition, they assert that there are two components of respect for persons, both of which are equally essential. One is justice, which is usually articulated in terms of rights, reciprocity, and equality. The other is benevolence or "active sympathy" for others (Kohlberg et al. 1990, p. 153). The former is best understood as the provision of protections against various action of others, while the latter is best understood as a sense of positive responsibility for the needs and welfare of others (Kohlberg et al., 1990, p. 156). O f course, it is impossible that potential conflicts between justice and benevolence can be overcome simply by unifying them under the title "respect for persons." If justice and benevolence can conflict with each other as two  principles,  they can also conflict with each other as competing aspects of respect.  Therefore, whether Stage 6 can really resolve all moral conflicts remains an open question. Kohlberg et al. believe that there is a sense in which benevolence is logically and psychologically prior to justice. To explain this they write: The way of regarding the other that we are calling benevolence views the other and human interaction through the lens of intending to promote good and prevent  90  harm to the other. It is an attitude that presupposes and expresses one's identification and empathic connection with others, or as Joan says, "is part of the responsibility of being a member of the human race." Thus, as a mode of interaction between self and others that manifests a Stage 6 conception of respect for persons, benevolence is logically and psychologically prior to what we are calling justice (Kohlberg et al. 1990, p. 157). The explanation is unclear. They reach the conclusion that benevolence is somehow prior to justice after stating the qualities required by benevolence, without considering the qualities required by justice or comparing two sets of qualities. They also claim that benevolence and justice constrain each other. Benevolence "constrains the momentary concern for justice to remain consistent with the promotion of good for all," (Kohlberg et al. 1990, p. 157). Justice constrains benevolence by requiring us to conceive others as individual autonomous agents when we promote the good for others (Kohlberg et al. 1990, pp. 157-8). This mutual constraints sounds plausible, but it still unclear how the two should be balanced when they come into conflict. Ideal role-taking seems especially important for understanding how Stage 6 resolves conflicts among Stage 5 values. It is the process of ideal role-taking in particular context that determines it, if anything does. Again, it is an important question whether the ideal role-taking can really resolve all moral conflicts. This question is closely connected with the problem of indeterminacy in moral philosophy, and we will return to this topic in Chapter 5.  The roles of principles and dialogue  Kohlberg et al. define a moral principle as a "general prescriptive proposition that guides individuals in making moral judgments about situations in which there are  91  conflicts between otherwise acceptable rules or norms" (Kohlberg et al. 1990, p. 158).32 They believe that the two Stage 6 subjects in the paper share the same supreme moral principle, namely the "principle of equal consideration of dignity or worth of all persons" (Kohlberg et al. 1990, p. 158). This formulation includes vague conceptions such as "equality" and "dignity," but the authors do not clarify them. Kohlberg et al. believe that it was the application of the same principle that made Joan say that the doctor should help the patient kill herself, and Judge D say that the people on the life boat should go for a lottery. Presumably what is included in Joan's response to the doctor's dilemma is the respect of the autonomy of the patient. What is included in Judge D's response is less clear, but we may suppose it is that all equivalent interests should be given the same weight. However, the principle of respect for persons remains very vague.  Besides the above principle, Kohlberg et al. also recognize the importance of dialogue. However, they believe that Judge D has exercised more typical Stage 6 reasoning than Joan because he was more aware that dialogue can break down, and therefore was willing to accept the lottery solution as the last resort in the lifeboat dilemma (Kohlberg et al. 1990, pp. 162-163). We have mentioned that Kohlberg believes Habermas is correct in emphasizing the importance of dialogue in the process of making a moral decision. But he believes that dialogue cannot replace the application of principle, for dialogue can break down while a valid principle cannot: principle is ultimate at least in this sense. But since, as we have seen, what actually determines the correct solutions in  This is similar to the early model, in which he says that a moral principle is "a rule or method of choosing between legitimate alternatives." See Kohlberg 1981, 220. 3 2  92  particular situations is ideal role-taking instead of the principles themselves, then we may say that ideal role-taking is ultimate. Formal  characteristics  of Stage 6 morality  Kohlberg et al. mention three main formal characteristics of Stage 6 morality: sympathy,  ideal role-taking,  and universality.  Sympathy here is not just a feeling or an  emotion. It requires at least two interrelated dimensions of social understanding. One dimension is the understanding of persons as "self-determining agents who pursue objects of interest to themselves." The other dimension is the understanding of general facts of the human condition within which persons exist and interact (Kohlberg et al. 1990, p. 165). Thus understood, sympathy "precludes assuming the validity of one's own conception of the actual interests of other persons a priori" (Kohlberg et al. 1990, pp. 165-6). The second characteristic of Stage 6 moral reasoning is the employment of ideal role-taking. The process of ideal role-taking is divided into three steps, which are essentially what the "second order application of the Golden Rule" requires: First, the subject takes the perspectives of others in the problematic situation in order to understand their interests, as expressed in their claims. Second, the intent to balance interests through ideal reciprocal role-taking involves the assumption that others are attempting to do the same thing. Third, the subject temporarily separates the actual identities of persons from their claims and interests in order to assess the relative merits of those claims and interests from the point of view of any person implicated in the dilemma. Kohlberg et al. believe that this process of role-taking can serve as a verification test for substantive  93  impartiality, and by doing this one can effectively balance the interests of different parties in a moral situation (Kohlberg et al. 1990, pp. 166-167). The third characteristic of Stage 6 moral reasoning is universalizability, which is most familiar to philosophers. Universalizability means two things. First, if one judges a certain kind of action to be right in a certain circumstance, one is committing oneself to judge all similar actions to be right in all similar situations. A Stage 6 subject realizes this, and therefore judges an action as a right one only when he is willing to commit himself to the same action in all similar cases. Second, when one makes a moral judgment, one attempts to find a solution that is acceptable to all. This explains why, in the process of ideal reciprocal role-taking, one has to separate oneself from one's actual identification. Therefore universalizability can be guaranteed by the process of ideal reciprocal role-taking. Dwight Boyd, one of the co-authors of "The Return of Stage 6," explicates the basic spirit of Stage 6 moral conception as follows. He maintains that justice for a Stage 6 reasoner is not a mechanistic application of abstract formulas. Rather, it involves (1) substantive ways of understanding persons and their interactions, and (2) a framework or process for resolving conflicts in a way acceptable to all parties. It recognizes that every person has the right to form his/her own ends, and strives to grant people as much freedom as possible to further their ends. The "motivational heart" of this structure is the  94  acknowledgment of "others as an equally unique subject of his or her own experience, including the experience of myself and my ends." 33 This "motivational heart," i.e., the recognition of individual persons as unique subjects of their experience, is an important idea. It seems that, if Kohlberg's emphasis on moral knowledge has any truth in it at all, it must have to do with this idea. That every individual is a unique subject of experience is indisputable. In Kohlberg's model, moral development presupposes development of social perspective. It implies that an adequate moral knowledge presupposes an adequate perspective of what a person is and how persons are related to each other. At Stage 6, there is a deep understanding of the subjective uniqueness of personal experience, and therefore the Stage 6 reasoner is most able to produce reversible moral decisions that are acceptable to all parties. This is done chiefly by ideal role-taking.  Section  4: Philosophical  implications  updated  In Chapter 1 we saw that the moral philosophy in Kohlberg's early model is a universalism that upholds justice as the sole moral value. It is also a rationalism, viewing that moral virtue is largely the ability to reason properly. In this way Kohlberg's theory is in line with traditional ethical theory, which aims at discovering a set of highly abstract and general principles that determines correct moral actions in particular situations. In  Dwight Boyd, "The Rawls Connection." In Moral Development, Moral Education and Kohlberg, ed. Brenda Munsey (Birmingham Alabama: Religious Education Press, 1980), 192. As we shall see later (Chapter 4, section 2.2), Kohlberg's theory is criticized as having a Western elitist bias. Boyd's explication helps us see why he invites such a criticism, for it involves the idea of maximizing individual freedom. This surely reflects the liberalist ideal. I do not mean to say that involving a liberalist ideal is itself an error. Whether it is an error depends on whether the liberalist ideal is a justified one, and we shall discuss this in Chapter 4.  3 3  95  light of the above explications of its recent development, we now need to summarize the major differences between the early and late models.  4.1 The differences between the late model and the early model  As we have mentioned, the redefinition of the moral stages involves sharpening the distinction between form and content of moral judgments. In the original definition, the chosen content is one of the elements determining the stage of a subject (Kohlberg, 1983, p. 42). In the late model, the stage of a person is determined by the "formal properties" of the person's reasoning. These formal properties include the social perspectives of a subject, the motive for doing the right thing, and the uses of justice operations, namely equality, equity, and reciprocity (Kohlberg, 1983, p. 42). These "formal" properties are still not moral forms in a strict philosophical sense, but at least we can say that moral stages in the late model are defined in more abstract terms than the early model.  Having separated the definitions of the stage from the content choices, Kohlberg introduces two moral types, namely Type A and Type B, that are more closely linked to moral content. Type B differs from Type A in that it is more autonomous than the latter. The criteria of Type B include a set of "Kantian" and a set of "Piagetian" criteria, both of which have been reviewed earlier in this chapter.  Kohlberg observes that when a change occurs in moral types, it is most likely a change from A to B. He finds that in Type B the justice reasoning occurs in an explicit way, while in Type A it occurs in an implicit way (Kohlberg et al. 1983, p. 45). In other words, Type B is better than Type A because, in part at least, Type B subjects are more  96  aware of the principles they employ than Type A reasoners. Although the stages themselves do not entail certain courses of action, there is a tendency for particular stages to be connected to certain action types. The same is also true for the moral types.  34  In the late model, Kohlberg retreats from the claim that morality is equivalent to justice. Instead of saying that morality covers only the realm of justice, he now recognizes that realms such as care and special obligations should also be covered (Kohlberg et al. 1983, pp. 19-21). He retreats from the claim that deontological justice is better than utilitarian morality. Rather, utilitarian morality, with careful refinement, can be 35  incorporated into the Stage 6 morality as a possible formulation of ideal role-taking (Colby et al. 1987 vol. 1, p. 31). The principle of utility was considered a part of Stage 5 in the early model, but is now regarded as belonging to Stage 6 when properly interpreted (Kohlberg et al. 1983, p. 63). On the other hand, he also admits that there are aspects of morality which his stage model does not cover. For instance, in The Measurement of  Moral Judgment, he and Colby divide moral judgment into three types: deontic judgment, aretaic judgment, and judgment about ideals of the good life. They claim that the major concern of the Kohlbergian stages is deontic judgments (Colby et al. 1987 vol. 1, pp. 2324).  This is problematic because, as we have seen, a subject's moral reasoning type is partially determined by his/her content choices for the dilemmas. So certainly there will be a connection between the type and the action chosen. Therefore this "tendency" for a type to be connected to a certain action is not very informative. 3 5 See Kohlberg, " A Current Statement on Some Theoretical Issues." In Lawrence Kohlberg: Consensus and Controversy, ed. S. Modgil and C. Modgil (Philadelphia & London: The Falmer Press, 1986), 521. 3 4  97 4.2 Claims that remains unchanged in the late model Though having abandoned several earlier claims, two central claims of Kohlberg's theory remain unchanged: that the stages represent a universal sequence o f development, and that the sequence marks a growth of moral maturity. If Kohlberg is correct at these two points, then one can conclude that comparison of (at least some) moral standards in terms of adequacy is possible, and we can accordingly dismiss some kinds of ethical relativism.  The other unchanged idea is that Kohlberg's ethics is universal (non-relative) but largely non-absolutist (Colby et al. 1987, v o l . 1, p. 35). Moral rules are not absolute, with a few exceptions such as that there should be no capital punishment. This means that most substantive values are relative to context. What distinguishes a higher moral stage from a lower stage is that the higher stage entails the recognition of the relativity of the lower-stage values. A t the preconventional level authority is an absolute moral standard, but at the conventional level it is not. A t the conventional level societal values are absolute, but at the postconventional level they are not. O n the other hand, it is also obvious that, in the process of moral development, people strive to maintain a unified vision of morality by subsuming all the recognized relativity o f moral standards or concerns under a more general and abstract moral principle or standard. A t the postconventional level, the subject becomes well aware o f the fact that these highly abstract principles or standards can hardly define every single right action for him. A t Stage 5, only a few basic values such as human rights are retained, but it is not clear whether a Stage 5 reasoner necessarily views human rights as absolute. A t Stage 6, it is  certain that even human rights cease to be absolute. What remains at Stage 6 is but a set of formal operations for moral decisions that are formulated under the postulate that all human beings have equal moral status, and are deserving of equal respect and dignity. In the previous chapter I pointed out that, though Kohlberg emphasizes that the moral adequacy is determined by "moral form," his theory does not amount to ethical formalism, for his "moral form" is different from the moral form as understood by philosophers. However, I also pointed out that his theory resembles ethical formalism, and indeed the entire traditional moral philosophy, in one way: it holds that there exist a set of highest moral principles that are most adequate and is able to determine correct moral actions in particular situations. This determinateness of Stage 6 moral principles are less explicit in the late model. Nevertheless, Kohlberg is still certain about what are the correct solutions to his hypothetical dilemmas. Heinz should steal the drug; people on the life boat should go for a lottery; and Joe should not give the money to his father. A l l these are still crystal clear, not a bit blurred by the redefinition of the Stage 6. This suggests that Stage 6 in the late model morality is as determinative as it has been in the new model.  The relation between "form" (moral reason) and content (choice) is weakened in the late model. In "Stage and Sequence" both the stage definitions and the scoring system presupposed some direct connection between the form of moral reasoning and the content of the choice. The moral stage of a person can be assessed by the content of his choice (Kohlberg 1983, p. 42). But now the form of moral reasoning is more sharply distinguished from the content. The formal properties of moral reasoning are, as we can  99  see from the table of definitions of the moral stages, confined to the social perspective, moral motive, the justice operations (equality, equity, and reciprocity), and the procedures like role-taking which support them (Kohlberg 1983, p. 42). Nevertheless, there is still some correlation between content choice and moral maturity. They are still related in the sense that (1) the higher the moral stage, the more likely that subjects will agree on a particular content choice in a moral dilemma; and (2) moral maturity is not only reflected in one's moral stage but also by its correlated type of reasoning which is defined partly in terms of the content of the choice. Kohlberg also retains what he calls the "cognitivist" claim in his late model, though perhaps in a weaker sense. Similar to the early model, higher stages are more mature partly in the sense that they are cognitively better. The hierarchy claim is supported in part by Walker's experiment, which finds that lower-stage subjects have difficulty in comprehending the higher stages. The Socratic claim that one who knows the good chooses the good, however, has quite surely been abandoned (compare Chapter 1, section 3, or Kohlberg 1981, p. 189). 36 The final claim which Kohlberg appears to retain in the late model is his principlism, i.e., the claim that the most adequate form of moral reasoning is a principled one. Stage 6 moral principles include the principle of justice or respect for human dignity and the principle of utility or benevolence. But the principle of benevolence involves the  Kohlberg does not make any assertion about this explicitly. However, he once mentions that there is a "horizontal sequence" from logical capacity to social perspective, and then to moral judgment, and finally to moral action (see Kohlberg 1984, pp. 171-172). This surely implies that it is possible that one can have attained the capacity of, say, postconventional moral reasoning, but not yet have the will power to carry out what she thinks to be right. 3 6  100  "attitude of universal human care and agape" (Colby et al. 1987 vol. 1, p. 31, my italics). Likewise, the inclusion of Agape in the realm of principles also loosens the notion of principle, and is hardly equivalent to what philosophers usually understand by the same term. The Categorical Imperative, however vague it is, is a formula that aims at defining what a right action (or at least a morally permissible one) is. It commands us not to do any action which cannot be universalized. Unlike the Categorical Imperative, what the so called 'Principle of Agape' commands is generalized love of others. No concept of action is directly involved in the command. What the command involves is rather more an attitude than a condition that tells the right action from the wrong. Therefore, though  Kohlberg may have no complaint about being named a "principlist," it is quite unlikely that he is really proposing a principlism in any strict philosophical sense, insofar as Stage 6 represents for him the supreme destiny of moral development. As I have repeatedly pointed out, what is really represented by Stage 6 morality is not only a set of very abstract principles, some of which are more properly thought of as attitudes, but also a set of rational procedures in the service of universalizability and justice. Therefore, Kohlberg's theory is at most a pseudo-principlism instead of a real one. By real principlism I mean the view that adequate moral reasoning is, ultimately, the application of the right principles. It is in this sense I say Kohlberg's theory is not a real principlism. There is, however, a weaker sense in which Kohlberg's theory can be regarded as a principlism, namely the view that there exists a set of general, abstract principles which jointly determined the moral truth and which are adequate, by means of procedures of application, to guide moral practice. But even so Kohlberg still owes us a complete list of these principles and their proper formulations.  101  As I have pointed out in the first chapter, Kohlberg's idea about the methodology of moral psychology and philosophy is one of reflective equilibrium between ethical theory and the empirical study, classification and assessment of people's moral reasoning. Confronted with anomalies in his empirical findings, Kohlberg makes a number of modifications to his theory. He sharpens the form-content distinction by redefining the moral incentives of the stages. He clarifies the meaning of moral stages by attaching to each of them a social perspective. He retreats from the claim that the existence of Stage 6 is verified, and introduces Type A and Type B moral reasoning into his model so that subjects ranked as Stage 6 in the early model are now ranked as Stage 3B, Stage 4B or Stage 5. He enlarges the moral domain to include care as well as justice, and redefines Stage 6 in terms of active sympathy, ideal role-taking, and universalizability. In spite of these changes, his theory remains a universalism holding abstract principles and values and a weak rationalism. He also retains the claim that Stage 6 reasoning is a principlism, but his understanding of principle is so broad and ambiguous as to rob this claim of much substance.  102  Chapter Three: Criticisms of Kohlberg's Theory Since one main theme of this dissertation is to examine the extent to which Kohlberg's theory supports ethical universalism, it is important that we examine the credibility of his theory. If his theory is itself incredible, then all attempts to support ethical universalism based on the theory will fail. Here, then, we shall examine the most systematically developed objections to the theory, and see how well the theory can survive these objections.  In the last chapter, we have seen that Kohlberg's new model has gained support from various studies. However, this does not mean that Kohlberg's theory is flawless. Today, many psychologists, including some of Kohlberg's main co-workers, believe some aspects of the theory to be mistaken or unsatisfactory. In this chapter I will take Carol Gilligan and Owen Flanagan as representative of Kohlberg's critics. I choose Gilligan because she is the most widely recognized opponent of Kohlberg's theory among psychologists. Flanagan is chosen because he has formulated the most systematic philosophical arguments against Kohlberg. Of course, Gilligan and Flanagan are not the only critics of Kohlberg's theory. I will refer to certain opinions of other critics when needed. But by discussing the Kohlberg's debates with Gilligan and Flanagan, we shall gain a fairly good idea about the merits and demerits of Kohlberg's theory.  Section  1: Gilligan  and the ethic  of  care  Gilligan is perhaps Kohlberg's most famous critic. She was first a student and then a colleague of Kohlberg, and has co-authored a paper with him on his stage theory.  103 However, as she went deeper into the issue of human moral development, she came to believe that the Kohlbergian model was not satisfactory. Gilligan's main criticism of Kohlberg's theory is that it is gender biased. This can also be interpreted as a charge of incompleteness. In her book In a Different Voice (abbreviated below as Gilligan 1982), Gilligan complains that for many years studies in moral psychology have been concentrated on male subjects.1 Theories have been developed based on these studies, and are presented as if they are applicable to all human beings. The problem with this approach is, she believes, that it fails to reflect the distinctive moral psychology of females, and for this reason the theories thus developed fail to apply universally. She then goes on to argue that Kohlberg's theory shares in this bias because the basic structure of his theory has been established on the basis of his original (1958) study, in which only male subjects were interviewed (Gilligan 1982, p. 18; Kohlberg et al. 1983, p. 122). Gilligan argues that the result of only male representation in Kohlberg's research is an unjustifiable concentration on the "orientation of justice or fairness" in moral reasoning, which Gilligan believes is largely a masculine one. She claims that what is thereby left out is the equally important orientation of care or responsibility, which is typically a feminine orientation in moral reasoning. By ignoring this feminine orientation, females are inevitably ranked as morally less developed in general when measured by Kohlberg's scale: Although Kohlberg claims universality for his stage sequence, those groups not included in his original sample rarely reach his higher stages. Prominent among those who thus appear to be deficient in moral development when measured by ' Carol Gilligan, In a Different Voice (Cambridge MA: Harvard University Press, 1982).  104  Kohlberg's scale are women, whose judgments seem to exemplify the third stage of his six-stage sequence (Gilligan 1982, p. 18). Gilligan adopts David McClelland's idea that the sex role is "one of the most important determinants of human behavior." With this idea in mind, she believes that psychologists who develop their theories based solely on the studies of males inevitably "regard male behavior as the norm and female behavior as some kind of deviation from that norm" (Gilligan 1982, p. 14). For this reason Kohlberg's theory is gender biased and ignores the fact that there exists a distinct moral language usually spoken by women, and consequently fails to estimate the moral maturity of women fairly. Gilligan suggests that, in order to remove the gender bias, one has to observe carefully how women make their moral decisions and how their decision-making processes differ from those of males.  1.1 T h e concept of i n d i v i d u a l person  According to Gilligan's observations, the word 'morality' tends to mean different things for males and females. A twenty-five-year-old male subject in Kohlberg's study said that for him morality essentially meant "recognizing the right of the individual." The individual rights thus recognized included, firstly, the "right to existence," and secondly, the "right to do as [one] pleases" (Gilligan 1982, p. 19). To put it simply, these are the right to life and the right to liberty, which are essential for postconventional moral concepts in the Kohlbergian model (Gilligan 1982, pp. 19-20). Comparing this subject's moral conception to that of a twenty-five year old female, the difference between genders becomes apparent. Like the above young man, the female subject rejected a strict relativistic view of morality, but with a different reason. Her reason was that "a person's life is enriched by cooperating with everybody  105  else, and to that end, there are right and wrong, there are things that promote that end and that move away from it." Even when an action affects nobody except the actor himself, she believed that it can still be wrong "to the extent it doesn't cohere with what [she] know[s] about human nature" (Gilligan 1982, p. 20). This is sharply different from the viewpoint of the male subject, who defined morality in terms of rights and noninterference. For a female, responsibility "signifies response, an extension rather than a limitation of action" (Gilligan 1982, p. 38). When the female subject was asked what made her take up such a moral view, and what she thought about herself in relation to it, she replied: Just seeing more of life, just recognizing that there are an awful lot of things that are common among people. There are certain things that you come to learn promote a better life and better relationship and more personal fulfillment than other things that in general tend to do the opposite, and the things that promote these things, you would call morally right. [I value] having other people that I am tied to, and also having people that I am responsible to. I have a very strong sense of being responsible to the world, that I can't just live for my enjoyment, but just the fact of being in the world gives me an obligation to do what I can to make the world a better place to live in, no matter how small a scale that may be on. (Gilligan 1982, p. 21) Gilligan emphasizes that this woman did not commit herself to such a moral conviction in any simple or direct manner. The subject had once been a relativist, believing that she should never pass any moral judgment on others. Her moral conviction was therefore a personal reconstruction of morality as a result of questions and doubts, just like all the male subjects at the higher stages in Kohlberg's studies. If this woman's case is typical of females in their moral conception, then there would be remarkable differences between the male and the female general conceptions of morality.  106 As we have already seen, the postconventional level in Kohlberg's model is marked by a conception of fairness. It seeks to resolve moral dilemmas by arriving at a choice which all rational persons could agree on. In the above example, the female subject obviously showed a concern for the responsibility she had for other individuals and the whole world. Morality existed because she cared for these people and because she cared for the world. Gilligan sees a moral orientation like this as significantly different from Kohlberg's ethic of fairness. Consider Heinz's dilemma for example. The orthodox interpretation of the dilemma is that it instantiates a conflict between the right to life and the right to property. According to Gilligan, while this interpretation is typical among males, females tend to interpret the case differently. Gilligan reports the response of Amy, an eleven-year-old girl, to the dilemma: " i f somebody has something that would keep somebody alive, then it's not right not to give it to them." Obviously, for Amy, the dilemma arose "not from the druggist's assertion of rights but from his failure of response" (Gilligan 1982, p. 28). Amy's solution to the dilemma was that Heinz should not steal the drug, but should talk it out with the druggist until they reach some compromise that avoids stealing (Gilligan 1982, pp. 28-29). Jake, an eleven-year-old boy, by contrast, gave a straightforward answer: Heinz should steal the drug, because "life is worth more than money" (Gilligan 1982, p. 26). Gilligan's interpretation of these answers is: Both children thus recognize the need for agreement but see it as mediated in different ways — he impersonally through systems of logic and law, she personally through communication in relationship. Just as he relies on the conventions to be shared, so she relies on a process of communication, assuming connection and believing that her voice will be heard (Gilligan 1982, p. 29, my italics).  107  Therefore, with this "different voice," a female's moral perspective is characterized by her view of the world as a world of relationships. It shows the subject's awareness of the connection between people, which in turn gives rise to a recognition of responsibility for others and a need for response. Since the subject views persons as basically connected to each other, the subject tends to resolve moral dilemmas by communication and contextual reasoning instead of applying abstract principles or by the ranking of values. Gilligan calls such a moral view the "ethics of care" (Gilligan 1982, pp. 30-32). By contrast, Kohlberg's ethic of justice defines morality in terms of fairness. Gilligan points out that Kohlberg's model "equate[s] moral development with the refinement of the idea of justice from its initial confusion with obedience to authority, through its equation with conformity to social roles and rules, to the recognition of its ideal form in the logic of equality and reciprocity. . . . In Kohlberg's system, the rational individual standing alone is the ideal moral agent" (my italics).  This difference between the moral conceptions of males and females can therefore be explained by the different imageries of human relationships between the sexes. We may say that the morality of justice is a morality of separation. Males tend to view persons (including themselves) as isolated entities. When asked to describe himself, Jake answered by presenting his abilities, beliefs, and physical appearance. Gilligan believes that his answer showed the tendency of males to understand themselves in terms of their own qualities apart from the world. This supports the notion that the idea of self for males  Carol Gilligan, "Do the Social Sciences Have an Adequate Theory of Moral Development?" In Social Science as Moral Inquiry, ed. Norman Haan et al. (New York: Cambridge University Press, 1983) 38-39. 2  108 is an isolated person. Females, on the other hand, tend to view persons as part of a larger entity. Unlike Jake, Amy described herself in terms of her interactions with others: "I want to help people." Gilligan believes that this is an example showing how females view themselves, i.e., they tend to view themselves through their positions in the world and their relations with others (Gilligan 1982, p. 34). If this contrast between males' and females' moral conceptions is correct, we can perhaps say that the male conception of morality is a negative or passive one. Assuming separation, they develop morally by establishing a conception of obedience to rules and laws so that individuals can be protected when they pursue their own goals. This notion of acting in accordance with rules inevitably gives rise to the categorical conception of morality. The disengagement of the self from his social world, on the other hand, gives rise to the idea of autonomy (Gilligan 1982, p. 46). The function of moral rules is mainly to limit interference and restrain aggression to minimize hurt, secure social contract, and protect autonomy (Gilligan 1982, pp. 37-38). In other words morality is only for the prevention of evils from happening. In contrast, starting with a view that persons are connected to each other, the moral conception of females is more positive: morality includes "doing what others are counting on her to do regardless of what she herself wants" (Gilligan 1982, p. 38). 1.2 Contextual reasoning and communication  A second but related difference between the moral reasoning of males and females is that males tend to adopt and apply general principles, while females tend to reason contextually. When Jake was asked how one should make his decision when his own interests and his responsibilities to others conflicted, Jake answered, " Y o u go about  109 one-fourth to the others and three-fourths to yourself (Gilligan 1982, p. 35). Gilligan's comment is that "Jake constructs the dilemma as a mathematical equation, deriving a formula that guides the solution" (Gilligan 1982, p. 37). In contrast to Jake's formula, Amy's answer was "it depends," and she rejected any general guide as the solution of the problem (Gilligan 1982, p. 38). In addition, Gilligan claims that females tend to resort to communication more than males do. We can see this tendency in the emphasis on the importance of communication in Amy's response to the Heinz dilemma: Heinz should not steal the drug but talk it out with the druggist. This emphasis on communication is perhaps related, firstly, to the contextualist conception behind females' moral reasoning. For without communication we cannot identify the particular needs of individuals in particular  situations. Secondly, it is also related to females' tendency to view individuals as related to, not isolated from, each other (Gilligan 1982, p. 29). To summarize, Gilligan attempts to demarcate two gender related orientations of moral reasoning, namely the orientation of justice and the orientation of care. The orientation of justice is typically exemplified in males' reasoning in terms of rights, fairness, and justice. The orientation of care is typically exemplified in females' reasoning in terms of responsibility and care (as contrasted with selfishness and hurt) as its core concepts (Gilligan, 1982, p. 73). The orientation of justice is characterized by applying abstract principles and rules in a categorical way, while the orientation of care is characterized by a contextual and narrative mode of thinking with an emphasis on communication in the resolution of moral dilemmas (Gilligan, 1982, p. 19).  110 1.3 Development of care reasoning  In her (1982) volume, Gilligan does not suggest that we should abandon Kohlberg's stage model of moral development as a whole. 3 Instead she wants to complement his theory by introducing another component of moral operation, namely care and responsibility. By interviewing pregnant women on how they decided whether they should abort or continue their pregnancy, she develops a three level model of development regarding the care orientation of moral reasoning. At the first level in Gilligan's model, the conception of care is centered on the self alone. The self is the sole object of caring and the goal is ensuring the survival of the self. At the second level, morality is represented by responsibility, which is understood within the framework of maternal morality. Being good is identified with care for others and self-sacrifice. At this level, caring for self is criticized as being selfish. At the third level, however, the self is recognized again to be a legitimate object of one's care. The tension between one's own need and others' are reconciled in the awareness that the self and others are interdependent. Morality at this level becomes a self-chosen morality of nonviolence, interpreted as a universal responsibility to exercise care and avoid harm (for both the self and others). This pattern of development in terms of care reasoning represents a "progressively more adequate understanding of the psychology of human relationships" (Gilligan 1982, p. 74).4  She later does propose that we should abandon the concept of developmental stages in the Kohlbergian sense. She suggests that development is a complicated process. It is "fraught with vulnerabilities", and "it entails both loss and gains." Therefore organizing human growth into stages and calling some of them higher than others does not make good sense of human development. Carol Gilligan, Annie Rogers, & Lyn Mikel Brown, "Soundings into Development." In Making Connections, ed., Carol Gilligan, Nona P. Lyons, & Trudy J. Hanmer (Cambridge, MA: Harvard University Press, 1990), 314-334, esp. 319-320. 4 This claim of the increasing adequacy is important for Gilligan. It is important for two reasons. First, it seems obvious that some forms of care are better than others. It is hard to see, for example, how care at 3  Ill 1.4 The relation between care and justice reasoning  Several questions can be asked about the two orientations. How are they related to each other? Are they utterly different or incompatible? What are the relative statuses of these orientations when compared to each other? Is one of them in any sense better than the other? Or are the different orientations each better in solving some particular kinds of moral problems?  Figure 1 (Joseph Jastrow, 1900) Gilligan sometimes uses the exan pie of ambiguous figures, such as a drawing that can be seen as a duck or a rabbit (Fig ure 1), to illustrate the nature of the contrast between the two orientations. Since it is impossible for one to see the figure simultaneously as a duck and a rabbit, the use of such analysis suggests that Gilligan thinks of the two orientations as in some sense incompatible, and indeed, she sometimes writes in this way:  level three in her (1982) model can fail to be better than at level 1. Second, she needs such a claim to make her theory a proper foundation for criticism of Kohlberg. For without a claim about the adequacy of different forms of care reasoning, there would be no problem for Kohlberg in leaving out the care orientation, for it would not make a person less mature to be uncaring. Therefore, even though she wants to abandon the concept of developmental stages in her (1990) paper (see the previous note), she still retains the claim that the pattern of development described in the (1982) book is a progression (p. 317). Whether it makes any sense to speak of progression while abandoning the concept of developmental stage (among which some are higher than the others) remains a question.  112  Within each perspective, the key terms o f social understanding take on different meanings, reflecting a change in the imagery of relationship and signifying a shift in orientation. A s the illustration of the ambiguous figure is perceived alternately as vase or faces, so there appear to be two ways o f perceiving the self in relation to others, both grounded in reality but each imposing on that reality a different organization. But, as with the perception of the ambiguous figure, when one configuration emerges, the other temporarily vanishes (my italics).  5  However, we have good reason to believe that Gilligan has overstated the incompatibility o f the two moral orientations within her own theory. In the same paper, a subject says that she and her friend "depend on each other in a way [they] are both independent." What she means by such a paradoxical expression is that she and her 6  friend are to a large extent independent, but neither can be completely so. Each needs the help and support of the other to survive. Such a view about interpersonal relationship suggests a possibility o f synthesizing the two moral perspectives, i.e., we can view a person as an individual who is connected with others. To the extent that they are individuals they are independent, but to the extent that they are connected with others they depend on others. This may explain why Gilligan softens her tone in a later paper when she talks about the tension between the two orientations: The analogy to ambiguous figure is useful here in several ways. First, it suggests that people can see a situation in more than one way, and even alternate ways o f seeing, combining them without reducing them — like designating the rabbitduck figure both rabbit and duck. Second, the analogy argues against the tendency to construe justice and care as opposites or mirror-images and also against the implication that these two perspectives are readily integrated or fused. 7  5  Carol Gilligan, "Remapping the Moral Domain: New Images of the Self in Relationship." In  Reconstructing Individualism: Autonomy, Individuality, and the Self in Western Thought, ed. Thomas C.  Heller (Stanford: Stanford University Press, 1986), 242. Ibid., 249. 7 Carol Gilligan, "Moral Orientation and Moral Development." In Women and Moral Theory, ed. Eva Feder Kitty & Diane T. Mayers (Totowa: Rowman & Littlefield, 1987), 30.  6  113  Though not utterly incompatible, the two moral conceptions do, for Gilligan, sometimes give rise to different solutions to the same moral dilemma. She gives an example: A boy and a girl were playing together. The boy wanted to play as pirates, but the girl wanted to play as next door neighbors. The solution the girl came up with was that the boy "be the pirate that live[d] next door." Gilligan calls this an "inclusive solution," which she contrasts with the typical "fair solution" of taking turns and playing each game for an equal period. The fair solution "leaves the identity of each game intact," but in the inclusive solution each child is invited to "enter the other's imaginative world" and "transform that world by his or her presence." This is the difference between the justice conception of morality and the care conception of morality. 8 In this sense there is a real difference between the two orientations. They are not just two formulations of the same moral conception, and neither of them can be reduced to the other. They seem to be equally valid in general, but in solving a particular moral problem one of them may be better than the other. Both orientations have their disadvantages. For Gilligan, the problem with justice reasoning is that it is apt to "confuse one's perspective with an objective standpoint or truth," and "define others in one's own terms by putting oneself in their place." 9 Gilligan does not explain why this happens, but her point is not hard to imagine. In Kohlberg's model, moral problems are solved by "ideal role-taking," which means that one has to imagine what he/she would have chosen if he/she were in the places of the other parties involved in the situation. However, when deciding what one would have chosen if one was in the place of the other, there is a distinct tendency for one to choose according to his/her own values and  8  Carol Gilligan, "Remapping the Moral Domain," 242-243.  114 his/her own preferences rather than adopting the values of the other. It might happen that the other party involved in the issue has different but equally reasonable values and preferences, leading to a different choice than the subject's, while the subject is unable to appreciate that. On the other hand, the potential error in care reasoning lies in the fact that it is easy for the subject to forget his/her own values and needs. By admiring the value of selflessness one is apt to "define oneself in other's terms." 1 0 But the danger involved in role-taking has been overstated. When one defines others in one's own terms, he/she does not put himself/herself 'in others' places; he/she puts others in his/her own place. There is certainly a risk of projecting one's own values onto others when one takes others' roles, but when one does so one's role-taking is by no means ideal. By attaching a set of social perspectives to the stage definitions, Kohlberg makes clear that higher stages are characterized by higher ability to appreciate the individuality of every person. Also, by adding active sympathy to Stage 6, he also makes clear that projecting one's values onto others in role-taking is itself a kind of injustice. The two orientations are not defined in terms of gender but in terms of theme. O f course, for Gilligan, the two orientations are associated with gender, but the association is not absolute. Gilligan and her colleagues observe that (1) both justice and response (or care) orientations are employed by both sexes, though (2) most females (75%) use response orientation predominantly, while most males (79%) use justice orientation predominantly." Therefore, it is a mistake to think that care orientation is unique to  Carol Gilligan, "Moral Orientation and Moral Development," 31. Ibid., 31. 11 See Kohlberg et al. 1983, vol. 1, p. 123. Original data is from Carol Gilligan, S. Langdale, N. Lyons, & J. M. Murphy, "Contribution of Women's thinking to Developmental Theory: The Elimination of Sex Bias in Moral Development Theory and Research," Final Report to National Institute of Education, 1982.  9  10  115 women and justice orientation is unique to men (Gilligan 1982, p. 2).  Moreover,  empirical data also show that gender is not the only factor affecting the moral orientation employed. The choice of orientation depends in part on the type of problem posed. In general, stories that highlight inequality and attachment result in higher rates of justice and care responses, respectively, for both men and women. But whatever the kind of moral question the story highlights, the proportion of females who choose care orientation remains higher than that of males. If each sex is capable of both kinds of moral reasoning, and if both forms of reasoning have their own limitations, then it seems the best answer to the question of superiority will be that neither is better than the other. Since different moral problems tend to generate moral reasoning utilizing different orientations, it seems that the best moral reasoning should integrate both orientations, perhaps with some kind of psychological division of labor. The idea that we should strive for a cooperation of the two moral orientations is very likely what Gilligan suggests.14 Unfortunately, no systematic proposal for integration has been formulated.15  Also see the section "Reply by Carol Gilligan" in Linda K. Kerber, Catherine G. Greeno & Eleanor E. Maccoby, Zella Luria, Carol B Stack, and Carol Gilligan, "On In a Different Voice: An Interdisciplinary Forum," Journal of Women in Culture and Society 11, no. 2 (1986), 327. 13 Carol Gilligan and Grant Wiggins, "The origins of Morality in Early Childhood Relationships." In The Emergence of Morality in Young Children, ed. Jerome Kagan & Sharon Lamb (Chicago & London: University of Chicago Press, 1987), 283-286. Also see Owen Flanagan, Varieties of Moral Personality (Cambridge MA & London England: Harvard University Press, 1991), 214. 14 Carol Gilligan, "Moral Orientation and Moral Development," 30. 15 Ibid., 27. 12  116 Section  2: Reply  from  Kohlberg  and his  school:  2.1 Incomplete?  As I mentioned in Chapter 2, Kohlberg agrees with Gilligan that an ethic of care usefully enlarges the moral domain, and thereby in the late model Stage 6 principles include not only the principle of justice, but also the principle of benevolence and Agape. In this sense, Kohlberg admits that his model, at least in its early formulation, is incomplete. However, as Gilligan points out, Kohlberg's attitude towards the relation between morality, justice, and Kohlberg's stages.remains ambivalent. In Kohlberg et al.'s (1983) book, Kohlberg and his colleagues respond to Gilligan's criticism by saying that they "now need to emphasize the nomenclature 'justice reasoning', since Kohlberg's stages have more typically been called stages of moral development" (Kohlberg et al. 1983, p. 17). This suggests that morality and justice are different, and his stages capture the development of justice reasoning but not moral reasoning. In other places, however, Kohlberg still calls his stages moral stages and his theory moral developmental theory.  2.2 G e n d e r biased?  Kohlberg admits that the charge of gender bias is "partially true of the original Kohlberg [1958] method," in which care orientation in moral reasoning tended to be scored Stage 3 (Kohlberg et al. 1983, p. 125). He even admits that "the 'principle' of altruism, care, or responsible love has not been adequately represented" in his work (Kohlberg et al. 1983, p. 20). However, Kohlberg insists, with the modified scoring method, his model is not gender biased.  117 In several of his publications, Kohlberg does suggest that females are usually morally less developed than males. 16 But for him this difference between the sexes is comparable to the difference between working class males and middle class males. Both are results of the difference in social roles. Since women are traditionally provided with less chance of social participation, particularly regarding the exercise of power, they have less opportunity for role-taking and hence their moral development is hindered (Kohlberg 1983, p. 122). Two questions are relevant to consider whether Kohlberg's model is gender biased. The first question is whether there are mean differences between males and females on the scores in Kohlberg's moral judgment tests. The second question is whether women's responses to the dilemmas can be scored in terms of the Kohlbergian stages so as to yield the same results of invariant sequence and structured wholeness as are found in the development of males (Kohlberg et al. 1983, p. 127). 17 To answer the first question, Kohlberg draws upon Walker's (1982) review of 54 studies using Kohlberg's moral judgment interview and 24 studies using Rest's (1979) Defining Issue Test. 18 According to this review, the only studies showing fairly frequent gender related differences are studies of adults among whom most female subjects are  Lawrence Kohlberg, "Stage and Sequence: the Cognitive-Developmental Approach to Socialization." In Handbook of Socialization on Theory and Research, ed., David A. Goslin (New York: Rand McNally, 1969); Lawrence Kohlberg and R. Kramer, "Continuities and Discontinuities in Childhood and Adult Moral Development," Human Development 12 (1969), 93-120. 17 For Colby et al.'s study see Anne Colby, Lawrence Kohlberg, John Gibbs, and Marcus Lieberman, " A Longitudinal Study of Moral Judgment." In New Research in Moral Development, ed. Bill Puka (New York and London: Garland Publishing , 1994), 1-124. 18 Lawrence J. Walker, "Sex Differences in the Development of Moral Reasoning: A Critical Review of the Literature," Paper presented in Can. Psychol. Ass., Montreal 1982. A more elaborated version of the paper is published as "Sex Differences in the Development of Moral Reasoning: A Critical Review," Child Development 55, 1984, 677-691. For Rest's Defining Issue Test see James Rest, Development in Judging Moral Issues (Minneapolis: University of Minnesota Press, 1979). 16  118 housewives. It is thus very likely that the differences are a result of the differences between education and job status of male subjects and female subjects in the sample. Indeed, Speicher-Dubin (1982) and Powers (1982) report that sex differences disappear when the variables of higher education level and job status are statistically controlled (Kohlberg et al. 1983, p. 129).19 In a more egalitarian social context, like that of the kibbutz, no differences are found between the sexes. Weisborth's (1970) study also found that, with the education and occupational status under control, there were no gender difference in the scores.20 Therefore, Kohlberg concludes that, the apparent gender difference is really a product of education and job opportunity (Kohlberg et al. 1983, pp. 129-130). Responding to the second question, Kohlberg claims that his moral judgment test and standard scoring system produce invariant sequence and structured wholeness for females just as it does for males. For example, the Snarey et al. (1982) longitudinal study found that both the development of males and females formed invariant sequences and also structural wholes. 21 Another piece of evidence is Erikson's (1980) study, which showed that females' development as measured by Kohlberg's method was always upward and there was no stage skipping.  19 B. Speicher-Dubin, "Parent Moral Judgment, Child Moral Judgment and Family Interaction: A Correlation Study" (Ph.D. diss. Harvard University, 1982). S. Powers, " Family Interaction and Parental Moral Judgment as a Context for Adolescent Moral Development: A Study of Patient and Non-Patient Adolescents" (Ph.D. diss. Harvard University, 1982). 20 S. P. Weisborth, "Moral Judgment, Sex and Parental Identification of adults," Developmental Psychology 2 (1970), 396-402. 21 A. Higgins, C. Power, L. Kohlberg, "The Relationships of Moral Atmosphere to Judgments of Responsibility." In Morality, Moral Behavior and Moral Development: Basic Issues in Theory and Research, ed. William M. Kurtines & Jacob L. Gewirtz (New York: John Wiley & Sons, 1984), 74-106. 22 V. L. Erickson, " The Case Study Method in the Evaluation of Developmental Programs." In Evaluating Moral Development, ed. Kuhmerker, Mentkowski, and Erickson (New York: Character Research Press, 1980).  119 For Kohlberg, it is untrue that males and females tend to solve moral dilemmas in terms of justice and care respectively. Higgins et al.'s (1983) study, for example, showed that the major differences among students in using the two orientations were mainly determined by two factors. First, it depends on the type of dilemma to be solved. "The dilemma about helping another student elicited caring considerations, while a dilemma about theft elicited justice concerns." On the other hand, it is determined by the sociomoral atmosphere of the school situation. The more familial or communitarian the sociomoral atmosphere, the greater the tendency to use the care orientation. 2.3 Lacking contextual sensitivity? Another charge against Kohlberg's theory is that his moral ideal ignores the importance of contextual considerations concerning special responsibilities and relationships in moral reasoning. In response, Kohlberg insists that even in situations which involve special relationships (such as family, friends, and groups of which the subject is a member), a "universalistic justice ethic of respect" is enough to solve moral problems, though he also admits that a "morality of particularistic relations" is also suitable for handling those situations (Kohlberg et al. 1983, p. 20). This claim of Kohlberg's may not itself be very convincing. But I think the crucial point is that emphasizing the importance of universal moral principles for an adequate understanding of morality does not imply that context is unimportant. He quotes the following statement from Nunner-Winkler to explain this idea:  Higgins et al, "The Relationships of Moral Atmosphere to Judgments of Responsibility." Also see Kohlberg etal. 1983, p. 132.  23  120 Therefore one cannot very well hold context orientations to be a feature to constitute contrasting approaches to morality. Context orientation is a prerequisite for all actual moral judgments (Kohlberg et al. 1983, p. 140). Nunner-Winkler's opinion is that Gilligan's ethic of care does not differ from Kohlberg's ethic of justice by its contextual sensitivity. The difference is in emphases: whereas care emphasizes the positive duty to help, justice emphasizes the negative duty of noninterference (Kohlberg et al. 1983, p. 139). When Kohlberg talks about the differences between A and B types of moral reasoning, he points out that Type B moral reasoning is not only more autonomous but also more "flexible in regard to rules and persons" (Kohlberg et al. 1983, p. 18). Since Type B moral reasoning is held to be more adequate than Type A , this further confirms the view that contextual sensitivity is a requirement of moral adequacy in Kohlberg's theory. 2.4 Two separate moralities? It is clear, then, that Kohlberg acknowledges the difference in the emphases between the ethic of care and the ethic of justice. Does he also admit that the two moral orientations are, as Gilligan puts, two separate moralities? The answer is a definite "no." Consider Heinz's dilemma. To solve the dilemma, the subject does not choose between care and justice. On the contrary, concern for the wife and the recognition of the wife's right to life are interwoven. The more Heinz cares for his wife, the stronger he feels that he has an obligation to protect her right to life. The same holds true even when the person who needs help is a stranger. One postconventional subject, for instance, said that in a significant sense all human beings are related to each other: "He felt human and that was enough of a bond" (Kohlberg et al. 1983, p. 135). It is thus obvious that one need not give  121 up the care orientation to reach a high Kohlbergian stage. On the contrary, "at the postconventional stages there is typically an effort to integrate concerns of benevolence and care on the one hand with justice concern on the other." After all, the principle of persons as ends in themselves is "common to both the ethic of care and the ethic of justice" (Kohlberg et al. 1983, p. 137). It is tempting to understanding the difference between justice and care, as Gilligan seemingly does, in terms of the difference between non-violation of rights and promotion of welfare. 24 But this understanding is not perfectly accurate. Kohlberg himself is responsible for this construal of his justice, for he says that morality is based on individual rights, and moral wrongness is understood in terms of the violation of these rights (Kohlberg 1981, p. 300). But looking at the definition of Stage 6, we see no reason to understand morality in this way. The most important element in Stage 6 morality is equal respect for persons, and the non-violation of rights is only part of it. (And since Stage 6 morality is universalistic but not absolutistic, we should not even suppose that non-violation of rights always has priority to the promotion of human well being.) In his (1981) anthology, Kohlberg presents an imaginary case in which a person is drowning in the river. He argues that a passerby should jump in to save the drowning person if the chance of success is great enough to compensate the risk of both drowning to death (Kohlberg 1981, p. 204). This means that Stage 6 morality also requires us to be responsive to others' needs. Similarly, when he argues that Heinz should steal the drug  The non-violation of rights is sometimes referred to as 'justice,' and the promotion of welfare as 'benevolence.' However, as I will show immediately, Kohlberg's ethic of justice is by no means limited to the non-violation of rights. Therefore I will avoid using 'justice' and 'benevolence' to refer to these two orientations in moral reasoning. 24  122 for his wife, he is talking not just about a negative obligation of not taking lives, but a positive duty to save lives. Therefore the ethic of justice and the ethic of care do not represent two "separate general moralities." But Kohlberg still emphasizes the importance of justice in morality, since without some orientation to justice one cannot even properly recognize most moral dilemmas. His reason is that moral dilemmas appear where there are conflicting claims among people, "all of whom should be cared about" (Kohlberg, 1983, pp. 21-22). The ethic of care is a distinct orientation from the ethic of justice only in that "special obligations of care presuppose but go beyond the general duties of justice" and hence "elicit care responses which supplement and deepen the sense of generalized obligations of justice" (Kohlberg, 1983, p. 21). Section  3.1  3: Reflections  on the Kohlberg-Gilligan  debate  Gender bias In my interpretation, incompleteness and gender bias are Gilligan's main  criticisms of Kohlberg's theory. Kohlberg insists that there is no gender bias in his theory. He is right in this respect. A necessary condition for Gilligan's claim of gender bias is that Kohlberg's model inaccurately down-scores females' moral reasoning. If neither gender has general superiority over the other in their moral maturity, then the existence of sex bias will mean that females are on average ranked lower than males according to Kohlberg's scheme. After reviewing a wide range of studies, Walker finds that there is little evidence for the claim that females are underscored by Kohlberg's method:  123 Walker (1984, 1986) reported a review of the literature that included all studies using Kohlberg's measure in which sex differences in the development of moral reasoning were, or could be, examined. M y review included 80 studies with 152 samples, and involved a total of 10,637 subjects. Of the 152 samples, a nonsignificant difference was reported for the vast majority, 130 (85.5 percent). Females had higher scores in 9 samples (5.9 percent), whereas males had higher scores in 13 samples (8.6 percent).... It is interesting that in every case where researchers controlled in some manner for education and/or occupation, sex difference disappeared. This is not an attempt to "explain away" these findings, but to illustrate Kohlberg's claim that such experiences are influential in stimulating moral development. The above is the result of the vote-counting, which involves simply counting the number of studies favoring or disfavoring a certain theoretical claim. Walker also reports the result of another analysis, which supposedly assesses a theoretical claim in a more accurate way by analyzing the results of various studies collectively, namely metaanalysis. The analysis also showed that Gilligan's claim is not true: Walker (1984, 1986) also reported a meta-analysis of the studies included in his review and found that the probability level for the overall pattern of findings did not reach significance. In subsidiary analysis the effect size was found to be extremely small: Sex explained only 1/20 of 1 percent of the variability in moral reasoning development. 6 3.2 Incompleteness However, if we concentrate too much on the problem of sex bias in Kohlberg's theory, we will miss the most interesting point in Gilligan's criticism. Gilligan claims that the two moral orientations (justice and care) are not defined by gender but by themes. A major reason why Kohlberg's theory is interesting is that it claims to be universally applicable. If there exists a moral orientation that is not reflected in Kohlberg's stage  Lawrence Walker, "Sexism in Kohlberg's Moral Psychology?" In Moral Development: An Introduction, ed, William M. Kurtines & Jacob L. Gewirtz (Boston & London: Allyn and Bacon, 1995), 100.  25  124  model, and if this orientation is no less adequate than the justice orientation, then Kohlberg's model would be insufficient for explaining human moral phenomena. As we have seen, Kohlberg admits that Gilligan's care orientation usefully enlarges the moral domain. This is in effect to admit that his early model is incomplete. In his late model, he extends Stage 6 morality so that it embraces principles other than justice. He seems to agree with Gilligan that justice and care are not merely two different formulations of the same morality. But for him the two orientations are not separate either. One of Kohlberg's most powerful arguments concerning interconnections between justice and care is that the more one cares the harder one will try to protect the rights of others. To care is to help others and to avoid harm to them, while to be just involves, chiefly, protecting the rights of others. But protecting rights is surely one way of avoiding harm. Moreover, rights can either be positive or negative. Negative rights are those which prohibit certain types of actions against others. For instance, the right to property is negative in the sense that it prohibits any person's taking over another's property by force. Positive rights prescribe certain positive actions regarding others. When we say people have a right to education, we do not mean only that a society should not interfere with the educational activities of its people, but also that society has a positive duty to provide education for its people. In this sense, positive rights are parallel to helping those in need, and negative right is parallel to avoidance of harm. According to Gilligan, Kohlberg's morality is largely one of non-interference. This would mean that Kohlberg gives primary place to negative rights over positive rights. However, it is difficult to decide whether Gilligan is correct. We can see that, in the Stage 5 moral conception, Kohlberg does emphasize the rights to life and liberty, both  125 being negative rights. But at Stage 6 morality, we find the Golden Rule, the principle of equal respect for all persons (treat every person as an end but not as means), and a procedural method of decision making (ideal role-taking). None of these suggests that non-interference is being emphasized to the point of relative exclusion of positive concern. So far, then, it would seem that Kohlberg's theory does not, as Gilligan argues, leave out the ethic of care. The problem is, however, that Stage 6 is not formulated in the new model in any empirically testable fashion. But Stage 5, the highest verifiable stage, does emphasize negative rights, and thus more definitive connection with the morality of non-interference. And if we remember that, for Kohlberg, the Stage 5 reasoner tends to agree on the same solutions that as the Stage 6 reasoner, it seems that the connection between morality of non-interference and Stage 6 is apparent. I think we can probably say this about Kohlberg's theory: as far as the stage  definitions are concerned, there is no obvious sign of ignoring the care orientation in moral reasoning. But when it comes to more substantive matters, Kohlberg does deemphasize the care orientation. This is evident when we look at the scoring manual. In the Heinz dilemma, not a single Stage 5 "criterion judgment" (i.e., sample judgment illustrating how moral judgments or justifications should be ranked) has anything to do with commitment, special responsibility, or compassion.27 The ethic of care is also supposed to be more innovative. Consider Gilligan's game example, in which a boy and a girl wanted to play two different games. Gilligan presupposes that using the justice orientation, the solution would be that the children take turns playing different games, while using the care orientation the solution would be  126  playing a new game that combines the two. But what principle of justice prescribes that the children should not play a game that combines the two games? Certainly, Gilligan does not mean that the justice orientation must forbid the solution reached by the care orientation. For her, the demarcation between justice and care by no means entails that justice is uncaring and care is unjust.28 But they are still two different approaches to solving moral problems. When compared to the justice orientation, the care orientation has "a greater tolerance, a greater tendency toward innovation in solving conflicts, a greater willingness to make exceptions to rules, and a lesser concern with legal elaboration."  When confronting a situation, the justice orientation will ask the question T A  "What is just?", while the care orientation will ask "How to respond?"  Presumably,  when a person asks this "how" question he is prepared to be more innovative in the deciding the right action. But is this true? Why cannot a just person ask instead "How to make a just decision?", and why cannot a caring person ask "What to do?" Distinguishing the two orientations in terms of "what" and "how" questions simply does not help. To begin a question with different words does not necessarily make two different questions. It is also a mistake to suppose that the difference in flexibility and innovativeness is intrinsic to the justice-care distinction. We can be just and flexible, just and inflexible, unjust and flexible, or unjust and inflexible (being just and inflexible is perhaps the least possible among the four). The same applies to the relation between justice and innovativeness, care and flexibility, and care and innovativeness. Consider an example quoted by Gilligan  27 28  Colby et al. 1987 vol. 2, pp. 11 -12, 63-64. Carol Gilligan, "Moral Orientation and Moral Development," 22.  127 herself. She describes the woman who "comes before Solomon and verifies her motherhood by relinquishing truth in order to save the life of her child" as an example of care (Gilligan 1982, p. 104-105). She does not mention Solomon. He does not relinquish the truth, but is he not innovative when he invents an ingenious method to discern the real mother and then does her justice? Why then has Kohlberg invited the criticism of being inflexible? I think the answer is twofold. First, it is due to the setting of the moral judgment interviews. For example, in Heinz's dilemma, the subject is forced to choose between letting the wife die or to steal the drug. When the subject raises other options, the interviewer will tell him/her that Heinz has tried that already, but it does not work. The setting forces the subject to choose between competing values so that scorers can see how (or according to what standards) he/she weighs different values. In this setting, not surprisingly, only those reasoning processes used to choose between readily available options will be elicited. Various possibilities for flexibility and innovativeness in one's reasoning remain undetected. The second reason Kohlberg has invited such criticism is his single-minded commitment to certain solutions for the dilemmas. Unfortunately, not all of his solutions make good senses to his readers. For instance, in the life boat dilemma, Kohlberg insists that the just solution is to go for a lottery. This is, of course, highly problematic. As 31  Carter has asked, why cannot the people on the life boat just join hands and sing? Kohlberg's solution does not seem to follow from his ideal role-taking. Indeed, we can 30 31  Ibid., 23. Robert E. Carter, "What is Lawrence Kohlberg Doing?" Journal of Moral Education 9, no. 2 (1990), 99.  128 quite forcefully argue against his solution from the Stage 6 principle of respect for individual persons. The principle requires us never to treat others merely as a means but always as an end. If forcing the old man to go overboard is to treat him as a means, then to determine who should go overboard by forced lottery equals determining who should be treated as a means by the lottery. Of course it is better if the three people have already agreed to use the lottery method. But then it is the agreement that makes it right, not the lottery itself. Therefore the just solution is to do whatever the three can agree on, not simply to go for the lottery. Taking the Stage 6 morality as one that emphasizes agreement seeking, everything Gilligan says about the care orientation gets in. To arrive at an agreement, there has to be communication. To arrive at a list of (good) options, there needs to be some innovative and flexible thinking. Gilligan contrasts the real mother in Solomon's story with Abraham (Abraham being taken as an example of someone using the justice orientation), who "prepared to sacrifice the life of his son in order to demonstrate the integrity and supremacy of his faith" (Gilligan 1982, p. 104-105). But as I have pointed out, Solomon is a just man, and he is no less flexible than the woman. And it is simply a mistake to say that Abraham is just, for what he tries to do is terribly unfair to his own son. When we look more closely at Gilligan's moral developmental model defined in terms of care, the correlation between the orientations of care and justice becomes even more evident. From the first level to the third level, the meaning of care does not change. Care always means to help the one who is in need and to avoid harm. What changes is the object of care. At the first level, the subject concern is only for herself. At the second  129 level, she cares for others but not the self, and therefore views self-sacrifice as the major value. At the highest level, the objects of care include both the self and others. The similarity between this scheme and Kohlberg's is remarkable. The first level corresponds to Kohlberg's preconventional level, in which right and wrong are defined in terms of self-interest. The second level also parallels Kohlberg's level 2, though the parallelism is less straight forward. By viewing self-sacrifice as the major moral value, the female adopts a particular role assigned by the society to her (i.e., as a nurturer). This perfectly instantiates Kohlberg's conventional morality, in which moral actions are defined in terms of social norms or societal expectations. At the highest level, the subject recognizes that she deserves her own care as much as others do. This is obviously an awakening to the call of equality: everybody is equal, and nobody should be assumed to have an obligation of self-sacrifice. At level 1 the subject is unfair to others; at level 2 unfair to herself; but at level 3 she tries to be fair  to everybody.  This level 3 is exactly  what the postconventional level is all about.32  Therefore, Kohlberg can quite reasonably claim that care and justice are two aspects of his model. These aspects cannot be reduced to each other, and are not different formulations of the same thing (which would mean that the two orientations are identical), but they are still not two separate moralities. They may not be in perfect  Even Gilligan herself would most likely agree with me on this point. In her (1990) paper she writes: Women's concerns about their own survival in the face of what they perceive to be abandonment by others, and their concerns about their appearance or "goodness" in the eyes of others, and their concern about truth — specifically truths about relationship and violence — suggested a pattern of increasingly sophisticated thinking about the nature of relationships between people. This pattern more or less conformed to the progression from an egocentric, through a normative or conventional, to an autonomous or reflective or critical position; the progression that most developmental psychologists have traced. ' See Gilligan, Rogers, & Brown, "Soundings into Development," 317.  32  130 harmony, but they sometimes do reinforce each other, as Heinz's dilemma shows. We may also say that the difference between two orientations is one in emphasis. Taking care for granted, we shall see justice as the prominent feature of morality; taking justice for granted we shall see care as prominent. 3.3 Contextual sensitivity and abstract principles  What about the problems of contextual sensitivity? Again I find Kohlberg's reply quite satisfactory. Kohlberg says that "[cjontext orientation is a prerequisite for all actual moral judgments" (Kohlberg et al. 1982, p. 140). Instead of upholding absolutist moral rules like "Never lie" or "Never break a promise," he upholds highly formal principles. Moral musical chairs also guarantees that adequate moral reasoning must be contextual. In ideal role-taking, the one imagines oneself in the position of others and attempts to decide on a solution that is acceptable to all the involved parties. The particular positions of all other parties involved is in turn determined by the particular situation in which the moral dilemma arises. Therefore it is a mistake to say that Kohlberg's theory lacks contextual sensitivity. Like the problem of care and justice, I think the problem of universality and contextual particularity is a matter of emphasis. Kohlberg's aim is to look for some universal destiny of human moral development that underlies the appearance of cultural and contextual relativity. What inevitably follows is that he can capture only those elements that are common to all adequate moral perspectives, and particular considerations cannot be included in his theory. However, it can still be a commonplace of all adequate moral decisions that they must be made after all relevant particular factors in the situation have been taken into consideration. We have to notice that Kohlberg does  131 not in general accept simple absolute moral rules. Principled morality "is not dictated by 'absolutes' or by rigid or exceptionless rules" (Kohlberg et al. 1983, p. 86). In other words, moral rules that are applicable regardless of situation are extremely rare or nonexistent. Thus, it might be correct to say that contextual relativity is not emphasized in Kohlberg's theory, but it is incorrect to say that Kohlberg's moral ideal is insensitive to contextual particularity.  3.4 Communication  According to Gilligan, Kohlberg's model of moral psychology leaves out the importance of communication in solving moral problems. To a certain degree this is true. Though Kohlberg acknowledges in his (1983) anthology that communication is important for solving moral problems, he does not spend much effort explicating the role of communication in solving moral problems. When Kohlberg does talk about communication at greater length in his (1990) paper, he gives his readers an impression that he has not taken the importance of communication very seriously. In the paper, although both of the two subjects interviewed (Judge D and Joan) are ranked at Stage 6, he still cannot resist the temptation to say that Judge D's moral judgment is more adequate than Joan's. Joan's judgment is said to be less adequate because she believes that the three people on the life-boat should keep communicating until they agree on a solution. She said she had difficulty imagining that they could never come to an agreement, and thus she found the lottery solution unsatisfactory. Compared to Judge D, who chose the lottery solution, her reasoning was less adequate because she failed to acknowledge that communication could breakdown (Kohlberg et al. 1990, pp. 161-162). But surely this criterion of judging the adequacy of  132 moral reasoning is ad hoc. It has nothing to do with the formal characteristics of equilibrated moral judgments. Kohlberg's comment signals a slight reservation about the importance of communication in solving moral dilemmas. Judge D said nothing about continuing the dialogue, and Kohlberg does not view this as a deficiency. But when Joan failed to acknowledge the lottery solution, her answer is said to be a less typical exemplification of Stage 6 reasoning. Surely his undermining the importance of dialogue is not intrinsic to his theory. Rather, I think by doing so he is somewhat inconsistent with his own claim that at Stage 6 there is an "attitude of recognizing the necessity of entering into dialogue" (Kohlberg et al. 1990, p. 160). Section  4: Flanagan  and  pluralism  The other critic I will discuss is Owen Flanagan. He is a philosopher who looks deeply into the research in moral psychology when reflecting on problems of moral philosophy. Few philosophers have paid as much attention to moral psychology as he has. In his book Varieties of Moral Personality (1991) (abbreviated as Flanagan 1991 below), Flanagan argues against Kohlberg's model, rejecting his principled approach. He agrees with Gilligan that principled thinking is not the only mode of moral thought. He then pushes further, arguing that morality is so variegated and multipurposed that it can hardly be analyzed in terms of one or two simple schemes. Even with justice and care blended together, it will not be enough to exhaust what we call moral personality (Flanagan 1991, p. 233). Below is a brief review of his arguments.  133 Argument 1. Objection to the cognitive criteria In an early paper, Flanagan raises three arguments against Kohlberg's claim that the higher stages are more adequate.33 The first argument is concerned with Kohlberg's "cognitive criteria," i.e., the greater differentiation and integration of the higher stages. As Flanagan understands it, greater integration of the higher stages means merely that the moral stages "fit into a neat logical hierarchy," while greater differentiation means that at higher stages "more subtle conceptual discriminations" are made. Flanagan argues that integration and differentiation in this sense can hardly guarantee adequacy. The statement "my brother and sister are wicked" is more integrated and differentiated than the statement "my sister is wicked." But if neither my sister nor my brother is wicked, the former statement will be less adequate than the latter because it is doubly false. Thereby, Flanagan concludes that "the mere logical facts of integration and differentiation are not sufficient to establish increasing adequacy."34  Argument 2. Objection to stage-preferences The second argument concerns the empirical evidence Kohlberg aduces. Kohlberg believes that the increasing adequacy of the stages is partly supported by the fact that people prefer the highest stage they comprehend. However, Flanagan says, since people cannot comprehend stages higher than their own, the preference for the highest comprehended stage is equivalent to the preference for one's own stage. This "says nothing about adequacy, only about tenancy."  Owen Flanagan, "Virtue, Sex, and Gender: Some Philosophical Reflections on the Moral Psychology Debate," Ethics 92 (1982): 481-517. 34 Ibid., 506.  33  134  A r g u m e n t 3. O b j e c t i o n to the n o r m a t i v e criteria  T h e t h i r d argument deals w i t h K o h l b e r g ' s n o r m a t i v e criteria, w h i c h i n c l u d e , c h i e f l y , r e v e r s i b i l i t y , u n c o n d i t i o n a l i t y , and u n i v e r s a l i z a b i l i t y . These c r i t e r i a o b v i o u s l y echo the K a n t i a n t r a d i t i o n , w h i c h stipulates that m o r a l i t y is d e o n t o l o g i c a l a n d that a m o r a l p r i n c i p l e m u s t be u n i v e r s a l i z a b l e . T h e process o f i d e a l role-taking also reflects the R a w l s i a n m e t h o d o l o g y o f m o r a l d e l i b e r a t i o n , w h i c h is u s u a l l y u n d e r s t o o d as part o f the K a n t i a n tradition. A s K o h l b e r g h i m s e l f r e c o g n i z e s , h i s n o r m a t i v e c r i t e r i a are " a l l i e d to the f o r m a l i s t i c t r a d i t i o n i n ethics f r o m K a n t to R a w l s . "  3 6  B u t since this K a n t i a n - R a w l s i a n  tradition i t s e l f is p h i l o s o p h i c a l l y questionable, K o h l b e r g o w e s us an e x p l a n a t i o n w h y w e have to f o l l o w it. W h y , asks F l a n a g a n , do w e not f o l l o w other traditions l i k e those o f N o z i c k , Q u i n e , Singer, H a r e , or H a b e r m a s ?  37  U n l e s s w e c a n p r o v e that the K a n t i a n -  R a w l s i a n t r a d i t i o n is the o n l y t r a d i t i o n w o r t h f o l l o w i n g , K o h l b e r g ' s f o r m a l criteria remain highly problematic.  F l a n a g a n agrees w i t h G i l l i g a n that j u s t i c e is not the o n l y f u n d a m e n t a l constituent o f m o r a l i t y . H e also agrees that care is an important element i n h u m a n m o r a l p s y c h o l o g y . H o w e v e r , he s t i l l f i n d s G i l l i g a n ' s theory unsatisfactory i n that it attempts to reduce m o r a l deliberation into t w o and o n l y t w o orientations, i.e., j u s t i c e a n d care. F l a n a g a n b e l i e v e s that m o r a l i t y is too m u l t i f a r i o u s to be r e d u c e d to o n l y t w o orientations. H e is sympathetic to the A r i s t o t e l i a n t r a d i t i o n w h i c h v i e w s m o r a l l i f e as one i n w h i c h people endeavor to realize v a r i o u s virtues that are g o o d f o r h u m a n beings qua h u m a n b e i n g s . H e does not ask " W h y not A r i s t o t l e ? " , but the question m a y best express his attitude.  Lawrence Kohlberg, "The Claim to Moral Adequacy of Highest Stage of Moral Judgment," Journal of Philosophy 70(1973): 633.  36  135  A r g u m e n t 4. T h e H e i n z ' s m a x i m s argument  In another paper co-authored w i t h Jonathan E . A d l e r , F l a n a g a n argues that K o h l b e r g ' s theory faces (as d o a l l f o r m a l m o r a l theories) the d i f f i c u l t y o f s p e c i f y i n g f r o m w h e n c e a f o r m a l u n i v e r s a l l a w gets its substantive c o n t e n t .  38  T a k e the p r i n c i p l e o f respect  f o r persons as a n e x a m p l e : " [ O J n what b a s i s , " asks F l a n a g a n , " d o w e m a k e a j u d g m e n t o f what counts as treating someone l i k e a means as o p p o s e d to a n e n d ? "  Consider Heinz's  d i l e m m a . There are w a y s o f d e a l i n g w i t h the d i l e m m a w h i c h are o b v i o u s l y inadequate but nevertheless c o n f o r m to Stage 6 m o r a l requirements perfectly. O n e p o s s i b l e w a y is to u n i v e r s a l i z e a m a x i m about acting for the interests o f o n e ' s w i f e . O r one can u n i v e r s a l i z e a m a x i m b y p l a c i n g the right to property and the right to l i f e at the same l e v e l , say, b y c o n c e i v i n g o f a p e r s o n ' s property as m o r a l l y i n d i s t i n g u i s h a b l e f r o m h i s b o d y . A t h i r d e x a m p l e i s to u n i v e r s a l i z e a m a x i m that n o one s h o u l d d i s o b e y the l a w .  F l a n a g a n admits that there are grounds to oppose these m a x i m s . H o w e v e r , F l a n a g a n argues, w h a t e v e r reasons w e use to reject these m a x i m s , w e have to b r i n g substantive b a c k g r o u n d k n o w l e d g e and values unto the evaluations. F o r K o h l b e r g , the d i l e m m a is best s o l v e d b y r e c o g n i z i n g the right to l i f e as s o m e t h i n g m o r e v a l u a b l e than the right to property. B u t this i n v o l v e s the assignment o f different values to different things. Therefore, e v e n i n K o h l b e r g ' s o w n reasoning, the d i l e m m a is not s o l v e d m e r e l y b y abstract p r i n c i p l e s l i k e the G o l d e n R u l e and the C a t e g o r i c a l Imperative, but is s o l v e d at least i n part b y the r a n k i n g o f particular values o r virtues associated w i t h t h e m . T h e  Owen Flanagan, "Virtue, Sex, and Gender," 506-507. Owen Flanagan & Jonathan. Adler, "Impartiality and Particularity," Social Research 50, no. 3 (1983): 576-596. 39 Ibid., 581.  37  38  136  upshot is that K o h l b e r g ' s v i e w about m o r a l reasoning is incorrect. K o h l b e r g b e l i e v e s that the correct w a y o f s o l v i n g m o r a l p r o b l e m s is to a p p l y a certain set o f m o r a l p r i n c i p l e s , not to realize a " b a g o f v i r t u e s . " B u t n o w it seems that "[t]he v e r y b a g o f virtues h e c l a i m s w e ought not teach is necessary to g i v e more abstract p r i n c i p l e s their c o n t e n t . "  40  A r g u m e n t 5. T h e depressed f r i e n d argument  In s t i l l another paper co-authored w i t h K a t h r y n J a c k s o n , F l a n a g a n m a k e s h i s A r i s t o t e l i a n appeal e x p l i c i t  4 1  F o l l o w i n g A n s c o m b e , he suggests that "the A r i s t o t e l i a n  tradition was the best place to l o o k for a r i c h e r and less s h a d o w y c o n c e p t i o n o f m o r a l agency than either u t i l i t a r i a n i s m or K a n t i a n i s m had p r o v i d e d . "  4 2  In this paper, F l a n a g a n  c r i t i c i z e s K o h l b e r g o n his c l a i m s that (1) a l l m o r a l j u d g m e n t s have the f o r m a l features o f p r e s c r i p t i v i t y and u n i v e r s a l i z a b i l i t y , and (2) the central f u n c t i o n o f m o r a l j u d g m e n t s and p r i n c i p l e s is to resolve c o n f l i c t s o f c l a i m s .  T o argue against the c l a i m that m o r a l j u d g m e n t s and p r i n c i p l e s m u s t be prescriptive and u n i v e r s a l i z a b l e , F l a n a g a n raises the e x a m p l e o f h e l p i n g a depressed f r i e n d . T o determine what is the best w a y to h e l p the f r i e n d , one has to c o n s i d e r m a n y factors w h i c h are p e c u l i a r to the situation, s u c h as his o w n personalities and the that o f his friend. H e then proceeds to say that " w h e r e f r i e n d s h i p or l o v e t r u l y exists, t h i n k i n g about what one i s o b l i g a t e d to d o c a n , as B e r n a r d W i l l i a m s has put i n a related context, i n v o l v e 'one thought too m a n y . ' "  4 3  Ibid., 582-583. Owen Flanagan & Kathryn Jackson, "Justice, Care, and Gender: The Kohlberg-Gilligan Debate Revisited," Ethics 97 (1987): 622-637. 42 Ibid., 622. 43 Ibid, 632 40 41  137  A r g u m e n t 6 : T h e argument f r o m other d i m e n s i o n s i n m o r a l l i f e  F l a n a g a n goes o n to argue that it is just not o b v i o u s that m o r a l i t y has the central f u n c t i o n o f r e s o l v i n g " c o n f l i c t s o f c l a i m s o r r i g h t s . " H e admits that r e s o l v i n g c o n f l i c t s o f c l a i m s i s a n important f u n c t i o n o f m o r a l i t y , but there are other d i m e n s i o n s o f h u m a n m o r a l i t y as w e l l . F l a n a g a n names three p l a u s i b l e d i m e n s i o n s o f m o r a l i t y other than r e v o l v i n g c o n f l i c t s o f c l a i m s : (1) self-improvement and the refinement o f character; (2) respectful interactions w i t h l o v e d ones, friends, and strangers; and (3) supererogation. A l l these d i m e n s i o n s are e x c l u d e d b y K o h l b e r g f r o m his m o r a l d o m a i n w i t h o u t justification.  44  '  A r g u m e n t 7. O b j e c t i o n to the n a r r o w range o f h y p o t h e t i c a l d i l e m m a s  F u l l e r d i s c u s s i o n o f K o h l b e r g ' s m o r a l theory b y F l a n a g a n c a n be f o u n d i n h i s b o o k Varieties of Moral Personality (abbreviated b e l o w as F l a n a g a n 1991). H e r e he argues that K o h l b e r g has a n a r r o w c o n c e p t i o n o f m o r a l g o o d that projects a telos o f h u m a n m o r a l development. T h i s telos, unfortunately, is s u s p i c i o u s l y oriented towards a p h i l o s o p h i c a l l y contentious n o r m a t i v e i d e a l associated w i t h l i b e r a l i s m ( F l a n a g a n 1 9 9 1 , p. 162). F i v e general objections are h i g h l i g h t e d b y F l a n a g a n i n h i s b o o k , and these objections c a l l into q u e s t i o n some o f the basic assumptions o f K o h l b e r g ' s theory.  T h e first o b j e c t i o n i s that the range o f K o h l b e r g ' s i m a g i n a r y h y p o t h e t i c a l m o r a l p r o b l e m s is too n a r r o w , and that h i s studies were conducted i n an a r t i f i c i a l w a y . T w o things then f o l l o w . F i r s t , it is unreasonable to s i m p l y extrapolate the o b s e r v e d results o f the tests so as to d r a w c o n c l u s i o n s c o n c e r n i n g a m u c h w i d e r range o f m o r a l p r o b l e m s  138  encountered b y the subject. S e c o n d , since the studies w e r e c o n d u c t e d i n an a r t i f i c i a l w a y , the response o f the subjects m a y not e v e n reflect real p s y c h o l o g i c a l activities i n their actual l i v e s ( F l a n a g a n 1 9 9 1 , p. 186).  A r g u m e n t 8. T h e argument f r o m w e a k e m p i r i c a l support  F l a n a g a n also argues that K o h l b e r g ' s e x p e r i m e n t a l results do not adequately support his theory. F i r s t o f a l l , F l a n a g a n reports, no subject's answers c o u l d a l l be scored as b e l o n g i n g to one single stage. M o s t subjects have o n l y t w o thirds o f their responses f a l l i n g under their p r e d o m i n a n t stages. A l s o , p o s t c o n v e n t i o n a l subjects w e r e too rare, and indeed n o Stage 6 subject w a s detected b y the M J I m e t h o d u s i n g the r e v i s e d m a n u a l .  4 5  So  it is i m p l a u s i b l e that the m o r a l reasoning o f h u m a n beings c a n be fitted neatly into K o h l b e r g ' s scheme ( F l a n a g a n 1 9 9 1 , p p . 186-7).  A r g u m e n t 9. T h e o b j e c t i o n to the use o f hypothetical d i l e m m a s  F l a n a g a n ' s t h i r d o b j e c t i o n is that K o h l b e r g too r e a d i l y assumes (1) that h i s standard tests s h o u l d accurately predict h o w i n d i v i d u a l s w o u l d respond to cases o f other k i n d s , and (2) that s u c h tests are suitable f o r assessing p e o p l e ' s levels o f m o r a l maturity ( F l a n a g a n 1 9 9 1 , p. 187). H e argues, o n the contrary, that K o h l b e r g ' s h y p o t h e t i c a l d i l e m m a s are not a v e r y sensitive counterfactual instrument f o r testing p e o p l e ' s m o r a l tendencies. T h e i m a g i n a r y cases are about a t h i r d party and therefore " d o not o b v i o u s l y  As we have seen, Kohlberg claims to have found two Stage 6 subjects in his 1990 paper, but Flanagan might not have had access to the paper when he wrote the book. Besides, Kohlberg also identifies a number of public figures such as Gandhi and Martin Luther King as Stage 6 reasoners. Even so, it remains true that Stage 6 reasoners are extremely rare.  45  139  d r a w subjects into t h i n k i n g about w h a t they themselves w o u l d d o i f they w e r e faced w i t h certain counterfactual s i t u a t i o n s " ( F l a n a g a n 1 9 9 1 , p. 188).  M o r e o v e r , F l a n a g a n argues, the a s s u m p t i o n that m o r a l p e r s o n a l i t y m u s t f o r m a structural w h o l e is b y n o means o b v i o u s . It is also d u b i o u s that a  unified  rationalization  is  necessary for m o r a l maturation. H o w e v e r , this is exactly what K o h l b e r g has presupposed w h e n he argues f o r the greater adequacy o f the h i g h e r stages b y p o i n t i n g to their greater degree o f integration. A l s o , he argues that w h a t constitutes a unitary f o u n d a t i o n f o r m o r a l reasoning is c u l t u r a l l y relative. " W h a t l o o k s l i k e an incoherent h o d g e p o d g e f r o m the outside m a y , f r o m an i n s i d e r ' s p o i n t o f v i e w , have s i g n i f i c a n t l y greater c o h e r e n c e " (Flanagan 1 9 9 1 , p. 191).  A r g u m e n t 10. T h e o b j e c t i o n to the h i e r a r c h i c a l c l a i m  F l a n a g a n m a i n t a i n s that e v e n i f the K o h l b e r g i a n stages r e a l l y f o r m an invariant sequence, it does not f o l l o w that they f o r m an order w i t h i n c r e a s i n g adequacy. T h i s objection stands e v e n w h e n the so c a l l e d l o g i c a l order o f stages is taken into consideration. H e argues: • [ E ] v e n i f one t h i n k s that each stage p r o v i d e s a coherent e n o u g h t y p o l o g y o f a k i n d o f c o g n i t i v e - m o r a l style w i t h a clear e n o u g h d e v e l o p m e n t a l order to warrant the c l a i m o f stage h o l i s m , one is under n o l o g i c a l c o m p u l s i o n also to t h i n k that the proper qualitative o r d e r i n g o f these stages corresponds to the order i n w h i c h they e m p i r i c a l l y u n f o l d . Resistance to any straight f o r w a r d inference f r o m the e m p i r i c a l order to the n o r m a t i v e order is p o s s i b l e e v e n i f there is a clear sense i n w h i c h attaining the competencies r e q u i r e d at stage n-1 is necessary f o r r e a c h i n g stage n. ( F l a n a g a n 1 9 9 1 , p. 191) F l a n a g a n illustrates his point b y the e x a m p l e o f a race car driver. It is certain that b e i n g able to drive is a necessary prerequisite for b e i n g a race car d r i v e n B u t it does not f o l l o w that it is better to be a race car d r i v e r than to be an o r d i n a r y driver. S i m i l a r l y , that l o w e r  140  stages are the prerequisites f o r h i g h e r stages does not entail that h i g h e r stages are better than l o w e r ones.  A r g u m e n t 11. T h e o b j e c t i o n f r o m the heterogeneity o f m o r a l i t y  T o argue f o r the supreme a d e q u a c y o f the highest stage, K o h l b e r g has to r e l y o n some particular k i n d o f n o r m a t i v e c o n c e p t i o n . H o w e v e r , F l a n a g a n argues, the n o r m a t i v e standard K o h l b e r g uses is h i g h l y p r o b l e m a t i c , as e x p l i c a t e d i n arguments 1-4. K o h l b e r g ' s standard h i g h l i g h t s j u s t i c e and fairness at the p r i c e o f underestimating m a n y goods that are also v e r y important i n our m o r a l l i v e s . " T h e heterogeneity o f the m o r a l is a deep and significant f a c t , " says F l a n a g a n . " O n c e w e p a y attention to the m u l t i f a r i o u s content o f m o r a l issues and t h i n k o f the v a r i o u s c o g n i t i v e and affective d i s p o s i t i o n s r e q u i r e d to meet t h e m , it seems s i m p l y u n b e l i e v a b l e that there c o u l d be a s i n g l e ideal m o r a l c o m p e t e n c e and a u n i v e r s a l and irreversible sequence o f stages a c c o r d i n g to w h i c h m o r a l maturity can be u n e q u i v o c a l l y p l o t t e d " ( F l a n a g a n 1 9 9 1 , p. 195).  Section 5: Kohlberg's reply K o h l b e r g ' s o n l y direct r e p l y to F l a n a g a n ' s c r i t i c i s m appears i n h i s " A R e p l y to O w e n F l a n a g a n and S o m e C o m m e n t s o n the Puka-Goodpaster E x c h a n g e . "  4 6  T h e paper i s  a response to F l a n a g a n ' s arguments i n h i s (1982) paper, w h i c h c r i t i c i z e K o h l b e r g ' s criteria o f the superiority o f the h i g h e r stages, and the e m p i r i c a l e v i d e n c e o f the c l a i m o f h i e r a r c h y ( A r g u m e n t s 1-3 i n the p r e v i o u s section).  Lawrence Kohlberg, " A Reply to Owen Flanagan and Some Comment on the Puka-Goodpaster Exchange," Ethics 92 (1982): 513-528.  46  141 K o h l b e r g agrees w i t h F l a n a g a n that integration and d i f f e r e n t i a t i o n themselves o n l y m a k e the h i g h e r stages m o r e sophisticated, not necessarily better ( A r g u m e n t 1. O b j e c t i o n to the c o g n i t i v e criteria). H e then m a k e s it clear that w h a t he r e a l l y believes i s that s u c h c o g n i t i v e s o p h i s t i c a t i o n i s o n l y a necessary c o n d i t i o n f o r m o r a l adequacy, not a sufficient o n e .  47  A s f o r F l a n a g a n ' s c l a i m about p e o p l e ' s j u d g m e n t preferences at different  stages, K o h l b e r g points out that F l a n a g a n i s s i m p l y w r o n g . Preference f o r the highest stage one c o m p r e h e n d s does not, as F l a n a g a n says, equal a preference f o r o n e ' s o w n stage ( A r g u m e n t 2. O b j e c t i o n to stage preferences). V e r y often p e o p l e c a n c o m p r e h e n d m o r a l j u d g m e n t s one stage h i g h e r than their o w n . Therefore, F l a n a g a n ' s c l a i m that e m p i r i c a l evidence o n l y supports tenancy i s a m i s t a k e .  In response to F l a n a g a n ' s c r i t i c i s m that he presupposes too r e a d i l y the adequacy o f R a w l s ' s ethical theory ( A r g u m e n t 3. O b j e c t i o n to the n o r m a t i v e criteria), K o h l b e r g replies that w h a t he has taken up f r o m R a w l s i s o n l y the f o r m a l structure o f m o r a l reasoning, not the substance o f it. T h i s f o r m a l structure requires i m p a r t i a l i t y and r e c o m m e n d s that the subject put h i m s e l f into the place o f each p e r s o n i n v o l v e d i n the situation. K o h l b e r g believes that this f o r m a l requirement o f m o r a l reasoning has gained w i d e support. It i s not o n l y supported b y K a n t i a n s or R a w l s i a n s , but i s " s u p p o r t a b l e w i t h a n u m b e r o f n o r m a t i v e theories o f m o r a l p r i n c i p l e s , whether d e o n t o l o g i c a l o r t e l e o l o g i c a l . " In essence K o h l b e r g stresses that what he has adopted f r o m R a w l s are features o f a " m o r a l p o i n t o f v i e w , " w h i c h i n c l u d e s i m p a r t i a l i t y , u n i v e r s a l i t y , r e v e r s i b i l i t y , and " p r e s c r i p t i v i t y " (or u n c o n d i t i o n a l i t y ) . " B o t h u t i l i t a r i a n and  Ibid, 522. Ibid, 523.  142  d e o n t o l o g i c a l n o r m a t i v e theorists c o m e to s i m i l a r f o r m a l characterizations o f m o r a l judgments."  4 9  K o h l b e r g disagrees w i t h F l a n a g a n ' s c l a i m that h i s emphasis o n u n i v e r s a l p r i n c i p l e s i n s o l v i n g m o r a l p r o b l e m s i m p l i e s ignorance o f contextual relativity. M o r a l p r i n c i p l e s , argues K o h l b e r g , c a n be " c o n t e x t u a l l y r e l a t i v e " i n that they c a n be sensitive to features o f particular situations. H e f o l l o w s H a r e i n d i s t i n g u i s h i n g m o r a l rules a n d p r i n c i p l e s . A m o r a l rule i s a p r o s c r i p t i o n or p r e s c r i p t i o n o f a certain class o f actions, s u c h as " D o not k i l l , " " D o not s t e a l , " and " L o v e y o u r n e i g h b o r . "  50  A m o r a l p r i n c i p l e , o n the  other h a n d , i s a m e t h o d o f m a k i n g a c h o i c e , a w a y o f p e r c e i v i n g a n d selecting m o r a l components o f a situation, and a s p e c i f i c a t i o n o f the m o r a l p o i n t o f v i e w . E x a m p l e s are K a n t ' s C a t e g o r i c a l Imperative, the G o l d e n R u l e , and R a w l s ' s o r i g i n a l p o s i t i o n . W h e r e a s the rule " D o not s t e a l " p r o h i b i t s H e i n z f r o m stealing the d r u g regardless o f the situation (even i f his intention i n stealing the d r u g is to save a l i f e ) , the p r i n c i p l e s m e n t i o n e d above surely a l l o w (or e v e n require) h i m to take a l l the particularities o f h i s situation into consideration w h e n he m a k e s a m o r a l d e c i s i o n .  51  Section 6: Flanagan's counter-arguments A s a counter-argument to K o h l b e r g ' s c l a i m that subjects' preference f o r the highest stage they c o m p r e h e n d does not equal a preference f o r their o w n stage, F l a n a g a n says:  Ibid., 523-524. 1 think Kohlberg has made a mistake here. "Love your neighbor" by no means prescribes any action at all. It prescribes an attitude. 51 Lawrence Kohlberg, " A Reply to Owen Flanagan and Some Comment on the Puka-Goodpaster Exchange," 520.  49  50  143  K o h l b e r g objects to this b y s a y i n g that " b o t h the theory and R e s t ' s e m p i r i c a l  f i n d i n g say that p e o p l e often c o m p r e h e n d and prefer the stage one stage a b o v e  their stage o f p r e d o m i n a n t u s e . " I r o n i c a l l y , it turns out that R e s t ' s data, to w h i c h  K o h l b e r g refers, s h o w that f e w e r than 50 percent o f h i s subjects' responses c o u l d be scored i n any one p r e d o m i n a n t stage, and that subjects often m a k e responses  w h i c h c a n be scored i n as m a n y as three or f o u r different stages. R e s t ' s response  data w o u l d seem to indicate that people c o m p r e h e n d several different stages and  prefer different ones at different t i m e s . W h e r e an i n d i v i d u a l has s o m e o v e r r i d i n g preference for s o m e m o r a l c o n c e p t i o n that is not h i s p r e d o m i n a n t one it is  i n v a r i a b l y one he at least s o m e t i m e s d e p l o y s . R e s t ' s data about preferences,  therefore, are entirely consistent w i t h an account i n terms o f ego identity and establish n o t h i n g about adequacy ... ( F l a n a g a n 1 9 8 2 , p. 531).  In response to K o h l b e r g ' s c l a i m that the f o r m a l features o f the m o r a l p o i n t o f v i e w are shared b y v a r i o u s ethical theories, F l a n a g a n argues that K o h l b e r g ' s defense is still an argument f r o m authority. Instead o f r e l y i n g o n R a w l s alone, K o h l b e r g n o w relies o n the consensus a m o n g H a r e , B a i e r , F r a n k e n a , R a w l s and others. M o r e o v e r , i f K o h l b e r g is right i n s a y i n g that the f o r m a l features he presupposes receive support f r o m b o t h utilitarian and non-utilitarian m o r a l p h i l o s o p h i e s , then "the m o r a l - p h i l o s o p h i c a l meat o n the bone o f K o h l b e r g ' s stages is t h i n " ( F l a n a g a n 1982, p p . 531-532).  Section 7: Reflections on the Kohlberg-Flanagan debate It seems that F l a n a g a n ' s most important c r i t i c i s m o f K o h l b e r g is that his c l a i m o f hierarchy for the stages is unsupportable. H e argues against b o t h K o h l b e r g ' s c o g n i t i v e and n o r m a t i v e criteria. H e also argues that studies r e g a r d i n g stage preferences indicate o n l y ego identity, not the superiority o f the h i g h e r stages. W e shall discuss these arguments one b y one.  F l a n a g a n ' s c r i t i c i s m o f K o h l b e r g ' s c o g n i t i v e criteria is o n l y partly correct. H e is correct i n that the integration o f the h i g h e r stages i m p l i e s that the m o r a l stages fit into a rather neat l o g i c a l h i e r a r c h y , and greater d i f f e r e n t i a t i o n i m p l i e s m o r e subtle conceptual  144  d i s c r i m i n a t i o n s ( F l a n a g a n 1 9 8 2 , p. 506). N e i t h e r o f these factors seems to have any significant n o r m a t i v e i m p l i c a t i o n . H o w e v e r , F l a n a g a n has o b v i o u s l y o v e r s i m p l i f i e d K o h l b e r g ' s c o n c e p t i o n o f differentiation. H a v i n g the a b i l i t y to m a k e subtle conceptual d i s c r i m i n a t i o n s means h a v i n g the a b i l i t y to d i s c o v e r relevant s i m i l a r i t i e s a n d difference between different m o r a l situations, and this a b i l i t y is certainly relevant to c o n s i s t e n c y i n m o r a l j u d g m e n t s . S i n c e c o n s i s t e n c y is an important c r i t e r i o n o f adequate m o r a l j u d g m e n t , there is e v e r y reason to say that d i f f e r e n t i a t i o n is relevant to m o r a l maturity. A l s o , greater integration and d i f f e r e n t i a t i o n entails greater a b i l i t y to handle c o m p l e x p r o b l e m s . I f w e b e l i e v e that the a b i l i t y to handle situational c o m p l e x i t y is r e q u i r e d b y m o r a l maturity, w e have further reason to b e l i e v e that K o h l b e r g ' s h i g h e r stages are m o r e adequate than the l o w e r ones.  F l a n a g a n charges that K o h l b e r g ' s n o r m a t i v e criteria too r e a d i l y presuppose the correctness o f R a w l s ' s ethical theory. F i r s t he charges that K o h l b e r g ' s reliance o n u n i v e r s a l p r i n c i p l e s l a c k s sensitivity to contextual particulars. T h e n K o h l b e r g responds that his p r i n c i p l e s are s u f f i c i e n t l y abstract and m e t h o d o l g i c a l to a c c o m o d a t e contextual c o m p l e x i t i e s . F l a n a g a n says that, i n that case, the h i g h e r stages m u s t lack substance. B u t s i m p l y s a y i n g this does not m a k e it so, a n d F l a n a g a n does not c a r r y out the task o f s h o w i n g that it is so.  W e n o w turn to the o b j e c t i o n to stage-preferences, w h i c h centers o n the e m p i r i c a l data that K o h l b e r g uses to support the h i e r a r c h i c a l c l a i m . U n f o r t u n a t e l y , F l a n a g a n ' s c r i t i c i s m starts w i t h a false c l a i m , that a subject's preference f o r the highest available stage equals a preference f o r h i s o w n stage. A f t e r K o h l b e r g has r e m i n d e d h i m that the highest available stage is quite often a stage h i g h e r than o n e ' s p r e d o m i n a n t stage,  145  F l a n a g a n f o l l o w s up b y s a y i n g that, since the subjects w h o c o m p r e h e n d the +1 stage at least d e p l o y that stage o c c a s i o n a l l y , this observation is still c o m p a t i b l e w i t h his c l a i m that the preference is j u s t a k i n d o f ego identity, rather than any s i g n o f superiority.  B u t e v e n i n the a m e n d e d f o r m F l a n a g a n ' s argument is still not c o n v i n c i n g . G i v e n that people o v e r w h e l m i n g l y prefer stages h i g h e r than their o w n p r e d o m i n a n t stages (as far as they c a n c o m p r e h e n d them), the question becomes w h y p e o p l e do not s i m p l y prefer their o w n p r e d o m i n a n t stages. If, as F l a n a g a n says, the preference o f + 1 stage i s consistent w i t h ego identity because the subject does d e p l o y t h e m , then it s h o u l d be equally p o s s i b l e that they prefer -1 stage or the p r e d o m i n a n t stage itself. H o w e v e r , this is not w h a t p s y c h o l o g i s t s have observed. W h y s h o u l d this be so? O b v i o u s l y F l a n a g a n ' s v i e w cannot e x p l a i n this. Therefore K o h l b e r g ' s theory, w h i c h presupposes that the stages are h i e r a r c h i c a l , r e m a i n s a better e x p l a n a t i o n than F l a n a g a n ' s .  M o r e o v e r , the c r e d i b i l i t y o f the h i e r a r c h i c a l c l a i m does not m e r e l y depend o n p e o p l e ' s preferences f o r different types o f m o r a l j u d g m e n t . A s I have argued i n C h a p t e r 1, it also depends o n the extent to w h i c h higher stages predict moral behavior. I t h i n k this is the most important evidence f o r the h i e r a r c h y c l a i m . Just l o o k at h o w p e o p l e at different m o r a l stages b e h a v e d i n the M i l g r a m situation. O n e can h a r d l y resist the i m p r e s s i o n that p o s t c o n v e n t i o n a l m o r a l i t y must i n some sense be better than l o w e r l e v e l s , unless one believes that p e o p l e w h o quit the experiment w e r e n o better than those w h o a p p l i e d electric shocks w e l l a b o v e the dangerous l e v e l . F r o m the nervousness s h o w n b y people w h o carried o n to h i g h voltage electric shocks (Chapter 1, s e c t i o n 2.2), it seems apparent that e v e n those w h o carried o n k n o w e d , h o w e v e r d i m l y , that it w a s w r o n g f o r t h e m to do so. T h e difference between c o n v e n t i o n a l reasoners and p o s t c o n v e n t i o n a l  146  reasoners was clear cut. P o s t c o n v e n t i o n a l reasoners w e r e m o r e l i k e l y to d i s o b e y w h e n they f o u n d that w h a t they w e r e ordered to d o was c l e a r l y w r o n g . E v e n F l a n a g a n h i m s e l f has to a d m i t that the result o f this e x p e r i m e n t is important: K o h l b e r g c l a i m s that higher-stage reasoners are most l i k e l y to resist M i l g r a m - l i k e c o e r c i o n and to engage i n s a m a r i t a n i s m . T h e o v e r a l l adequacy o f K o h l b e r g ' s stage theory to one side, this is an important c l a i m i f true ( F l a n a g a n 1 9 9 1 , p p . 313-314). It seems that F l a n a g a n s t i l l has reservations, as he o n l y says the c l a i m is important  if true.  B u t I cannot f i n d any reason to doubt this c l a i m . It is a c l a i m about recorded e m p i r i c a l observations. W h e r e is the p o i n t i n d o u b t i n g s u c h a statement unless a c c o m p a n i e d w i t h a b e l i e f that those w h o have reported the results are l y i n g or are under an i l l u s i o n ? A l s o , F l a n a g a n resists assessing the relevance o f s u c h e m p i r i c a l f i n d i n g s to K o h l b e r g ' s stage theory. If he h a d done s o , I cannot i m a g i n e h o w he c o u l d a v o i d d r a w i n g the same c o n c l u s i o n I have j u s t d r a w n . K o h l b e r g ' s theory m i g h t be, as G i l l i g a n b e l i e v e s , i n c o m p l e t e or biased. B u t e v e n i f this is so, the theory must at least be correct to the extent that p o s t c o n v e n t i o n a l reasoning is i n general m o r e adequate than the l o w e r levels.  W e have n o w d i s c u s s e d F l a n a g a n ' s first three arguments. K o h l b e r g has not responded to the rest o f F l a n a g a n ' s arguments. B u t w e can still try to r e p l y to these arguments f r o m a K o h l b e r g i a n p o i n t o f v i e w . B y d o i n g so, w e can estimate h o w p o w e r f u l F l a n a g a n ' s arguments are, and have a m o r e t h o r o u g h understanding o f the merits a n d demerits o f K o h l b e r g ' s theory. B e l o w I discuss F l a n a g a n ' s A r g u m e n t s 4-11 one b y one (except f o r arguments 8 and 10, w h i c h need lengthy replies and w i l l be left to the next chapter).  A r g u m e n t 4. T h e H e i n z ' s m a x i m s argument  147  T h i s argument states that to reason adequately one does not s i m p l y a p p l y h i g h l y abstract m o r a l p r i n c i p l e s , s u c h as the p r i n c i p l e o f u n i v e r s a l i z a b i l i t y . H e i n z can e a s i l y formulate v a r i o u s m a x i m s that are certainly m o r a l l y b a d , but w h i c h can be u n i v e r s a l i z e d w i t h o u t c o n t r a d i c t i o n . T o a v o i d this, one has to return to virtues or values that are m o r e particularistic and diverse than the ultimate p r i n c i p l e . A b o u t this I t h i n k F l a n a g a n is correct. B u t this need not entail that K o h l b e r g is w r o n g i n e m p h a s i z i n g general ethical p r i n c i p l e s i n d i f f e r e n t i a t i n g d e v e l o p m e n t a l stages, a l t h o u g h he has s e e m i n g l y exaggerated their p o w e r i n d e t e r m i n i n g the adequacy o f m o r a l reasoning.  F l a n a g a n ' s argument focuses o n the i m p o r t a n c e o f the k n o w l e d g e o f substantive values i n the actual d e r i v a t i o n o f acceptable m o r a l d e c i s i o n s f r o m Stage 6 m o r a l p r i n c i p l e s . H o w e v e r , I t h i n k h i s argument w o r k s at l o w e r stages as w e l l . F o r e x a m p l e , let us c o m p a r e the f o l l o w i n g j u d g m e n t s : S I : H e i n z s h o u l d steal the d r u g because people must be w i l l i n g to save others i f society is to s u r v i v e ( C r i t e r i o n j u d g m e n t 2 4 , C o l b y et a l . 1987 v o l . 2, p. 9). S 2 : H e i n z s h o u l d steal the d r u g because the right to l i f e supersedes the right to property ( C r i t e r i o n j u d g m e n t 3 6 , C o l b y et a l . 1987 v o l . 2, p. 11). S I is a Stage 4 j u d g m e n t because it j u s t i f i e s H e i n z ' s stealing the d r u g i n terms o f the maintenance o f s o c i a l order. S 2 , o n the other h a n d , is Stage 5, since it j u s t i f i e s the b e h a v i o r i n terms o f h u m a n rights. P s y c h o l o g i c a l studies s h o w that p e o p l e w h o c o m p r e h e n d b o t h statements prefer S2 to S 1 .  B u t let us c o n s i d e r the f o l l o w i n g statements, w h i c h argue against H e i n z ' s stealing the d r u g :  148  S 3 : H e i n z s h o u l d not steal the drug because a l t h o u g h some l a w s m a y not be c o m p l e t e l y fair or are u n f a i r i n some cases, they are still set up b y society to p r o m o t e s o c i a l welfare or the c o m m o n g o o d , or to p r o v i d e benefits and protection to a l l m e m b e r s o f society ( C r i t e r i o n j u d g m e n t 2 6 , C o l b y et a l . 1987 v o l . 2, p. 62). S 4 : H e i n z s h o u l d not steal the d r u g because the right to property supersedes the right to l i f e ( m y f a b r i c a t i o n , not f o u n d i n the s c o r i n g m a n u a l ) . S3 i s , l i k e S I , a Stage 4 j u d g m e n t because it is based o h the s o c i a l order. S 4 is a j u d g m e n t w i t h e x a c t l y the same structure as S 2 , except that it ranks the right to property o v e r the right to l i f e . S i n c e the m o r a l stages are d e f i n e d i n terms o f f o r m instead o f content, S 4 s h o u l d be a Stage 5 j u d g m e n t as w e l l . I w o n d e r whether p e o p l e w h o c o m p r e h e n d b o t h S3 and S4 w o u l d s t i l l prefer S4.  It is easy to see w h y S4 cannot be a Stage 5 j u d g m e n t , f o r Stage 5 shares w i t h Stage 6 most o f their substantive values. G i v e n that respect f o r persons is a central value at Stage 6, it is p r a c t i c a l l y i m p o s s i b l e f o r the Stage 5 reasoner to b e l i e v e that the right to property s h o u l d o v e r r i d e the right to l i f e . B u t then it means that, w h e n K o h l b e r g says the concept o f h u m a n rights is one o f the core c o n c e p t s o f Stage 5 m o r a l i t y , he already has i n m i n d s o m e t h i n g substantive about w h a t h u m a n rights are, and h o w they s h o u l d be ranked. B u t this still means that an adequate understanding o f m o r a l i t y does not m e r e l y i n c l u d e the understanding o f the f o r m o f m o r a l r e a s o n i n g , but o f various substantive values as w e l l . W i t h o u t s o m e agreement o n basic substantive values ( s u c h as the v a l u e o f l i f e over property i n general), the h i g h e r adequacy o f the h i g h e r stages cannot be established.  A r g u m e n t 5. T h e depressed f r i e n d argument  149  In this argument, F l a n a g a n argues that i n some m o r a l b e h a v i o r , l i k e h e l p i n g a depressed f r i e n d , the Stage 6 f o r m a l characteristics not o n l y do not seem to p l a y any significant role but i f appealed to c a n , i n fact, seem to u n d e r m i n e appropriate m o r a l d i s p o s i t i o n s . W h e n f r i e n d s h i p or l o v e t r u l y exists, to t h i n k about o b l i g a t i o n can sometimes be to t h i n k too m u c h . T h i s is correct. Suppose m y w i f e is i n danger and needs m y help. It is not v e r y n i c e for m e to hesitate and consider w h e r e m y o b l i g a t i o n lies b y c o n s i d e r i n g whether h e l p i n g her w i l l do m o r e g o o d i n this c i r c u m s t a n c e than h e l p i n g others w h o are nearby a n d i n s i m i l a r d i f f i c u l t i e s . B u t i f a l l f o u r p e o p l e i n need are strangers such considerations w i l l be fine.  T h e a b o v e e x a m p l e s h o w s that p r i n c i p l e s or fixed m o r a l c o n c e p t i o n s d o not h e l p m u c h i n m o r a l deliberations i f they are insensitive to context. V a l u e s s u c h as fairness and i m p a r t i a l i t y must be f l e x i b l e e n o u g h to handle v a r i o u s personal relationships and the special o b l i g a t i o n s i n v o l v e d , or otherwise they c o u l d not p o s s i b l y represent an adequate m o r a l standard. I f K o h l b e r g ' s theory were a p r i n c i p l i s m i n the sense that adequate m o r a l reasoning is merely  a matter o f a p p l y i n g the set o f v a l i d m o r a l p r i n c i p l e s into a m o r a l  situation, it w o u l d face the k i n d o f d i f f i c u l t y that a l l p r i n c i p l e d ethics share: the d i f f i c u l t y i n e x p l a i n i n g h o w h i g h l y abstract p r i n c i p l e s c a n derive a l l the m o r a l solutions w e need i n different situations.  B u t it seems that K o h l b e r g ' s theory is not an ethical p r i n c i p l i s m i n this sense. Stage 6 p r i n c i p l e s i n c l u d e the p r i n c i p l e o f A g a p e , w h i c h is a c t u a l l y m o r e an attitude than a p r i n c i p l e . T h e r e is also the p r i n c i p l e o f role-taking, w h i c h is a p r o c e d u r a l m e t h o d o f d e c i s i o n m a k i n g a n d does not s i m p l y u p h o l d a single m o r a l v a l u e that determines a l l m o r a l actions. R o l e - t a k i n g is important i n the r e s o l u t i o n o f a l l m o r a l p r o b l e m s i n that it  150  a l l o w s us to capture, to the finest details, the particularities o f the m o r a l situation. T o decide h o w to h e l p a depressed f r i e n d , it surely is useful to put o n e s e l f into the f r i e n d ' s situation and i m a g i n e w h a t one w o u l d need most i n the f r i e n d ' s situation. O n the other h a n d , A g a p e also a l l o w s f o r the r e c o g n i t i o n that, i n some situations, l o v e , i n order to be l o v e , e x c l u d e s c a l c u l a t i o n s i n terms o f fairness i n the f o r m o f strict i m p a r t i a l i t y .  A r g u m e n t 6: T h e argument f r o m other d i m e n s i o n s i n m o r a l l i f e  A r g u m e n t 6 states that r e s o l v i n g c o n f l i c t s is not the o n l y d i m e n s i o n o f m o r a l i t y . T h e r e are other d i m e n s i o n s s u c h as (1) self-improvement a n d the r e f i n e m e n t o f character; (2) respectful interactions w i t h l o v e d ones, friends, and strangers; and (3) supererogation. (2) and (3) o b v i o u s l y p a r a l l e l G i l l i g a n ' s ethic o f care. I f m y arguments i n section 3 are s o u n d , then K o h l b e r g ' s e x p a n s i o n o f the m o r a l d o m a i n w i l l be e n o u g h to c o v e r these d i m e n s i o n s . E v e n so, (1) r e m a i n s a b l a n k i n K o h l b e r g ' s theory. T h e f o l l o w i n g quotation f r o m R o b e r t Carter can h e l p us see m o r e c l e a r l y the insight i n F l a n a g a n ' s argument: K i e r k e g a a r d m a y serve as one e x a m p l e o f someone w h o a s s u m e d that m o r a l r e a s o n i n g (the E t h i c a l stage) i s i n t r i n s i c a l l y i n f e r i o r to the R e l i g i o u s stage o f F a i t h , at w h i c h , and o n l y at w h i c h , matter c a n b e r e s o l v e d p r o p e r l y . O r , i f C h r i s t i a n a g a p e i s i n v o k e d , a l l the r e a s o n i n g i n the w o r l d , w i t h o u t l o v e , i s next to n o t h i n g , and  agape, e v e n w i t h modest systematic r e a s o n i n g , y i e l d s a l l . O r  consider the Z e n master w h o b e c o m e s m o r a l b y m e d i t a t i o n w h e r e b y he discerns the state o f inner h a r m o n y , integration and peace, a n d endeavors to express this d i l i g e n t l y i n the countless situations o f life. W e m a y  retrospectively c l a i m that these people are reasoning m o r a l l y , and e v e n that they display j u s t i c e r e a s o n i n g . H o w e v e r , they have either transcended s u c h reasoning ( K i e r k e g a a r d ) , o r have b y passed i t b y l e a r n i n g to l o v e , o r to b e h a r m o n i o u s l y receptive i n i d e n t i f y i n g w i t h another p e r s o n i n his o r her unique situation. It m a y be a j u s t act, but a n i n t e r v i e w need not r e v e a l that the state o f m o r a l i t y was a r r i v e d at t h r o u g h m o r a l r e a s o n i n g at a l l .  5 2  Robert Carter, "Does Kohlberg Avoid Relativism?" In Lawrence Kohlberg, Consensus and Controversy, 15-16.  52  151  T h o u g h self-perfection remains a b l a n k i n K o h l b e r g ' s theory, the i d e a is not strictly i n c o m p a t i b l e w i t h K o h l b e r g ' s theory. K o h l b e r g has admitted i n the late m o d e l that j u s t i c e does not exhaust the m o r a l d o m a i n . N e v e r t h e l e s s , I t h i n k , j u s t i c e c a n still be regarded as the central core o f m o r a l j u d g m e n t s . S e l f - i m p r o v e m e n t and refinement o f character presuppose that w h a t I aspire to becomes acceptable f r o m an i m p a r t i a l p o i n t o f v i e w . O t h e r w i s e it is not refinement at a l l . In this w a y , j u s t i c e is central to m o r a l i t y i n that it set l i m i t s to w h a t is m o r a l l y p e r m i s s i b l e and a d m i r a b l e r e g a r d i n g ideals o f s e l f i m p r o v e m e n t or self-realization.  A r g u m e n t 7. O b j e c t i o n to the n a r r o w range o f h y p o t h e t i c a l d i l e m m a s  A r g u m e n t 7 addresses the v a l i d i t y o f extrapolating f r o m i n t e r v i e w s c o n c e r n i n g a n a r r o w range o f i m a g i n a r y situations to the m u l t i f a r i o u s reality o f e v e r y d a y m o r a l l i f e . T h e argument has a certain p o w e r . S u r e l y a theory gains m o r e support i f it is tested i n different and various settings. S i n c e the range o f hypothetical d i l e m m a s used b y K o h l b e r g is not e s p e c i a l l y w i d e , one c a n t h r o w doubts o n the p o w e r o f the e m p i r i c a l support f o r the theory. N e v e r t h e l e s s , there are t w o points w e c a n m a k e o n b e h a l f o f K o h l b e r g . F i r s t , a l l s c i e n t i f i c theories i n v o l v e extrapolation. P h y s i c s tells us that the v o l u m e o f a gas or a m i x t u r e o f gases is p r o p o r t i o n a l to its absolute temperature w h e n the pressure is kept constant. It is i m p o s s i b l e for p h y s i c i s t s to test the theory f o r every c o m b i n a t i o n o f different gases, i n a l l p o s s i b l e proportions o f t h e m , at a l l p o s s i b l e temperatures and under a l l p o s s i b l e pressures. A n d i n general, the range o f settings i n w h i c h experiments can be carried out is d o o m e d to be n a r r o w w h e n c o m p a r e d to the m u l t i p l i c i t y o f c o n d i t i o n s the theory is supposed to cover. H o w e v e r , w e still t h i n k w e have g o o d reason to b e l i e v e that m a n y theories, l i k e that c o n c e r n i n g the b e h a v i o r o f  152  gases, are true. S e c o n d l y , the p o w e r o f the o b j e c t i o n f r o m the n a r r o w range o f h y p o t h e t i c a l d i l e m m a s is l i m i t e d i n a n y case. T h e reason i s t w o f o l d . O n the one h a n d , there is n o standard m e t h o d to j u d g e whether the range o f settings i s n a r r o w . A s already indicated, a l l e x p e r i m e n t a l p r o o f s o f s c i e n t i f i c theories i n v o l v e e x t r a p o l a t i n g the observations i n a r e l a t i v e l y n a r r o w range o f settings to a m u c h (and u s u a l l y i n f i n i t e l y ) larger range o f c o n d i t i o n s w h i c h the theory i s intended to cover. O n the other h a n d , e v e n i f w e a l l agree that a range o f settings is too n a r r o w , the f i n d i n g s i n these settings d o not thereby cease to support the theory. A l l w e c a n c o n c l u d e i s that the e m p i r i c a l evidence i s not as strong as w e w o u l d l i k e .  B e y o n d a l l these, there i s p o s i t i v e support f o r the general a p p l i c a b i l i t y o f K o h l b e r g ' s theory. R e c e n t research indicates that p e o p l e ' s m o r a l stages, estimated i n relation to real l i f e d i l e m m a s , matches quite c l o s e l y their m o r a l stages estimated b y hypothetical d i l e m m a s .  53  T h i s suggests that K o h l b e r g ' s theory i s a p p l i c a b l e to p e o p l e ' s  everyday m o r a l l i f e .  A r g u m e n t 9. T h e o b j e c t i o n to the use o f h y p o t h e t i c a l d i l e m m a s  A r g u m e n t 9 throws doubts o n the l e g i t i m a c y o f the p r e s u p p o s i t i o n that the w a y people reason about h y p o t h e t i c a l d i l e m m a s reflects h o w they reason about their p e r s o n a l m o r a l d i l e m m a s i n real l i f e . I n response to A r g u m e n t 7,1 have p o i n t e d out that p e o p l e ' s responses t o real l i f e m o r a l d i l e m m a s c l o s e l y m a t c h their responses to h y p o t h e t i c a l d i l e m m a s . S o , again, this suggests that the use o f h y p o t h e t i c a l d i l e m m a s i n m o r a l  Lawrence J. Walker & John H. Taylor, "Stage Transitions in Moral Reasoning: A Longitudinal Study of Developmental Process," Developmental Psychology 27, no. 2 (1991): 230-337.  53  153  j u d g m e n t interviews is legitimate and is quite reliable i n e l i c i t i n g the subject's pattern o f m o r a l reasoning i n general.  F l a n a g a n argues that the a s s u m p t i o n o f structural w h o l e n e s s f o r m o r a l r e a s o n i n g is by no means o b v i o u s . It is true, o f course, to the extent that the a s s u m p t i o n is not o b v i o u s i n itself. T h e real question here i s , h o w e v e r , not whether the a s s u m p t i o n is o b v i o u s i n itself, but whether it has gained substantial support f r o m e m p i r i c a l research. A s w e have seen i n C h a p t e r 2, w i t h the m o d i f i c a t i o n s made i n the late m o d e l , the a s s u m p t i o n o f structured w h o l e n e s s is w e l l c o n f i r m e d b y e m p i r i c a l data. F l a n a g a n is not c o n v i n c e d that the recent data support the a s s u m p t i o n because most subjects do not present m o r a l j u d g m e n t s at a single m o r a l stage. B u t his requirement is unreasonable, f o r w e s h o u l d at least a l l o w p e o p l e to be between stages as they are p r o g r e s s i n g f r o m one stage to another. T h u s , f o r the a s s u m p t i o n o f structural w h o l e n e s s to be true, the reasonable expectation is that p e o p l e s ' response s h o u l d be w i t h i n one stage or w i t h i n t w o adjacent stages and this is c o n f i r m e d b y e m p i r i c a l data (Chapter 2, section 2).  F l a n a g a n ' s f i n a l p o i n t is that u n i t y o f r a t i o n a l i z a t i o n is c u l t u r a l l y relative. U n f o r t u n a t e l y , he is b e g g i n g the q u e s t i o n w h e n he says this, f o r the q u e s t i o n K o h l b e r g wants to answer is p r e c i s e l y whether there are any u n i v e r s a l t y p o l o g i e s each o f w h i c h c a n be regarded as a u n i f i e d f o u n d a t i o n o f m o r a l r a t i o n a l i z a t i o n . K o h l b e r g ' s answer is that there are s i x s u c h t y p o l o g i e s , and he s h o w s us h i s e m p i r i c a l evidence. F l a n a g a n s i m p l y does not respond to K o h l b e r g ' s evidence at a l l , and insists instead that since u n i t y is relative K o h l b e r g m u s t be w r o n g . B u t this is mere assertion.  A r g u m e n t 11. T h e o b j e c t i o n f r o m the heterogeneity o f m o r a l i t y  154  It m a y be m i s l e a d i n g f o r m e to c a l l this a n argument. T h e o b j e c t i o n starts w i t h the c l a i m that the heterogeneity o f m o r a l i t y is a deep and s i g n i f i c a n t fact, a n d then j u m p s directly to the c o n c l u s i o n that the stage m o d e l can h a r d l y be correct. A g a i n , this is m o r e assertion than argument. A n d , i n general, whether things c a n be g r o u p e d into a finite n u m b e r o f sets depends m o s t l y o n h o w the sets are d e s i g n e d , not h o w heterogeneous these things are. L i v i n g creatures are e x t r e m e l y heterogeneous, but they c a n be quite neatly g r o u p e d under t w o categories: plants and a n i m a l s . T h e r e is n o reason i n p r i n c i p l e w h y basic m o r a l c o n c e p t i o n s cannot be grouped into a finite n u m b e r o f types.  Section 8: Summary In this chapter w e r e v i e w e d G i l l i g a n ' s and F l a n a g a n ' s c r i t i c i s m s o f K o h l b e r g ' s theory. G i l l i g a n argues that (1) there is a gender bias i n K o h l b e r g ' s theory; (2) K o h l b e r g ' s account o f the f o r m s o f m o r a l reasoning is i n c o m p l e t e ; (3) K o h l b e r g ' s m o r a l theory ignores the i m p o r t a n c e o f c o n t e x t u a l considerations i n adequate m o r a l r e a s o n i n g ; and (4) he ignores the i m p o r t a n c e o f c o m m u n i c a t i o n i n m o r a l d e c i s i o n . F l a n a g a n , o n the other h a n d , challenges K o h l b e r g r e g a r d i n g h i s research m e t h o d , h i s p r e s u p p o s i t i o n s , and various theoretical c l a i m s , s u c h as the c l a i m o f hierarchy o f the m o r a l stages, the d e f i n i t i o n o f m o r a l i t y as a d e v i c e f o r r e s o l v i n g c o n f l i c t s , and the c l a i m that m o r a l adequacy is determined b y the f o r m o f m o r a l reasoning. R e f l e c t i n g o n the debates between these t w o important critics and K o h l b e r g , I argue that most o f their arguments are not s o u n d , but w i t h the f o l l o w i n g important e x c e p t i o n s :  1. G i l l i g a n is correct i n c l a i m i n g that j u s t i c e does not exhaust the m o r a l d o m a i n . B y a d d i n g the orientation o f care to the picture o f m o r a l i t y w e c a n understand the nature o f  155  m o r a l i t y better. F l a n a g a n pushes a step further and c l a i m s that j u s t i c e and care are still not enough. E u d a i m o n i a c m o r a l ideals, for instance, are left out.  2. A s G i l l i g a n argues, K o h l b e r g has not done j u s t i c e to the i m p o r t a n c e o f c o m m u n i c a t i o n i n the process o f m o r a l d e c i s i o n . T h o u g h K o h l b e r g later v e r b a l l y a c k n o w l e d g e s that c o m m u n i c a t i o n is w h a t Stage 6 reasoning requires, he nevertheless partly u n d e r m i n e s this c o n c e s s i o n b y s a y i n g so little about c o m m u n i c a t i o n i n his d i s c u s s i o n o f particular m o r a l dilemmas.  3. F l a n a g a n is right that K o h l b e r g overestimates the p o w e r o f m o r a l f o r m to determine right actions i n particular situations. Substantive values and c o n t e x t u a l considerations are a l w a y s important i n adequate m o r a l reasoning.  156  Chapter Four: From Moral Psychology to Moral Philosophy W e have n o w r e v i e w e d m a n y o f the most important c r i t i c i s m s o f K o h l b e r g ' s theory. I argued that m o s t o f the c l a i m s against the theory d o not s e r i o u s l y threaten it, at least not s e ri o u s l y e n o u g h to m a k e us a b a n d o n the theory as a w h o l e . T h e next q u e s t i o n is whether K o h l b e r g ' s theory is relevant to m o r a l p h i l o s o p h y at a l l . K o h l b e r g certainly t h i n k s it is a n d I b e l i e v e he is correct. I n this chapter w e shall see h o w a n d w h y . M o r e p a r t i c u l a r l y , w e shall see h o w one c a n j u s t i f y the c l a i m o f i n c r e a s i n g adequacy o f the K o h l b e r g i a n stages, r e l y i n g p a r t i a l l y o n the e m p i r i c a l f i n d i n g s . T h e n w e shall discuss the c l a i m f o r u n i v e r s a l i t y o f Stage 6.  Section  1: The relevance  of psychology  to moral  philosophy  1  1.1 The "is-ought" gap S o m e p h i l o s o p h e r s t h i n k that e m p i r i c a l research i n p s y c h o l o g y has little relevance for m o r a l p h i l o s o p h y . T h i s v i e w has its basis i n H u m e ' s d e m a r c a t i o n betw een " i s " and " o u g h t , " and M o o r e ' s idea o f the naturalistic f a l l a c y . H u m e asserts that no c o m b i n a t i o n o f " i s " statements c a n l o g i c a l l y i m p l y a n " o u g h t " statement. T h e same basic idea i s further elaborated b y M o o r e i n h i s d i s c u s s i o n o f the " n a t u r a l i s t i c f a l l a c y , " w h e r e i n M o o r e c l a i m s that it is a f a l l a c y to attempt to define the w o r d " g o o d , " o r any other n o r m a t i v e e x p r e s s i o n , i n p u r e l y descriptive terms (Chapter 1, section 3.1). O n the basis o f this 2  l o g i c a l d i v i s i o n between d e s c r i p t i o n a n d e v a l u a t i o n , S i e g e l c l a i m s that K o h l b e r g ' s p r o o f o f the greater adequacy o f the h i g h e r stages i s i n v a l i d :  By psychology here I mean the empirical study of human behavior and human thought. In this sense, Kohlberg's theory is not purely psychological, but is a mixture of psychological theory and normative theory. " G. E. Moore, PrincipiaEthica, 12-17. 1  157  I f h i g h e r stages are i n d e e d m o r e m o r a l l y adequate than l o w e r stages, then they w o u l d be so e v e n i f the facts o f m o r a l d e v e l o p m e n t w e r e different than K o h l b e r g c l a i m s they are. Suppose p e o p l e d e v e l o p e d d i f f e r e n t l y than K o h l b e r g ' s research indicates — suppose, that i s , they d e v e l o p e d t h r o u g h a r a n d o m sequence o f stages, or that they d e v e l o p e d sequentially f r o m Stage 6 t h r o u g h Stage 1, or that d e v e l o p m e n t w a s not u n i v e r s a l so that people h a d different sequences o f d e v e l o p m e n t f r o m one another — i f the c l a i m to m o r a l adequacy c o u l d be j u s t i f i e d , w e w o u l d s t i l l have to say that Stage 6 w a s m o r a l l y superior to Stage 5, Stage 5 t o Stage 4, a n d so o n .  3  B u t e v e n i f there i s a l o g i c a l gap between facts and values a n d , therefore, e v e n i f this point o f S e i g e l ' s i s correct, does it f o l l o w that K o h l b e r g ' s d e v e l o p m e n t a l theory i s irrelevant  to m o r a l p h i l o s o p h y ? N o . T h e gap betw een values and facts i s a l o g i c a l one.  N o set o f factual statements l o g i c a l l y entails any value statement, but i t does not f o l l o w that no set o f factual statements supports or j u s t i f i e s any value statement. T h e fact that a l l birds w e have observed are w a r m - b l o o d e d does not entail that a l l b i r d s are w a r m - b l o o d e d either. N e v e r t h e l e s s , the fact that a l l birds observed before are w a r m b l o o d e d still i n some w a y supports the general statement about the warm-bloodedness o f a l l birds i n general. T o e x p l a i n h o w the f o r m e r supports the latter i s the m a i n task o f p h i l o s o p h y o f science, w h i c h I w i l l not d i s c u s s i n detail here. W h a t I w a n t to p o i n t out is o n l y that, entailment and support are t w o different matters. E n t a i l m e n t i s sufficient, but not necessary, for a support.  1.2 The second step: moral psychology and moral philosophy Is m o r a l p s y c h o l o g y relevant to m o r a l p h i l o s o p h y ? It w o u l d certainly seem so. A s M a r k J o h n s o n argues, w h e n w e study m o r a l p h i l o s o p h y , w e are s t u d y i n g human m o r a l i t y . S u c h a m o r a l i t y m u s t be " a m o r a l i t y directed to h u m a n c o n c e r n s , r e a l i z a b l e b y  Harvey Seigel, "On Using Psychology to Justify Judgments of Moral Adequacy." In Lawrence Kohlberg: Consensus and Controversy, ed. Sohan Modgil and Celia Modgil (Philadelphia and London: The Felmer Press, 1986), 71.  3  158  h u m a n creatures l i k e ourselves, and a p p l i c a b l e to the k i n d s o f p r o b l e m a t i c situations w e encounter i n o u r l i v e s . " W e s i m p l y cannot k n o w w h a t is the best t h i n g to do i f w e k n o w n o t h i n g about the h u m a n m i n d , i n c l u d i n g h u m a n e m o t i o n s . F o r h i m , the relevant questions o f m o r a l p s y c h o l o g y f o c u s o n personal identity, h u m a n ends a n d m o t i v a t i o n , m o r a l d e v e l o p m e n t , c o n c e p t u a l i z a t i o n , r e a s o n i n g , and affect.  4  W h a t interests m e m o s t is J o h n s o n ' s emphatic r e m i n d e r that w h a t w e are d i s c u s s i n g i n ethics is h u m a n m o r a l i t y . S i n c e an adequate m o r a l standard m u s t be a reasonable one, it f o l l o w s that w h a t is adequate must be s o m e t h i n g reasonable for  human  beings. B u t w h a t is reasonable f o r h u m a n beings depends o n w h a t h u m a n reason or rationality is l i k e . A n d this last q u e s t i o n about the features o f h u m a n r a t i o n a l i t y is at 5  least partly a p s y c h o l o g i c a l q u e s t i o n , a n d thus partly an e m p i r i c a l one.  F l a n a g a n argues f o r the relevance o f p s y c h o l o g i c a l studies to m o r a l p h i l o s o p h y f o r a s i m i l a r reason, a l t h o u g h he stands strongly against K o h l b e r g ' s theory. H i s reason is s i m p l e : f o r a m o r a l i d e a l to be v a l i d , it s h o u l d at least be p o s s i b l e that h u m a n beings can achieve s u c h an i d e a l ( h o w e v e r d i f f i c u l t it i s ) ; but d i s c o v e r i n g w h a t is p o s s i b l e f o r h u m a n beings to achieve is part o f the j o b o f p s y c h o l o g y ( F l a n a g a n 1 9 9 1 , p. 26). F l a n a g a n i s t a l k i n g about a p l a i n truth: " o u g h t " i m p l i e s " c a n . " I f a person is utterly unable to do X , then it does not m a k e sense to say he/she ought to do X . I f a s c h i z o p h r e n i c p e r s o n cannot  Mark Johnson, "How Moral Psychology Changes Moral Theory." In Mind and Morals, ed. Larry May and others (Cambridge, MA & London, England: MIT Press, 1995), 49-50. 5 The term 'reasonable' is ambiguous here. On one hand, it can mean that it is reasonable for the individual to do the morally right thing. On the other hand, it can mean that it is reasonable for the human species to have a certain moral practice. My own view is that these two meanings are connected to each other: what is reasonable for the human species to value is reasonable for me to value. But this is not the question I am discussing here. The most important thing here is only that, whatever 'reasonable' means, a reasonable moral standard must be reasonable for human beings, and therefore must be responsive to the psychological make-up of human beings. 4  159  help s c o l d i n g p e o p l e around him/her, it does not m a k e sense to say he/she s h o u l d not d o so. L i k e w i s e , i f it i s i m p o s s i b l e f o r people to be p e r f e c t l y i m p a r t i a l , then it does not m a k e sense to say they s h o u l d be so. W h e t h e r it is p o s s i b l e f o r s c h i z o p h r e n i c persons o f v a r i o u s types to c o n t r o l their b e h a v i o r , or f o r persons i n general to be p e r f e c t l y i m p a r t i a l , can certainly be v e r i f i e d b y p s y c h o l o g i c a l studies. A n adequate understanding o f h u m a n m o r a l i t y requires an adequate understanding o f h o w the h u m a n m o r a l p s y c h e actually w o r k s . I f I p r o p o s e s o m e m o r a l i d e a l that the h u m a n c o n s c i e n c e c a n never approve, then s u c h an i d e a l is o b v i o u s l y unsuitable f o r h u m a n beings.  Therefore, m o r a l p s y c h o l o g y is relevant to m o r a l p h i l o s o p h y i n at least t w o w a y s . First, w h a t i s m o r a l l y reasonable is reasonable for h u m a n beings. S e c o n d , w h a t is regarded as m o r a l l y o b l i g a t o r y must be p o s s i b l e for h u m a n beings to achieve. C o n s e q u e n t l y , K o h l b e r g i s certainly correct i n b e l i e v i n g that e m p i r i c a l studies i n m o r a l p s y c h o l o g y c a n offer s o m e t h i n g o f general relevance to m o r a l p h i l o s o p h y . H e also has h i s o w n m o r e particular v i e w s about h o w m o r a l p s y c h o l o g y is relevant, to m o r a l p h i l o s o p h y . W e s h a l l n o w e x a m i n e these.  1.3 A third step: moral development and moral philosophy W h a t i s K o h l b e r g ' s m o r e particular answer to the question about the relevance o f m o r a l p s y c h o l o g y to m o r a l p h i l o s o p h y ? W e are not c o n c e r n e d w i t h " b r i d g i n g " the " i s o u g h t " gap (in the l o g i c a l sense) but w i t h d i s c o v e r i n g certain p o s i t i v e e v i d e n t i a l relations between K o h l b e r g ' s e m p i r i c a l f i n d i n g s and his c l a i m s c o n c e r n i n g the superiority o f certain d e v e l o p m e n t a l stages. K o h l b e r g e x p l a i n s as f o l l o w s : T o b e g i n w i t h , there are t w o f o r m s o f the naturalistic f a l l a c y I a m not c o m m i t t i n g . T h e first is that o f d e r i v i n g m o r a l j u d g m e n t s f r o m p s y c h o l o g i c a l , cognitivep r e d i c t i v e j u d g m e n t s or pleasure-pain statements, as is done b y naturalistic  .160  n o t i o n s o f m o r a l j u d g m e n t . M y analysis o f m o r a l j u d g m e n t does not assume that  prescriptive and sui generis. T h e second naturalistic f a l l a c y I a m not c o m m i t t i n g i s that o f a s s u m i n g that m o r a l i t y or m o r a l m a t u r i t y is part o f b i o l o g i c a l h u m a n nature o r that the b i o l o g i c a l l y older i s the better. T h e t h i r d f o r m o f the naturalistic f a l l a c y , w h i c h I am c o m m i t t i n g , is that o f asserting that any c o n c e p t i o n o f w h a t m o r a l j u d g m e n t ought to be must rest o n an adequate c o n c e p t i o n o f w h a t it is'. T h e fact that m y c o n c e p t i o n o f the m o r a l " w o r k s " e m p i r i c a l l y i s important f o r its p h i l o s o p h i c adequacy. B y this I m e a n first that any c o n c e p t i o n o f w h a t adequate or ideal m o r a l j u d g m e n t should be rests o n an adequate d e f i n i t i o n o f w h a t m o r a l j u d g m e n t is i n the m i n d s o f people ( K o h l b e r g 1 9 8 1 , p p . 177-178). m o r a l j u d g m e n t s are r e a l l y s o m e t h i n g else, but insists that they are  A s D w i g h t B o y d p o i n t s out, it i s clear that K o h l b e r g ' s v i e w is not a r e d u c t i o n i s m o f m o r a l v a l u e s . T h i s i s evident f r o m the above quotation, i n w h i c h K o h l b e r g says that 6  m o r a l j u d g m e n t is not reductive and is sui generis.  T h e same p o i n t i s also seen w h e n  K o h l b e r g c l a i m s that "the concept o f m o r a l i t y is i t s e l f a p h i l o s o p h i c a l (ethical) rather than a b e h a v i o r a l c o n c e p t " ( K o h l b e r g 1 9 8 1 , p. 102). I t h i n k B o y d i s right. K o h l b e r g i s not u n f a m i l i a r w i t h the p h i l o s o p h i c a l p r o b l e m about the " i s - o u g h t " gap. Indeed, as w e have seen i n C h a p t e r 1, K o h l b e r g used the naturalistic f a l l a c y to reject the arguments f o r ethical r e l a t i v i s m raised b y some s o c i a l scientists. O n the other h a n d , B o y d also argues that K o h l b e r g ' s theory is not " a naive m o v e f r o m ' i s ' to ' o u g h t . ' " B y this he means that K o h l b e r g i s not t r y i n g to derive the c o n c l u s i o n that higher stages are superior f r o m the mere fact that p e o p l e do m o v e u p w a r d i n the stage sequence. A s w e can see i n the 7  quotation above, K o h l b e r g ' s theory does not assume that "the b i o l o g i c a l l y o l d e r is the better," so w h y w o u l d he t h i n k the b i o l o g i c a l l y later is better f o r the reason that it is later?  W h i l e the first k i n d o f " n a t u r a l i s t i c f a l l a c y " K o h l b e r g talks about i s doubtless a f a l l a c i o u s f o r m o f argument, w h y the second f o r m o f " n a t u r a l i s t i c f a l l a c y " i s i n d e e d  Dwight R. Boyd, "The Ought of Is: Kohlberg at the Interface between Moral Philosophy and Developmental Psychology." In Lawrence Kohlberg: Consensus and Controversy, 45 &51.  6  161  f a l l a c i o u s requires some e x p l a n a t i o n . I n Principia  Ethica, M o o r e e x p l i c i t l y a l l o w s that  there m a y be some u n i v e r s a l relations between certain facts and certain values. I n other w o r d s , he admits it could be, that a l l g o o d things, and o n l y g o o d things, are i n fact X , where X i s some natural property. S i m i l a r l y , it m a y be that a l l m o r a l stages that emerge later are better than their predecessor stages. B u t e v e n i f this is the case, "better s t a g e " does not mean " l a t e r stage," any m o r e than their b e i n g " g o o d " means b e i n g " X , " and therefore b e i n g a later stage p r o v i d e s i n i t s e l f n o reason to b e l i e v e that s u c h a stage is a better one. S o to attempt to derive value s i m p l y f r o m the c h r o n o l o g i c a l order o f the emergence o f the stages i s to c o m m i t the naturalistic f a l l a c y .  It i s a little strange for K o h l b e r g to say that there is a t h i r d f o r m o f naturalistic f a l l a c y that he is c o m m i t t i n g . T o c o m m i t a f a l l a c y , there must be some fault i n o n e ' s argument or r e a s o n i n g , but this is surely not what K o h l b e r g means. In this sense, K o h l b e r g surely does not b e l i e v e the " n a t u r a l f a l l a c y " he c o m m i t s is a f a l l a c y at a l l . W h a t he means b y c o m m i t t i n g a f o r m o f natural f a l l a c y is s i m p l y that he is t r y i n g to use studies o f p s y c h o l o g i c a l d e v e l o p m e n t to support a c l a i m o f superiority for a certain general f o r m o f m o r a l reasoning. A p p a r e n t l y , K o h l b e r g i s reacting to what he takes to be a general d i s p o s i t i o n i n p h i l o s o p h y to s i m p l y reject or d i s m i s s p s y c h o l o g i c a l studies as h a v i n g no relevance f o r ethics, w h i c h general d i s p o s i t i o n he regards as a l e g a c y o f M o o r e .  T o see h o w p s y c h o l o g y is related to p h i l o s o p h y i n K o h l b e r g ' s v i e w , let us see w h a t K o h l b e r g says about i t : W h a t I a m c l a i m i n g about the r e l a t i o n o f " i s " to " o u g h t " i n m o r a l d e v e l o p m e n t c o m e s to this:  7  Ibid, 52.  162  1. T h e s c i e n t i f i c facts are that there is a u n i v e r s a l m o r a l f o r m s u c c e s s i v e l y  e m e r g i n g i n d e v e l o p m e n t and centering o n p r i n c i p l e s o f j u s t i c e .  2. T h i s K a n t i a n m o r a l f o r m is one that assumes the fact-value d i s t i n c t i o n ;  that i s , m o r a l p e o p l e assume that their m o r a l j u d g m e n t is based o n c o n f o r m i t y to an ideal n o r m , not o n c o n f o r m i t y to fact.  3. S c i e n c e , then, c a n test whether a p h i l o s o p h e r ' s c o n c e p t i o n o f m o r a l i t y  p h e n o m e n o l o g i c a l l y fits the p s y c h o l o g i c a l facts. S c i e n c e cannot go o n to j u s t i f y that c o n c e p t i o n o f m o r a l i t y as w h a t m o r a l i t y ought to be.... M o r a l a u t o n o m y is k i n g , a n d values are different f r o m facts for m o r a l discourse. S c i e n c e cannot  p r o v e o r j u s t i f y a m o r a l i t y , because the rules o f s c i e n t i f i c discourse are not the rules o f m o r a l d i s c o u r s e .  4. L o g i c o r n o r m a t i v e ethical analysis c a n , h o w e v e r , p o i n t out that a  certain type o f m o r a l p h i l o s o p h y — for e x a m p l e , Stage 4 — does not handle or resolve certain p r o b l e m s that it a c k n o w l e d g e s to be p r o b l e m s that it ought to  handle, whereas another type o f m o r a l i t y (for e x a m p l e , Stage 5) c a n do so. H e r e , factual i n v e s t i g a t i o n o f p e o p l e ' s beliefs must support internal l o g i c a l analysis o f  w h y the d e v e l o p m e n t a l l y h i g h e r p h i l o s o p h y c a n handle p r o b l e m s not h a n d l e d b y  the l o w e r ones. S c i e n c e , then, c a n contribute to a m o r a l discourse as to w h y one m o r a l theory is better than another.  5. T h e s c i e n t i f i c theory as to w h y people f a c t u a l l y do m o v e u p w a r d f r o m  stage to stage, and w h y they factually do prefer a higher stage to a l o w e r , is  b r o a d l y the same as a m o r a l theory as to w h y p e o p l e should prefer a h i g h e r stage to a l o w e r . . . . ( K o h l b e r g 1 9 8 1 , pp. 178-179)  W e can see that statement (1) is about the p s y c h o l o g i c a l p h e n o m e n a K o h l b e r g c l a i m s to have d i s c o v e r e d t h r o u g h h i s studies. Statement (2) is about the d i s c r e p a n c y between facts and v a l u e s as p e r c e i v e d b y n o r m a l p e o p l e , and this d i s c r e p a n c y is l a r g e l y the same as w h a t m a n y (or most) m o r a l p h i l o s o p h e r s have i n m i n d . In spite o f s u c h a discrepancy, K o h l b e r g nevertheless suggests i n (3) that i n some w a y m o r a l p s y c h o l o g y is relevant to m o r a l p h i l o s o p h y . T h e statement echoes w h a t I reported a f e w paragraphs ago, that the c o n c e p t i o n o f w h a t an adequate m o r a l j u d g m e n t s h o u l d be m u s t c o m p o r t w i t h w h a t such j u d g m e n t s are or can be i n p e o p l e ' s m i n d s . That i s , (3) p r o v i d e s us w i t h an interpretation o f h o w c o n c e p t i o n s o f ideal m o r a l j u d g m e n t must rest o n capacities o f actual j u d g m e n t s . T h i s is an e x a m p l e o f w h a t H a b e r m a s c a l l s the ' c o m p l e m e n t a r i t y thesis' o f the r e l a t i o n b e tw e e n p s y c h o l o g y and p h i l o s o p h y , w h i c h c l a i m s that  163  p s y c h o l o g i c a l studies " f u n c t i o n as a c h e c k o n the n o r m a t i v e v a l i d i t y o f h y p o t h e t i c a l l y reconstructed m o r a l i n t u i t i o n s . "  8  (3) states that p s y c h o l o g y c a n o n l y f u n c t i o n as a c h e c k o n a certain m o r a l c o n c e p t i o n , but cannot  prove its adequacy. N e v e r t h e l e s s , (4) states that s c i e n c e c a n  "contribute to a m o r a l discourse as to w h y one m o r a l theory is better than another." T h i s suggests that (a) p s y c h o l o g y c a n support certain aspects o f m o r a l t h e o r y ; and (b) this support is w e a k e r than a p r o o f . (4) a n d (5) together p r o v i d e us w i t h an e x p l a n a t i o n o f h o w this support is p o s s i b l e . (4) states that, i n m o r a l d e v e l o p m e n t , a subject gives up a certain m o r a l c o n c e p t i o n a n d takes up another because the p r e v i o u s c o n c e p t i o n cannot handle certain p r o b l e m s that need to be h a n d l e d . T h e p e r s o n thus replaces one m o r a l c o n c e p t i o n w i t h another w h i c h is m o r e adequate to the p r o b l e m s m o r a l i t y deals w i t h . T h i s is the p s y c h o l o g i c a l e x p l a n a t i o n o f w h y p e o p l e do m o v e f r o m one stage to another. A t the same t i m e , (5) states that the reason w h y people do m o v e u p w a r d and do prefer higher stages is b r o a d l y the same as the reason w h y they should prefer t h e m . That is, the f u n c t i o n a l superiority o f the h i g h e r stages  explains the fact that p e o p l e prefer t h e m and  m o v e u p w a r d as they are able to. That i s , w e can diagnose the d i f f i c u l t i e s w i t h the l o w e r stages and describe the advantages o f the higher. A s w e shall see shortly, these advantages are connected essentially w i t h c o n s i s t e n c y and comprehensiveness. A p h y s i c a l l a w e x p l a i n s a p h y s i c a l p h e n o m e n o n , and the observed p h e n o m e n a i n turn support the l a w . S i m i l a r l y , i f the pattern o f h u m a n m o r a l d e v e l o p m e n t t h r o u g h K o h l b e r g i a n stages is e x p l a i n e d b y the f u n c t i o n a l superiority o f h i g h e r stages then,  Jiirgen Habermas, "Interpretive Social Science vs. Hermeneutics." In Social Science as Moral Inquiry, ed. N. Haan et al. (New York: Columbia University Press, 1983), 266. 8  164  o b v i o u s l y , to this extent the h i g h e r stages are better, a n d p e o p l e ' s preferences f o r t h e m , based o n these f i n d i n g s , supports the c l a i m o f superiority.  S e e n i n this light, (5) is a p l a u s i b l e c l a i m . B o y d w o u l d most l i k e l y agree. H e argues that one o f the tasks o f d e v e l o p m e n t a l p s y c h o l o g y is to e x p l a i n w h y p e o p l e ' s m o r a l c o n c e p t i o n s change s y s t e m a t i c a l l y i n a particular d i r e c t i o n . T o e x p l a i n the pattern o f these changes one s h o u l d not treat t h e m as mere p h e n o m e n a , for d o i n g so o v e r l o o k s the fact that the subjects are themselves constructive m o r a l agents. B o y d e v e n c l a i m s 9  that " w h a t the m a t u r i n g m o r a l subject does and w h a t the m o r a l p h i l o s o p h e r does are essentially the same c o g n i t i v e a c t i v i t y , the same orientation t o w a r d the c o n s t r u c t i o n o f m o r a l m e a n i n g and d e t e r m i n a t i o n o f m o r a l truth, c a r r i e d o n at different l e v e l s o f abstraction."  10  I a m sympathetic to B o y d ' s i d e a about the s i m i l a r i t y between the c o g n i t i v e activities o f the p h i l o s o p h e r and the l a y p e r s o n i n their reflections and c o g n i t i v e constructions i n m o r a l r e a s o n i n g . L a y p e o p l e surely do p h i l o s o p h i z e about m o r a l p r o b l e m s too. O t h e r w i s e h o w c o u l d non-philosophers ever b e c o m e p h i l o s o p h e r s ? W e need not suppose that e v e r y b o d y is e q u a l l y sophisticated about m o r a l p r o b l e m s ( K o h l b e r g does not suppose this either), but it is a m i s t a k e to suppose that the w a y the l a y p e r s o n strives to construct h i s o w n m o r a l m e a n i n g is irrelevant to w h a t m o r a l reasoning s h o u l d be.  9  10  Dwight B o y d , "The Ought o f Is,"  Ibid., 5 5 .  60.  165  1.4 The "empirical" implications of statements of value K o h l b e r g suggests that p s y c h o l o g y supports a m o r a l theory b y testing its consequences, and an adequate m o r a l theory must fit the p s y c h o l o g i c a l facts about h u m a n beings. B u t h o w do w e j u d g e whether a theory fits these p s y c h o l o g i c a l facts o r not? A n d w h a t k i n d s o f the consequences are o f c o n c e r n here? I f there is a m o r a l theory w h i c h c l a i m s that some general f o r m o f m o r a l r e a s o n i n g is better, w h a t s h o u l d w e expect to observe?  Suppose there are t w o things, A and B, and A is o b j e c t i v e l y better than B. W h a t do w e expect? D o w e expect, f o r e x a m p l e , that e v e r y b o d y prefers A to B ? N o . I f this were a reasonable expectation then w e s h o u l d expect everyone to prefer B e e t h o v e n to M a d o n n a . ( M y p o i n t is s i m p l y that e v e n those w h o b e l i e v e that B e e t h o v e n is better than M a d o n n a do not expect e v e r y b o d y share this v i e w . ) S h o u l d w e appeal to the preference o f the majority? S h o u l d w e expect that most people w i l l prefer A to B i f A i s o b j e c t i v e l y better than B ? A g a i n , no. W e a l l k n o w that people m a k e m i s t a k e s , particular i n d i v i d u a l s d o , a n d so do the majority.  A m o r e p l a u s i b l e answer c o n c e r n i n g expectation and the a s s u m p t i o n o f objective goodness is this: the o b j e c t i v e l y better is s o m e t h i n g that a f u l l y rational agent prefers, p r o v i d e d he has a l l the relevant k n o w l e d g e . B y ' f u l l y r a t i o n a l ' I m e a n f u l l y consistent and coherent i n o n e ' s r e a s o n i n g . M o r e o v e r , w e s h o u l d expect a converg ence o f preferences a m o n g p e o p l e to the extent that they are rational and f u l l y i n f o r m e d . In other w o r d s , the m o r e c o m p l e t e l y rational and f u l l y i n f o r m e d are a set o f p e o p l e , the greater the  166  tendency f o r agreement o v e r values that are o b j e c t i v e .  11  L e t us c a l l this the C R P  (convergence o f rational preferences) c r i t e r i o n o f o b j e c t i v i t y o f values.  N o t e that b y s a y i n g a l l this I a m m a k i n g n o c l a i m about h o w objective values exist. Perhaps they exist as P l a t o n i c f o r m s independent o f h u m a n existence and h u m a n p s y c h o l o g y , o r perhaps they exist p r e c i s e l y as a c o n s t r u c t i o n f r o m the intersubjective, rational convergence itself. I n either case the C R P c r i t e r i o n w o r k s . T h e reason it w o r k s for the case where v a l u e i s a h u m a n c o n s t r u c t i o n is clear: i f values d o not exist independently but o n l y as a h u m a n p r o j e c t i o n , then the converg ence o f rational preference under f u l l k n o w l e d g e defines these values. That objective values are those o n w h i c h rational preferences c o n v e r g e (as far as p e o p l e are f u l l y i n f o r m e d ) b e c o m e s a k i n d o f conceptual truth, for this i s a l l they can be. B u t i f values exist independently o f the h u m a n m i n d , then it m i g h t seem that there i s n o reason w h y i n f o r m e d rational p e o p l e s h o u l d necessarily prefer  t h e m . ( I f their existence does not d e p e n d o n our existence, w h y  s h o u l d there be a n y c o n n e c t i o n between t h e m and our preferences?) B u t this p r o b l e m i s i l l u s o r y . W h a t w e are d i s c u s s i n g are human values, things that are v a l u a b l e for us. B u t i f there are things that are o b j e c t i v e l y g o o d f o r us, i n this mind-independent w a y , this  goodness  w i l l be i n e x p l i c a b l e i f f u l l y i n f o r m e d , rational p e o p l e d o not r e c o g n i z e these  values. F o r w h a t c o u l d p o s s i b l y e x p l a i n this believe? S o i f people are f u l l y i n f o r m e d (about their o w n true interests and e v e r y t h i n g else, i n c l u d i n g m e t a p h y s i c a l o r r e l i g i o u s truths i f y o u l i k e ) and r a t i o n a l , their preference s h o u l d converge o n these objective (human) values.  11  I owe this idea to Earl Winkler, who suggested this to me in a personal discussion.  167  T h e C R P c r i t e r i o n is not p u r e l y e m p i r i c a l . T o see w h a t is o b j e c t i v e l y g o o d , w e d o not o n l y observe w h a t p e o p l e actually prefer, w e also e x a m i n e h o w rational they are and h o w m u c h they k n o w o f a l l the things relevant to a g i v e n j u d g m e n t . B u t criteria o f rationality and k n o w l e d g e are at least partly n o r m a t i v e , f o r there are n o r m a t i v e standards that w e use to j u d g e w h o i s rational a n d w h o has proper k n o w l e d g e . F o r instance, to b e rational one has to be consistent and coherent i n o n e ' s reasoning. H o w e v e r , to be consistent i n o n e ' s r e a s o n i n g one has to f o l l o w v a r i o u s n o r m s o f l o g i c a l a n a l y s i s , s u c h as "It i s w r o n g to c o m m i t the f a l l a c y o f a f f i r m i n g the c o n s e q u e n t " and "It i s correct to infer that Q f r o m ' I f P then Q and P.'" S i m i l a r l y , to establish proper k n o w l e d g e w e have to f o l l o w n o r m s l i k e " I f a theory predicts P and w e observe P, then w e have some reason to believe that the theory i s t r u e . " B u t the c r i t e r i o n o f rational converg ence i s not p u r e l y n o r m a t i v e either. It does predict some e m p i r i c a l p h e n o m e n a . T o see h o w i t does so, let us i m a g i n e the f o l l o w i n g scenario.  1.5 The "Adobe Photoshop" Analogy. Imagine y o u w a n t to b u y a c o m p u t e r p r o g r a m f o r photo m a n i p u l a t i o n . S o y o u want to k n o w w h i c h p r o g r a m is the best i n the market. N o w y o u learn that a l l the most k n o w l e d g e a b l e buyers w h o w a n t a f u l l range o f m a n i p u l a t i o n c a p a b i l i t y b u y A d o b e P h o t o s h o p . W h a t does this p h e n o m e n o n suggest? D o e s it suggest that the A d o b e P h o t o s h o p i s the best p r o g r a m c i r c u l a t i n g i n the market?  It seems that it does suggest so. A n d this is c o m p a r a b l e to K o h l b e r g ' s f i n d i n g s about p e o p l e ' s preferences f o r h i g h e r m o r a l stages. There are four important observations that support the superiority o f h i g h e r stages: (1) higher-stage m o r a l reasoning requires greater c o g n i t i v e a b i l i t y ; (2) stage d e v e l o p m e n t i s i r r e v e r s i b l e ; (3) p e o p l e prefer the  168  highest stage they c a n c o m p r e h e n d ; and (4) higher stages predict m o r a l b e h a v i o r . T a k e n together, (1) and (3) c o m e pretty close to the fact that people w i t h p r o p e r k n o w l e d g e prefer the A d o b e P h o t o s h o p i h o u r a n a l o g y . N o t e that i n the A d o b e P h o t o s h o p case it is not the m a j o r i t y ' s preference i t s e l f that suggests that A d o b e P h o t o s h o p is the best p r o g r a m . It is rather the fact that k n o w l e d g e a b l e buyers prefer it that suggests it is the best. Therefore, that h i g h e r stages require greater c o g n i t i v e a b i l i t y does s e e m to support the h i e r a r c h i c a l c l a i m .  B y e x a m i n i n g w h a t e x p l a i n s stage p r o g r e s s i o n w e can p r o v i d e e v e n m o r e s o l i d f o u n d a t i o n f o r a c c e p t i n g K o h l b e r g ' s h i e r a r c h i c a l c l a i m . C o n s i d e r c o g n i t i v e a b i l i t y . If there is any m o r a l reasoning at a l l , then m o r a l maturity must, to some extent, be connected w i t h c o g n i t i v e maturity. T h e r e f o r e , a m o r a l l y mature p e r s o n s h o u l d s h o w a stronger a b i l i t y to deal w i t h the c o m p l e x i t i e s o f m o r a l situations. H e s h o u l d also be able to d i s t i n g u i s h between v a l u e concepts i n a systematic w a y . These are w h a t K o h l b e r g c a l l s the c o g n i t i v e c r i t e r i a (see C h a p t e r 1, section 1.3 f o r details). T h e r e f o r e , that h i g h e r stages require greater c o g n i t i v e a b i l i t y , i.e., (1), is a natural consequence o f the h i e r a r c h i c a l c l a i m . S i n c e (1) is e m p i r i c a l l y c o n f i r m e d b y e m p i r i c a l studies, these studies p r o v i d e indirect support f o r the h i e r a r c h i c a l c l a i m .  O f course, g i v e n a l l this, the support for the h i e r a r c h i c a l c l a i m remains weak. T o be a p r o f e s s i o n a l k i l l e r requires greater c o g n i t i v e a b i l i t y than to be an average gangster, but this does not m e a n that b e i n g a p r o f e s s i o n a l k i l l e r is o b j e c t i v e l y better than b e i n g a gangster. (3) is therefore e s p e c i a l l y important. O n one h a n d , (3) is relevant to the hierarchical c l a i m i n that the p e r s o n w h o understands o n l y Stages 1 -4 is c o g n i t i v e l y i n f e r i o r to the p e r s o n w h o understands also Stage 5, at least as far as their abilities to  169  c o m p r e h e n d different m o r a l c o n c e p t i o n s is concerned. T h e r e f o r e , (3) supports the objective superiority o f the h i g h e r stages i n a w a y s i m i l a r to (1). M o r e i m p o r t a n t l y , (3) also supports the s u p e r i o r i t y o f the h i g h e r stages i n a n important w a y i n w h i c h (1) does not. T h i s has to do w i t h the fact that w h e n w e assess the relevance o f p e o p l e ' s preferences to the o b j e c t i v i t y o f values, considerations o f e p i s t e m i c adequacy c o m e into p l a y as w e l l . S i n c e it appears that n o one w h o articulates a preference f o r a l o w e r stage over a h i g h e r stage understands the h i g h e r stage correctly, there is r e a l l y n o b o d y w h o prefers a l o w e r stage to a h i g h e r stage. C o n s e q u e n t l y , there is r e a l l y n o disagreement o n whether a h i g h e r stage is better than a l o w e r stage. A l l the preferences p e o p l e have are preferences for h i g h e r stages to l o w e r ones.  T h o u g h w e do not c l a i m that the objective superiority o f A o v e r B e m p i r i c a l l y i m p l i e s a general agreement o n p e o p l e ' s preference for A to B, the existence o f agreement does u n d e r m i n e r e l a t i v i s m i n an indirect w a y . R e l a t i v i s m seems to predict that there are different and c o n f l i c t i n g m o r a l standards a m o n g p e o p l e , and that there is n o w a y to resolve these. B u t this is not w h a t w e f i n d b y o b s e r v a t i o n o n the q u e s t i o n o f the preferability o f h i g h e r to l o w e r K o h l b e r g i a n stages. W h a t w e f i n d is rather that there is little real disagreement at a l l i n preferences f o r m o r a l standards as represented b y these stages. A t least here, then, r e l a t i v i s m is less p l a u s i b l e than it appears, a n d the less c r e d i b l e r e l a t i v i s m i s , the m o r e c r e d i b l e is o b j e c t i v i s m or u n i v e r s a l i s m .  T h e relation b e tw e e n (2) and the objective superiority o f h i g h e r stages is s i m i l a r . T h e i r r e v e r s i b i l i t y o f stage d e v e l o p m e n t means that n o b o d y w h o has adopted the m o r a l c o n c e p t i o n o f a h i g h e r stage w i l l ever, under n o r m a l c i r c u m s t a n c e s , g i v e it up and replace it b y a l o w e r stage. P r e s u m a b l y , a person w h o has adopted a certain m o r a l c o n c e p t i o n  170  understands it s u f f i c i e n t l y to be able to a p p l y it i n m a k i n g m o r a l j u d g m e n t s . Irreversibility therefore amounts to a k i n d o f p e r f o r m a t i v e agreement a m o n g p e o p l e to the effect that, relative to the stages they c a n e m p l o y , a h i g h e r stage is a l w a y s better than a l o w e r one. T h i s again u n d e r m i n e s the relativist c l a i m that different m o r a l standards are equally adequate.  L a s t l y w e have to see h o w (4) is relevant to the objective superiority o f the h i g h e r stages. S i n c e w e expect that m o r a l maturity predicts m o r a l b e h a v i o r , the m o r a l l y mature person s h o u l d have a greater tendency to do the right t h i n g i n a m o r a l situation. T h u s , i f i n situation S the m o r a l l y r i g h t t h i n g to d o is X , the m o r a l l y mature p e r s o n s h o u l d have a greater tendency to do X i n situation S than the i m m a t u r e person, other things equal. T h i s c r i t e r i o n is p h i l o s o p h i c a l l y e v e n m o r e c o n v i n c i n g than the above t w o , f o r w h a t w e do i n e m p l o y i n g this c r i t e r i o n is not to infer an "oughf'-statement f r o m an "is"-statement. Q u i t e the contrary, w e are i n f e r r i n g an "oughf'-statement f r o m an "ought"-statement. T h e inference is l i k e this: since Stage 5 subjects are m o r e l i k e l y to do w h a t they ought to do than Stage 4 subjects, Stage 5 r e a s o n i n g is m o s t l i k e l y better than Stage 4.  A q u e s t i o n arises c o n c e r n i n g this l i n e o f argument. T h e argument starts w i t h the a s s u m p t i o n that there are things that are o b v i o u s l y right, s u c h as q u i t t i n g the e x p e r i m e n t i n the M i l g r a m situation. B u t i f this course is really so o b v i o u s , then e v e n the less mature s h o u l d f i n d it w r o n g to continue j u s t as the m o r e mature d o . It w o u l d thus seem u n l i k e l y that the less mature w o u l d d i f f e r m u c h i n their tendency to quit f r o m the m o r e mature. B u t since they do d i f f e r , the i d e a that " t o quit is r i g h t " must not be as o b v i o u s as it seems, at least not to the less mature.  171  It is true that, f o r the less mature, to quit is less o b v i o u s l y the right t h i n g to d o . B u t this has n o t h i n g to d o w i t h whether it is right to quit. Indeed, w h e n I argue f o r the superiority o f h i g h e r stages o n the basis o f the difference betw een subjects at different stages i n the M i l g r a m situation I assume that the reader already agrees w i t h m e that it is right to quit. T h e q u e s t i o n I w a n t to ask is not whether it is right to quit but, rather, since it is right to quit, a n d since most higher-stage subjects quit w h i l e m o s t lower-stage subjects do not, whether it is reasonable to suppose that higher-stage subjects are i n d e e d m o r a l l y m o r e mature. M o r e o v e r , the fact that w h a t is right i n the s i t u a t i o n is less o b v i o u s to the lower-stage subjects i t s e l f suggests that these subjects are less mature, f o r w h a t is o b v i o u s l y right is less o b v i o u s l y right f o r t h e m .  O n the other h a n d , that q u i t t i n g is less o b v i o u s l y right f o r lower-stage subjects does not m e a n that they do not b e l i e v e or have some strong sense that it is right to quit. M o s t o f the subjects w h o c o n t i n u e d s h o w e d strong signs o f a n x i e t y , and they c o n t i n u e d to a p p l y the s h o c k s o n l y w h e n f o r c e d b y the experimenter to do so. There are t w o points w e can d r a w f r o m these facts. F i r s t , there is no reason to doubt the p r e s u p p o s i t i o n that it is o b v i o u s l y right to quit the M i l g r a m experiment. S e c o n d , a l t h o u g h a l m o s t a l l participants feel that it is w r o n g to c o n t i n u e , higher-stage subjects are m o r e i n c l i n e d to quit than lower-stage subjects, suggesting that higher-stage r e a s o n i n g p r o v i d e s stronger m o t i v a t i o n or c o n f i d e n c e f o r a subject to do w h a t he/she k n o w s to be right. D i f f e r e n t f o r m s o f m o r a l r e a s o n i n g appear therefore to have different m o t i v a t i n g p o w e r s , w i t h higher K o h l b e r g i a n stages p r o v i d i n g stronger m o t i v a t i o n f o r m o r a l a c t i o n . A s suggested above, s u c h a c o r r e l a t i o n between m o r a l stage and m o r a l b e h a v i o r m a y also be related to self-confidence. If y o u understand y o u r o w n reasons w e l l y o u do not defer so m u c h to  172  authority. Therefore w e can c o n c l u d e that the h i g h e r stages are better o n the basis that they p r o v i d e stronger or m o r e secure m o t i v a t i o n f o r a p e r s o n to do the right t h i n g .  1 2  T h e a b o v e d i s c u s s i o n c a n be regarded as a r e p l y to F l a n a g a n ' s A r g u m e n t 10 against K o h l b e r g ' s h i e r a r c h i c a l c l a i m . F l a n a g a n ' s p o i n t is that invariant sequence does not i m p l y h i e r a r c h y , a n d he is d e f i n i t e l y correct. That i s , the sequentiality o f the stages i t s e l f does not suggest m u c h . H o w e v e r , w h e n t a k i n g other e m p i r i c a l f i n d i n g s into consideration, the h i e r a r c h i c a l nature o f these stages gains s i g n i f i c a n t support.  1.6 The C R P criterion revisited N o w let us return to the C R P c r i t e r i o n . W h a t does m y interpretation o f K o h l b e r g ' s theory have to do w i t h the C R P criterion? T h e C R P c r i t e r i o n states that the o b j e c t i v e l y better t h i n g is one o n w h i c h f u l l y rational agents' preferences c o n v e r g e , p r o v i d e d that they have f u l l k n o w l e d g e about the things they are c o m p a r i n g , and regardless o f whether our conceptions o f o b j e c t i v i t y f o r h u m a n values is realist or constructivist.  I have p o i n t e d out that the C R P c r i t e r i o n has some e m p i r i c a l content that is o f relevance i n assessing K o h l b e r g ' s theory. T h i s content c o m e s f r o m the fact that w e h u m a n beings are rational beings. T h o u g h not e v e r y b o d y is e q u a l l y r a t i o n a l , w e strive to be as rational as w e c a n w h e n w e reflect o n any k i n d o f p r o b l e m . S i n c e most o f us, or a l l o f us, do reflect o n things, it is most l i k e l y that our a b i l i t y to solve p r o b l e m s r a t i o n a l l y w i l l sharpen t h r o u g h t i m e . It is absurd to suppose that there is n o trend whatsoever i n the c o g n i t i v e d e v e l o p m e n t as p e o p l e g r o w u p . S i m i l a r l y , it is e x t r e m e l y i m p l a u s i b l e to  1 do not pretend that this is a logically conclusive argument. Nevertheless, I find it compelling, and all the premises in the argument are either true or highly plausible. 12  173  suppose that p e o p l e ' s m o r a l m a t u r i t y does not g r o w a n d d e v e l o p at a l l . O f course s o m e people m a y s i m p l y b e c o m e better able to r a t i o n a l i z e and c o g n i t i v e l y distort the matters they encounter. B u t it seems that they w i l l m o r e l i k e l y be e x c e p t i o n s than n o r m s . O t h e r w i s e experience a n d reflections w i l l be entirely or a l m o s t entirely irrelevant to g r o w t h , w h i c h is h i g h l y i m p l a u s i b l e . Therefore the general pattern one f i n d s i n h u m a n m o r a l d e v e l o p m e n t cannot be p h i l o s o p h i c a l l y i n s i g n i f i c a n t . There also r e m a i n s the p o s s i b i l i t y that the c h r o n o l o g i c a l order o f the emergence o f m o r a l stages c a n be e x p l a i n e d b y other reasons else. T h e r e f o r e F l a n a g a n and S e i g e l are correct i n c l a i m i n g that sequentiality does not i m p l y hierarchy. B u t they cannot p r o v i d e us w i t h a n y other p l a u s i b l e e x p l a n a t i o n either. In this absence, i n c r e a s i n g adequacy remains the o n l y and best e x p l a n a t i o n , e s p e c i a l l y w h e n buttressed w i t h analyses o f increased adequacy i n terms o f greater c o n s i s t e n c y and c o m p r e h e n s i v e n e s s .  C h r o n o l o g i c a l order is not the o n l y " e v i d e n c e " w e have. W e also observe that the e m p l o y m e n t o f m o r a l stages is related to c o g n i t i v e a n d rational a b i l i t y . T h e C R P c r i t e r i o n requires that evaluators have f u l l k n o w l e d g e about the things they c o m p a r e i n order to j u d g e w h i c h is better. K o h l b e r g discovers that a l l disagreements about the c o m p a r i s o n o f t w o stages are a c c o m p a n i e d b y an e p i s t e m i c a s y m m e t r y . A l m o s t a l l w h o understand t w o stages prefer the h i g h e r stage, and almost a l l w h o " p r e f e r " a l o w e r stage to a h i g h e r m i s u n d e r s t a n d the nature o f the higher stage. T h i s strongly suggests that h i g h e r stages are o b j e c t i v e l y better than l o w e r stages a c c o r d i n g to the C R P c r i t e r i o n . T h e same l i n e o f argument can also be a p p l i e d to the i r r e v e r s i b i l i t y o f d e v e l o p m e n t as w e l l .  T h e effect o f free d i s c u s s i o n s o n stage d e v e l o p m e n t is e v e n more supportive o f our c o n c l u s i o n . A c c o r d i n g to the C R P c r i t e r i o n , objective v a l u e is necessarily connected  174  to convergence o f rational preference, either quite d i r e c t l y o n a constructivist account o f o b j e c t i v i t y , or m o r e i n d i r e c t l y o n a realist account. W h e n w e are engaged i n free d i s c u s s i o n s about values and the determination o f m o r a l rightness the v e r y c o n d i t i o n s o f d i s c u s s i o n i m p o s e constraints w h i c h force us to a v o i d arbitrariness i n p r e f e r r i n g some p e o p l e ' s interests o v e r others. T h e abstract tendency o f these constraints is s u c h as to lead everyone towards a r e c o g n i t i o n o f the equal basic m o r a l status and d i g n i t y o f a l l persons. Therefore, i n this m o r e substantive w a y a l s o , the m o r e rational w e are the m o r e l i k e l y w e shall agree o n m o r a l standards o f the sort represented b y the h i g h e r K o h l b e r g i a n stages. These considerations — w h i c h , a g a i n , i n v o l v e a substantive matter connected w i t h equality and j u s t i c e — p r o v i d e a d d i t i o n a l support for the c l a i m that h i g h e r stages are better than the l o w e r ones.  W h e n d e f i n i n g rationality, w e say that b e i n g rational means b e i n g consistent and coherent i n o n e ' s reasoning. A s e m p h a s i z e d throughout this d i s c u s s i o n , these criteria are p r o m i n e n t l y reflected i n K o h l b e r g ' s d e v e l o p m e n t a l m o d e l . T h e related concept is that o f d i s e q u i l i b r i u m . A stage is better than its p r e v i o u s stage because, w i t h its greater r e v e r s i b i l i t y , it is less l i k e l y to lead to d i s e q u i l i b r i u m (Chapter 1, section 1.4). A n d , o f course, this is a major reason p e o p l e m o v e to a h i g h e r l e v e l . A s they d i s c o v e r that their m o r a l c o n c e p t i o n leads to irreversible solutions to m o r a l d i l e m m a s they b e c o m e less and less c o m f o r t a b l e w i t h their o w n m o r a l standard, and they try to revise their m o r a l c o n c e p t i o n to m a k e it m o r e equilibrated. E s p e c i a l l y under the pressures o f m o r a l discourse a m o n g p e o p l e o f d i f f e r i n g race, b a c k g r o u n d , gender, s o c i a l p o s i t i o n , etc., they  175  try to extend their realms o f m o r a l c o n c e r n , to understand better the i n d i v i d u a l situation o f e v e r y p e r s o n , and to treat different persons as e q u a l l y as p o s s i b l e .  13  1.7 A clarification of K o h l b e r g ' s hierarchical claim There i s an important p o i n t to note so that w e do not m i s u n d e r s t a n d K o h l b e r g ' s h i e r a r c h i c a l c l a i m . W h a t K o h l b e r g endeavors to s h o w is that higher-stage r e a s o n i n g i s i n general m o r e adequate than that o f lower-stages. H e is not t r y i n g to s h o w that higher f o r m s o f r e a s o n i n g always g i v e rise to better m o r a l d e c i s i o n s . In H e i n z ' s d i l e m m a , a Stage 4 reasoner m a y decide that H e i n z s h o u l d not steal the d r u g so that the l e g a l system is protected; w h i l e a Stage 2 reasoner m a y decide to steal the d r u g because h i s w i f e w i l l then do h i m a favor i n return. T h i s does not m a k e the c h o i c e o f r e f r a i n i n g f r o m stealing better than stealing. F o r K o h l b e r g , it is clear that stealing the d r u g is the right c h o i c e .  S i m i l a r l y , that t w o persons are at the same stage does not necessarily m e a n that they have e q u a l l y adequate m o r a l b e l i e f s . Therefore w e c a n r e p l y to objections s u c h as the one raised b y L a u r e n c e T h o m a s . " I f K o h l b e r g is to be taken l i t e r a l l y , then people w h o at Stage 4 came to embrace N a z i anti-Semitic i d e o l o g y were e x h i b i t i n g as m u c h proper m o r a l d e v e l o p m e n t as c o n t e m p o r a r y C a n a d i a n s w h o at Stage 4 c o m e to embrace the egalitarian i d e o l o g y o f C a n a d i a n s o c i e t y . "  14  T h e k e y to p u z z l e s l i k e this l i e s i n the  d i s t i n c t i o n between the adequacy o f the " f o r m " o f m o r a l reasoning a n d the h i g h l y substantive values h e l d b y a subject. T h e fact that b o t h the N a z i and the egalitarian are i n Stage 4 m a y suggest their f o r m s o f m o r a l reasoning are e q u a l l y adequate. H o w e v e r , b y  13 The hypothesis that disequilibrium stimulates stage progression is also tested empirically (Chapter 1, section 2.2). 14 Laurence Thomas, "Morality and Psychological Development." In A Companion to Ethics, ed. Peter Singer (Oxford, Blackwell, 1993), 471.  176  n o means does it i m p l y that a v a l u e system d r i v i n g p e o p l e to massacre a n d persecute others m u s t be as g o o d as one r e q u i r i n g t h e m to l i v e p e a c e f u l l y w i t h others. K o h l b e r g does not c l a i m that a better f o r m o f m o r a l reasoning a l w a y s guarantees a better m o r a l d e c i s i o n . T h i s is e x a c t l y the reason he distinguishes m o r a l " f o r m " a n d m o r a l content. O n e m a y have c h o s e n a m o r a l l y " c o r r e c t " a c t i o n b y l u c k , so to speak, as has the egalitarian i n C a n a d a . I say that he has c h o s e n the correct m o r a l content b y " l u c k " because, p r o v i d e d that he has c h o s e n it because he wants to m a i n t a i n the s o c i a l order b y f o l l o w i n g c o n v e n t i o n a l n o r m s , he/she w o u l d have c h o s e n the N a z i n o r m had he/she l i v e d i n N a z i G e r m a n y . A n d , o f course, one c a n also reason i n a f a i r l y adequate w a y but still m a k e a w r o n g d e c i s i o n (perhaps except Stage 6 reasoners). T h e r e f o r e , it is not s u r p r i s i n g f o r a N a z i w h o s e m o r a l reasoning is f o r m a l l y as adequate as the C a n a d i a n egalitarian's to m a k e m o r a l c h o i c e s that are w o r s e , and i n d e e d far w o r s e , than the C a n a d i a n ' s .  M o r e o v e r , it is w r o n g to suppose that a l l reasoners at the same stage d i s p l a y e q u a l l y adequate m o r a l r e a s o n i n g . A s w e have seen i n C h a p t e r 2, K o h l b e r g does not measure m o r a l maturity i n terms o f m o r a l stages o n l y , but also i n terms o f m o r a l types (Chapter 2, section 1.3). A Stage 4 N a z i w o u l d almost certainly be a T y p e A reasoner, f o r he/she fails to r e c o g n i z e the v a l u e o f c o n s c i e n c e o v e r f i x e d l a w s ( a s s u m i n g that conscience p r o h i b i t s massacre), and therefore his/her reasoning violates the criterion  hierarchy.  of  L i k e w i s e , he/she c o u l d not l i k e l y satisfy the c r i t e r i o n o f a u t o n o m y , w h i c h  requires him/her to r e c o g n i z e that he/she is an autonomous m o r a l agent. If one had s u c h r e c o g n i t i o n , one c o u l d h a r d l y a v o i d f o l l o w i n g o n e ' s o w n c o n s c i e n c e instead o f o b e y i n g the government. He/she c o u l d not satisfy the criterion  of intrinsicalness,  w h i c h requires  him/her to s h o w an " i n t r i n s i c respect f o r p e r s o n s ; " and he/she w o u l d have f a i l e d to satisfy  177  the criterion  of moral respect, w h i c h requires him/her to r e c o g n i z e that the other parties  as a u t o n o m o u s m o r a l agents too. He/she c o u l d not p o s s i b l y satisfy the criterion  constructivism,  of  w h i c h requires him/her to c o n s i d e r the v i e w points o f a l l the.other parties.  Therefore, the Stage 4 N a z i c o u l d o n l y b e a Stage 4 A reasoner, but the C a n a d i a n egalitarian w o u l d quite l i k e l y f a l l under Stage 4 B . In this sense, there is s t i l l a s i g n i f i c a n t difference i n the adequacy o f their m o r a l reasoning or values ( f o r the c r i t e r i a o f type B m o r a l r e a s o n i n g refer to C h a p t e r 2 , s e c t i o n 1.3).  F u r t h e r m o r e , it is e x t r e m e l y p r o b l e m a t i c whether a n y Stage 4 reasoner c a n t r u l y believe i n N a z i s m . W h e n I say this I d o not m e a n that Stage 4 r e a s o n i n g and N a z i i d e o l o g y are l o g i c a l l y i n c o m p a t i b l e w i t h each other. I a m s a y i n g that as a matter o f fact it is extremely u n l i k e l y f o r a Stage 4 reasoner to embrace s u c h a v a l u e at a l l . T h e Stage 4 m o r a l c o n c e p t i o n is not one o f obedience to authority o r government. A l s o , n o r m s f o r a Stage 4 reasoner serve to " p r o m o t e c o o p e r a t i o n o r s o c i a l c o n t r i b u t i o n a n d act as regulations d e s i g n e d to a v o i d disagreement a n d d i s o r d e r " ( K o h l b e r g 1984, p. 632). D o e s genocide a v o i d disagreement a n d disorder? D o e s it promote c o o p e r a t i o n ? A Stage 4 reasoner m a y t h i n k it does, but this is p o s s i b l e o n l y i f he/she s eri ous l y m i s j u d g e s the s o c i a l situation. Indeed b y 1 9 3 0 ' s most G e r m a n s h a d l i v e d w i t h the J e w s f o r a l o n g t i m e a n d h a d g o o d reasons to b e l i e v e that the J e w s w e r e part o f their society. E v e n i f they d i d not t h i n k so, they m i g h t w e l l t h i n k that the best w a y o f m a i n t a i n i n g their society w a s to l i v e w i t h outsiders p e a c e f u l l y . S i n c e the core m o r a l c o n c e p t i o n o f Stage 4 is that o f maintenance o f the societal system as a w h o l e , it is not to be expected that a Stage 4 reasoner w o u l d endorse e v e r y single n o r m e n f o r c e d b y their society. Y e s , there m i g h t still be a f e w extreme cases i n w h i c h a Stage 4 reasoner w o u l d embrace the anti-Semitic  178  i d e o l o g y , but this s t i l l does not damage K o h l b e r g ' s theory because Stage 4 f o r m a l reasoning does not guarantee correct m o r a l d e c i s i o n s , and K o h l b e r g has never c l a i m e d that it does.  In this s e c t i o n I argued that K o h l b e r g ' s studies p r o v i d e s i g n i f i c a n t support f o r the n o r m a t i v e c l a i m that Stage 6 is o b j e c t i v e l y better than the l o w e r stages. T h i s defense o f K o h l b e r g ' s h i e r a r c h i c a l c l a i m has integrated v a r i o u s considerations related to the d e v e l o p m e n t a l patterns o f K o h l b e r g i a n m o r a l p s y c h o l o g y , i n c l u d i n g (1) h i g h e r m o r a l stages require greater c o g n i t i v e a b i l i t y ; (2) the stages f o r m a u n i v e r s a l invariant sequence; (3) subjects o f l o w e r stages t y p i c a l l y cannot c o m p r e h e n d h i g h e r stages; (4) free d i s c u s s i o n s tend to stimulate stage p r o g r e s s i o n s ; and (5) h i g h e r stages predict better m o r a l behavior.  F o r us this is an important c o n c l u s i o n . O n e o f o u r m a i n purposes is to e x p l o r e the relation between K o h l b e r g ' s theory and ethical u n i v e r s a l i s m . A l t h o u g h up to this p o i n t w e have not e x a m i n e d p r e c i s e l y w h a t k i n d o f u n i v e r s a l i s m h i s theory supports, w e can already refute an extreme f o r m o f r e l a t i v i s m . T h i s extreme r e l a t i v i s m c l a i m s that different m o r a l standards cannot be c o m p a r e d . N o m o r a l standard c a n be said to be o b j e c t i v e l y better than others. I f the arguments above are s o u n d , then surely some m o r a l standards are o b j e c t i v e l y better than others, independently o f any particular culture or v a l u e system. T h i s is an important step towards e s t a b l i s h i n g a c r e d i b l e u n i v e r s a l i s m .  Section 2: Supremacy and universality of Stage 6 W e have j u s t seen h o w e m p i r i c a l studies o f m o r a l d e v e l o p m e n t c o m b i n e d w i t h n o r m s o f rationality p r o v i d e s support f o r the n o r m a t i v e c l a i m that the K o h l b e r g i a n stages  179  are i n c r e a s i n g l y adequate. W e n o w turn to the q u e s t i o n o f whether K o h l b e r g ' s m o d e l represents a c r e d i b l e v e r s i o n o f universalist ethics. U l t i m a t e l y I w o u l d l i k e to suggest that there c a n be different v e r s i o n s o f u n i v e r s a l i s m . B u t first let m e define a strong and perhaps t y p i c a l f o r m o f it, w h a t I w i l l c a l l ' p a r a d i g m u n i v e r s a l i s m . ' A c c o r d i n g to p a r a d i g m u n i v e r s a l i s m there exists a set o f p r i n c i p l e s that are absolute a n d a l l e m b r a c i n g , but yet perfectly determinate w h e n a p p l i e d i n particular situations. K o h l b e r g ' s theory tends to suggest a u n i v e r s a l i s m i n this p a r a d i g m sense. I w i l l e x a m i n e K o h l b e r g ' s universalist c l a i m s i n this section.  In section 1 I argued that e m p i r i c a l f i n d i n g s , together w i t h n o r m s o f rationality, f a v o r the h i e r a r c h i c a l c l a i m i n K o h l b e r g ' s theory. H o w e v e r , the h i e r a r c h i c a l c l a i m alone says n o t h i n g about whether Stage 6 is r e a l l y the best w h e n c o m p a r e d to all p o s s i b l e f o r m s o f m o r a l r e a s o n i n g . S e c t i o n 1 o n l y offers a general s c h e m a for understanding h o w a theory o f m o r a l d e v e l o p m e n t l i k e K o h l b e r g ' s c a n be relevant to m o r a l p h i l o s o p h y and a defense o f K o h l b e r g ' s h i e r a r c h i c a l c l a i m relative to the K o h l b e r g i a n stages. W e do not k n o w i f rational people w o u l d find some other p o s s i b l e f o r m o f m o r a l r e a s o n i n g e v e n m o r e attractive than Stage 6. E v e n granted that K o h l b e r g ' s m o r a l stages are o f i n c r e a s i n g adequacy, w e cannot therefore c l a i m that Stage 6 m o r a l r e a s o n i n g is the best f o r m o f m o r a l reasoning p o s s i b l e . T h e r e f o r e , to further evaluate K o h l b e r g ' s theory, w e need to discuss his c l a i m that Stage 6 m o r a l reasoning is not o n l y o b j e c t i v e l y superior to a l l other stages but is also the g o a l or destiny o f mature m o r a l d e v e l o p m e n t .  2.1 The rarity of postconventional reasoners K o h l b e r g ' s c l a i m that the Stage 6 m o r a l c o n c e p t i o n is o b j e c t i v e l y m o r e adequate than a l l other stages o f his m o d e l is related to u n i v e r s a l i s m i n that the superiority o f Stage  180  6 is independent o f any particular culture or v a l u e system. B e y o n d this, K o h l b e r g also believes Stage 6 to be the supreme g o a l o f m o r a l d e v e l o p m e n t . Y e t , as w e have seen i n C h a p t e r 2, w i t h the r e m o v a l o f Stage 6 f r o m the (1987) s c o r i n g m a n u a l , K o h l b e r g a n d his colleagues c a n n o l o n g e r prove the existence o f Stage 6 reasoners. D o w e , then, still have any reason to b e l i e v e that Stage 6 is r e a l l y the destiny o f m o r a l d e v e l o p m e n t ? T h i s question i s , i n effect, w h a t F l a n a g a n ' s argument 8 is a l l about.  B e f o r e I r e p l y to this q u e s t i o n about the reality o f Stage 6, s o m e t h i n g needs to be said o n b e h a l f o f K o h l b e r g . M o r a l d e v e l o p m e n t a l p s y c h o l o g y i s , after a l l , a y o u n g d i s c i p l i n e . T h e first major w o r k i n the f i e l d w a s P i a g e t ' s  Moral Judgment of the Child, .  p u b l i s h e d i n 1932. M o r e o v e r , research i n m o r a l d e v e l o p m e n t was v i r t u a l l y d i s c o n t i n u e d after this, and d i d not resume u n t i l K o h l b e r g single-handedly r e v i v e d the entire d i s c i p l i n e i n the 1 9 6 0 ' s . T h e r e f o r e it is fair to say that the d i s c i p l i n e is less than 4 0 years o l d . N a t u r a l science, w h i c h is generally regarded as the p a r a d i g m f o r s c i e n t i f i c enterprises, has taken several centuries to d e v e l o p into the integrated f o r m it has today. It w o u l d be unrealistic to expect K o h l b e r g to achieve a n y t h i n g c o m p a r a b l e to this integration a l l b y h i m s e l f and w i t h i n s u c h a short p e r i o d .  Indeed, K o h l b e r g has never pretended that h i s establishment o f a d e v e l o p m e n t a l m o d e l o f m o r a l i t y is a f i n i s h e d project. H i s last (1990) paper is not intended to be the l a u n c h i n g o f h i s theory about the highest stage i n m o r a l d e v e l o p m e n t . Rather, it is an attempt to reintroduce Stage 6 into h i s m o d e l after h a v i n g retreated f r o m c l a i m i n g its actual occurrence f o r a decade. B u t even i n the m i d 80s, w h e n he a d m i t s that there is n o evidence f o r the actual occurrence o f Stage 6 subjects, he keeps m a i n t a i n i n g that w e need  181  to postulate Stage 6 as the end p o i n t o f m o r a l d e v e l o p m e n t a n d as a d e v i c e to define the m o r a l d o m a i n . B e l o w w e s h a l l deal w i t h this c l a i m .  In section 1,1 argued that K o h l b e r g is j u s t i f i e d i n u s i n g e m p i r i c a l data as indicators o f the objective adequacy o f different f o r m s o f m o r a l r e a s o n i n g . T h e m a i n question then is whether the e m p i r i c a l data available so far lead us to K o h l b e r g ' s c o n c l u s i o n about the supreme adequacy o f Stage 6. S t r i c t l y s p e a k i n g , the a n s w e r is no. F o r w e have to note that K o h l b e r g ' s argument is based o n e m p i r i c a l o b s e r v a t i o n : I f a certain f o r m o f m o r a l r e a s o n i n g X is o b j e c t i v e l y better than another f o r m Y , then w e s h o u l d observe that p e o p l e ' s r e a s o n i n g m o v e s f r o m Y to X but not v i c e versa, etc. W i t h the recent re-development o f the m o d e l a n d the research m e t h o d , the existence o f Stage 6 reasoning is n o l o n g e r c l a i m e d to be v e r i f i e d b y such o b s e r v a t i o n , and it is not e v e n i n c l u d e d i n the (1987) s c o r i n g m a n u a l . U p to 1990 o n l y t w o Stage 6 subjects have been i n t e r v i e w e d , and this is an i n s u f f i c i e n t f o u n d a t i o n f o r any general c l a i m about Stage 6 reasoning. D o a l l Stage 6 reasoners d e v e l o p t h r o u g h a l l the l o w e r stages? W i l l Stage 6 reasoners give u p Stage 6 later? D o a l l subjects w h o c a n understand Stage 6 prefer it to Stage 5? D o Stage 6 reasoners behave better than Stage 5 reasoners i n situations i n w h i c h w e k n o w the right t h i n g to do? S i n c e the (1987) s c o r i n g m a n u a l has e l i m i n a t e d Stage 6, w e do not e v e n have a p r o t o c o l f o r detecting Stage 6 m o r a l r e a s o n i n g , and none o f the above questions can be answered o n the basis o f K o h l b e r g ' s later e m p i r i c a l research.  Nevertheless one m a y p a r t i a l l y defend K o h l b e r g ' s c l a i m about Stage 6 as f o l l o w s . It is not s u r p r i s i n g that Stage 6 reasoners are so rare. A f t e r a l l , i f K o h l b e r g ' s theory is correct, Stage 6 reasoners are real experts i n m o r a l reasoning. U n l i k e experts i n other fields l i k e c o m p u t e r p r o g r a m m i n g , w e do not have any systematic t r a i n i n g f o r m o r a l  182  reasoning.  15  O n e w i l l be surprised to f i n d a p r o g r a m m e r w h o c a n d e s i g n a software as  sophisticated as W i n d o w s 95 w i t h no f o r m a l t r a i n i n g b e h i n d h i m at a l l . A c c o r d i n g to K o h l b e r g ' s theory, c o g n i t i v e a b i l i t y and sensitivity to d i f f e r i n g i n d i v i d u a l and s o c i a l perspectives are t w o prerequisites o f m o r a l development. It is o n l y to be expected then that Stage 6 subjects w i l l be quite rare. C o n s e q u e n t l y , the s m a l l n u m b e r o f Stage 6 subjects f o u n d does not necessarily m e a n that there is any serious p r o b l e m i n K o h l b e r g ' s theory.  M o r e o v e r , as a result o f r e v i s i n g the s c o r i n g m a n u a l , there o c c u r r e d a d o w n w a r d shift o f the scores f o r the h i g h e r stages. In particular, reasoning w h i c h w a s earlier categorized as Stage 6 w a s m o s t l y categorized as Stage 5 i n the (1987) s c o r i n g m a n u a l . H o w e v e r , l o o k i n g b a c k to the earlier studies, s i g n i f i c a n t d i f f e r e n c e s b e t w e e n the (  performance o f Stage 5 subjects and Stage 6 subjects can be noted. Stage 6 subjects s h o w e d clear superiority i n their m o r a l b e h a v i o r i n the M i l g r a m situation and the Free S p e e c h M o v e m e n t study. T h i s i s a g o o d reason to conjecture that K o h l b e r g h a d done m o r e than he needed w h e n he c o l l a p s e d Stage 6 into Stage 5 i n the late m o d e l . Perhaps there w e r e w a y s b y w h i c h K o h l b e r g c o u l d have kept Stage 6 i n the s c o r i n g m a n u a l w h i l e o v e r c o m i n g the a n o m a l i e s he encountered.  In the a b o v e d i s c u s s i o n I attempted to e x p l a i n the rarity o f p o s t c o n v e n t i o n a l reasoners i n terms o f c o g n i t i v e demands and the absence o f t r a i n i n g i n s o c i a l understanding t h r o u g h a p o s t c o n v e n t i o n a l perspective. I have also suggested that the rarity o f Stage 6 subjects m a y be exaggerated b y the fact that Stage 6 appears to be  15 It is interesting to note that, according to the early writings of Kohlberg, all philosophers he has interviewed are postconventional reasoners.  183  c o l l a p s e d unnecessarily into Stage 5 i n the (1987) s c o r i n g m a n u a l . Researchers i n m o r a l p s y c h o l o g y have recently p r o p o s e d other h i g h l y p l a u s i b l e explanations. A n e x a m p l e i s R i c h a r d S h w e d e r , w h o suggests that, i n K o h l b e r g ' s m o r a l j u d g m e n t i n t e r v i e w s , subjects are r e q u i r e d to "generate arguments, v e r b a l l y represent c o m p l e x concepts, a n d talk l i k e a m o r a l p h i l o s o p h e r . " S i n c e " k n o w l e d g e o f concepts often precedes their self-reflective representation i n s p e e c h , " s u c h a research m e t h o d tends to underscore the r e a s o n i n g o f the s u b j e c t s .  16  M o r e o v e r , w h e n w e l o o k c l o s e l y at K o h l b e r g ' s (1987) s c o r i n g m a n u a l , w e cannot h e l p suspecting that the m a n u a l is not an accurate representation o f h i s d e f i n i t i o n s o f the m o r a l stages. F o r e x a m p l e , i n response to H e i n z ' s d i l e m m a , a subject is scored as a Stage 4 reasoner i f he/she says H e i n z s h o u l d steal the d r u g because human life is m o r e important than property rights, or because h u m a n l i f e has i n t r i n s i c value ( C o l b y et a l . 1987 v o l . 2, p. 9). H o w e v e r , one is r a n k e d as Stage 5 i f one answers that the right to l i f e transcends the right to property ( C o l b y et a l . 1987 v o l . 2, p. 11). T h e difference i n these t w o answers i s surely arbitrary, as it turns m e r e l y o n the fact that the latter uses the t e r m 'right to l i f e , ' w h i l e the f o r m e r uses instead the t e r m ' v a l u e o f l i f e . ' E x a m p l e s o f this k i n d tend to s h o w that S h w e d e r is right i n s a y i n g that K o h l b e r g has r e q u i r e d the subjects to talk l i k e p h i l o s o p h e r s . I w o u l d e v e n say that K o h l b e r g has r e q u i r e d h i s subjects to talk l i k e p h i l o s o p h e r s who share his terminology.  I f a p h i l o s o p h e r does not t h i n k there i s  m u c h difference, i n the relevant context, between s a y i n g "the v a l u e o f l i f e " a n d "the right  Richard A. Shweder et al, "Culture and Moral Development." In The Emergence of Morality in Young Children, ed. Jerome Kagan and Sharon Lamb (Chicago and London: University of Chicago Press, 1987), 16. 16  184  to l i f e , " he/she m a y be scored as Stage 4 because he chooses to use the f o r m e r expression.  T h e i m p l i c a t i o n is that the (1987) s c o r i n g m a n u a l m a y have d o w n s c o r e d s o m e p o s t c o n v e n t i o n a l subjects. I f this is true, then w h a t is partly responsible f o r the rarity o f p o s t c o n v e n t i o n a l subjects is the s c o r i n g m a n u a l instead o f the concepts o f the stages themselves. O n e p o s s i b l e w a y to v e r i f y this c l a i m is to d e v e l o p m o d i f i e d v e r s i o n s o f the s c o r i n g m a n u a l , and test whether there i s any v e r s i o n w h i c h (1) f a i t h f u l l y reflects the m o r a l c o n c e p t i o n o f e a c h stage a n d , (2) produces the same pattern o f e m p i r i c a l data, except (3) it detects m o r e cases o f p o s t c o n v e n t i o n a l subjects. I f these results w e r e obtained under one or m o r e m o d i f i e d v e r s i o n o f the s c o r i n g m a n u a l , then the r e s p o n s i b i l i t y f o r the present appearance o f scarcity o f Stage 6 w o u l d go to the (1987) m a n u a l . O t h e r w i s e , one w i l l have reason to b e l i e v e that there are some i n t r i n s i c p r o b l e m s w i t h the m o r a l stage m o d e l or that Stage 6 is a k i n d o f i d e a l f o r m o f r e a s o n i n g that is rarely a c h i e v e d i n practice. S o r t i n g a l l this out is a task that needs further effort o f professional p s y c h o l o g i s t s .  2.2 A philosophical argument in favor of Kohlberg's supremacy claim W h a t w e n o w need, h o w e v e r , is p o s i t i v e support f o r the c l a i m o f the supreme adequacy for Stage 6. I n h i s latest f o r m u l a t i o n , K o h l b e r g defines Stage 6 r e a s o n i n g i n terms o f active s y m p a t h y f o r others, i d e a l role-taking, and u n i v e r s a l i z a b i l i t y ( K o h l b e r g et a l . 1990, p. 153). These f o r m a l properties are i n t u i t i v e l y attractive as the characteristics o f correct m o r a l r e a s o n i n g .  185  M o r e o v e r , the p o i n t o f postulating Stage 6, K o h l b e r g says, is that the f o r m a l characteristics i n Stage 6 are the f o r m a l characteristics that m a k e each stage better than its p r e v i o u s one ( K o h l b e r g et a l . 1990, p. 152). L e t us first c o n s i d e r u n i v e r s a l i z a b i l i t y . F r o m Stage 1 to Stage 5, there is an o b v i o u s e x p a n s i o n o f the r e a l m o f m o r a l c o n c e r n . A t the p r e c o n v e n t i o n a l l e v e l , the s e l f is v e r y m u c h the o n l y m o r a l c o n c e r n o f the subject. A t the c o n v e n t i o n a l l e v e l , c o n c e r n is m a i n l y directed to people i n the agent's o w n group. T h e c o n v e n t i o n a l subject can care about p e o p l e outside his/her g r o u p , but he/she is unable to c o n s i d e r m o r a l p r o b l e m s i n terms o f values that transcend his/her o w n s o c i a l setting, and therefore the u n i v e r s a l i z a b i l i t y o f his/her m o r a l r e a s o n i n g is still l i m i t e d . A t the p o s t c o n v e n t i o n a l l e v e l , the subject f i n a l l y acquires the a b i l i t y to reason i n terms o f values independent of, and therefore not l i m i t e d b y , the s o c i a l setting a n d the s o c