Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Moral climate and the development of moral reasoning: the effects of dyadic discussions between young… Taylor, John Harrison 1995

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

Item Metadata


831-ubc_1995-983566.pdf [ 3.42MB ]
JSON: 831-1.0088850.json
JSON-LD: 831-1.0088850-ld.json
RDF/XML (Pretty): 831-1.0088850-rdf.xml
RDF/JSON: 831-1.0088850-rdf.json
Turtle: 831-1.0088850-turtle.txt
N-Triples: 831-1.0088850-rdf-ntriples.txt
Original Record: 831-1.0088850-source.json
Full Text

Full Text

MORAL CLIMATE AND THE DEVELOPMENT OF MORAL REASONING:  THE EFFECTS OF DYADIC DISCUSSIONS BETWEEN YOUNG OFFENDERS by John Harrison Taylor B.A., The University of British Columbia, 1969 M.Ed., The University of Alberta, 1982  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF ThE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in  THE FACULTY OF GRADUATE STUDIES Department of Psychology  UNIVERSITY OF BRITISH COLUMBIA March 1995 © John Harrison Taylor, 1995  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.  (Signature)  Department of  PQyn11 ngy  The University of British Columbia Vancouver, Canada Date -  DE-6 (2J88)  /  11  ABSTRACT  Cognitive-developmental  theory  claims  that  moral  reasoning  ordinarily  progresses through distinct stages, and that such development can be stimulated by discussion with others, especially discussions involving exposure to higher-stage reasoning. The concern of this study was the social/contextual factors that interact with cognitive processes involved in the development of moral reasoning. Two types of such factors were studied:  namely, sociometric status and intensity of moral education  program. The first of these could be studied because the participants were residents of a facility for young offenders (a total institution), characterized by an obvious and rigid hierarchical peer status system within the culture. The second factor could be studied because the participants were drawn from three residential units within the larger center, which varied significantly in terms of their program activities (specifically, unit meetings), and hence their moral climates. A total of 101 young offenders served as participants. They were assessed for moral reasoning, their perceptions of moral and institutional climate, and also through behavioral ratings  -  all at the pretest and at the 1-month posttest. The three levels of  program were reflected in the institutional and moral climate measures. As well, better climates were associated with improvements in behavior and lesser climates with reductions in prosocial behavior. It was concluded that moral climate represents a valid measure of the factors which predict behavior within and following release from institutional settings. In order to study the effects of peer status, 40 participants served as target subjects who engaged in moral dilemma discussions with one other subject, each day for 3 consecutive days. According to cognitive-developmental theory, a dyadic intervention such as the one used here would be expected to stimulate the moral reasoning competence of the participant who is lower in that ability. However, the dyads were  U’  formed in such a way that some of the high stage participants (who would be expected to have an influence on their partner) were of significantly lower peer status. It was found that both exposure to higher-stage reasoning and higher peer status were necessary but not sufficient elements within this developmental process, consistent with the Piagetian notions regarding peer interaction and disequilibration.  iv  TABLE OF CONTENTS ABSTRACT  .  •.  .  ii  LIST OF TABLES  vi  LIST OF FIGURES  vii  ACKNOWLEDGEMENTS  viii  INTRODUCTION Review of Relevant Research Piaget’s Theory Kohlberg’s Moral Stage Theory Levels of Reasoning Stages of Reasoning The Measurement of Moral Reasoning The Validity of the Moral Stage Model Moral Reasoning and Delinquency Developing the Moral Reasoning of Offenders Moral Climate Moral Climate Methodology Institutional Climate Peer Status Moral Reasoning and Behavior Summary and Hypotheses  1 4 6 8 11 12 14 16 19 23 27 30 35 38 42 44  METHOD Sample Procedure Intensity of Moral Education Program Peer Status Correctional Institutions Environment Scale Moral Reasoning Moral Climate Intervention Behavioral Measures Summary  46 46 47 49 50 52 52 54 55 57 58  RESULTS Moral Reasoning: Pretest Measures Climate Institutional Climate Moral Climate Stage of Norm Stage of Community Moral Climate and Improvement in Moral Reasoning Moral Climate and Improvement in Behavior Moral Climate and Moos CIES Effects of Peer Discussion on Moral Reasoning Moral Reasoning Across Content and Context Hypothetical versus Real-Life Dilemmas  60 60 61 61 62 62 64 66 67 68 70 73 73  V  Judgments of Responsibility and Practicality Moral Reasoning and Moral Intentions Moral Reasoning and Behavior  75 78 79  DISCUSSION Moral Judgment, Moral Climate, and Behavior Real-life/Hypothetical Issues and the Concept of Structure Moral and Nonmoral Issues Equilibration Theory Implications for Educational and Correctional Programs  82 82 85 90 91 94  REFERENCES  99  APPENDIX A: Moral Judgment Interview  111  APPENDIX B: Moral Climate Interview  114  APPENDIX C: Scoring Criteria for Stage of Community  118  APPENDIX D: Correctional Institutions Environment Scale: Subscale Descriptions  119  APPENDIX E: Behavior Frequency Rating  120  APPENDIX F: Dilemmas Used for Dyadic Discussions  122  APPENDIX G: Sample Transcript of Dyadic Discussion  128  vi  LIST OF TABLES Table 1: Hypothesized Effects of Dyadic Intervention  48  Table 2: Synopsis of Procedure  51  Table 3: CIES Subscales across the Three Programs (Collapsed over Time)  62  Table 4: Stage of the Norm Scores across Three Programs (Collapsed over Time).  64  Table 5: Correlations between Moral Climate (Stage of Community) and dES Subscales (Posttest)  69  Table 6: Effects of Higher Status and Higher Stage Combined  71  Table 7: Effects of Dyadic Intervention  72  Table 8: Moral Reasoning in Hypothetical and Real-Life Dilemmas (Collapsed over Time)  74  Table 9: Judgments of Responsibility, Practicality, Others, and the Ideal (Collapsed over Time)  76  Table 10: Relations between Moral Intentions and Moral Reasoning (Expressed in WASs) Table 11: Correlations between Moral Reasoning and CBCL Subscales  79 81  VII  LIST OF FIGURES Figure 1: Direct and Indirect Factors in the Development of Moral Reasoning Figure 2: Stage of Community Scores (for Three Norms) within Three Programs.  4 66  Figure 3: Accommodation in Reasoning about Helping by Judgment Type and Program  77  vi”  ACKNOWLEDGEMENTS This research project was not completed easily or in a short period of time. It is my only hope that it may prove to be valuable. I am indebted to many people, including Maria Calderwood and Heather McVicar who conducted most of the interviews. Shiela Longpre also volunteered many hours of her time to assist with interviewing.  I am  deeply indebted to Beryl Andison who gave freely of her time to transcribe virtually hundreds of interviews; without her help and her positive attitude this project would have faltered. I would also like to thank my wife Carol, who was so supportive and patient around my “absence,” and my 5-year-old son Jordan, who has been so effective in cheering me up. This project could not have been possible without a 2-year fellowship from the Alberta Heritage Scholarship Fund. I especially wish to honor and acknowledge Lawrence Kohlberg who, despite illness, travelled with Ann Higgins from Boston to Alberta (the site of this research) in 1986, to consult to several correctional programs, and who encouraged me to attempt to make correctional facilities and schools better developmental contexts. Finally, Larry Walker, my advisor, teacher and friend, is a calm and positive supporter who has guided me with remarkable patience and understanding through the pastS years. He may not know the many ways in which he has taught me.  1.  INTRODUCTION In what has been considered a classic review within the field of criminology, Martinson (1974) summarized the outcome research concerned with programs for adult and juvenile offenders and concluded that “nothing works.”  Since then, Martinson’s  original conclusion has been substantiated by a number of researchers (Hackler, 1978; Whitehead & Lab, 1989), even though he (Martinson, 1979) later revised his position. His original discouragement has been reinforced and this trend had significant impact on government fiscal policy concerning offender treatment programs. However, some other researchers in the field (Arbuthnot, Gordon, & Jurkovic, 1987; Cüllen & Gendreau, 1989; Gendreau & Ross, 1980; Ross & Fabiano, 1981; Ross & Gendreau, 1984) have argued that most research reviews have failed to highlight and explore the few programs that did indicate positive results. In addition, they suggest that the ineffective programs involved only one type of treatment intervention, that the research involved too few outcome variables, that there was little effort to match types of programs with types of offenders, and that the programs did not integrate community resources with the treatment intervention (Gendreau & Ross, 1980; Gottfredson, 1979). Ross argues that not only are there some programs which have been found to be effective (e.g., Arbuthnot & Gordon, 1986; Chandler, 1973; Chandler, Greenspan, & Barenboim, 1974), but that these programs have factors in common. He proposes that the effective programs are those that understand the offender as evidencing developmental problems, particularly in the domain of cognition and social and moral reasoning (Gibbs, 1991; Izzo & Ross, 1990). At least one common factor has been recognized (among effective conectional programs): cognitive development. A component analysis revealed that effective and ineffective programs differed significantly in terms of the presence of techniques which would foster the development of the offender’s interpersonal problem-solving skills, teach him to consider the consequences of his behavior before he acts, and enhance his understanding of other people’s behavior and feelings. (Ross & Gendreau, 1984, p. 20) The studies that Ross cites include such interventions as problem-solving and  2.  interpersonal skills training, interpersonal cognitive problem-solving skills, interpersonal negotiation skills, role-playing (Arbutlmot, 1975), perspective-taking (Chandler, 1973), and reasoning skill training.  Given that Kohlberg (Colby & Kohlberg, 1987a) and  several others (Walker, 1980) have described moral reasoning as essentially a role-taking or perspective-taking competence, Ross might well have included studies demonstrating the effectiveness of moral reasoning with offenders (Arbuthnot, Gordon, & Jurkovic, 1987; Gibbs, Arnold, Ahlborn, & Cheesman, 1984; Rosenkotter, Landman, & Mazak, 1980) in his list.  The interventions described by Ross are best classified as  educational/developmental, and are based on the view of the offender as evidencing (at least in part) developmental delays in social/moral/cognitive functioning.  They are  distinguished from treatment-oriented approaches and represent a departure from the more traditional medical model (referred to as “the ‘sick’ premise” by Ekstedt & Griffiths, 1988, p. 204), which would tend to view offenders as suffering from an illness or a condition. All of this is an introduction to the argument that (a) the field of juvenile corrections is a valid arena for developmental research, and (b) that moral reasoning, at whose core is the perspective taking (Selman, 1980; Walker, 1980) discussed by Ross, represents a description of the way in which offenders construct relations with others and solve problems (Arbuthnot, 1984; Nelson, Smith, & Dodd, 1990).  It is this cognitive  dimension which underlies many, if not all, of the programs highlighted by Ross and others. There is strong evidence that offenders use consistent and distinctive cognitive structures in approaching socio-moral problems (Blasi, 1980; Buchanan, 1992; Jurkovic, 1980; Nelson, Smith, & Dodd, 1990; Samenow, 1984).  It is argued here that these  structures represent a useful and comprehensive method of describing offenders, in contrast to explanations invoking social disorganization, personality factors, subcultural socialization, neglect or abuse, etc. Certainly, histories of abuse and/or deviant socialization help to explain the etiology of delinquent behavior; however (etiology  3.  aside), identifying a developmental delay is to specify the nature of a problem, not its possible cause. Understanding the nature of delays in the area of socio-moral reasoning lead one to specific remedial interventions, with some reason for optimism (Arbuthnot, Gordon, & Jurkovic, 1987; Gibbs, Arnold, Ahlborn, & Cheesman, 1984; Rosenkotter, Landman, & Mazak, 1980). This dissertation was based on a central developmental concept, the Piagetian notions of equilibration and disequilibration and, in particular, the application of such notions to the development of socio-moral reasoning as outlined by Kohlberg (1969; Blatt & Kohlberg, 1975; Colby & Kohlberg, 1987a). Disequilibration is usually named as the result of exposure to, or interaction with, or conflict with higher or better forms or stages of reasoning (a direct factor in Figure 1). However, this research was concerned with the real-life social context that can either enhance or constrain such developmental processes. Even though Piaget (Chapman, 1988) acknowledged that social factors are an integral part of the equilibration process, there has been little research investigating the effects of social organization on this developmental process  -  especially those  social/contextual factors tied to the organization of adolescent peer groups. Two such factors were studied. The first of these had to do with the rated peer status (or what shall later be termed as credibility), termed a direct factor in Figure 1. It was hypothesized that exposure to a higher stage or form of reasoning would only be possible if the person representing such reasoning was of sufficient status or credibility  -  a claim that Piaget made in 1932 but which has not been adequately demonstrated. The second social factor rests on the claim that primary groups can be described and analyzed in terms of their moral climate (an indirect factor in Figure 1). It was hypothesized that groups higher in moral climate would be more supportive of developmental changes than those lower in moral climate. It was because this study drew its participants from a large residential center for young offenders, with its cliques and subcultures, that an excellent opportunity to study  4.  Figure 1 Direct and Indirect Factors in the Development ofMoral Reasoning  these social/contextual factors was available.  However these factors were studied  because it was felt that these are the dimensions most prominent and influential for most adolescents; that is, the way others in one’s reference group report their reasoning, and the status of the person(s) to whom one is listening. Review of Relevant Research In a comprehensive review of the research dealing with the relationship between moral reasoning and moral action, Blasi (1980) claimed that there is one essential distinction within all of the approaches to research on moral behavior, one which divides them into two theoretical positions. The first of these positions (which I will call the personological approach) describes moral behavior as the outcome of a complex interplay of many factors, some motivating, some inhibiting, some contextual, and some  5.  resulting from the individual’s perception of the situation. Such factors include concepts as diverse as guilt, superego, and reinforcement history, but within this approach, cognition is, at best, a process which serves to provide rationalization of behavior. The basic assumption of this approach is that the organism is essentially passive and reactive to forces external to it. The second approach (cognitive-developmental theory) assumes that cognitive processes are a part of a natural tendency of the individual to make sense of and to operate actively upon the world which he or she encounters. Whereas the personological approach would tend to view the individual’s behavior as largely the result of, or shaped  by, internal and external factors, the second approach sees behavior (and cognition) as part of a constructive process, which is to say that the individual actively transforms or operates on his or her environment and experience. However, this is not to say that the individual is independent of the environment. Indeed, Piagetian theory has frequently been misunderstood in this regard (e.g., Donovan & McIntyre, 1990, p. 4). For Piaget, the organism is constantly “in relation” (Youniss, 1980), and is not a solitary learner (Chapman, 1986). The kinds of experiences and stimulation available to the individual facilitate, enhance, and limit the rate (and perhaps even the direction of development, cf. Sanvitale, Saltzstein, & Fish, 1989); but the meaning of these experiences is determined by the individual, and cognitive processes provide this meaning. Some theories within the personological approach propose guilt, for example, as the motivating affective state, or a condition of anxiety related to tendencies of self-punishment, and the result of socialization. Alternatively, the cognitive-developmental approach suggests that guilt is “conscious cognitive judgment” (Kohlberg, 1984, p. 34). It claims that “cognitive and moral judgments and rational argumentation are central to moral psychology” (Colby & Kohlberg, 1987a, p. 1) and are meaningful in their own right, reflecting the individual’s view of the world, rather than reflections of other, less rational processes. It is the cognitive-developmental approach which forms the theoretical foundation  6.  of this research and serves as a theoretical base for more specific hypotheses. The theory (Colby & Kohlberg, 1987a; Kohlberg, 1969, 1981, 1984) proposes a sequence of qualitative changes (stages) in moral reasoning. It does not propose to fully explain individual differences in moral behavior (such as resistance to temptation, or honesty), nor does it attempt to predict moral traits (e.g., guilt), but it does argue that there is a relation between moral cognition and moral action (Buchanan, 1992).. It is based on the work of Piaget (1974/1980a) who, in turn, grounded his notions on biological systems and the way in which they adapt (Chapman, 1988). Piaget’s Theory Perhaps the most distinctive feature of PiageCs theory of cognition has to do with the active role ascribed to the developing individual. Piaget held that in adapting to the environment, organisms actively respond to environmental constraints, conditions, or intrusions” (Piaget, 1974/1980a, p. 101). 11  As noted earlier, such a position was in  contradiction to the current psychodynamic and behavioristic theories of psychology, which prescribed a more passive role to the organism. It was also in contradiction to contemporary Darwinian biology, which invoked the processes of natural selection and mutation (Chapman, 1988, p. 273). According to the latter view, these complementary processes explain evolutionary development by means of random genetic variations (mutation) which are then selected by virtue of their viability in the particular environment. Within this model, the environment provides the conditions for adaptation. Piaget, in contrast, proposed that the active adaptation of an organism (phenotype) precedes genetic mutation and can contribute to it (Chapman, 1988, p. 274). The key distinction between the two positions has to do with the active role ascribed to the organism in Piaget’s theory. Piaget applied the same model to the cognitive realm. In the same way that new genotypes are formed as a partial result of the action of the organism, the individual  7.  child’s action on the external physical world is said to ultimately lead to the construction of internal cognitive structures representing that world (Chapman, 1988). The child’s understanding of his or her world depends, then, upon his or her interaction with and action upon the environment. Thus, it does not depend entirely upon the contingent reinforcement which trial and error behavior produces, or the socialization to which the individual has been subjected. Piagetian theory proposes that, throughout the developmental process, mental structures are progressively reorganized into better, more complex structures. Cognitive development can be described as a sequence of stages, with each stage representing a better, more complex organization of mental structures. There is a natural tendency towards equilibrium, and developmental change (or reorganization) occurs through the process of disequilibration, which is facilitated when the individual’s experience conflicts with his or her current cognitive structures.  Therefore, the concepts of conflict, or  cognitive conflict, or opposition, or contradiction (Piaget, 197411980b) are essential to this theory. Optimizing equilibration (an improved level of equilibrium; Chapman, 1988, p. 280) involves reorganization into both new and more coherent ways of thinking, and ways of interacting with the environment (Turiel, 1974, p. 15). Hence, this model of development is not simply an additive “layer-cake” one (Turiel, 1974, p. 16) in which competencies are simply added to those acquired earlier.  However, with each  progression each stage subsumes and elaborates the prior stage so they still exist within the new one. One of the empirical problems faced by this theory has to do with demonstrating that people actually experience disequilibration or cognitive conflict (Kupfersmid & Wonderly, 1982). Murray (1982, 1983) noted that the many studies which demonstrated the effects of cognitive conflict proved only that there is conflict in the design of the experiment not necessarily in the minds or experience of the children who served as subjects. (The same definitional problem, it should be noted, is faced by reinforcement  8.  theories.) Most of the research in this area has structured some form of conflict for the subjects, for example, writing an essay containing conflicting opinions on a topic (de Vries & Walker, 1986), disagreement with a peer (Ames & Murray, 1982), exposure to higher stage reasoning (Walker, 1983).  Murray (1983) claimed that it is not known  which aspect of these potentially conflictual events actually represented or stimulated cognitive conflict in individual subjects. Neither Piaget (nor Kohlberg) suggests that equilibration theory is the antithesis to social-learning theory, but rather a complement to it (see McCann & Prentice, 1981). Rather, Piaget proposes four factors in development (Chapman, 1988, p. 281): maturation, physical experience, and social influence, in addition to equilibration. Equilibration is important because it represents an explanation of why children progress from less adequate to more adequate ways of knowing (e.g., from nonconserving to conserving) and not the other way round, and why children can, with certainty, look back on their previous ways of understanding and see them as “silly,” lacking in sensibility. Social-learning theory, on the other hand, would hypothesize that behavioral changes progress only in the direction of the model. It is conceivable, according to the sociallearning approach, that, given the right model and contingencies, a child could move, for example, from conservation to nonconservation, a phenomenon which has not been observed. Kohlberg’s Moral Stage Theory Piaget used these basic notions to explain development in many domains including, of course, moral reasoning (Piaget, 1932/1965).  Kohlberg elaborated his  ideas. One of his accomplishments was to bring the discussion of moral development out of the purview of religion and philosophy and make it a legitimate focus for psychology. However, in doing so he found himself swimming against the tide of opinion that “morality was simply a matter of attitudes, customs, norms, and values that were  9.  culturally relativ&’ (DeVries, 1991, P. 7). Kohlberg claimed that we cannot determine whether an action is moral without knowing the private moral judgment of the actor. He argued that morality should be understood from the actor’s perspective, not compared to an external standard.  For  example, striking a person or disobeying a military order can be quite different moral acts, depending upon the reasons of the actor. One can assault another for self gain, or to prevent injury to another. One could disobey an order out of fear of harm to the self, or because the order was thought to be contrary to one’s principles. Thus, Kohlberg’s claim 1 is that moral judgment or reasoning is central to the study of morality. Kohlberg proposed six hierarchical stages of cognitive-structural ability, which correspond to, and involve moral reasoning structures which are isomorphic to Piaget’s stages of cognitive development. Indeed, Walker and Richards (1979) have provided strong support for the thesis that stages of cognitive development are prerequisite for development of moral reasoning. Kohlberg’s theory of moral stages departs from Piaget’s moral theory, however. Piaget (1932/1965) proposed two stages, or ideal types, of moral development: heteronomous (unilateral respect of external authority) and autonomous reasoning (mutual respect among peers and equals).  It is sometimes thought that  Kohlberg simply added the remaining stages, attempting to describe the structures of moral reasoning beyond childhood, through adolescence into adulthood.  In fact,  Kohlberg found that Piaget’s stages did not cluster together the way he felt a true “hard” stage model requires (Kohlberg, 1984). Rather, he came to believe that the Piagetian types did not fully separate content (i.e., the moral decision made) from form or structure (i.e., the reasons for the decision), nor by his own admission, did his own earlier system (Colby & Kohlberg, 1987a). Only after several revisions of his scoring system, could It should be emphasized however, that Kohlberg would admit that reasons or reasoning can be produced after the fact, as rationalizations, but that “at least sometimes, the truth related explanations that one gives for one’s actions are genuine motives for doing them” (Blasi, 1983, p. 185). 1  10.  Kohlberg make the claim that he was measuring the structure of reasoning separate from the content. Thus, it is clear that the notion of structure is essential in Kohlberg’s theory. By structure we mean general organizing principles or patterns of thought rather than specific moral beliefs or opinions. That is, we assume that concepts are not learned or used independently of one another but rather are bound together by common structural features. (Colby & Kohlberg, 1987a, p. 2) Content refers to the specific values or opinions held by an individual. Values can be espoused or held for a variety of reasons, but it is the reasons for taking a certain value position that are indicative of the structure of reasoning. The emphasis is on structure because it is structure that exhibits developmental regularity and is predictable within individuals and across groups of similar ages. Before proceeding further to describe Kohlberg’s stages in more detail, it is first necessary to unfold the philosophical and ethical basis of his theory; for it is one thing to propose a developmental course for moral reasoning, but a much bolder claim to say that ethical principles are the end point of this developmental course for all individuals of all religious views, or in all cultures. Even though (as noted above) Kohlberg brought the study of morality into psychology, he rejected the idea that moral development can be dealt with completely from a psychological position.  He argued that one cannot  undertake a theory of moral reasoning without bridging psychology and moral philosophy. It is the difficulty in making this bridge that has led some researchers to abandon the field altogether, with the conclusion that it belongs solely in the philosophical or religious domain. Kohlberg (1981, pp. 104-105) argued that one of his essential tasks was to justify his claim that higher stages are more adequate stages. He rejected the notion that moral values can be dealt with from the position of cultural or ethical relativism, a position which claims that there can be no a priori logic or standard justifying any value or  11.  principle as better than any other (Rorte, 1994).2 That is, he argued that it is not correct to say that because people do have different value orientations, all people should have their own moral values. He found that persons in different cultures progress through the same stages, in the same order (Snarey, 1985). Even though cultures vary in terms of specific moral values and level of moral development, most cultures (except the most tribal ones) have some individuals at the highest stages. Kohlberg stressed that each stage represents a distinct viewpoint, or socio-moral perspective, from which moral judgments are made.  Selman (1980), a student of  Kohlberg’s, has separated out the stages of perspective-taking that underlay Kohlberg’s stages. Again, Walker (1980) has demonstrated that stages of social perspective-taking ability (Selman, 1980), like cognitive stages, are prerequisite for progression in moral reasoning. Levels of Reasoning Kohlberg’s stages represent the movement from exclusive consideration of the concrete and physical (Stage 1) to the use of abstract concepts and, ultimately, highly abstract moral principles (Stages 5 & 6).  They are grouped into three levels of  development, the preconventional, conventional, and postconventional; and such levels are useful in explaining Kohlberg’s theory.  What he calls the preconventional level  (subdivided into Stages 1 and 2) of moral development is characterized by reasoning which ‘has not yet come to really understand and uphold socially shared moral norms and expectations” (Colby & Kohlberg, 1987a, p. 16). It is reasoning which is dependent upon the physical and concrete consequences of actions, and therefore tends to refer only to expectations which are external to the individual, and it is this level of reasoning  2  Although not strictly a relativist, as a representative of an alternate position, Rorte (1994) refutes the idea that there are a priori principles apart from historical and familial context, and particularly that some principles are better than others. As a philosopher, he seems to argue for a psychological basis for ethics.  12.  which charaóterizes most children under the age of 9, and many adolescent and adult criminal offenders. The conventional level represents internalized expectations of others. Reasoning at this level involves reference to the importance of trying to do what is good or of not being selfish, obeying laws so that order is maintained, and it is this level of reasoning which has been described as a “cognitive buffer” (Gibbs, Arnold, Ahibom, & Cheesman, 1984, p. 37) to illegal behavior. The idea that Stage 3 reasoning represents a “cognitive buffer” to illegal behavior is an important one, and will be discussed more fully, later in this paper. The connection forms one of the strongest arguments for a meaningful relationship between moral reasoning and moral behavior.  The  postconventional level represents reasoning based on principles, such as the utilitarian principle or the principle of contract, and although in general, such reasoning is in accord with the conventions of society, it may not always be so. The principles referred to here are those on which the individual feels a good society should be based, hence they are what Kohlberg calls “prior to society” (Colby & Kohlberg, 1987a, p. 17). A person reasoning at this level would, for example, argue for the necessity of defying a law, or a command of a superior military officer, in order to act consistently with a higher principle, for example, the intrinsic value of human life. Stages of Reasoning When asked if it is wrong to steal, a child at the first stage of development of moral reasoning is likely to focus on the power of a person in authority (“It’s wrong to steal because you don’t have permission”), or on the inevitability of adverse consequences, for example, being taken to jail. At this stage (which Kohlberg has called Heteronomous Morality), the child makes no distinction between his or her own perspective and that of others. As development proceeds, the focus of the child becomes less concrete, less egocentric, and more abstract. At the second stage (Individualistic, Instrumental Morality), egocentrism is  13.  diminished. The child recognizes that others have their own needs and wants, but cannot simultaneously coordinate another’s perspective and his or her own. Therefore, he or she will reason in terms of the concrete consequences for the self, but with the knowledge that the behavior and needs of others must be somehow considered or handled. Unlike Stage 1 reasoning, which is characterized by the inevitability of consequences, Stage 2 reasoning involves the calculation of the possibility of consequences. Stealing is “not worth the risk.”  The individual understands relationships in reciprocal terms.  Caricatured reasoning at this stage is: “You scratch my back and I’ll scratch yours.” Some individuals at this stage would argue that it is acceptable to steal if the victim “deserved” to be stolen from, or was “careless enough to leave money where it could be stolen.” At the third stage (Interpersonally Normative Morality) the ability to deal with abstractions has begun to develop (as found in the Beginning Formal Operational Stage, Walker, 1980). Most importantly, the child can simultaneously hold two perspectives, and therefore is not only aware of his or her needs, but also can consider and operate upon the expectations of others.  A person at this stage can take the third person  perspective, seeing what group membership requires; he or she has a loyalty to the norms of the group to which he or she belongs. Hence, a child could reason that it is wrong to steal because others (parents, friends) would disapprove or, alternately (and very rarely), that it is acceptable to steal because theft is valued by members of the reference group (or gang) to which he or she is loyal. It is Stage 3 reasoning which has described as the “cognitive buffer” to illegal behavior. There are two reasons why Stage 3 might be considered a baseline for mature moral judgment. First, most eighteen-year-olds reason at either Stage 3 or at the transition between Stages 3 and 4. Few are found either at pure Stage 2 or at pure Stage 4. Second, the development from Stage 2 to Stage 3 is perhaps the single most important transition in the entire sequence of moral stages. (Power, Higgins, & Kohlberg, 1989, p. 29) At the fourth stage (Social System Morality), the individual is even better able to  14.  think abstractly and can construe choices in terms of implications for society in general. Realizing that many people can have many different perspectives and needs, and that the larger group (society) can also be represented as “having needs,” a Stage 4 reasoner would make reference to options such as, “the law is made by the majority of people and you have to consider what’s good for the majority.” Hence, this stage is often referred to as the law-and-order stage. If Stage 4 reasoning is “society maintaining,” then Stage 5 is “society creating” (Kohlberg, 1984, P. 634).  This reasoning (called Human Rights and Social Welfare  Morality) represents an even greater abstraction, by reference to universal rights and principles. The rights of all members of society are considered at this stage, and in the process, the possibility of contradiction of the society’s law may arise, by virtue of considering rights and principles. Scoring criteria for Stage 6 are not included in the scoring manual (Colby & Kohlberg, 198Th), in part because none of the longitudinal subjects in the studies upon which the theory is based developed beyond Stage 5.  Regardless, Stage 6 has been  outlined by Kohlberg (1984) and he has identified individuals whose reasoning is characteristic of Stage 6. Such reasoning involves an unqualified respect for individuals (Kohlberg, Boyd, & Levine, 1990), taking every possible perspective of a moral problem, and giving each equal consideration (in what Kohlberg calls “moral musical chairs”). It goes beyond Stage 5 in that it represents a deliberate use of the principles of justice in the process of moral decision-making, so that consistency between judgment and action becomes an imperative. The Measurement of Moral Reasoning The method used by Kohlberg to assess moral reasoning competence (sometimes referred to as a clinical method) involves presenting the individual with a series of hypothetical stories, each containing a moral dilemma.  Each of these hypothetical  15.  dilemmas (Appendix A) forces a choice between two conflicting issues; for example, whether it is more important to obey a parent or honor a contract; obey the law or save a person’s life. The use of hypothetical dilemmas has been criticized (Baumrind, 1978; Gilligan, 1982; Haan, 1977) for failure to tap the ability to reason about real or actual moral problems.  These critics claim that Kohlberg’s interview measures reasoning about  irrelevant issues, within which the respondent is uninvolved.  In fact, in comparing  reasoning about real-life moral issues and hypothetical ones, some researchers (Power et al., 1989; Trevethan & Walker, 1989; Walker, de Vries, & Trevethan, 1987) have found reasoning about hypothetical issues to be higher than reasoning about real-life issues. This is an important issue for this research project. However,, the variability found in reasoning about different kinds of problems does not pose a problem for Kohlberg’s theory. Kohlberg has argued (Colby & Kohlberg, 1987a) that one should not expect an exact correspondence between reasoning about the hypothetical and real-lif issues because there are always mediating nonmoral factors (as this study will demonstrate) which interact with the ability measured by the Moral Judgment Interview. In conducting the interview, after each dilemma is read to the subject, a series of standardized questions or prompts are given, for example, “Should Heinz steal the drug? Why?”  The purpose of the questioning is to draw out the individual’s best possible  reasons for the course of action he or she has chosen. If the individual fails to elaborate reasons sufficiently, the interviewer is required to prompt again.  The responses are  usually recorded and transcribed, for scoring purposes. During the scoring process, the reasons provided are matched with criterion judgments from a comprehensive coding manual (Colby & Kohlberg, 198Th), and subsequently assigned a stage score on a 9-point scale; 1, 1i2, 2, 2/3,  ...,  5. These stage  scores are then used to determine a global stage score (GSS), which can be on a 5- or 9point scale. The 5-point scale uses only references to Stage 1, 2, 3, etc., while the 9-point  16.  scale allows for additional classifications into intermediate or transitional stages, such as 1/2 or 3/4. Alternately, these stage scores can be translated into a more quantitative weighted average score (WAS), as was done in this study, which is based on the percentage of reasoning evidenced at each stage, and ranges from 100 to 500 (with 100 equivalent to a pure Stage 1 reasoning, 150 equivalent to transitional Stage 1/2 reasoning, 500 equivalent to pure StageS reasoning). There are three forms of the MJI (A, B, and C) that can be used, each with three dilemmas.  Interrater, retest, and alternate form reliability of the three forms are  acceptable (Colby & Kohlberg, 1987a; Walker, 1988).  Normally, a moral judgment  interview is comprised of three dilemmas; however, using only two dilemmas has been found to provide an equally valid measure (Trevethan & Walker, 1989), and in this study only two dilemmas were used. The Validity of the Moral Stage Model Kohlberg has characterized his model as a strict stage” model. He has proposed that there are three empirically testable criteria for such a model: 1. He insists that the progression through the stages would always be the same (i.e., Stage 3 always follows Stage 2; there is no possibility of skipping stages or regression), regardless of varying environmental conditions. A review of several studies by Walker (1986) found very few instances of regression (0-17%) or stage skipping (04%). This rate of violations of this sequence criterion is within the range of measurement enor. 2. Kohlberg claims that each stage represents an organized, structured whole, what Piaget referred to as “structure operative d’ensemble.” That is, each stage replaces the previous stage through a process of re-organization (described earlier as equilibration), and involves an internally consistent structure within the reasoning of an individual, which is the same in different situations, at different times, and with different  17.  moral problems, or content. According to the theory, a person is either “in” a stage, which would mean that the majority of the reasoning would be at a modal stage, or “in transition,” which means that the reasoning is divided between two adjacent stages. In re analyzing data on Kohlberg’s original longitudinal subjects, Dolby, Kohlberg, Gibbs, and Lieberman (1983) found that (with the most recent scoring system) 68-72% of their reasoning was scored at the modal stage, and a total of 97-99% was at the modal and adjacent stages. Walker (1988), in re-examining several of his own studies (again using the most recent scoring system) found very similar patterns, concluding that there is considerable consistency in responses to the Moralfudgment Interview (MJI). This structure criterion also holds that an individual’s reasoning should be consistent across content, that is, across differing moral dilemmas.  This is usually  examined by calculating the percent agreement of responses between forms of the Moral Judgment Interview. Colby et al. (1983), using the 9-point GSS scale (referred to in the section above), reported a 75% agreement between the use of different forms within the same interview. Walker (1983) found similar levels of consistency. With respect to consistency across a time interval, Colby et al. (1983) reported 70-77% agreement ktween testings 3 6 weeks apart, and Walker (1983) found a similar -  pattern (2 months apart), with 95% agreement in modal stage scores. In addition to being found consistent across content, moral reasoning has also been found to be consistent across different types of moral problems. To illustrate, the MJI has been described (Colby & Kohlberg, 1987a) as a test of a person’s moral reasoning competence or his or her ability to reason about hypothetical, abstract situations. Walker, de Vries, and Trevethan (1987) found that 91% of 240 participants used the same or adjacent stage reasoning (on the 9-point GSS) when dealing with what they had identified as a personal, real-life moral problem, as they did when dealing with the hypothetical dilemmas of the Moral Judgment Interview. Therefore, even though there is some variability in reasoning about different types of moral problems (sometimes  18.  at the lower adjacent stage, sometimes at the higher), there is sufficient consis tency to support the argument that, in general, people reason at the same stage. Walker (1988) has made reference to the fact that certain conten ts pulled reasoning in specific directions. However, even more interesting is whether specific social situations also demonstrate an ability to pull for higher or lower moral reasoning. Although it does not represent a challenge to the structure criterion of the stage model, there is, in fact, a body of evidence that indicates that moral reasoning does vary with social or subcultural contexts. This topic will be explored more fully in the Moral Climate section. The final (and most difficult) criterion for the moral stage model has already been referred to, and is the requirement that each stage is a more adequate or hierarchical re organization of the previous stage. As noted earlier, it is impossible to make such a claim about a stage model of moral reasoning without being accused of makin g nonscientific, philosophical assumptions. There is some evidence (Rest, 1973), howev er, that the progression through these stages is based upon the ability of persons to recognize the superiority of (and hence perhaps be drawn toward) higher stages, even though they do not yet use such higher stage reasoning. In order to add weight to this eviden ce, Walker, de Vries, and Bichard (1984) presented statements to subjects in pairs, each statement representing an example of reasoning at any one of the five stages of reasoning. These pairs were comprised of all possible combinations of the stages, for examp le, Stage 3 versus Stage 1, Stage 3 versus Stage 5. Subjects were also asked to paraphrase each statement in order to demonstrate that they understood it. Walker et al. (1984) found that the subjects recognized as better those statements which were one or two stages higher than the stage they were at (when they understood such statements), even though they did not spontaneously generate such reasoning (during the MJI). When they did not understand the statements, they showed no clear pattern of preference, and when the statements were of lower stage, even though they demonstrated unders tanding, they  19. showed no preference for such lower stag e statements. Walker (1988) summar izes the findings of this important study as follows : These findings clearly support the hier arch of moral reasoning involve more than y criterion. Kohlberg’s stages there are differences in meaning acro differences in language; indeed perceive. Individuals do regard succes ss the stages that subjects can sive increasingly adequate frameworks for mak stages of moral reasoning as ing moral decisions. (p. 47) Therefore, there seems to be strong sup port for the validity of the stage model, using the three criteria of stage sequence, stage structure, and stage hierarchy, set out by both Kohlberg and Piaget. Moral Reasoning and Delinquency Several research projects and reviews (Arnold, 1989; Blasi, 1980; Hams, 198 4; Hams & Miller, 1980; Lockwood, 197 8; Nelson, Smith, & Dodd, 1990) hav e found significant differences in stage of moral reasoning between offender and nonoffe nder groups. Offenders tend to score low er than nonoffender groups, that is, to be developmentally delayed at Stage 1 or 2 (see Kegan, 1984). For example, Treveth an and Walker (1989) found both delinquent and psy chopathic youth to be significantly low er in moral reasoning scores than a matched gro up of normal youth (matched for educati onal level and parents’ SES). Jurkovic and Prentice (1977) found that psychopath ic delinquents were significantly lower than neurotic delinquents or controls (of the same age) in moral reasoning scores. Bear and Richards (1988) found that elementary school students who were rated by their teachers as antisocial, scored lower in moral reas oning than normals, although other students with other types of behavior problem s (i.e., attention problems, hyperactivity), wer e not significantly lower than normal s. In reviewing 30 studies dealing with this topic, Blasi (1980; see also Arnold, 1989, and Buchanan, 1992) found that most of the stud ies he reviewed found delinquents at a lower stage of reasoning than nondelinquents. This consistent pattern, finding offende rs at significantly lower levels of moral reasoni ng, represents the best established beh avioral  20.  correlate of moral reasoning. Therefore, it is of critical importance for this research project to highlight the distinctions between Kohlberg’s Stage 2 and Stage 3.  Because Stage 3 reasoning  summons references of loyalty to groups or immediate friends, of not disappointing the expectations of others, and of disapproval of that which would hurt others, the structure of this reasoning is less compatible with behavior that involves breaking the law or violating strong social conventions. That is why this reasoning has been described as a “cognitive buffer” (Gibbs et al., 1984) to offending behavior.  Alternately, Stage 2  reasoning (“it’s alright to steal because you might not get caught”), is more amenable to violating the law; if one is concerned only with the consequences of action for the self (Samenow, 1984; Yochelson & Samenow, 1976), and is operating from within a social context that does not censor such behavior with sanctions of some sort, the possibility of delinquency increases. Other research dealing with the treatment of offender groups has developed the notion of “thinking errors” (Meichenbaum, 1977; Samenow, 1984). Such thinking (or what has also been called “private speech” (Kohlberg, Yaeger, & Hjertholm, 1968), enables or sets the cognitive groundwork for illegal behavior (Gibbs, 1991; Kahn, 1990; Samenow, 1984). It  is  exemplified by statements (or self-talk) as “No one cares for me,  so why not?” or “That teacher is out to get me, so I’ll get him!” and, in addition to the thinking errors, often contains elements of Stage 2 thinking. This literature derives from a quite different theoretic position, but has produced yet more support for Kohlberg’s notion that moral judgment (thought) is linked to action, and that delinquency is, in part, explained by a developmental arrest in sociomoral reasoning. In a review similar to that of Blasi (1980), Jurkovic (1980) drew parallel conclusions, but noted several studies with anomalous findings. Jurkovic (1980) noted that repeat and/or institutionalized offenders, as well as older offenders, evidenced more Stage 3 reasoning than less experienced ones. Varying opinions have been offered  21.  (Arbuthnot, Gordon, & Jurkovic, 1987) to explain the continued delinquency of these individuals, despite their Stage 3 competence. For example, samples of older delinquents have been found to contain significantly more drug users (Kohlberg & Freundlich, cited in Jurkovic, 1980) than the younger delinquent subsamples. Presumably, drug addiction is a noncognitive factor strong enough to override Stage 3 reasoning competence. Another exception to this trend was found by Thornton and Reid (1982), who in examining types of offenses, found that “prudent” offenders (those that used more planning in their offenses) were lower in moral reasoning than imprudent ones (see also Schonert & Cantor, 1991). Another explanation of why a person of Stage 3 competence would engage in delinquent behavior (and a critical one for this research) invokes the social (or institutional) context in which these delinquents found themselves. That is, despite the fact that some offenders demonstrated higher levels of moral reasoning competence in response to hypothetical dilemmas (on the MJ1), the social group or system within which they regularly found themselves may not have tolerated or recognized certain behaviors or reasoning, and may have exerted a press towards Stage 2 problem solving. Thus, it seems that there may be factors which allow one to make a distinction between moral performance. Competence, as noted above, reasoning competence and moral reasoning 3 is that which is best measured by the Moral Judgment Interview, in response to abstract and hypothetical problems, and which “tests the limits” of ability. As I shall report, competence “cannot be enlisted as a way of accounting for individual differences on The distinction between performance and competence here is not consistent with that proposed by Chandler (1991) and others (Chandler & Chapman, 1991). Chandler rejects the notion that competence has a causal relation with performance, but instead argues that competence is represented by those formal qualities or patterns that provide the meaning to performance. Kohlberg’s assumption is that one’s basic moral reasoning competence (responses to the MJI) are causally related to moral reasoning in other areas or contents (performance). I will later propose that the relation between moral reasoning competence (on hypothetical issues) and responses to issues closer to the real-life fabric faced by these offenders is explained in part by competence and in part by the moral climate of the context.  22.  cross-situational variables” (Chandler & Chapman, 1991, p. 2).  Performance is that  which is used in real-life situations, and reflects competence to a degree. The distinction between performance and competence in some ways parallels the relation between moral reasoning and moral behavior. Therefore, in order to further explore all of these distinctions, this study measured both moral reasoning competence and performance (reasoning about moral issues which were real and salient for the participants), and as well, indexed the moral choices and behaviors which can be 4 All of this is discussed further in the section construed as having moral implications. below on moral climate. A delay in sociomoral reasoning has also been linked with family backgrounds of power assertiveness and disharmony (Gibbs et al., 1984), with the suggestion that contexts characterized by punitive and/or physical reactions to behavior (Alexander & Parsons, 1980), or family dynamics low in affective bonding (Kerr & Bowen, 1988), or those characterized by disengagement (Brennan, Huizinga, & Elliot, 1978; Stierlin, 1973) would reinforce sociomoral reasoning concerned only with the consequences of behavior for the self (Stage 2), and/or would fail to stimulate identification with societal and/or family norms (Stage 3) (Jurkovic & Prentice, 1974). The parallel between the structure of the power-assertive/abusive parent-child relationships and the structure of the child’s Stage 2 sociomoral reasoning is an important one. It relates very closely, to several of the hypotheses in this study. I hypothesized that undesirable social situations or climates are less conducive to moral development than better ones. What is most interesting about all of the above discussion, however, is the challenge the findings represent to more traditional views of delinquency. Not only do they present delinquency in a developmental frame, but they also focus attention on the  purposes of simplicity, I will use these two notions of competence and performance in a manner similar to that employed by Colby and Kohlberg (1987a) and will later review these same notions in light of the findings.  4  23.  structure of the reasoning of delinquent persons, rather than the content (Jurkovic, 1980) of their particular value orientations or attitudes, which is the traditional focus of sociologists (Gibbons, 1976; Nettler, 1978) in explaining delinquency (cf. Gibbs & Schnell, 1985). Developing the Moral Reasoning of Offenders According to Piaget, as the individual encounters his or her environment, there always exists the possibility of some form of conflict with that environment, conflict which may require some form of internal cognitive reorganization. This conflict can be either spontaneously generated by the individual as he or she copes or solves problems, or it can be socially induced (Berkowitz & Gibbs, 1983, 1985; Crockenberg & Nicolayev, 1979). The earliest research concerned with this phenomenon was undertaken by Turiel (1966) who had normal school children play the role of the main protagonist of the classical Kohlbergian dilemmas. The protagonist was required to seek the “advice” of a friend played by the experimenter. This advice took the form of reasoning at various stages above and below that of the subject, depending upon experimental condition. He found that advice one stage above the subject’s was effective in producing a significant change. Similarly, Blatt (1969) conducted moral dilemma discussions with groups of elementary students, once a week for a period of 12 weeks and witnessed a developmental change of one full stage in moral reasoning for 65% of the students (as measured by the assessment procedure available at that time). Blatt and Kohlberg (1975) later carried out a similar procedure and observed an average developmental change of one-third of a stage. In both of these studies, the students were encouraged to argue with each other in groups which were designed to contain a mixture of levels of competence of moral reasoning. Higher stage statements were deliberately amplified and clarified by the discussion leader for the benefit of the lower stage participants. The changes that occurred were shown to be lasting, through follow-up testing 1 year later. Since that time  24.  there have been several studies which replicated BlatCs findings (see reviews by Jurkovic, 1980; Maag, 1989; and Walker, 1982, 1988). In addition to exposure to higher stage reasoning, there appears to be other factors which can serve to heighten conflict (see Hersh, Paolitto, & Reimer, 1979).  Using  delinquent and predelinquent boys as subjects, Niles (1986) produced a significantly greater effect (in comparison to a values clarification group and a no-treatment group) with large-group discussion of hypothetical dilemmas. These group discussions, which occurred twice a week for 16 weeks, emphasized the attainment of consensus. Rosenkotter, Landman, and Mazak (1980), again using forced consensus, found an effect in seven weekly sessions, and Arbuthnot and Gordon (1986) found an effect after 12-16 weekly group sessions. Gibbs et al. (1984) produced a significant effect for lower-stage subjects after only eight group sessions, over a period of 2 months, by maximizing or highlighting the differences between the views of subjects. There is also evidence to support the thesis that effects can be evident with very short-term interventions. For example, when participants were screened for cognitive and perspective-taking prerequisites, Walker (1983) achieved a significant effect by having his subjects observe two adults role-play an argument, a process lasting approximately 30 minutes.  Arbuthnot (1975) produced a significant effect (in  comparison to a control group) with subjects who engaged in a single 15-20 minute role play with a more mature ‘opponent.”  Similarly, Berkowitz, Gibbs, and Broughton  (1980) had similar results from five dyadic discussions (over a period of weeks) between college students. Interestingly, they found that those members of dyads whose partner was less than a full stage higher than they were, developed more in moral reasoning than those paired with a partner who was at the same stage or a full stage above them. In other words, there may be limits to the degree of stage disparity that facilitates development. Kruger (1989) found that peers engaging in one dyadic discussion (using an immediate posttest) benefitted more than children who had a similar discussion with  25.  their mother. Although this difference was just marginally significant (p  =  .053), the fact  that only one dyadic discussion was used (here three such discussions were used) and that the posttest came immediately after the intervention 5 was taken as support for the expectation of an effect of short-term dyadic discussion. Therefore, it appears that when participants of group or dyadic discussions encounter higher stage reasoning and are encouraged/pressured to experience more conflict, they are likely to improve in moral reasoning and that such improvements can be achieved in relatively short-term interventions. However, none of the short-term interventions described here were used with delinquents. The argument could be made that delinquents would be less likely than normals to benefit from such an intervention. However such an argument would be based on the assumption that delinquents, admittedly lower in moral reasoning, are more resistant to change. The position taken here, however, was that delinquents are lower in moral reasoning because they simply have had less exposure to better reasoning, not that they were more resistant to it. Consequently, it seems reasonable to test the hypothesis that they could benefit from such a short intervention. The research concerning dyadic discussion is critical for this study because it used such an intervention (see Lyman & Selman, 1985; Miller & Brownell, 1975; Selman & Demorest, 1984). Here, two offenders sat in a room, across a table from each other, and engaged in debate about both real-life and hypothetical dilemmas.  The dyads were  organized in such a way that the target subject was either of higher moral reasoning (and therefore not expected to change) or at a lower stage of reasoning than his partner (and therefore expected to change). The nature of what seems effective in such discussions has been explored in detail by Berkowitz and Gibbs (1983) who re-analyzed the data from their 1980 study Most interventions employ a delayed posttest, of at least one week, because of evidence that there is a ‘sleeper effect” (Sullivan & Beck, 1975).  26.  (Berkowitz, Gibbs & Broughton, 1980).  They examined the dialogue between the  discussants and found that those dyads which were effective in producing moral reasoning development contained more challenging kinds of statements, which they  called “transactive.” Transactive statements were ones which served to either dispute and/or critique the statement of another (operational transacts); or to elicit, clarify or paraphrase the statements of the other (representational transacts).  Powers (1983)  modified their coding system (Berkowitz & Gibbs, 1979) by adding categories of interactions that involved emotional support or interference.  In studying families  participating in moral dilemma discussions, she analyzed their “speeches’ in a 2 X 2 matrix  -  whether they were cognitive or affective, interfering or stimulating.  Powers  found that for the family context, supportive (affectively stimulating) speeches such as praise, in addition to the “transactive” or “challenging” statements noted by Berkowitz and Gibbs, were an important aspect of the interaction of families that contained more highly developed children. Walker and Taylor (1991), using 2-year longitudinal data, reached a similar conclusion about the role of support, but also noted the importance of what Berkowitz and Gibbs (1983) called “lower-order transacts,” which include interactions such as a paraphrase, or a perception check.  Such speeches may be  important for an atmosphere of acceptance within family dialogue.  These findings,  therefore, are yet more support for the importance of social context for the process of structural change (that is the concern of this study). The view taken here is that these factors are better handled as contextual variables (i.e., climate), rather than interactional or affective ones (Villenave-Cremer & Eckensberger, 1985). To summarize to this point, it is claimed here that the moral reasoning of delinquent youth appears to be delayed and that, therefore, moral reasoning represents a reasonable arena for intervention or rehabilitation. Further, there is strong evidence that group and/or dyadic discussion of moral dilemmas (real and hypothetical) is an effective vehicle for improving the moral reasoning competence of offenders, especially if  27.  consensus is encouraged. Most importantly, there is evidence, that climate or context plays a role in this process, and it is these concepts which are explicated in the following sections. Moral Climate In attempting to stimulate the development of moral reasoning, the earlier applications of Kohibergian theory (noted above) usually involved group discussion of hypothetical moral dilemmas. Even though these methods frequently proved effective in improving moral reasoning competence, Kohlberg and his students came to realize that this was a process essentially outside the real moral worlds of the participants, whether students or offenders (cf. Crockenburg & Nicolayev, 1979; Gilligan, 1980). Therefore, via the medium of community meetings in jails and schools, he replaced the use of hypothetical dilemmas by using the real moral problems currently facing the discussants, both as individuals and as a group. He felt that this approach was more cognizant of the pressures faced by students, 6 particularly those pressures that were part of the subculture or hidden curriculum (Kohlberg, 1983, 1986; Kohlberg & Higgins, 1987; Mosher, 1980). He also felt that the democratic nature of discussing such problems was an essential aspect of the developmental process. Thus, Kohlberg faced group subculture head-on, with the belief that this variable was an important factor in understanding the processes of the development of moral reasoning. In doing so, Kohlberg introduced the idea of moral climate (Higgins, Power, & Kohlberg, 1984; Power et a!., 1989; Power & Reimer, 1978) as a part of a model (Kohlberg & Candee, 1984b) of the relation between moral reasoning and moral action. This model proposes several social-psychological factors that interact with pure moral reasoning competence in detennining what actual moral decision a person will make  6  Kohlberg used a similar approach in the Niantic Prison for Women, Connecticut (Kohlberg, 1986).  28.  within a specific context. This model makes a distinction between what the individual thinks he or she should do (a prescriptive choice), and what he or she would do (a deliberative decision). Similarly, Leming (1973), in addition to studying responses to classical hypothetical dilemmas, derived what he called practical moral dilemmas which he used to ask both “should” and “would” follow-up questions.  He found that his  subjects, 7th to 12th grade students, scored highest on the hypothetical dilemmas, lower on the “should” questions about the practical dilemmas, and lowest of all in response to what they would do, in response to practical dilemmas.  Thus, Leming observed a  progressive lowering of quality of reasoning as his subjects moved from the purely hypothetical to moral problems which had situational complexities, and as they moved from purely isolated consideration to actual social contexts. Recognizing this variability in moral reasoning, Higgins, Power, and Kohlberg (1984) argued that a critical variable in determining the degree to which reasoning changes when moving from hypothetical situations to real-life ones is the moral climate of the group within which a person is functioning at the time. The moral climate of the group serves as a “scaffold” (Vygotsky, 1934/1986; Wertsch, 1984, 1985) for the development of members of that group. Thus we can see that 1) the peer group is an important reference point for adolescents and (perhaps) particularly adolescent offenders, and 2) that the group’s moral climate is a critical variable in determining whether the influence of the group is developmental or conservative. (Higgins, Power, & Kohlberg, 1984, p. 75) Echoing the work by Leming discussed above, Scharf (Hickey & Scharf, 1980; Scharf, 1973) found that prison inmates used a much lower stage of reasoning about real life prison dilemmas than they did about hypothetical dilemmas (Jennings, Kilkenny, & Kohlberg, 1983). In reviewing Scharfs research, Higgins et al. (1984) claimed that this lower reasoning was an adaptation to the predominantly Stage 2 moral climate of the prison. In our view, the Stage 2 practical reasoning of the prisoners with Stage 3  29. competence in classical moral judgment was more a function of the prison environment than of the prisoners as personalities. We would characterize the real environment of prison guards and inmate peer groups as a Stage 2 environment or moral atmosphere, and inmates’ Stage 2 practical judgments were a realistic adaptation to it. (p. 81) The Just Community approach was Kohlberg’s way of addressing the subculture (Kohlberg, Kauffman, Scharf, & Hickey, 1974). In describing the approach, Higgins et al. (1984) argued that given a minimal level of moral climate, with “the opportunity for democratic discussion and decision making” (p. 103), groups of students will make decisions of responsibility which are more consistent with their competence (their prescriptive choices in response to hypothetical dilemmas) than would students who are part of a less mature moral climate. In fact, this approach suggests that unless the climate of a program can be characterized as being of a stage higher than, for example, Stage 2, individual developmental changes beyond that stage are less likely (Kohlberg, 1986). Kohlberg argues that residents in any rehabilitative environment, whether they be inmates or adolescents living in group homes, must “have a say” in the operation of the facility, the assumption being that this would contribute towards a Stage 3 context, one characterized by a degree of belonging and mutual commitment. Whether Kohlberg’s argument about democratic participation is accurate or not, any environment which is better than the “highly power-assertive, disharmonious homes which preclude role-taking opportunities” (Gibbs et al., 1984, p. 37), (and from which a majority of subjects in this study originated), would be more likely to have a developmental effect. The challenge is to produce such a context within an institution. This challenge is made even greater by the fact that the residents of a young offender facility tend to re-create their punitive context within the conectional setting (Fahlberg, 1990), if not directed or guided to do otherwise by staff or more mature peers. In retracing the actual development of the moral climate within a group, Kohlberg and Higgins (1987) theorized that, initially, it was the more active advocacy of the teachers about norms of caring, fairness and trust within a community that led the way  30.  for the students. Kohlberg initially noted little enthusiasm about voting and respectful dialogue. They had moral sympathies for the teacher, and for children with leukemia, without our Just Community program, but they were immobilized by what we call counter-norms, peer norms or expectations that violate not only conventional moral norms but the capacity to empathize with each other or take each other’s viewpoint. (Kohlberg & Higgins, 1987, p. 110; italics in original) Later, with the persistence of the teachers, enabling “a construction dependent upon teacher direction” (p. 111), the students gradually developed the sense of community that was targeted. The destructive result if this intervention by institutional staff does not take place has been well documented by Poisky (1962; Poisky & Claster, 1970). Some of the participants in this study were drawn from two residential units (within a larger Young Offender Center) in which an approximate form of what Kohlberg advocates in his Just Community was practised. Daily unit meetings were held in these two units in order to discuss common problems and the residents were challenged and  enabled to help each other and resolve their conflicts. One unit had begun meetings 3 months prior to the onset of the study, the other had been conducting them for over a year.  In the third unit, no such meetings were held, but the daily routine and basic  behavioral expectations of the residents were the same. These three units were therefore  considered three levels of intensity of moral education program. Moral Climate Methodology •  Higgins et a!. (1984), Power (1980), and Power et a!. (1989) have produced the  only published research on moral climate, having developed a complex method of measurement. Because of differences in the sources of data, only a portion of their method was used here. In general, Higgins et a!. (1984; Higgins, 1980) and Power et a!. (1989) believe that groups develop collective norms that belong only to the group. They insist that moral climate is not represented by the average moral reasoning competence of the  31.  members of the group. Instead, they argue for the assessment of “collective norms” as a measure of the moral climate of the group. Collective norms, which define what is expected from group members as group members in their actions and attitudes, serve as the “leading edge” for the group’s development. We thought a Stage 3 collective norm is not the same thing as an average of individuals’ stage judgments at Stage 3. In a school promoting moral development, collective norms would be formulated at a stage that was the leading edge for the group, and adapting to these norms would stimulate those students whose individual stage was slower to grow. (Power et al., 1989, p. 76) Higgins et al. (1984) outlined a progression for the development of collective norms within a group. They argued that a collective norm would go through the same progression of stages as does individual reasoning, although unlike the reasoning of an individual, a collective norm could emerge at any stage (depending upon the competence of the members of the group), and could also show regression (again, depending upon changing composition and dynamics of the group).  They scored the stage of the  collective norm using the same scoring manual (Colby & Kohlberg, 198Th) used to score individual reasoning. Their data (Power et al., 1989) supported the notion of a stage progression of collective norms. In their study, using four school-related dilemmas, measures were made of the following: (a) what each student thought should be done, and why; (b) what each student would do, and why; (c) each student’s perception of what other students think should be done, and why; and (d) what each predicted others would do in each of the situations, and why. Samples of 20-30 students were selected from four high schools, two of which represented experimental groups, and two representing comparison groups. Each subject was administered the MJI and also interviewed using the four school-related dilemmas (Appendix B). Transcripts of community meetings were used to originally construct the variables involved in this study.  32.  Higgins et a!. (1984) also derived a 15-point scale of Degree of Collectiveness of each norm, ranging from total rejection of the norm (“No one can make a rule or agreement about this school that would be followed or taken seriously”) to total collective acceptance (“We, the members of this group, think that we should act in accordance with the norm”).  They also developed a measure of the Phase of each  collective norm, which indicates the source of the norm, that is, whether it is stipulated and preferred by the staff group, or whether it emerged from within the group itself. They stress that it is not only important to measure the moral stage of a collective norm, but also its strength or the degree to which members of the group are committed to it. For example, the subjects in the study almost all agreed that they should not “rat” on each other, although some had different reasons (stage structure) for doing so. Norms that are widely shared and strongly enforced may be unfair or woefully lacking in their moral significance. Examples of repressive totalitarian states, fanatical cults, violent gangs, and organized crime come to mind when we think of groups with strong but morally defective collective norms. These negative examples lead to certain caution or wariness about collectivities in general. In our view, this caution may be exaggerated if one fails to differentiate the collective stage from the other two dimensions of norms (degree of collectiveness and phase). Understanding these dimensions is especially relevant when considering adolescent moral education. Peer pressure should not be viewed as a problem per se but only as a problem if the pressure supports low-stage norms or norms that would be regarded as immoral from the perspective of a highest stage of morality. (Power et al., 1989, p. 142) For the Kohlberg group, each of the four dilemmas represented a possible norm for the group, for example, “We should not steal from each other.” Higgins et al. (1984), Power (1980), and Power et al. (1989) used the variables described above (namely Stage ofNorm, Degree of Collectiveness, Phase of the Norm, Stage of Community), and another (called Level of Institutional Valuing). In this study I used a modification of the School Dilemmas Interview, referred to as the Moral Climate Interview (cf. Appendix B), which involves three norms: helping, stealing, and accepting contraband. I scored these data on the two variables most relevant to the purposes of this study: the Stage of the Norm and the Stage of Community (Appendix C), both reflected in the discussion of each norm.  33.  What follows is a detailed description of each of these two variables, and an explanation of the importance of each for each other and the notion of moral climate. The Moral Climate Interview (Appendix B) asks questions of the subject about why he should and would do a specified action. The Stage of the Norm represents the reasoning about why others would do what they would do, in response to the three situations posed to each subject. Part of the definition of moral climate here, then, rests upon the perceptions of members of a group about the reasoning of most other members of that group. To illustrate, after asking what each participant would do in each situation, the interviewer then asked what the participant thought others would do in that same situation, and what their reasons would be for doing so. Although some respondents claimed that they didn’t know why others would or would not behave in certain ways, most had rather confident responses; for example, “Most other guys on this unit would accept it [the contraband], and they would do it simply because they want the drugs” (Stage 2), or “Most other guys on this unit would not take it because the heat brought down on the Unit wouldn’t be worth it” (Stage 2). (These responses were scored using the Standard Issue Scoring Manual, in the same way that Moral Judgment Interviews are scored, with matches being made between responses and criterion judgments contained within the manual.) As a measure of the perception of the reasoning of others, the opportunities to challenge the integrity of this variable are many. For example, one could claim that to ask such a question as “What would others do and think about  is to simply invite  downward comparisons, that is, the tendency to see others in a less favorable light or condition, in order to feel more comfort about one’s own position. Alternately, it could be claimed that the answers to such questions, especially of immature subjects, are limited by the preconventional perspective-taking ability of such subjects, that is, they would tend only to see the reasoning of others as being the same as theirs. Or finally, one could make the contention that a person’s ability to describe the reasoning of another  34.  person is limited (as a ceiling effect) by his or her own reasoning capacity. Although Power et al. (1989) did not anticipate such challenges, here such critiques were countered as follows: (a) The responses of these participants simply did not seem to be generally negative or self-serving such as downward comparisons would be. Instead, they appeared quite valid just as likely to acknowledge others’ differences -  and higher level responses and choices as inferior ones. (b) The perception of another’s reasoning is not limited by one’s own level of moral development, because persons can recognize superior forms of reasoning even though they cannot spontaneously produce such reasoning. 7 If they can recognize it, it follows that they can report it, which is what is required of them in the Moral Climate Interview. The second dimension of moral climate employed here (Stage of Community) is a rating of the degree of solidarity or shared understanding of community. More than just a measure of the degree to which all residents agreed about a specific norm (what Power et al., 1989, would call the degree of collectiveness), this variable represents the degree to which residents have common goals which would benefit all members of the group, or the degree to which they feel a part of a group with common goals. Ratings range from  till’,  where the respondent sees himself and others as simpiy “doing time,” others only being persons with whom he happens to have to share a residence, to “5”, which represents a relatively strong sense of togetherness, and commitment to helping each other (Appendix C). Although it may not appear on the surface to have any relation to moral climate, in fact this dimension was included because it underscores the importance of a sense of a bond between members. The two variables represent two vital aspects of the process of moral development: the cognitive and the affective. The stage of the norm represents the perception of others’ reasoning about the norm, the stage of community represents the et al. (1984) found that participants preferred higher over lower stage reasoning if they had sufficient understanding of this reasoning. Understanding was assessed by the ability to correctly paraphrase moral reasoning statements and level of understanding typically exceeded moral reasoning production by one stage.  35.  perception of group commitment to that norm. These variables are central to the moral education proposals of Kohlberg and his colleagues, and pivotal for his model of the relation between moral reasoning and moral action. Kohlberg (Power et al., 1989) argued that schools should consciously foster their moral climate, not only to promote social and moral development, but also to thereby provide the motivation to act on one’s moral judgment. According to the model proposed by Kohlberg (Kohlberg & Candee, 1984a; Power et al., 1989), it is only at postconventional reasoning (Stage 5 and higher) that consistency between moral judgment and action becomes an imperative. At conventional levels (e.g., Stage 3) one could almost as easily justify cheating on a test (for example, by making reference to disappointment that parents would experience with a poor grade), as one could reason about not cheating (because of loyalty to parents’ standards). Therefore, the socio-moral context is a significant factor in determining which choice is made. Kohlberg (1980) at this point introduces the power of the collective the moral authority of the group to provide a support system for adolescents to act on their higher stage modes of reasoning. If students who operate at a stage 2 or stage 3 level lack the consistent internal motivation to act on what they judge to be right, then the group or community can provide the external motivation for such action. (Power et a!., 1989). -  -  Institutional Climate In an effort to strengthen construct validity for the moral climate measures (since there has been little published work to date), another measure of climate was also used, the Correctional Institutions Environment Scale (dES; Moos, 1975). The distinctions  between these two types of measures can be highlighted by referring to Taguiri’s (1968, cited in Power et al., 1989) four dimensions of climate: (a) ecology, (b) milieu, (c) social system, and (d) culture. The ecological dimension describes the physical and material resources, the milieu encompasses the characteristics of the staff and student body (their education and achievement levels, staff salary levels, etc.), while the social system dimension includes  36.  the organizational struct!4re and operational procedures and policies.  The culture  dimension, according to Taguiri (1968), refers to collective values and norms of the staff and student body. The Moos Correctional Institutions Environment Scale evaluates the strength of preselected social system and culture dimensions and norms, while moral climate is concerned only with culture; specifically the strength of specific norms, and the rationale for such norms provided by individual members of the group. Parallel to Kohlberg’s claim (Power et al., 1989) that the moral development of groups can be described in the same way that individuals can, Moos (1975) claims that many types of groups, including family groups (Moos & Moos, 1986), can be characterized as having a “personality” profile. He argues that, rather than focusing on either intrapersonal variables, or situational factors (as Hartshorne & May, 1928), or the pseudo-question of the relative strength of each of these types of variables, instead the most important conclusion is that both social settings and person-by setting interactions consistently account for substantial portions of the variance in a wide range of individual behaviors. (Moos, 1975, p. 9) The Moos Correctional Institutions Environment Scale (CIES) embodies nine subscales: Involvement Support Expressiveness  1 J  Autonomy Practical Orientation Personal Problem Orientation  1  Order and Organization Clarity Staff Control  1  ‘  Relationship Dimension  Program Dimension  j System Maintenance Dimension  J  Moos (1975) has found that, depending on program goals, different profiles emerged, for example, treatment-oriented correctional program had generally higher scores on most subscales than custodial programs. He also determined that subscale scores were also affected by the number of residents in each unit and the staff-resident ratio.  Most  importantly, in programs which evidence a strong oppositional subculture (what was  37.  earlier referred to as hidden curriculum), there were significant differences between the scores of the staff members and those of the residents, with the residents scoring lower than the staff on all scales except Staff Control. The scale is based upon the notion of environmental press first articulated by Murray (1938), who conceptualized the environment as either constraining or enabling the satisfaction of personal needs. Making a philosophical leap, Stern (1970) and others have gone beyond Murray’s approach and emphasized that only the perceived climate (as opposed to the “real” climate) can be measured. Although the CIES can also be used to measure the ideal climate, the perceived climate is the essential function of the Moos dES. When there is a high degree of consensus in perceptions of a particular aspect of the environment, this represents a directional influence upon the behavior of members of the group. The argument made here is that, regardless of the degree of consensus, both instruments used in this study, as measures of perception of others, represent expectations of others, and that it is essentially expectations of others that comprise the rules and roles of culture. Moos (1975) has made three general conclusions that are relevant to this undertaking: (a) There were very weak correlations between any of the subscale scores and individual characteristics of the respondents.  Thus, CIES scores do not reflect  personal characteristics, or background variables (such as length of incarceration, number of times incarcerated).  (b) Programs with more positive climate scores released  offenders who showed greater improvement in self-rated attitudinal measures (Jesness, 1966) and self-rated behavioral measures (Jesness, 1971). (c) Only some items from several different subscales predicted success after release (that is, length of time after release without parole violation). It was hypothesized in this study that there would be a positive correlation between  the  Relationship  CIES  dimensions  (Involvement,  Support  and  the  Expressiveness) those which could be construed to have a moral dimension and moral -  -  38.  climate scores. That is, it was expected that those residents who felt a higher degree of support and involvement with others (both staff and other residents in their program) and the opportunity to speak freely, would tend to perceive the moral climate in higher stage terms.  It was also expected that the mean CIES scores for each program would  conespond to the mean moral climate scores for the same program, that is, if a program was higher overall on several subscales (particularly Support, Involvement and Expressiveness), it would also reflect higher moral climate scores. Peer Status For both Piaget (1932/1965) and Kohlberg (1984), the distinction between peer interaction and parent-child interaction was critical. Both theorists claimed that because peer interaction is more egalitarian, it is therefore a better context for developmental change.  In contrast, unilateral (Kohlberg, 1969) family or adult-child interactions  involve constraint (Piaget, 1932/1965) and tend to produce conformity through reinforcement. Although they both argued that both types of socialization are necessary, cognitive-developmentalists have held that it is largely peer interaction which provides the opportunity for children to discover “whose meaning new meaning  -  -  one’s own, the other’s, or a  is most workable” (Youniss, 1980, p. 7). In fact, there is support for this  notion; in comparing peer interaction during moral dilemma discussions and parent-child interactions of a similar nature, Kruger and Tomasello (1986) found that, indeed, peer interactions included significantly more self-generated challenging or transactive (Berkowitz & Gibbs, 1979) statements than discussions between child and parent, where children used more responsive and clarifying statements. This project attempted to flesh out the meaning of peer interaction by examining the effects of an extreme disparity of status between peers involved in an interaction. It proposed that the specific structure of this kind of relationship is a factor that could constrain the disequilibration process. That is, in the same way that a parent might have  39.  less impact on a child than a peer (Kruger, 1989; Kruger & Tomasello, 1986) because of the way in which he or she is perceived (i.e., less egalitarian, more unilateral), it was hypothesized here that within a social network of peers there are some individuals with significantly less credibility and influence than others.  More specifically, this study  explored the structure of a particular type of peer interaction  -  that of conflictual verbal  interaction between peers of significantly disparate status, because it is felt that such an -  interaction is one lacking in mutual credibility or “social valuation or respect” (Chapman, 1986, p. 183). Such interactions were compared to those between peers of relatively equal status. In this study, peer status was determined by ratings by each participant of other participants, one of whom he would eventually be paired with to engage in a moral dilemma discussion. One of the earliest examinations of peer status disparity was completed by Polsky (1962; Polsky & Claster, 1970), and carried out within a treatment institution for youth. As a participant observer in one specific “cottage” within a large, residential treatment facility for youth, he was able to describe the hierarchical structure of the group of residents of this unit. Polsky provided a rich and detailed description of the informal rules of the group, and how these served to maintain the hierarchical status structure of the cottage and the various roles within it (cf. Goffman, 1961).  The 16 residents  described themselves as either “toughs” (the top clique), “bushboys” (the larger middle group), or as “queers” (the smaller group of scapegoats). All residents usually worked to maintain their respective positions, and the homeostasis of the system. Those individuals at the lowest end of the hierarchy were almost totally without the respect or regard of all others in the system. Most importantly, this social structure appeared to work against any effort on the part of any individual within the group to change or derive benefit from the individual therapy that was provided. It is in light of this last point that the author proposed that attainment of rehabilitative goals for individuals within such programs could be limited by the  40.  developmental level of other residents who have the greatest status and credibility within the group, and further, that even for those residents of higher developmental level but of low status, a desired developmental effect on others would not be evidenced. Of special importance for this study was Poisky’s observation that the group had two subgroups at the extremes of the status hierarchy, and a middle group which was comprised of middle status “bushboys.” Because of Poisky’s finding and the intent here to maximize the effect of status differential, in this study dyads (see Dimant & Bearison, 1991; Lyman & Selman, 1985; Selman & Demorest, 1984) were formed of status disparity from the extremes of the groups within the sample. Peer status has most often been used as a correlate of social competence (Foster, Bell-Dolan, & Berler, 1986) or as an outcome measure in evaluating social skills training (Coie, 1985). It has also been used to describe children along a dimension of popularity, or to classify them into groups of accepted, rejected, neglected, or controversial groups of children (Coie, Dodge, & Coppotelli, 1982). These categories have been most frequently used with nondelinquent populations, in classrooms, etc., but may also be descriptive of groups of offenders living together in a custody facility. Accepted children would be those thought by most to be popular and liked. For the group studied here, controversial children are perhaps similar to those that Poisky called the “top clique,” that is, those that have influence but were not well liked, but perhaps valued for their ability to provide protection. Such youth are therefore occasionally aggressive. The rejected are those that are extremely unpopular and whose behavior is disturbing and disruptive to their peers  -  Poisky’s “scapegoats” (Sabornie & Kauffman, 1985), while the neglected are those often not noticed by their peers, tending to be “loners.” A major difficulty in using peer ratings of status is that slight variations in sociometric method can produce different groupings of children. For example, different groupings of accepted, rejected, neglected, and controversial children were found when subjects were asked to rate only classmates of the same sex or those of both sexes (Foster  41.  et al., 1986). Similarly, different groupings were found when children were asked to nominate those they would most/least like to work with, or play with (Foster et al., 1986).  It is clear that peer status has meant many different things to researchers, depending upon whom is asked which question. For the purposes of this study, given that my major interest was in the ways adolescent peers do or do not affect each others’ moral reasoning, status was defined in terms of influence, and the question asked of the participants and staff members was: “Excluding yourself, who is the most influential person, or the person with the most status in this group? Name the five guys here with the most status and the five with the least status or influence.” Only five choices were allowed because of my interest in tapping the extremes.  Although one could use the word “power”  synonymously with “status,” I was careful not to use this word with these subjects  because of the possibility that they would interpret “power” exclusively as “physically strong” or “aggressive.” This is not to say, however, that some subjects did not believe that those with the most status or influence were the strongest or most able to intimidate. In this study I could only assume, that, for such participants, aggressive or strong individuals also had the greatest influence or credibility. The question about whether social status (as it is defined in this study) and moral reasoning competence are confounded needs to be considered. That is, in order for the planned analyses to be carried out, a convincing argument for the independence of these two variables needs to be produced. Dodge, Schlundt, Schocken, and Delugach (1983),  for example, found that popular (high status) children were more socially competent in making entry into an activity involving unfamiliar peers.  Keasey (1971) found that  adolescents that were rated either by self, peers, or teachers as more popular, and engaging in greater social participation, were also superior moral reasoners. However, in Kohlberg’s longitudinal study (Colby et al., 1983), after classifying all of the subjects at the first time of measurement as either socially integrated or socially isolated (much the same method of classification as was used in this study), Kohlberg found weak  42.  correlations (none of which were significant) between sociometric status and moral 8 In order to operationalize the dimension of status, four subtypes of dyads judgment. were formed: (a) high status participant (target) paired with a low status participant, (b) a low status participant (target) with a high status participant, and (c) dyads in which the participants are of equal high status, and (d) dyads in which the participants are of equal low status. Within each of these types, subjects were also classified as either higher or lower in moral reasoning (consistent with the +1 paradigm discussed earlier). Thus, in total, eight types were formed (cf. Table 1). All of the preceding has hopefully provided an explication of the social contextual variables which are thought to be involved in the process of development of moral reasoning. However, what import does such development have if it does not ultimately lead to improved moral action?  What follows tackles this issue, one that has been  difficult for Kohlbergian theory. Moral Reasoning and Behavior  Kohlbergian moral development theory has been often criticized for its failure to predict behavior in a linear fashion, that is, Stage 1 reasoners should behave in a Stage 1 fashion, Stage 3 in a characteristic Stage 3 fashion, etc. Alternate theories (sometimes called two-track theories) hold that there is no consistency to be found between moral thought and moral action that young people learn to recognize what is right and wrong, -  and they also learn when, and when not to act (morally) in certain situations, depending upon the circumstances and contingencies the two being quite independent. -  In contrast, because he believed that cognition is a part of action, Kohlberg argued that there is a relation between on&s reasoning and moral actions, but not necessarily a directly causal one. That is, those who reason at Stage 4 or Stage 1 should  8  In this study, a similar pattern was determined; status was negatively but insignificantly correlated (from .02 to -.22) with moral reasoning.  43.  not be expected to act according to specific Stage 4 or Stage 1 standards, respectively. The model proposed by Kohlberg has already been discussed. His claim holds moral action to be a function both of one’s moral reasoning competence and other factors, in part, the moral atmosphere of the group. For example, he notes that many of the soldiers who killed noncombatant women and children at My Lai, during the Viet Nam War, did so not because they were low in moral reasoning, but because they participated in what was essentially a group action based on group norms. The moral choice made by each individual soldier who pulled the trigger was embodied in the larger institutional context of the army and its decision making procedures. The soldiers’ decisions were dependent in large part upon a collective definition of the situation and of what should be done about it. In short, the My Lai massacre was mOre. a function of the group “moral atmosphere” that prevailed in that place at that time than of the stage of moral development of the individuals present. (Kohlberg, 1984, pp. 263-264) It was hypothesized here that both moral reasoning, and more importantly, moral climate would predict the incidence of certain behaviors. Two behavioral measures were used: the Achenbach Child Behavior Checklist (which includes two dimensions of behavior, Internalizing and Externalizing, derived from ratings on 113 items), and a daily behavior frequency recording procedure (Appendix E), which logged 11 categories of behaviors or events. The Child Behavior Checklist (CBCL) uses a 3-point rating scale of a list of 113 behaviors. It must be completed by a person who knows the child well, but who is not required to be trained in its administration. It encompasses a broad range of behavior, and because it involves ratings by others, does not fall prey to the methodological problems inherent with self-report data, particularly with populations (such as this one) that may tend to misrepresent their actions. Although separate forms for both parents and teachers exist, it was the Parent Form that was selected for this study, with the belief that the close daily exposure of the staff members to the offenders more closely approximated that of the parent than teacher. Achenbach and Edeibrock (1983) have found that CBCL scores are stable in a  44.  test-retest analysis, but sensitive to changes in behavior in response to a residential treatment program for children. Children seen on an out-patient basis from mental health centers also showed improved scores following treatment. Test scores have been found to be free of any influence of race and socioeconomic status (ibid. p. 60). The second behavioral measure involved direct observation on a daily basis. The decision to use this method was made because it was felt that a form of direct observation should be used to complement the more subjective ratings involved with the CBCL. The form used is included in Appendix E. Summary and Hypotheses This study focused on a key aspect of cognitive-developmental theory  -  the  relation between social context and moral reasoning development. Two social contextual variables (thought to be relevant for most adolescents) were manipulated: the relative peer status of members of dyads involved in moral dilemma discussions, and a broader measure, what could best be described as level of intensity of moral education program, which effected the moral climate of the group. It was hypothesized that these contextual variables would interact in facilitating or constraining the effects of peer interaction. Equally important, this study also included behavioral measures in order to further the investigation of the relation between moral reasoning, moral climate, and moral action. Specifically, it was hypothesized that: 1. Participants with higher moral reasoning competence would engage in less disruptive and antisocial behavior than those of lower moral reasoning competence. 2. Participants who were involved in the moral education programs of greater intensity (Programs 1 and 2) would engage in less disruptive and antisocial behavior. 3. Participants of dyadic discussions who were partnered with another individual of higher or same stage moral reasoning would show greater improvement in moral reasoning than participants partnered with someone of lower moral reasoning. In this  45.  hypothesis, the developmental effects hypothesized for target subjects paired with one of “same or higher” stage were consistent with Kohibergian theory, and more specifically the conclusions of Walker (1983) who found: (a) that target subjects exposed to higher stage reasoning showed significant increase in their moral maturity; (b) that subjects exposed to the same stage reasoning also developed, but to a lesser degree; and (c) that subjects exposed to lower stage reasoning did not evidence any developmental change.  However, the developmental effect of exposure to higher stage reasoning would be weakened or constrained in those cases where the “other” participant was of a lower peer status rating (see Table 1). 4. Whether or not they participated in a specific dyadic intervention, participants residing in programs of greater intensity would improve in moral reasoning more so than participants from less intense programs. In addition to these central ones, two other secondary hypotheses were made:  5. Participants from programs of greater intensity (which involved unit meetings) would show higher moral climate scores and higher CIES scores than participants from other programs. 6. There would be positive correlations between the Relationship subscales of the CIES (Involvement, Support, Expressiveness) and the Stage of Community (moral climate) measure. 9  A relationship between Stage of Norm scores and CIES subscales was not hypothesized because of the differences between these two variables; one measures the degree to which participants agree with, or perceive certain prosocial values within their respective units (the CIES), the other reflects the structure of reasoning about that value (Stage of Norm). One variable has to do with the strength of values, the other has to do with the justification for values.  46.  METHOD  Sample The total sample was comprised of 101 young offenders, all males, and all living in any one of three residential units within a large young offender facility (with a total of six units) in Alberta. Only males were used because of the small number of females available to this study.’° The average age of the sample was 16.5 years; the range was from 14 to 18 years. Delinquent histories were similarly varied, ranging from three convictions of property offenses, to over 50 prior property offenses and several (4) offenses against persons. No participants were convicted only of status offenses (see Moran, 1988). Many of the 101 participants (approximately 60%) were also awaiting trial on other charges. Only those offenders who had dispositions (sentences) of 1 month or more were included in the study, therefore providing sufficient time for pretesting, intervention (which lasted 3 days), and posttesting approximately 15-20 days following the intervention. Of the total group, 40 participants served as target subjects for the intervention, 40 served as dyad partners for the intervention, and 10 comprised the control group. The remaining 11 participants either could not be paired or had incomplete data. Eighteen staff members also technically served as participants. They ranged in age from 22-40 years, were predominantly (65%) female and, in terms of educational level, had either a 2-year community college certificate or a university undergraduate degree. Each staff member worked predominantly in one program, and therefore knew the participants in that program well. It was important that they were familiar with both the general maturity of each offender as well as his status within the program group, because they were asked (see below) to estimate the peer status and the moral reasoning  10  Because only male offenders were involved as subjects, masculine pronouns are used throughout the remainder of the report.  47.  ability of each participant. Most staff members (from two of the three programs) had received 3-day training sessions (from the author) which familiarized them with Kolhberg’s theory of moral reasoning. Procedure Voluntary participants were nominated by staff from the on-going resident list of each of three residential units.  Nominations were made within the first 3 weeks of  admission. Equal numbers were drawn from each of the units. Dyads were initially formed (on paper) with the help of staff members, who were asked to nominate pairs on the basis of disparate or similar peer status (high, low, both high, both low) within the unit group, and of their apparent 11 moral maturity (high or low) (Table 1).  These  assignments were later compared to the ratings of status by the participants themselves and, for moral reasoning, the scores derived from the Moral Judgment Interview. Given the total number of subjects available and appropriate for selection, this process of dyad formation continued until 40 dyads had been determined and tested. This process took several months. Once nominated by staff members, each participant was approached by one of the experimenters (who were known to the participants but not considered staff). The study was explained and they were told that their participation was voluntary and that refusal to participate would not affect their program or privileges within the institution. They were given consent forms, which were also explained.  Most (97%) agreed to participate,  perhaps because the interviews usually occurred during a 1-hour period of the day (viz, “shift change”) when all residents were required to remain in their rooms, reading or working on school assignments. They preferred the interviews, which they reported to find interesting, to remaining in their rooms.  11  Training provided by the author enabled staff members (in consultation with each other) to make estimates of the moral reasoning ability of each of the participants.  48.  Table 1 Hypothesized Effects ofDyadic Intervention  Type  “Other”  #1 (n  =  =  high status higher (or same) stage  low status lower (or same) stage  yes  high status low stagea  low status high stage  no  3)  low status higher (or same) stage  high status lower (or same) stage  no  7)  low status low stagea  high status high stage  no  high status lower (or same) stage  yes  5)  high status higher (or same) stage high status low stagea  high status high stage  no  4)  low status higher (or same) stage  low status lower (or same) stage  yes  8)  low status low stagea  low status high stage  no  4)  #3 (n  =  #4 (n = 4) #5 (n  =  #6 (n  =  #7 (n  =  #8 (n a  =  Change  5)  #2 (n  “Target”  The other participant was at least 20 WAS points lower than the target.  The control group was formed by pretesting 10 other offenders, randomly selected over the duration of the study, in about equal numbers from each of the three programs. These participants were also required to have dispositions of 30 days or more. They did not experience any intervention and were posttested after the same 30-day interval as those involved in dyadic discussions.  49.  Intensity of Moral Education Program The three units from which the participants were drawn were similar in many aspects (physical environment, basic rules, and daily routines) but different in important and complex ways.  Unit #1 (hereafter called Program 1) had been conducting unit  meetings for approximately 2 years prior to the onset of this research. This meant that the staff members were skilled and comfortable with intricacies in conducting such meetings and, over the period of development of the program, had developed a common language between themselves and with the residents.  The staff had learned, most  importantly, to articulate the reasons for certain actions, for example, helping others within the unit, and such public reasoning provided a “scaffold” (Wertsch, 1984) for the residents. It was assumed that because these practices (unit meetings, staff language, expectations, etc.,) had been in place for over 1 year, this program represented the highest or most intensive level of moral education program, similar to that described by Power et a!. (1989), and not dissimilar to what has been described as a “therapeutic milieu” (Fahlberg, 1990). Unit #2 (now called Program 2) had more recently been trained in the same techniques (Hersh et al., 1979; Lickona, 1991; Vorrath & Bendtro, 1985) as those used by the staff members of Program 1. They had been conducting unit meetings for 2-3 months prior to the onset of data collection. However, it is fair to state that they were not as practised as the staff of Program 1. Program 2 represented a “moderate” level of moral education program. Unit #3 (Program 3) had received no training and conducted no unit meetings. This program was largely custodial. Having explicated these three levels of program intensity, it is necessary to suggest a possible qualification. That is, it is important to add that the effects of program level may not have come as a result of exposure to the program for the 30-day observation interval.  It is possible, but not tracked here, that some participants had  resided in their respective units at some earlier time, during a previous period of  50.  detention. It is also possible that they could have been placed in a different unit on prior occasions and therefore subject to those “better” or “lesser” programs. Therefore, one cannot make the claim that the dosage ofprogram was for one program only and during only one 30-day period. However, it is my view that the climate of a group or program was quickly picked up by each participant and that this is generally true of adolescents as they enter a group and that, further, any effects found represent a kind of accommodation to the climate of the group. Each participant was first administered the following measures (see Appendices): the Moral Judgment Interview (MJI), the Moral Climate Interview (adapted from Power et al., 1989), and the Correctional Institutions Environment Scale (dES). In total, the interviewing process took 45-75 minutes, with some participants more elaborate in their responses than others. The Moral Judgment Interview and the Moral Climate Interview were audio-recorded, and later transcribed for scoring purposes. Except a minority that had been transferred to other centers or units, all participants were administered the same tests after an interval of about 30 days. The procedure is outlined in Table 2. Peer Status At any convenient point during the interviews listed above, each participant was asked to rank the status of other residents in the unit. They were asked: “Excluding yourself, who is the most influential person, or the person with the most status in this group? Name the five guys here with the most status and the five with the least status or influence.” As noted earlier, research by Polsky and Claster (1970) and Coie et al. (1982) suggest that status is not distributed in a linear fashion throughout such a group but categorically, or in extremes. In this way, the ratings of peer status by the staff members could be validated by the subjective ratings by the residents, either “high” or “low.’  If  the staff had indicated that the participant was of low status and his designated “other” participant had also rated him low, the staff rating was validated. However, if the staff  51.  Table 2  Synopsis ofPrQcedure  ASSIGNMENT TO PROGRAM LEVEL (#1, #2, or #3) PRETEST MEASURES:  Moral Judgment Interview Moral Climate Interview • Stage of Norm • Stage of Community Correctional Institutions Environment Scale  Day 1 7 -  Behavior Ratings • Child Behavior Checklist • Behavior Frequency Rating Peer Status Rating DYADIC INTERVENTION  POS’flEST MEASURES:  Day 6  -  8  Moral Judgment Interview Moral Climate Interview • Stage of Norm • Stage of Community Day 25  -  30  Correctional Institutions Environment Scale Behavior Ratings • Child Behavior Checklist • Behavior Frequency Rating  members had rated him low status, but the 11 other” had not rated him in the low group, the status was not confirmed and, although the case was not dropped from the study, it was not used in the intervention. In this way, each participant never directly rated himself (see Hymel & Franke, 1985) relative to the partner that he would be paired with. With very few exceptions  52.  (6%), the sociometric ratings of the staff members were validated by the rating of the participants. The participants were not asked to rank themselves because of indications in previous research that such self-perceptions are subject to the influence of actorobserver attributional biases (Dubin & Dubin, 1965) and other subject variables (Hymel & Franke, 1985). Correctional Institutions Environment Scale The CIES was the first test administered. This test involves 90 true/false items. In several cases, these items were read aloud, because the participant had difficulty with reading. Those participants who could read the test were encouraged to ask any question about the meaning of specific words. Moral Reasoning Because it was not possible to score the Moral Judgment Interview immediately, the dyads were initially formed on the basis of the staff informal estimates (highv or “low”) of moral reasoning (see Kohlberg et al., 1974, pp. 1-2). Later, in order to verify which member of each proposed dyad was higher or lower in moral reasoning, the pretest moral reasoning scores were compared to the estimates. Where necessary, dyads were reclassified, or an alternate decision made about which participant represented the “target,” or the “other,” according to the more accurate measurement. These decisions were made blind to the posttest scores, and this process continued until all cells were filled. Unfortunately, all eight cells were not able to be filled in equal numbers, because of the difficulty in finding the exact combinations of high or low status and moral reasoning scores. The cell frequencies are outlined in Table 1. If the pretest scores of the “other” participant was higher than that of the “target” or if the pretest scores of both were within 20 WAS points the “other” participant was considered as “same or higher” (Types 1, 3, 5, 7; Table 1). This decision to combine same- and higher-stage dyad partners was based on research by Walker (1983) who  53.  found that developmental effects could be achieved as a result of discussion with others of the same stage, as well as from discussion with another of higher stage. If the pretest scores of the “other” participant was more than 20 WAS points lower than the “target”, then the target was considered “higher” (Types 2, 4, 6, 8; Table  1).12  This scoring was done by first transcribing the Moral Judgment Interviews from the tape-recording, then matching the scorable responses to the criterion matches within the scoring manual (Colby & Kohlberg, 198Th). Only one participant was used in these analyses because the scores for the target participant were dependent upon the scores of the “other” participant. Rather than the standard three-dilemma interview, only two dilemmas were used: two from Form A, two from Form B (Appendix A), the use of each form alternating between pre- and post-test. The selection of the Form (A or B) for the pretest was done randomly. Walker (1988) reports sufficient reliability between these two forms (90-95% agreement) to justify this procedure, and Colby and Kohlberg (1987a) also report reliabilities between these two forms of .95. Also, Trevethan and Walker (1989) have found that use of two dilemmas with young offenders has reliably produced consistent results. Colby and Kohlberg (1987a, p. 66) report similar reliability in the use of two dilemmas. Therefore, to reduce testing time and the possibility of subject fatigue, only two dilemmas were used to assess moral reasoning competence. All scores of moral reasoning were expressed in WAS points, which can potentially range from 100 to 500, and which is calculated by weighting and averaging the proportion of judgments made at each stage. Reliability was established using a trained, independent rater, blind to classification of the subject, on a 20% random sample of the subject pooi. For the moral  12 noted earlier, Berkowitz, Gibbs, and Broughton (1980) found that differences of 1/3 stage (approximately 33 WAS points) between discussants was optimally effective in stimulating developmental change.  54.  reasoning scores this was found to be r = .92. Moral Climate During the same interview, each participant was administered a modification of  the Practical School Dilemmas Interview (Power et al., 1989), here called the Moral Climate Interview (MCI). This consisted of two versions of three additional real-life dilemmas, which had been chronically a concern for all the participants in this study. One of these three was a helping dilemma, similar in form to that of Power et al. (1989). However, rather than setting the dilemma in a school, it presented an unpopular youth in a custody facility who, in desperation, is about to either hurt himself, or run away. (Running away is considered a serious act, and is an indictable offence, certain to only seriously compound the problems of this youth.) The question was whether to help this youth. Another dilemma, again similar to the Kohlberg group’s, was a stealing dilemma, that is, whether to take a valued ghetto blaster or cassette tape that was left lying in a common living area.  The third had to do with whether a resident should accept  contraband drugs which were brought into the facility by a youth who had earned a weekend pass (see Appendix B). The three dilemmas of this interview generated scores for three norms (“helping,” “stealing,” and “contraband”). Two variables represented the measurement of moral climate: (a) Stage ofNorm. This variable was scored in a fashion parallel to the scoring of individual moral reasoning.  The responses scored were those given to the question,  “What do you think most others in this unit would do in this situation, and why?” (Appendix B).  Each scorable response was then matched with a criterion judgment  contained within the Standard Issue Scoring Manual (Colby & Kohlberg, 1987b) and yielded a score ranging from 100 to 500. (b) Stage of Community. This variable was an overall rating of the entire moral  55.  climate interview, for each norm separately, assessing the degree to which the respondent perceived his living group to be a collective working together towards positive goals, or, simply a group of individuals who happen to be living in the same physical unit, each doing their “own time.”  This variable produced a global score  ranging from 1 to 5 (Appendix C). As noted earlier these two  measures represented two  important,  but  complementary aspects of moral climate. The Stage of Norm reflects the quality of reasoning supporting a group norm, while the Stage of Community indicates the degree of group commitment expressed in discussion around that norm. Interrater reliability was established using a process corresponding to that used with the MJI scores. Reliability was similar to that for stage of moral reasoning, ranging (for Stage of Norm scores) from r  =  .99 (for the contraband norm) and .92 (for the  helping norm) to .67 (for the stealing norm), and percent agreement in Stage of Community scores ranging from 79% (for helping) to 75% (for contraband). In addition to these variables, responses to other questions asked in the interview were also scored. The responses to the question, “What do you think  should do?  Why?” were labelled Judgments of Responsibility. Responses to the question, “What would you do?  Why?” were labelled Judgments of Practicality, following Leming  (1973). The import of these variables will be discussed separately in both the Results and Discussion sections of this dissertation. Intervention Within 1 or 2 days of the initial interviews, each participant was told the name of his partner and asked if he was willing to continue with the study.  Each pair of  participants then discussed two dilemmas each day for 3 consecutive days. One dilemma was a real-life dilemma based on actual problems that most residents had encountered, and the other was a Kohlbergian hypothetical dilemma, one not previously used during  56.  the MJI. A different set of dilemmas was used each day (see Appendix F). These discussions were coordinated by either the author or one of three other female student/volunteers who were considered part of the volunteer staff group. By virtue of their status as volunteers or, in my case, an “outsider,” the cooperation of the participants was very high. As noted above, these discussions took place during a period of the day when they would otherwise have to remain in their rooms, reading, doing homework, or listening to radio. The participants were informed that the purpose of the “debate” (as well as the prior interviews) was so that we could “figure out how kids think about problems, solve problems, and argue about them.” In order to heighten the possibility of disagreement, each participant was instructed to focus on, or “think about” only one of the two characters described in the dilemmas during the discussion. They were also instructed to try to come to an agreement about the best solution to the problem (a condition which has been found to heighten the effects of the intervention, Maitland & Goldman, 1974), even though each participant, by virtue of being assigned to take the perspective of one of the two characters in the dilemma, was technically in conflict with the other.  Each  discussion (described as a debate) took place within a small room in a separate wing of each residential unit. Participants were seated, separated by a table, with a tape-recorder between them. All discussions were tape-recorded, and transcribed. 13 An example of such a discussion is provided in Appendix G. Prior to commencement, the experimenter read the dilemma aloud in order to ensure that both participants understood the conflict. The experimenter did not remain in the room, but stood outside and was visible to the participants, so that he/she could be available if any questions arose. After the first dilemma, the experimenter re-entered the room and read the second dilemma aloud, and the discussion continued again without his or her presence. Each discussion period was  13  The quality and style of the discussion was not a focus of the current study.  57.  about 20-40 minutes. This series of dyadic discussions was a relatively short-term intervention. Most other similar interventions with delinquent groups have been more protracted. However, there is sufficient evidence (reviewed earlier) to justify undertaking such an intervention. Therefore, with the polarization of views structured within the task, and the requirement that they come to a consensus, and with three such dilemma discussions (two dilemmas each), on 3 consecutive days, it was hypothesized that an effect in terms of improved moral reasoning was a reasonable expectation. Approximately 25 days later, each target subject (and most “other” subjects) were posttested using Form A or B (depending upon which was used during pretest) of the Moral Judgment Interview (Colby & Kohlberg, 1987b).  This interview was tape-  recorded and later transcribed for scoring purposes. Behavioral Measures It was hypothesized that level of moral reasoning would be negatively correlated with disruptive behavior ratings. It was also hypothesized that moral climate would be related to behavioral changes. Therefore, in order to ensure that any behavioral relationships were captured, two types of measures of behavior were undertaken (testing Hypotheses 1 and 2). The first of these (Appendix E) represents the degree of compliance to the institutional routine, number of “incidents,” peer conflicts, defiance towards staff members, etc. This rating (done after participants had been nominated and paired) required staff members to indicate the incidence of 11 types of behaviors for an 8-hour period, twice each day for 7 days.  Scores were expressed in terms of total of incidents of two types: (a)  resisting/noncompliance; and (b) aggression/intimidation. It was felt that this measure represented a thorough assessment of the actual, observable behavior of each participant. This form was completed in conjunction with  58.  another internal form which determined what program privileges were awarded, through which staff members exercised their control over the unit. The second instrument reflects more general or subjective appraisal of behavioral trends; the Achenbach Child Behavior Checklist, for children aged 4 to 18 years. This test required a rating by a staff member (who was familiar with, and assigned to, the youth for case management purposes) of the frequency with which 113 behaviors and/or symptoms occur (“never,” “sometimes,” “often”) and yields scores on eight subscales (for this age group), 14 either Externalizing or Internalizing. Only some of these subscales were expected to correlate with moral reasoning or to be sensitive to the variation in behavior evidenced by these subjects. (Because of limitations of the staff time it would have required to establish reliability ratings for both behavioral measures, such reliabilities were not undertaken.) Summary In summary, the general purpose of this study was to explore those factors which interfered with or enhanced the developmental process initially described by cognitivedevelopmental theorists (Chapman, 1988; Kohlberg, 1969; Piaget, 1932/65: Turiel, 1974, 1977). To do this, the author used a correctional facility for youth, with the knowledge that within such an institution, there existed a number of micro-cultures which might provide the necessary variations in program intensity to explore the effects of such programs.  In addition, the author conjectured that within three residential programs  (from which all subjects were drawn), there did exist a social structure that, in turn, would reflect ratings of peer status, the other social contextual factor of interest. It was these two factors that were explored in asking the following questions: Are differences  14  The authors of the CBCL derived different numbers and types of subscales (or factors) through principal components analysis, for different age groups. For example, girls aged 4-6 have nine subscales, some of which are not part of the profile yielded by this test for boys aged 4-6.  59.  in moral education program intensity reflected in climate scores? Are better climates associated with lower levels of misbehavior? Can a 3-day dyadic intervention have demonstrable effects? Do differences in status affect this process?  60.  RESULTS There were two data sets used in the analyses to be presented. One set included the participants who served as the targets of the dyadic intervention (n members of the control group (n  =  =  40), and  10). This data set formed the basis for the results to  be described in the Intervention section. The second data set  (ii =  101) included the same  50 intervention and control group subjects, as well as 51 others who served as “other” subjects, or who were tested but not included in the dyadic intervention. This second data set was used in testing hypotheses regarding correlations between moral and institutional climates, behavior ratings, and different types of judgments and dilemmas. The analyses to follow begin with a consideration of the second data set. Moral Reasoning: Pretest Measures The mean moral reasoning weighted average score (WAS) of the full complement of 101 pretest subjects (i.e., M  230.6, SD  related studies of juvenile delinquents.  =  31.4) was consistent with results of other  Trevethan and Walker (1989), for example,  reported a mean WAS of 248 for their sample of incarcerated delinquents (in contrast to a mean of 282 for their nondelinquent sample). Moran (1988): M  =  Similar differences were reported by  245 (delinquent), 290 (nondelinquents); and in Blasi’s (1980)  15 review. As noted above, the participants were drawn from three distinct programs that differed in terms of their intensity. At the time of institutionalization, participants were randomly assigned to two of the three programs, Programs 2 and 3, depending upon the  15  One recent Canadian study (Larose, Dionne, & Larivee, 1992) quoted a higher mean WAS of 273 for their delinquent sample. These higher scores may be due to the fact that the sample included youth who were being “treated” under a provincial Youth Protection Act in addition to those serving custody under the Young Offenders Act. Therefore a significant proportion may have been more accurately described as “emotionally disturbed” rather than “delinquent.” In the present study all participants had been adjudicated through the Young Offenders Act.  61.  availability of space. On some occasions, Program 1 was reserved for offenders with special behavior problems, therefore the claim that all subjects were randomly assigned to all three program cannot be made. However, the mean moral reasoning WASs of the subjects from each of the three programs were not significantly different, F(2,89)  =  .67,  ns. They also did not differ in their age or delinquent history. (There were, however, some differences in the behavior patterns of the participants from Program 1. In terms of their CBCL scores, they had higher scores on Immaturity than the residents of Programs 2 or 3, and scored higher on Aggressive and Hyperactive subscales that those in Program 3.) Climate Institutional Climate Having determined that the participants from each program did not differ in their reasoning about hypothetical moral issues, it was next important to ascertain that their reasoning about more real-life circumstances (i.e., climate issues) did differ, because of the above noted differences in program intensity. One indication of the differences in the participants’ perceptions of the three  programs was found in the Correctional Institutions Environment Scale (CIES) scores (Table 3). A 3 (Program) X 2 (Time) analysis of variance (ANOVA), for each of the dES subscales, with repeated measures on the last factor, and CIES subscale scores as the dependent variable, was used. A main effect for program was found for Involvement, Support, Expressiveness, Practical Orientation, and Personal Problem Orientation. Subsequent Tukey Multiple Range Tests (with a  =  .05) scores revealed that Program 1  was higher for all five of these subscales (Table 3), and that Program 2 was higher than Program 3 on the Expression subscale. Thus, the respondents were able to observe differences between programs within the first 7 days and, 30 days later, were reliably able to do so again (i.e., there was no  62.  Table 3  CIES Subscales across Three Programs (Collapsed over Time)  Subscale  Program #1  Program #2  Program #3  Involvement  6.09a  Support  6.16a  Expressiveness  4.22a  3.60b  2. 90c  Autonomy  4.73  3.77  3.97  Practical Orientation  6.OOa  b 0 5 4  b 440  Personal Orientation  4.96a  3.28b  .6lb 2  Order  5.97  5.07  5.12  Clarity  5.17  4.05  4.37  Staff Control  4.47  4.72  4.63  377b  8b 4 . 3  4.42b  Note: Within subscales, different superscripts indicate significant differences between units.  effect for time for any of these subscales). It can therefore be argued that there do exist some reliable differences between programs in at least one type of measure of climate (dES). The participants from Program 1 felt more involved and active in day-to-day program activities, more supported by staff members and encouraged to be supportive of others, more likely to express emotions, and more encouraged to explore future plans and personal problems. Moral Climate Stage of Norm As described in the Method section, the moral climate of each of the programs was assessed in two ways. The first of these (Stage of Norm) involved an attempt to index and score the reasoning that supported group norms operating in each of the  63.  programs. This was accomplished by soliciting from each subject answers to a series of questions regarding the general reasoning within the unit, about “helping,” “stealing,” and “contraband.” More specifically, after a problem involving one of each of these three norms was presented to them, the participants were asked, “What would most others on this unit do in this situation?” and (more importantly), “Why would they do  “(see  Appendix B). The responses to the latter question were scored in the same manner that responses to the Moral Judgment Interview, by matching the verbal response to the criterion judgment contained within the Standard Issue Scoring Manual (Colby & Kohlberg, 1987b). That is, the scores on this variable also were weighted average scores (WAS), ranging from 100 to 500.16 Table 4 displays the data for the Stage of Norm variable, for the norms of helping, stealing, and contraband.  In order to determine if participants acknowledged the  differences between programs in the same way that they did in their response to the dES, a 3 (Program) X 2 (Time) ANOVA was carried out for each norm separately, with repeated measures on the last factor, and with Stage of Norm scores as the dependent variable. The respondents found Program 1 to be at a higher level of reasoning about the helping norm, but not the stealing or contraband norms (i.e., an effect for program was found for helping only, F(2,54) tests (with cx  =  =  11.84, p  <  .001), and subsequent Tukey Multiple Range  .05) indicated that Program 1 was higher than both other programs. As  with the dIES scores, these WAS measures of moral climate were consistent across two points of measurement. The one exception to this pattern was the level of reliability for the stealing norm within Program 2, where a statistical effect of time, F(2,56)  =  4.68, p  <  .05, was found, indicating higher stage terms at posttest than pretest (from 195 to 218). 16  Unlike scores from the Moral Judgment Interview, which are almost always based on several matches between responses and criterion judgments within the manual, the scores representing the stage of norm were often based on only one match. Because it was usually possible to achieve only one match, it was also frequently the case that responses could be scored for one norm but not another, thus creating unequal numbers of missing data. This will be evident in the tables throughout this section.  64.  Table 4  Stage ofNorm Scores across Three Programs (Collapsed over Time)  Program  Helping Norm  Stealing Norm  Contraband Norm  #1  243.1  (40.9)  208.1  (28.3)  196.8  (20.1)  #2  212.9  (30.2)  206.3  (32.7)  199.9  (11.8)  #3  208.5  (24.7)  196.7  (28.4)  202.4  (15.9)  Total  221.5  (32.7)  203.7  (29.6)  199.7  (15.8)  Note: Standard deviations are indicated in parentheses.  However, in general the residents of Program 1 tended to see the reasoning of others  about helping in higher stage terms than the participants from the other two programs; most participants were able to catalogue the group norms within the first 7 days of their residency, and persist in that view for the next 30 days. These scores may reflect the particular efforts of the staff of Program 1 to provide a scaffold for helping. Although Program 2 had been conducting unit meetings for a period of 2 3 months, such efforts were not reflected in these scores; thus, Hypothesis 5 -  was only partially supported in that both CIES scores and Stage of Norm scores (for helping) were higher only for the program which had used unit meetings for an extended period. Stage of Community The second moral climate variable, called Stage of Community was a global rating (from 1 to 5, see Appendix C) based on each subject’s response to the series of questions posed about the moral climate (Appendix B). In general, this variable represents an independent rating of each subject’s sense of collectiveness with and mutual commitment  65.  to the other residents of his living unit. Like the Stage of Norm variable, scoring of this variable was done separately for each norm because there were important distinctions in the way the participants expressed this collectiveness. The differences between programs in these Stage of Community scores are detailed in Figure 2. As can be seen from an inspection of this figure, differences were most evident with respect to the helping norm. This was determined by a 3 (Program) X 3 (Norm) X 2 (Time) ANOVA, with repeated measures on the last two factors, using Stage of Community scores as the dependent variable.  This analysis revealed the  predicted main effect of program on Stage of Community ratings, F(2,87) .001; a main effect for norm, F(2,350)  =  16.84, p  program and norm, F(4,350)  <  .001.  =  7.88, p  <  =  21.95, p  <  .001; and an interaction between  The locus of the interaction between  program and norm was determined by an analysis of the simple main effects of program for each norm separately. The analyses for the helping norm and the stealing norm were significant: F(2,69)  =  21.99, p  <  .001; and F(2,85)  Subsequent Tukey Multiple Range Tests (with cc  =  =  9.93, p  <  .001, respectively.  .05) found that for the helping norm,  Program 1 was significantly higher than Program 2, and Program 2 higher than Program 3. For the stealing norm, Programs 1 and 2 were similar but both higher than Program 3. There were no differences among programs for the contraband norm (Figure 2). These data added more support for Hypothesis 5 which claimed that programs of greater intensity would have higher moral climate scores. Unlike the Stage of Norm scores, for this variable, Program 2 (with the shorter history of unit meetings) was higher than Program 3 (which had conducted no unit meetings) for two of the three norms. For the contraband norm, for both Stage of Norm and Stage of Community, the residents of all three programs were “of like minds,” that is, there were no significant differences found between programs. It could be argued that for these participants, the contraband norm (particularly) is not a moral issue, but rather a social convention. This distinction between moral and nonmoral issues will be considered in more detail in the  66.  Figure 2 Stage of Community Scores (for Three Norms) within Three Programs  4 •Program 1 EProgram 2 Eliprogram 3 3  z  2  1  0  Helping  Stealing  Contraband  NORM Discussion section.  Moral Climate and Improvement in Moral Reasoning Having established that the participants of the three programs differed in their perceptions of the moral climate operating in their units (so that we can make reference  67.  to “better” moral climates), it seems essential to determine what influences such differences in these climates had. To this end, Hypothesis 4 held out the prospect that those participants from programs with better climates would improve more in level of moral reasoning than those from programs with less positive climates. A 3 (Program) X 2 (Time) ANOVA was performed, with repeated measures on the last factor, using WAS (pre-, post-test) scores as the dependent variable. Although many participants improved in their level of moral reasoning during the 30-day interval (i.e., a significant effect for time was found, F(1,78)  =  30.99, p  <  .001), the level of improvement demonstrated did  not differ from program to program (i.e., there was no significant interaction between program and time, F(2,78)  =  .04). Hypothesis 4 was not supported.  Moral Climate and Improvement in Behavior If the level of participant’s moral reasoning did not improve as a function of the quality of the moral climate, what, one might well ask, would be the expected relation between the quality of moral climate characteristic of each participant’s program and their actual behavior? Hypothesis 2 stated that participants from programs with higher moral climates would also engage in less disruptive and antisocial behavior. To explore the effect of climate on behavior, a 3 (Program) X 2 (Time) ANOVA, with repeated measures on the last factor, and Child Behavior Checklist (CBCL) scores 17 as the dependent variable, was carried out. This was done separately for each of the three CBCL subscales (Delinquency, Aggression, and Hostile/Withdrawn).  A significant  interaction of program and time was found for all three subscales: Hostile/Withdrawn, F(2,77) =  =  8.47, p  13.24, p  <  <  .001; Delinquency, F(2,77)  =  9.92, p  <  .001; and Aggression, F(2,77)  .001. The subsequent analysis of simple main effects of time for each  program (with a  =  .05) revealed that scores on all three subscales improved for Program  1; Program 2 improved in Hostile/Withdrawn and Delinquency only; while scores in 17  behavior frequency ratings failed to distinguish in any way among the programs.  68.  Program 3 for all three behavioral subscales worsened. Hence it can be argued, at least on the basis of the CBCL ratings (unlike the direct observation scores), that Program 1 with the higher moral climate scores (and to a lesser extent, Program 2) was more effective in reducing delinquent and antisocial behavior within the institution. Moral Climate and Moos CWS Since levels of program intensity were reflected in both CIES scores and the two moral climate variables, it would reasonable to expect that there might also be a correlation between these two types of measures. Hypothesis 6 stated this specifically by arguing that there would be a positive correlation between the CIES Relationship subscales of Involvement, Support, and Expressiveness, and the Stage of Community scores. The first of these subscales, Involvement, reportedly measures how active and energetic the residents are in day-to-day functioning (e.g., “interacting and developing pride and group spirit in the program,” Moos, 1975, p. 41). The second, Support, is meant to reflect “the degree to which residents are encouraged to be helpful and supportive toward other residents, and how supportive staff members are toward residents” (Moos, 1975, p. 41). Finally, Expressiveness is said to measure the degree to which the expression of emotions (including anger) is encouraged. The Stage of Community scores for the helping norm produced the strongest of the correlations observed (Table 5). It is of interest that, of the three norms, the helping norm, as measured by the Stage of Community variable, produced the strongest correlations with the CIES subscales, particularly when using the posttest scores. That is, the degree to which the participants saw each other as working together in some form of community, especially around issues of helping, significantly predicted scores on four CIES subscales. The degree to which they thought they worked together to resolve or respect property issues (i.e., the stealing norm) still predicted two of the four CIES  69.  Table 5 Correlations between Moral Climate (Stage of Community) and CIES Subscales (Posttest) Norm  CIES Subscale Support Involvement Expressiveness *  p  <  Helping 53* •54* •43*  Stealing  Contraband  .15  -.15  .30*  .00  .11  -.21  *  .05  scores. Finally, the degree to which unit members saw themselves working together to avoid problems around contraband drugs and cigarettes (i.e., the contraband norm), did not predict these scores. In short, Hypothesis 6 did receive some support, in that the several measures of moral climate adopted here (about which no reports have yet appeared in the research literature) did relate strongly to the dimensions indexed by the CIES. However, the data for the stealing and contraband norms seem to indicate that the dynamics of this sample are quite different around those norms, and that analyses such as this, across several norms, are necessary to provide a richer appraisal of climate. The significance of the variation of scores between norms will be considered in the Discussion section. The reader should remember that the focus of Programs 1 and 2 perhaps came closest to what is tapped here by questions about the helping norm. That is, by regularly encouraging and teaching the residents to help others when in need, the staff were demonstrating that this norm was important to them. They assumed, along with Vorrath and Bendtro (1985), that encouraging this norm (as a kind of “hook’) would effect a  70.  spread of hoped for treatment effects to other norms, and counteract other antisocial influences. This assumption received only partial support for the stealing norm and less still for the contraband norm  -  norms which may reflect what was earlier referred to as  “the hidden curriculum” of the group (Mosher, 1980). Up to this point, then, it has been demonstrated that the three programs can be distinguished in terms of their moral climate and institutional climate (CIES) scores, and further, that although there was no evident effect of the quality of the climate on the level of moral reasoning, the programs did have differential behavioral effects. Climate was the first type of contextual variable investigated here; the second concerned the impact of peer status on moral discussion. The results involving this variable are reported in the next section, which in turn will be followed by other findings, important but not directly related to the hypotheses. Effects of Peer Discussion on Moral Reasoning As outlined earlier, dyadic discussions involving cognitive conflict in moral reasoning have previously been found to promote higher levels of reasoning for those participants who were paired with someone who reasoned at higher levels than themselves. One purpose of this research was to replicate such findings. To do this, the first data set, comprised of 40 target subjects and 10 members of a control group, was used. Hypothesis 3 stated that those participants whose discussion partner was of the same or a higher stage of moral reasoning, would demonstrate a significant improvement in their moral reasoning between pre- and post-test. It also stated that the effects of exposure to higher stage reasoning would be constrained by the status of the “other” participant if the status of the other was lower, and the positive intervention effect was expected only for those subjects paired with a similar or high status partner (see Table 1). In order to test Hypothesis #3, the types of dyads (Table 1) were classified into  71.  Table 6 Effects ofHigher Status and Higher Stage Combined  Group  Pretest WAS  “Most likely”  (ii =  “Least likely” (n ‘  Control (n  =  10)  =  Posttest WAS  Gain  18)  215.6  238.7  23.1  22)  253.4  259.8  6.4  235.7  236.3  0.6  three groups: (a) those for whom change was thought most likely, that is, the target was of lower or equal status, and of lower (or at the same) stage in moral reasoning (Types #1, 5, and  7);18  (b) those for whom change was thought least likely, that is, the target  participant was paired with a dyad partner of lower moral stage, regardless of status (Types #2, 4, 6, and 8), or (in the case of Type #3) a dyad partner of similar or higher stage but lower status; and (c) the control group (see Table 6). Using a 3 (Group) X 2 (Time) ANOVA, with repeated measures on the last factor, and moral reasoning as the dependent variable, an overall significant interaction between group and time was found, F(2,47)  =  3.62, p  <  .05. The locus of the interaction  between group and time was determined by analyses of the simple main effects of time for each group separately. Only one effect was significant, for the group thought most likely to change (Dyad Types #1, 5, 7), F(1,28)  =  11.32, p  <  .01.  Thus, when the  variables of stage disparity and peer status were considered together, the combined effect represented a significant improvement over time. Neither the control group nor the group comprised of target subjects whose level of moral reasoning was higher than that of their  18 grouping was thought to represent the combined effects of high status and high stage of the target.  72.  Table 7 Effects ofDyadic Intervention  Dyad Type #1 (n  Status of Target  Stage of Target  5)  lower  Pretest WAS  Posttest WAS  Gain  lower/same  201.2  236.4  35.2  #2 (n  =  3)  lower  higher  266.7  273.7  7.0  #3 (n  =  7)  higher  lower/same  228.1  231.0  2.9  #4 (n  4)  higher  higher  267.2  281.0  13.8  #5  5)  higher/same  lower/same  225.6  247.0  21.4  (ii =  #6 (n  =  4)  higher/same  higher  280.0  290.0  10.0  #7 (n  =  8)  lower/same  lower/same  218.4  234.9  16.5  #8 (n  =  4)  lower/same  higher  254.0  253.7  -.3  235.7  236.3  0.6  236.3  247.5  11.2  Control (n Total (n  =  10)  50)  dyad partner improved over the time interval. Table 7 indicates the actual changes in moral reasoning scores for each of the eight dyad types. In order to determine if separate effects of status and stage could be isolated, another ANOVA was carried out a 2 (Stage: target lower/same, or higher) X 4 -  (Status: target higher, lower, higher/same, or lower/same) X 2 (Time), with repeated measures on the last factor and moral reasoning scores (WASs) as the dependent variable. This analysis revealed a main effect for time, F(1,41)  =  15.20, p  <  .001, but no  significant interaction between stage and time, or status and time, or status, stage, and time. Thus, it cannot be argued that for this sample exposure to higher stage alone had any developmental effects.  Similarly, it cannot be argued that those paired with  73.  participants of same or higher peer status improved any more than participants paired with those of lower status. Hypothesis 3 concerning peer status was supported by these findings (that is, although there was no separate main effect for peer status, these results do lend support to the notion that peer status, or credibility, exercises a real effect on the likelihood that interaction with a peer of higher moral reasoning competence will have an effect on one’s own moral reasoning level). The preceding sections have noted the significant differences in stage of reasoning of the stealing and particularly contraband norms. Such variability has been the concern of other researchers (Haan, Aerts, & Cooper, 1985; Krebs, Vermeulen, Carpendale, & Denton, 1991; Walker et at., 1987), but is usually construed as differences in reasoning between hypothetical dilemmas and real-life ones. The following section looks more closely at such variability found in this sample. Such findings are interesting in their own right, but are indirectly related to the hypotheses. Moral Reasoning Across Content and Context Hypothetical versus Real-Life Dilemmas With a variety of theoretical axes to grind, several research groups have examined differences in reasoning about hypothetical and real-life problems, particularly between the classic Kohibergian dilemmas and those generated by subjects from personal experience. The trend is that people use lower structures in reasoning about real-life issues than when confronting purely hypothetical ones. Table 8 compares the moral reasoning scores from the Moral Judgment Interview with reasoning about the real-life helping, stealing, and contraband dilemmas (i.e., scored responses to the question in the Moral Climate Interview “What should one do...? Why?”). All of these real-life dilemmas were thought to be common situations for the  participants. A 3 (Program) X 4 (Dilemma Type) X 2 (Time) ANOVA with repeated  74.  Table 8 MoralReasoning in Hypothetical and Real-Life Dilemmas (Collapsed over Time) Real-Life Program  Hypothetical  Helping  Stealing  Contraband  Mean  #1  236.1  256.4  221.1  210.2  230.9  #2  241.2  239.9  221.7  204.4  226.8  #3  231.8  243.7  207.4  205.5  221.1  Total  238.5  246.3  216.8  208.8  227.6  measures on the two last factors, determined a statistical main effect for dilemma type, F(3,540)  =  44.05, p  <  .OOi.’ Subsequent Tukey multiple range tests (with a  =  .05)  revealed that, in general, all programs were higher on hypothetical Kohibergian dilemmas than stealing or contraband issues (consistent with previous research), but also that reasoning about real-life helping was higher than reasoning about hypothetical dilemmas. Again, there are significant fluctuations in reasoning depending upon the dilemma. There were no other significant effects. These data about the helping dilemma involve greater variability in reasoning across content than has been previously documented. The trend is in contradiction to that found by Leming (1973), Walker et al. (1987), and Trevethan and Walker (1989), and challenges the notion that reasoning about the hypothetical (responses from the Moral Judgment Interview) represents the limit beyond which moral reasoning cannot exceed  19  There was also an interaction between dilemma type and time. The locus of this interaction was determined by an analysis of the simple main effects of time for each dilemma type separately. The analysis for reasoning about hypothetical issues only was significant, F(1,90) = 22.04, p < .001, indicating that such reasoning was the only one to improve over the testing interval, a finding that has been discussed earlier.  75.  (see Bearison, 1986, P. 143). However, the reasoning about stealing and (even more so) accepting contraband was consistent with previous research. Such reasoning was at a significantly lower level than reasoning about the hypothetical dilemma. Judgments of Responsibility and Practicality Further to this discussion, Leming (1973, 1976) found a progressive lowering of stage scores as subjects moved from reasoning about responsibility issues to more practical issues. What Leming called judgments of responsibility are the responses to the question: “What should the character in this dilemma do, and why?” Judgments of practicality are those statements in response to the question “What would you do, and why?” Judgments about others (discussed earlier as the Stage of Norm variable) are the responses to the question, “What would most other guys on this unit do, and why?” All three types of judgment are part of the Moral Climate Interview. However, in the Moral Climate Interview (following the lead of Power et a!., 1989), a fourth question was asked: “Do you think there should be an understanding or agreement on this unit about helping/stealing/contraband? Why?” This question, about the ideal, elicited responses which, when scored, were generally higher than any of the other responses. Table 9 indicates a progressive lowering of WASs for three of these four types of responses. A 3 (Program) X 4 (Judgment Type) X 2 (Time) ANOVA with repeated measures on the last two factors, was carried out, for each norm separately. A main effect for judgment type was found for the helping and stealing norms, F(3,404) p  <  .001; and F(3,476)  =  34.56, p  <  =  22.23,  .001, respectively. Characteristically, there were no  differences in judgment type for the contraband norm, F(3,404)  =  .46, ns.  For the stealing norm, subsequent Tukey multiple range tests (with a  =  .05)  revealed that judgments about the ideal were higher than all other types of judgments, and that judgments of responsibility and practicality (what should and would be done)  76.  Table 9 Judgments ofResponsibility, Practicality, Others, and the Ideal (Collapsed over Time)  Norm Judgment Type  Helping  Stealing  Contraband  Judgments of Responsibility (“What should be done?”)  246.3  (41.9)  216.8 (32.6)  208.8  (19.6)  Judgments of Practicality (“What would you do?”)  239.1  (40.4)  209.8  (24.3)  208.8  (19.6)  Judgments for Others (“What would others dO?”)a  222.8  (35.9)  203.8  (28.3)  205.9  (24.2)  Judgments for the Ideal (“Should there be an understanding?”)  256.6  (53.7)  236.0 (33.8)  226.5  (40.0)  Note: Standard deviations are indicated in parentheses. a This variable is also called the Stage of the Norm, as noted earlier.  were not different from each other, but higher than judgments of what the others would do. This same pattern did not hold true for the helping norm. For helping, there was an interaction between judgment type and program, F(6,404)  =  3.5, p  <  .01, which  proved relevant to the hypotheses concerned with program differences and moral climate (see Figure 3). The locus of the interaction between judgment type and program was determined by analyzing the simple main effects of judgment type for each program separately. Only the results for Programs 2 and 3 were significant, Fs(3,135) 22.5, p  <  =  12.9 and  .001, respectively. Participants from Programs 2 and 3 made judgments of the  reasoning of others in lower stage terms than they made judgments of responsibility and practicality (similar to the pattern shown in Table 8).  However, participants from  77.  Figure 3 Accommodation in Reasoning about Helping by Judgment Type and Program  260  250  240  230  220  210  Responsibility  Practicality  Others  Ideal  JUDGMENT TYPE  Program 1 made consistent judgments across all types. In other words, the participants from the better moral climate of Program 1 perceived the reasoning of others in significantly higher stage terms, similar to their own reasoning about should and would questions, and the participants from Programs 2 and 3 -  even though they were capable of higher stage reasoning  -  estimated the reasoning of  others at a significantly lower level, as much as 28 WAS points, approximately 1/3 of a  78.  stage.  This adds to the validity to the moral climate (Stage of Norm) variable by  demonstrating that estimates of the reasoning of others are not simply a “projection’ of each participant’s own reasoning. It also lends to the support for the concept of the Stage of Norm reported earlier (Table 4). Moral Reasoning and Moral Intentions Although not part of the initial hypotheses, another important finding of this study has to do with the relationship between the actual decisions or intentions 20 expressed by the participants and the structure of their reasoning. It was found that regardless of the intent stated (i.e., whether one should. or would help, steal, or accept contraband) there were no differences found in moral reasoning about hypothetical issues. That is, those who scored higher on Kohlberg’s standard measure did not tend to more often state prosocial intentions than those who scored lower. However, with only one exception (Table 10), when the reasoning about more real-life issues (i.e., specific norms) is examined, the reasoning about prosocial intentions is higher than reasoning about more antisocial ones. That is, when justifying their stated intentions about what they should or would do, the participants whose deliberations were prosocial used higher stage reasoning than those who indicated that they should or would steal or accept contraband. Even though there were no differences in moral reasoning competence between those who intended to help and those who would not help (or steal, or accept contraband), there were differences in what Kohlberg would call moral reasoning performance. Thus reasoning about real issues tended to be a better predictor of moral choices that the reasoning measured by the MJI.  20  Reference here is made to “intentions” rather than “choices.” In order to use the word “choice” or “decision,” one would have to have observed the participant engage or follow through in a specific act. Here, answers to the question, “What would you do?” are considered normal practices, plans, or deliberations, and have been labelled “intentions.”  79.  Table 10  Relations between Moral Intentions and Moral Reasoning (Expressed in WASs)  Norm  Judgment Type  Prosocial Choice  Helping  Judgment of Responsibility (“Should one...?”)  260.6 = 61)  185.4 (n = 12)  47.2**  (n  Judgment of Practicality (“Would you ...?“)  248.5 (n = 51)  192.3 (n = 18)  31.7**  Judgment of Responsibility (“Should one ...?“)  215.6 = 79)  206.2 (n = 8)  .7  (n  Judgment of Practicality (“Would you ...?‘)  209.3 (ii = 69)  203.3 (n = 9)  4.1*  Judgment of Responsibility (“Should one ...?“)  212.9 (n = 41)  200.0 (n = 32)  6.6  Judgment of Practicality (“Would you ...?“)  208.4 (ii = 31)  198.8 (n = 42)  5.1*  Stealing  Contraband  *  . . .  Antisocial Choice  F value  **  **  05 p<.  01 p<.  Moral Reasoning and Behavior Having established that moral climate (reasoning about others) is a predictor of behavior and further, that reasoning about real-life issues is a similarly better predictor of moral intentions then MJI scores, it is of at least equal importance here to determine if a similar trend exists for the relation of moral reasoning and behavioral measures. As noted earlier, there were two measures of behavior used in this study: a direct count of disruptive behaviors by staff members (Appendix E:  Behavior Frequency  Rating) for a period of 7 days, and a general rating of categories of behaviors (the Achenbach Child Behavior Checklist, CBCL), also completed by staff members. The behavior frequency rating involved counting the frequency of 11 behaviors or events.  80.  These were subdivided into two categories: either resisting/noncompliance (swearing, not following a rule), or aggression/intimidation (fighting, provoking, challenging staff, etc.). Hypothesis 1 proposed that those residents higher in moral reasoning would engage in less disruptive or antisocial behavior than those lower in moral reasoning. In order to test this hypothesis, a correlational analysis of CBCL scores and moral reasoning scores was undertaken (Table 11). None of the correlations between the MJI (WASs, of hypothetical dilemmas) and any of these behavioral measures was significant. Similarly, none of the types of reasoning about the helping norm covaried with any of the behavioral measures. However, the reasoning of the participants about the stealing norm and the contraband norm proved much more predictive of how they actually behaved.  Table 11 documents the relation between moral reasoning performance and three subscales of the CBCL. 21 As stated earlier, reasoning about stealing and particularly contraband seemed much less affected by the level of intensity of the program. Table 11 indicates a trend of stronger correlations between reasoning about these kinds of issues and behavioral measures than reasoning about more hypothetical issues (MJI scores) or reasoning about issues that were important to the staff (helping). The same table also indicates that the higher the stage in reasoning about contraband were, the lower (less deviant) the scores on the Hostile-Withdrawn and Aggression CBCL subscales were. Hypothesis 1 was not directly supported by these data however, in that it refers to moral reasoning competence as measured by the MJI. particular issues  -  Nevertheless, it does appear that moral reasoning about  issues that are more entrenched within the peer subculture of the unit  (as suggested earlier), or more closely related to the focus of the behavioral measure is -  more predictive of behavior than reasoning about other issues. 21  None of the other subscales of the CBCL were found to covary with moral reasoning, nor were they expected to. Also, it is these three subscales that best reflect the goals of the overall program to reduce aggression, passive-aggressive resistance, and antisocial behavior. -  81.  Table 11 Correlations between MoraiReasoning and CBCL Subscales  CBCL Subscale Type of Moral Reasoning  Delinquency  Aggression  Hostile/Withdrawn  -.15  .23*  -.12  -.19  -.15  .12  .02  -.18  .07  Wouldyou steal?  .41*  .24*  Should there be an agreement?  .25*  .25*  -.16  -.21  -.24  -.40  Moral Judgment Interview Helping Norm Would you help? Should there be an agreement? Stealing Norm  -.05  Contraband Norm Would you accept  .  it?  Should there be an agreement? *05  **  p<.Ol  *  *  -.21  -.60  *  **  82.  DISCUSSION This study was based on the assumption that wherever there is a group whose members live together or have frequent contact and some degree of mutual dependency, there is also a culture and, while others may disagree, a culture with moral dimensions. Whether they are within a classroom or a correctional facility, group members inevitably come to know each other and develop relationships; and I have earlier argued, as have Kasen, Johnson, and Cohen (1990), Poisky (1962), Power et al. (1989), and Vorrath and Bendtro (1985), that the possibility always exists that the culture (or moral climate) can influence members of the group in an unproductive, even harmful manner. The alternate prospect is that, if encouraged and guided (presumably by adults), the moral climate of such a group can facilitate development and guide prosocial behavior. In order to demonstrate the correlates of moral climate, it was seen as important here to first determine whether there were indeed better or worse climates operating within the institution. In this respect, the Moos Correctional Institutions Environment Scale (CIES) proved useful because it was sensitive to the responses of the participants in the three different programs. In particular, Program 1 (with its greater intensity of moral education) was consistently and significantly higher than the two others on most of the subscale scores of the dES. Similarly, these same programs differed on both moral climate measures (Stage of Norm and Stage of Community) for the helping norm (and to some extent, the stealing norm). As such, this study is the first to evidence supporting validity to the concept of moral climate, and to demonstrate positive correlations between specific subscales of the Moos CIES and the Stage of Community variable. The fact that such correlations appeared for the helping norm only proved to be informative, and is discussed in more detail later in this section. Moral Judgment, Moral Climate, and Behavior Next, it was seen to be important to determine what effects such climates had on  83.  the development of moral reasoning and on behavior. Some readers of Kohlberg’s theory have expected a direct relation between the stage of one’s moral reasoning and the quality of one’s moral actions. Simplistic connections have not been found however (Buchanan, 1992), nor in fact did Kohlberg hypothesize them, although a positive statistical relationship has frequently been established (Arnold, 1989; Buchanan, 1992; Nelson, Smith, & Dodd, 1990). In his early review of these correlational studies that examined the relation between moral reasoning and behavior, Blasi (1980) argued that it should be possible to go beyond the search for mathematical correlations and enquire about the roles that different elements play in the overall scheme in which moral action is the natural end product. (p. 2; italics added) In the present study, moral reasoning (as measured by the MJI) did not predict any of the behavioral measures, but other kinds of moral reasoning  -  reasoning about  real-life issues and the perceptions of others’ moral reasoning did predict the behavioral -  22 That is, some of the evidence presented did provide some ground upon which scores. to argue that moral climate is one of the elements to which Blasi referred. Participants from the two programs in which unit meetings were conducted (which also had a better moral climate) were also shown to become significantly lower in antisocial behavior ratings over the 30-day interval, in comparison to those from the program that had no meetings, who became higher in such ratings. Therefore, it may be that moral climate does help explain the nature of the relation between moral reasoning and moral behavior.  In his effort to explain this  relationship, Kohlberg (Kohlberg & Candee, 1984a) proposed a four-function model: (1) interpretation of the situation (i.e., social perspective taking);  22  my knowledge, using narrowly defined samples of delinquents such as this one, no one has found a relation between MJI scores and behavioral ratings such as the CBCL, although some research has found relations between type of offense and moral reasoning (for example, Moran, 1988, who found that status offenders and sex offenders were higher in moral reasoning than other types.)  84.  (2) decision making (deontic choice, i.e., one should...); (3) follow-through (moral judgment of responsibility or obligation; therefore I will ...); (4) follow-through (nonmoral skills, i.e., intelligence, attentional capacity). Kohlberg and Candee claim that deontic choices are not usually made at the lower stages (Stages 1, 2 and 3), thus, at the lower stages, for persons such as those in this study, there may be only three functions (I, III, and IV). Based on this research, I would propose that Function I includes factors such as moral climate, and further that moral judgment can be a complex process that includes some parts of Function I (interpretation of the situation) and Function III, that is, simultaneous interpretation of the moral climate (what others would do and why) as well as personal judgment. In some cases, moral statements or judgments cannot be adequately interpreted without reference to the moral climate of the group. Kohlberg makes a similar claim, in referring to what he called “the structure of the outside world.” The cognitive-developmental assumption is that basic mental structure is the result of an interaction between certain organismic structuring tendencies and the structure of the outside world, rather than a reflection of either one directly. (Kohlberg, 1984, p. 13) Such ideas are not dissimilar to Vygotsky’s notion of “scaffolding” (Wertsch, 1984) and portray moral development and moral judgment as much more a socio cognitive (Light & Perret-Clermont, 1989) or transactive process (Bearison, 1986) than purely a reflective one. Because program intensity (and moral climate) did not, as predicted, have an effect on the development of moral reasoning, the question arises, why might it be that exposure to a better moral climate would in fact successfully produce better behavior ratings but not better moral reasoning?  One interpretive possibility is that if, as  suggested by Piaget (1974/1980a) and others (Chapman, 1988), cognitive changes are always the result of newly interiorized actions and so should follow rather than precede behavioral changes, then behavior and behavioral changes play a role in the development  85.  of moral reasoning. In fact, some theorists (Youniss & Damon, 1992, P. 277) have argued that Kohlberg had taken the position of Piaget and turned it on its head, suggesting that Kohlberg’s success at isolating the purely structural, cognitive aspect of moral judgment inevitably led him to the conclusion that such cognition is prior to moral action. It could be argued that what was documented here were beginning improvements in behavior, which are related to moral climate and would have become developmental changes in the structure of the moral reasoning, if optimal conditions continued for a longer period. Real-Life/Hypothetical Issues and the Concept of Structure The findings of this study support the notion that not only does reasoning about real-life moral dilemmas predict morally relevant patterns of behavior more effectively than does reasoning measured by the abstract Moral Judgment Interview, but that such reasoning also better predicts moral intentions (cf. Table 10). Why would reasoning about real-life, rather than hypothetical issues, correlate more strongly with behavior and intentions? The view proposed here is that the stealing and contraband dilemmas tapped normal or everyday moral reasoning (which includes interpretation of the moral climate  -  Function I), and that it is this reasoning which is  more likely to impact moral decision-making. It is of note that the WASs for these dilemmas (stealing and contraband) did not correlate with the CIES (Correctional Institutions Environment Scale). One of the criticisms of the CIES (Moos, 1975) has been its poor ability to predict behavior either within an intervention program or after release. It may be that this measure fails to tap the real but “hidden” issues of climate. That is, the CIES appeared (in the writer’s view) to be more truly a measure of the attitudes of interest to the professional staff, on the periphery of the inmate culture, whereas the Stage of Norm scores for stealing and contraband represented a more authentic, inmate-centered view of the structure of the informal peer culture within the  86.  institution. It follows from these prospects that previous investigators have “missed the boat° in attempting to predict behavior through staff-focused measures of atmosphere or climate with instruments like the dES, which are essentially measures of attitudes (or content). Instead, it seems (from these data) that the measurement of the structure of reasoning, particularly about commonplace issues, is a much more productive avenue to pursue. For a variety of theoretical reasons, several research groups (Gilligan & Attanucci, 1988; Haan, Aerts, & Cooper, 1985; Krebs, Vermeulen, Carpendale, & Denton, 1991; Smetana & Killen, 1991; Walker et al., 1987) have examined such differences. The trend shown by the data is that people reason at lower stages, and/or display different orientations (Gilligan & Attanucci, 1988), in reasoning about real-life issues than when confronting purely hypothetical ones. Trevethan and Walker (1989) found, with both juvenile offender and normal subjects, that reasoning about real-life issues was 47 WAS points lower than on hypothetical dilemmas. Such differences have been understood to be a function of the content of the dilemmas about which moral judgments were made. Although Walker and Taylor (1991) produced similar findings in their analysis of family triads, their data (i.e., lower moral reasoning of parents on real-life issues) was interpreted as a deliberate accommodation on the part of the parents to the specific social context (i.e., level of reasoning of their child, with whom they were discussing moral problems). This influence of content and context might go some distance in answering the issue raised by Haan (1985) who states: People’s stages and levels of performance shift from one action situation to another, apparently in response to the difficulty of different action situations, but actually little is now known about why some practical moral situations are more difficult than others. (p. 104) The data here are compatible with (but not entirely) a model of moral reasoning  87.  proposed by Krebs et al. (1991), who hold that instead of moral reasoning structures being transformed and integrated into higher stage structures, individuals retain earlier structures and use them as required by the problem encountered. 23 Krebs makes this argument because, consistent with the results described here, he (Krebs et al., 1991) found that moral reasoning varies according to situation. For example, in examining the issue of impaired driving, they found that when reasoning about whether one should drive while impaired by alcohol, subjects scored over 75 WAS points (3/4 of a stage) lower than they did when reasoning about Kohibergian hypothetical dilemmas. They concluded that the pattern was consistent with the idea that high stage competence on Kohlberg’s test is necessary but not sufficient for high stage performance on other dilemmas: no subject who obtained a major Stage 2 score on Kohlberg’s test scored higher than Stage 2 on the Impaired Driving dilemma, but 40% of the subjects who demonstrated Stage 4 competence on Kohlberg’s test (i.e., scored at Stages 3/4, 4, 4/5) made Stage 2 (or 2/3) judgments on the Impaired Driving dilemma. (p. 151) Such results are also in agreement with much earlier studies summarized by Kohlberg and Candee (1984b, pp. 545-552). Krebs and his colleagues (1991) surmise that their Impaired Driving dilemma is one for which one choice (not, to drive and drink) is more highly weighted (by virtue of television media campaigns and other cultural forces) than the other. It is therefore a “closed” dilemma, unlike the classical Kohlbergian ones. They also point out that the reasoning in the media about this dilemma is lower stage, that is, one might be punished, lose their driver’s license, get a jail term, etc. They concluded that moral judgment is an interaction between the structure of the stages of moral reasoning and the nature of the specific dilemma or situation that is encountered, and go on to claim that “people are more flexible than Kohlberg assumes, and moral judgment is more structurally plastic” 23  Colby and Kohlberg (1987a) refer to this kind of model as the additive or “layer cake” model, versus the transformational model which holds that in the developmental process, earlier structures are transformed and integrated into higher level structures. Each of these models imply different views of the “structured whole” assumption.  88.  (p. 155). The position taken here is that it is more accurate to describe phenomena like those reported here as an interaction between an individual’s stage and the context, rather than “structural plasticity,” which seems to be a notion fraught with conceptual contradictions. The notion of structure involves internal consistency across contents and situations the opposite to the idea of plasticity. -  Although the variability in the present sample was not as great as that found by Krebs and his colleagues, clearly the contraband dilemma demonstrated a specific pull, 24 and may also qualify as a closed dilemma, in the sense that Krebs et al. intended. Not only did it pull for a specific choice (cf. Table 10), it pulled for lower (Stage 2) reasons. It is my contention that, like the dilemmas used by Krebs, this dilemma had within it, within this specific culture, a structural bias that accounted for lower stage reasoning. Up to this point, none of what has been discussed has posed any serious problems for Kohlbergian theory. What needs to be brought out clearly about Kreb’s claim in favor of so-called “structural plasticity” is that it is not indifferent with regard to the direction that moral reasoning might be moved by various contextual factors. Krebs is not just as content to see moral reasoning scores rise above as fall below the upper-limit index set by the performance on the MJI. However, this is precisely the problem that does arise from the data presented in this study.  Reasoning about helping was higher than  reasoning about decontextualized hypothetical issues on the MJI. If the MJI does not measure the upper limits, then, where does that leave the notion of structure, given that it was assumed by Kohlberg that when measuring a subject’s best responses, one was measuring his or her moral reasoning structural competence?  In order to address such a question, it is first necessary to review and  elaborate on Kohlberg’s notion of structure, the structured whole criterion (Walker, 1988), and Kohlberg’s use of the terms performance and competence. For an interesting discussion of the effects of “closed” dilemmas and their “pull” for specific responses, see Berndt, McCartney, Caparulo, and Moore (1984).  89.  The structured whole assumption, as understood by Kohlberg, requires same stage usage (or with only “very slight decalage,” Colby & Kohlberg, 1987a, p. 8) across a variety of contents. That is, it has been understood that the best way to empirically determine or verify this assumption was to confirm that persons use predominantly only one stage in facing a variety of dilemmas.  However, Kohlberg argues that the  assumption applies only when measuring what he calls competence)- The assumption does not apply to comprehension and preference (Rest, 1973) of levels of moral reasoning (i.e., it is acknowledged that persons understand reasons higher or lower than the ones they spontaneously produce or could state a preference for reasoning higher or lower than their “own” stage; see Walker et al., 1984). The assumption also does not apply to what he calls performance. That is, Kohlberg acknowledges that persons will use lower stage reasoning in some contexts (e.g., correctional facilities) in an effort to be more “psychologically appropriate” (Colby & Kohlberg, 1987a, p. 8). Perhaps one way to resolve this theoretical difficulty is to abandon or rethink Kohlberg’s definitions of competence and performance. If one assumes a causal relation between competence and performance (as Kohlberg did, implicitly), then one is in trouble if performance (the result of competence, i.e., reasoning about helping) is better than the cause (responses to the MJI).  If, as an alternative position, one views  competence as purely formal, inferred, and that any measurement of it is a performance measure, then one is in less theoretical difficulty.  If competence is understood as a  noncausal, formal, “pattern of skills or talents” (Chandler & Chapman, 1991, p. xi), then perhaps one is freer to acknowledge such variability as is found here, and the debate becomes centered around which test or method is the best indicator of competence (see  The reader is reminded that the competence/performance distinction here is one not universally accepted and not consistent with the definition of competence proposed by Chandler and Chapman (1991). What Kohlberg means by competence may be best summarized as moral reasoning about decontextualized hypothetical problems unencumbered by affective or personal considerations.  90.  Chandler & Chapman, 1991). Another perspective is that of Youniss and Damon (1992) who claim that Kohlberg had fundamentally misunderstood Piaget’s (1932/1965) original claim regarding moral development. Kohlberg seems to have assumed that moral reasoning was, like logical reasoning was to problem-solving, prior to and ultimately causative of moral behavior. Youniss and Damon argue that Piaget felt that moral development was much more socially constructed (or co-constructed) than Kohlberg assumed and therefore was constrained by the social context. Relying upon the notion of the “structured whole” and the pressures of “cognitive dissonance,” Kohlberg assumed that reasoning about some issues at a high stage would exert pressure upon reasoning about other issues at a lower stage. Moral and Nonmoral Issues A distinction between moral and nonmoral issues (or domains) has been proposed by Turiel (1983), Nucci (1982), Nucci and Weber (1991). They claim that nonmoral conventions are arbitrary norms that regulate members of a group or social system towards the goals of that group or system.  Such nonmoral social conventions are  regarded by many people as relativistic, contingent on rules, subject to the commands of authority and thus alterable and nonuniversal.  Moral rules and principles are  generalizable, prescriptive, independent of authority dictates, and regarded as more important than nonmoral conventions. Berkowitz, Guerra, and Nucci (1991) and Nucci (1982) use this distinction to explain why, for example, some forms of drug use are not correlated with moral reasoning many adolescents regard drug use as a conventional or -  strictly personal event rather than a moral transgression. However, for Kohlberg, the distinction between moral and nonmoral dilemmas is blurred in moral climate research; he would argue that violations of what Nucci and Turiel would call nonmoral issues could cause the same hurt or disappointment that violation of a “moral” convention like  91.  fairness would (Power et al., 1989).  It is difficult to determine if the rules about  accepting contraband in this study were moral or nonmoral conventions. Certainly the author assumed them to be moral questions when designing the study, but they could be otherwise in the minds of the participants, that is, in these cultures, it could be simply expedient to accept or decline contraband drugs (with no possibility of psychological hurt to others that contravention of other norms might entail). Certainly, like many real-life dilemmas, there were costs to the individual involved in one or both sides of the dilemma. Krebs et al. would describe the contraband dilemma as “closed.” Regardless, it is interesting to observe how consistent the reasoning about contraband, especially, was between and within units. Even in the unit with the better moral climate scores, reasoning about contraband was at the same stage as the other units. This norm seemed impervious to the attempts of the staff to improve the moral climate. Perhaps what can be drawn from these findings is the recommendation that the focus of the program be broadened from helping, to include a more intensive emphasis on other specific concerns, such as property issues and drug use. More will be said of this later in this section. Equilibration Theory Although the review of the literature (concerning the effects of discussion with a peer of higher stage moral reasoning competence) indicated that there was reason to believe that an intervention as brief as the one used here (two dyadic discussions for 3 consecutive days) would prove effective, stage disparity alone did not produce a significant improvement in moral reasoning (in fact, even when those cases where stage disparity was significantly higher more than 20 WAS points the separate effect of such -  -  differences in disparity could not be teased out). That is, contrary to what one would expect from the so-called +1 paradigm, stage differences did not account for changes. Status differences appeared strong, but alone, also did not produce significant results.  92.  Had they, such a result would have proposed a problem for equilibration theory and the +1 paradigm,” which claims that at least exposure to reasoning at a higher stage is required for developmental change. The combination of these two variables, making a group which has been called “most likely to change,H did produce a significant effect. 26 This points to the role of peer status, or credibility, as exercising a real effect on the likelihood that interaction with a peer of higher moral reasoning competence will have an effect on one’s own moral reasoning level. These results may echo the findings of others: for example, those of Damon and Killen (1982) who claimed that it was not merely disagreement between peer discussants, but subjects who were able to transform and accept the statements of others (that is, those who saw the other as credible) who showed the greatest developmental gains. These results are not discouraging, however clearer they could have been in separating out the distinct effects of status and stage.  Had there been a longer  intervention, or a greater number of subjects, the separate effects of each of these variables may have been evident. It should also be noted that pretest disparity in moral reasoning does not guarantee an identical disparity in level of moral reasoning used by the participants in the dyadic discussion. No research has yet been reported which has examined variability in moral reasoning between an interview and a discussion among peers (although Walker & Taylor, 1991, did examine such variability in a parent/child context). It is interesting that Berkowitz and Gibbs (1983) reported that the extent of “transactive” dialogue behavior in a discussion between peers was more predictive of moral development than was moral stage disparity (although both were significant predictors). There are other methodological limitations in this study. For example, in this  26  It is possible that a naturally occurring effect of regression to the mean could explain, in part, this effect because the groups were formed on the basis of their pretest scores and the group that gained the most initially had the lowest pretest scores on average.  93.  study (as in others) status was difficult to operationalize.  Status appears to mean  different things in different settings. Participants and staff members were asked about who had “the most or least status,” but this could mean power, influence, credibility, or physical strength, or many other factors. Its meaning may have varied according to the moral climate of the unit; that is, in lesser moral climates, status could have been more likely to represent physical strength, while in better climates, it could represent influence. Regardless, I feel that this measure of status had a great deal to do with credibility, an awareness of reasons about why the target should attend to the other, which proved to be important enough to support the contention that there is a factor which has been ignored, assumed, or taken for granted in past research in this field, and which needs to be more closely considered  -  particularly when planning or directing actual school, correctional,  or peer counselling (Carr, 1988) programs.  That is, we may need to ask if those  individuals who are expected to have the influence, also have the credibility, in the eyes of those being targeted for change? To understand what is going on in such an interaction, we first of all notice that the subjects are confronted simultaneously with two domains of interpretation: The domain of discourse as intended by the other person, and the domain of social inferences about the other person. (Davidson, 1992, p. 26) Even though they may have had difficulty articulating it, the offenders and the staff in this study could both, with relative ease, recognize the status measured here. Therefore, this study pointed to the importance of a factor obvious to all in this situation  -  a factor  which interacts with the variables about which developmental theorists currently know so much more. It underscores the notion that a “peer” is not always a peer (Forrester, 1992), and adds to the research in equilibration theory by supporting the notion that the superior reasoning structures of others that effect equilibrium, is a necessary but not sufficient condition for development of moral reasoning (cf. Davidson, 1992; Turiel, Smetana, & Killen, 1991, p. 314). This is not inconsistent with cognitive-developmental theory and Piaget’s assertion that peer interaction (with its egalitarian quality) is vital, but perhaps all  94.  of the preceding discussion is an argument for removing social contextual factors from their secondary role within equilibration theory. It is also supportive of the argument made by Haan (1985; Haan et a!., 1985) who argues that disequilibration is less cognitive than social. Morality is seen, not as judgmental competence, but as a social, emotional dialectic of practical reasoning among people. Its distinctive feature and its ground is the attempt people make to equalize their relationship during disputes and in their conclusions.... The skills and tangible resources that allow conflict to be resolved not the moral understanding in itself may be the essential ingredient of moral development. (Haan, 1985, p. 997) -  -  -  -  Implications for Educational and Correctional Programs Although there are prices to be paid, there are many benefits to research conducted “in the field”  -  at least that was the author’s hope for this study.  The  correctional facility in which the research was conducted did prove rich in variations in culture and subculture, and was therefore a unique opportunity to study incarcerated adolescent subjects in their literal environment while, at the same time, achieving some control over key variables. The polarized social structure of the residential programs provided an opportunity to investigate the issue of peer status. However, it may be that there are limitations in the degree to which one can generalize from this population to others.  It is possible that the degree to which status is polarized or organized  categorically (Polsky, 1967) is not as great in schools and classrooms as it is in institutional facilities for disturbed or delinquent youth (studied here), or it may also be the case that peer status in residential programs is not the same as it is in day programs such as. schools. Having said this, however, I will still make the argument that what was learned here about peer status has both theoretical and practical implications. The Moral Climate Interview employed here sampled reasoning about moral issues which were close to the issues faced daily by these subjects, and is offered here as a guide to similar research, for example that investigating behavior patterns (e.g., drug  95.  use, gang violence) in school settings.  The price may have been that many of the  hypothesized effects would have been stronger had there been more participants, a longer intervention, and the degree of control over variables that ordinarily accompanies research in laboratory settings, for example, screening participants for psychopathy, low IQ, and previous admissions to the center. The final words here are directed towards the implications such research has for correctional and educational programs. Ruback and Innes (1988) have argued that there are specific reasons why much psychological research is irrelevant to correctional policy makers and administrators.  One such reason is that research often does not use  dependent variables that are employed and meaningful in the field, at the local level (e.g., incidence rates of disruptive behavior) and instead use variables that require inferential leaps of the policy maker. Although it is my hope that this research has some utility for youth correctional programs, it is with the admission that the variables used here do not meet some of the practical standards proposed by Rubak and Innes. For example, the measurement of moral reasoning competence and moral climate is not done easily. However, it does not require too great an inferential leap to acknowledge the potential benefits of moral education/community building practices (unit meetings, classroom meetings) such as those proposed by Hersh et al. (1979), Power et al. (1989), and Vorrath and Bendtro (1985).  There has been no shortage of publicity concerning the deleterious effects of incarceration (Caron, 1979; Ross & Gendreau, 1980, 1984; Scott & Trent, 1982) and residential treatment on offenders, particularly young offenders (Malarek, 1984; Vonath & Bendtro, 1985). Much of what is described can be characterized as the effects of a Stage 2 moral climate, involving both inmates and staff. The prison society operates its own civilization in a careful and observant way. A continuous battle is waged against the major culture, which includes custodial officers, keepers and wardens. Severe rules will evoke strong reaction from the inmate culture. Lax and poorly enforced regulations excite the subculture into seeking more privileges and  96. demanding more rights.... On the other side of the picture, an inmate may be given certain privileges by the custodial staff with the understanding that he will pass on relevant information on prison activities.... The inmate culture abhors weakness and indecision on the part of custodial forces. (Scott & Trent, 1982, pp. 24-25) Given this, it is not surprising that treatment programs have so frequently been found ineffective. What seems to be indicated here is that, in addition to programs directed at individuals, there also needs to be much more effort directed towards the dynamics between individuals, and the moral and structural aspects of such dynamics (i.e., moral climate), because of their developmental implications. I suggest this despite the pejorative implications the word “mora1 has within current correctional research. The power of the peer group, especially for adolescents, is now commonplace knowledge; but this study goes further to suggest that a peer group which is negative in its influence (and which does not involve cooperation and collaboration) can be assessed and identified in terms of its moral structure and can be held responsible for negative behavioral outcomes, which in turn can be prevented by programmatic intervention. The basic ideas of Positive Peer Culture (PPC) (Vonath & Bendtro, 1985) and moral education programs (Hersh et al., 1979; Power et al., 1989) propose an alternative to the Stage 2 moral climate.  PPC is, essentially, a Stage 3 moral climate, focused on  reciprocal helping and caring among group members. However, as noted earlier, it is assumed by Vorrath and Bendtro (1985; Tannehill, 1987, p. 115) that focusing on helping behavior with such youth will “hook” them and other antisocial attitudes will be influenced subsequently. The general concept is this. Delinquent behavior is often accompanied by a romanticizing terminology that reinforces such behavior. Staff in a PPC program are alert to these terms and attempt to lower the attractiveness of the behavior by calling it a name that is undesirable to the youth. Likewise, all reference to positive, helping behavior should be made with labels that are desirable. This will produce a state of dissonance that will motivate youth to develop negative attitudes about negative behavior and positive attitudes about positive behavior. (Vorrath & Bendtro, 1985, p. 31) The assumption made by Vorrath and Bendtro (1985) is similar to that made by  97.  “context-independent  ...  global theories of social development” (Nucci & Weber, 1991, p.  254), such as Kohlberg’s, which argue for a single developmental system of morality. The data here concerning the contraband norm, particularly, did not support this assumption; in fact gave support to the idea that the subculture retained a firm grip on the judgments and reasoning regarding contraband, as if it were a separate domain, perhaps the equivalent of a convention (Turiel, 1983; Turiel, Smetana, & Killen, 1993). Thus the PPC approach, although appropriate, needs to be refined in light of these findings, but neither does the moral education approach directly focus on the importance of addressing each important domain directly and specifically (Nucci & Weber, 1991), that is, focus on fostering decalage. In the same way that Kohlberg’s subjects, who could demonstrate empathy for a woman with leukemia but fail to recognize the discomfort of a peer living next to him, these participants could justify helping another resident in high stage terms but fail to see the injustice in stealing from a fellow resident. Moral climate cannot be .  described or measured in a unidimensional fashion. Obviously there are several dimensions to the broader concept institutional climate (one of which is its moral dimension)  -  including physical environment, staff  morale and training level, etc., and there are also many elements to an effective educational or correctional program. It would be naive to assume that moral climate is the single most important factor. The ideal correctional or educational program would foster the development of socio-moral reasoning but also use peer interaction/pressure within a moral climate which involved both shared norms at a high stage (Stage of Norm variable) and a sense of group with shared goals (Stage of Community), both important but distinct dimensions of moral. climate. It would be broad in scope, recognizing the complexity of the process, the importance of addressing a range of important domains or norms, and also address other cognitive factors (i.e., cognitive distortions; Gibbs, 1991). To achieve this, there would need to be a shift in emphasis away from the treatment model of corrections (so frequently found ineffective) and the punishment model (the  98.  default), to one best described as a socio-cognitive model. Educational and correctional programs would espouse goals of development and be described as developmental  contexts.  99.  REFERENCES Achenbach, T. M., & Edeibrock, C. (1983). Manual for the child behavior checklist and revised behavior profile. Burlington, VT: University of Vermont. Alexander, J. F., & Parsons, B. V. (1980). Short-term behavioral intervention with delinquent families: Impact on family process and recidivism. In R. R. Ross & P. Gendreau (Eds.), Effective correctional treatment (pp. 129-141). Toronto: Butterworths. Ames, G., & Murray, F. (1982). When two wrongs make a right: Promoting cognitive change by social conflict. Developmental Psychology, 18, 894-897. Arbuthnot, J. (1975). Modification of moral reasoning through role playing. Developmental Psychology, 11, 319-324. Arbuthnot, J. (1984). Moral reasoning development programmes in prison: Cognitivedevelopmental and critical reasoning approaches. Journal of Moral Education, 13, 112-123. Arbuthnot, 3., & Gordon, D. A. (1986). Behavioral and cognitive effects of a moral reasoning development intervention for high-risk behavior disordered adolescents. Journal of Consulting and Clinical Psychology, 54, 208-216. Arbuthnot, J., Gordon, D. A., & Jurkovic, G. (1987). Personality factors in delinquency. In H. C. Quay (Ed.), Handbook of juvenile delinquency (pp. 139-193). New York: Wiley. Arnold, M. L. (1989). Moral cognition and conduct: A quantitative review of the literature. Unpublished manuscript, Harvard Graduate School of Education, Cambridge, MA. Baumrind, D. (1978). A dialectical materialist’s perspective on knowing social reality. In W. Damon (Ed.), New directions for child development: Moral development (No. 2, pp. 61-82). San Francisco: Jossey-Bass. Bear, G. G., & Richards, H. C. (1988, November). Sociomoral reasoning and antisocial behaviors. Paper presented at the meeting of the Association for Moral Education, Pittsburgh. Bearison, D. J. (1986). Transactional cognition in context: New models of social understanding. In D. J. Bearison & H. Zimilies (Eds.), Thought and emotion (pp. 129-146). Hillsdale, NJ: Eribaum. Berkowitz, M., & Gibbs, J. (1979). A preliminary manual for the coding of transactive features of dyadic discussion. Unpublished manuscript, Marquette University. Berkowitz, M., & Gibbs, J. (1983). Measuring the developmental features of moral discussion. Merrill-Palmer Quarterly, 29, 399-410.  100. Berkowitz, M., & Gibbs, J. (1985). The process of moral conflict resolution and moral development. In M. W. Berkowitz (Ed.), New directions for child development: Peer conflict and psychological growth (No. 29, pp. 71-84). San Francisco: Jossey-Bass.  Berkowitz, M., Gibbs, J., & Broughton, J. (1980). The relation of moral judgment disparity to developmental effects of peer dialogues. Merrill-Palmer Quarterly, 26,341-357. Berkowitz, M. W., Guerra, N., & Nucci, L. (1991). Sociomoral development and drug and alcohol abuse. In W. M. Kurtines & 3. L. Gewirtz (Eds.), Handbook of moral behavior and development (Vol. 3, pp. 35-53). Hillsdale, NJ: Erlbaum. Berndt, T., McCartney, K., Caparulo, B., & Moore, A. (1984). The effects of group discussions on children’s moral decisions. Social Cognition, 2, 343-359. Blasi, A. (1980). Bridging moral cognition and moral action: A critical review of the literature. Psychological Bulletin, 88, 1-45. Blasi, A. (1983). Bridging moral cognition and moral action: A theoretical review. Developmental Review, 3, 178-210.  Blatt, M. (1969). The effects of classroom discussion upon children’s level of moral development. Unpublished doctoral dissertation, University of Chicago. Blatt, M., & Kohlberg, L. (1975). The effects of classroom moral discussion upon children’s moral judgment. Journal ofMoral Education, 4, 129-161. Brennan, T., Huizinga, D., & Elliot, D. (1978). The social psychology of runaways. Lexington, MA: Heath. Buchanan, T. (1992, November). Why is the literature examining the moral cognitionmoral action relationship inconsistent? A meta-analytic investigation of five moderating variables. Paper presented at the meeting of the Association of Moral Education, Toronto Caron, R. (1979). Go-boy! Don Mills, ON: McGraw-Hill. Carr, R. A. (1988). The city-wide peer counselling program. Services Review, 10, 217-232.  Children and Youth  Chandler, M. J. (1973). Egocentrism and antisocial behavior: The assessment and training of social perspective-taking skills. Developmental Psychology, 9, 326333. Chandler, M. J. (1991). Alternative readings on the performance-competence relation. In M. 3. Chandler & M. Chapman (Eds.), Criteria for competence: Controversies in the conceptualization and assessment of children’s abilities. Hilisdale, NJ: Erlbaum. Chandler, M. J. & Chapman, M. (Eds.). Criteria for competence: (1991). Controversies in the conceptualization and assessment of children’s abilities. Hillsdale, NJ: Eribaum.  101. Chandler, M. 3., Greenspan, S., & Barenboiffi, C. (1974). Assessment and training of role-taking and referential communication skills in institutionalized emotionally disturbed children. Developmental Psychology, 10, 546-553. Chandler, M. J., & Moran, T. (1990). Psychopathology and moral development: A comparative study of delinquent and non-delinquent youth. Developmental Psychopathology, 2,227-246. Chapman, M. (1986). The structure of exchange: Piaget’s sociological theory. Human Development, 29, 181-194. Chapman, M. (1988). Constructive evolution: Origins and development of Piaget’s thought. New York: Cambridge University Press. Coie, 3. (1985). Fitting social skills intervention to the target group. In B. H. Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children’s peer relations: Issues in assessment and intervention (pp. 141-156). New York: Springer-Verlag. Coie, J., Dodge, K., & Coppotelli, H. (1982). Dimensions and types of social status: A cross-age perspective. Developmental Psychology, 17, 557-571. Colby, A., & Kohlberg, L. (1987a). The measurement of moral judgment: Theoretical foundations and research validation (Vol. 1). New York: Cambridge University Press. Colby, A., & Kohlberg, L. (198Th). The measurement of moral judgment. Standard issue scoring manual (Vol. 2). New York: Cambridge University Press. Colby, A., Kohlberg, L., Gibbs, J., & Lieberman, M. (1983). A longitudinal study of moral judgment. Monographs of the Society for Research in Child Development, 48, 1-124. Crockenberg, S. B., & Nicolayev, J. (1979). Stage transition in moral reasoning as related to conflict experienced in naturalistic settings. Merrill-Palmer Quarterly, 25, 185-192. Cullen, F. T., & Gendreau, P. (1989). The effectiveness of correctional rehabilitation: Reconsideration of the nothing works debate. In L. Goodstein & D. Mackenzie (Eds.), The American prison: Issues in research and policy (pp. 122-133). New York: Plenum. Damon, W., & Killen, M. (1982). Peer interaction and the process of change in children’s moral reasoning. Merrill-Palmer Quarterly, 28, 347-368. Davidson, P. M. (1992). The role of social interaction in cognitive development: A propaedeutic. In L. T. Winegar & J. Valsiner (Eds.), Children’s development within social context. Hillsdale, NJ: Erlbaum. de Vries, B., & Walker, L. J. (1986). Moral reasoning and attitudes toward capital punishment. Developmental Psychology, 22, 509-513. DeVries, R. (1991). The cognitive-developmental paradigm. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development (Vol. 1, pp. 7-12). Hilisdale, NJ: Erlbaum.  102. Dimant, R., & Bearison, D. 3. (1991). Development of formal reasoning during successive peer interactions. Developmental Psychology, 27, 277-284. Dodge, K., Schlundt, D. G., Schocken, I., & Delugach, 3. D. (1983). Social competence and children’s sociometric status: The role of peer group entry strategies. Merrill-Palmer Quarterly, 29, 309-336.  Donovan, D. M., & McIntyre, D. (1990). Healing the hurt child: A developmental contextual approach. New York: Norton. Dubin, R., & Dubin, E. (1965). Children’s social perceptions: A review of research. Child Development, 36, 809-838.  Ekstedt, J. W., & Griffiths, C. T. (1988). Corrections in Canada: Policy and practice (2nd ed.). Vancouver: Butterworths. Fahlberg, V. (Ed.). (1990). Residential treatment: A tapestry of many therapies. Indianapolis, IN: Perspectives Press. Forrester, M. A. (1992). The development of young children’s social-cognitive skills. Hove, UK: Eribaum. Foster, S. L., Bell-Dolan, D., & Berler, E. S. (1986). Methodological issues in the use of sociometrics for selecting children for social skill research and training. In R. J. Prinz (Ed.), Advances in behavioral assessment of children and families (Vol. 2, pp. 227-248). Greenwich, CT: JAI Press. Gendreau, P., & Ross, R. R. (1980). Effective correctional treatment: Bibliotherapy for cynics. In R. Ross & P. Gendreau (Eds.), Effective correctional treatment (pp. 463-489). Toronto: Butterworths. Gibbons, D. C. (1976). Delinquent behavior. Englewood Cliffs, NJ: Prentice-Hall. Gibbs, J. C. (1991). Sociomoral developmental delay and cognitive distortion: Implications for the treatment of antisocial youth. In W. M.. Kurtines & 3. L. Gewirtz (Eds.), Handbook of nwral behavior and development: Application (Vol. 3, pp. 95-110). Hilisdale, NJ: Erlbaum. Gibbs, J. C., Arnold, K. 0., Ahlborn, H. H., & Cheesman, F. L. (1984). Facilitation of sociomoral reasoning in delinquents. Journal of Clinical and Consulting Psychology, 52, 37-45. Gibbs, 3. C., & Schnell, S. V. (1985). Moral development vs. socialization: A critique. American Psychologist, 40, 1071-1080. Gilligan, C. (1980). The effects of social institutions on the moral development of children and adolescents. Bulletin of the Menninger Clinic, 44, 498-523.  Gilligan, C. (1982). In a different voice: Psychological theory and women’s development. Cambridge, MA: Harvard University Press. Gilligan, C. & Attanucci, 3. (1988). Two moral orientations: Gender differences and similarities. Merrill-Palmer Quarterly, 34, 223-237.  103.  Goffman, E. (1961). Asylums: Essays on the social situation of mental patients and other inmates. Garden City, NY: Anchor. Gottfredson, M. R. (1979). Treatment destruction techniques. Journal of Research in Crime and Delinquency, 16, 39-54. Haan, N. (1977). Coping and defending: Processes of self-environment organization. New York: Academic Press. Haan, N. (1985). Processes of moral development: Cognitive or social disequilibrium? Developmental Psychology, 21, 996-1006.  Haan, N., Aerts, E., & Cooper, B. (1985). On moral grounds: The search for practical morality. New York: New York University Press. Hackler, J. C. (1978). The prevention of youthful crime: The great stumble forward. Toronto: Methuen. Rains, A. A. (1984). Variables in social cognitive development: Moral judgment, roletaking, cognitive processes, and self-concept in delinquents and non-delinquents. Journal ofEarly Adolescence, 4, 65-74. Rains, A. A., & Miller, D. J. (1980). Moral and cognitive development in delinquent and nondelinquent children and adolescents. Journal of Genetic Psychology, 137, 21-35.  Hartshorne, H., & May, M. A. (1928). Studies in the nature of character: Studies in deceit (Vol. 1). New York: MacMillan. Hersh, R., Paolitto, D., & Reimer, J. (1979). Promoting moral growth: From Piaget to Kohlberg. New York: Longman. Hickey, J., & Scharf, P. (1980). Towards a just correctional system. San Francisco: Jossey-Bass. Higgins, A. (1980). Research and measurement issues in moral education interventions. In R. Mosher (Ed.), Moral education: A first generation of research and development (pp. 92-107). New York: Praeger. Higgins, A., Power, C., & Kohlberg, L. (1984). The relationship of moral atmosphere to judgments of responsibility. In W. M. Kurtines & J. L. Gewirtz (Eds.), Morality, moral behavior, and moral development (pp. 74-108). New York: Wiley. Hymel, S., & Franke, S. (1985). Children’s peer relations: Assessing self-perceptions. In B. H. Schneider, K H. Rubin, & J. E. Ledingham (Eds.), Children’s peer relations: Issues in assessment and intervention (pp. 75-92). New York: Springer-Verlag. Izzo, R. L., & Ross, R. R. (1990). A meta-analysis of rehabilitation programs for juvenile delinquents: A brief report. Criminal Justice and Behavior, 17, 134142.  104. Jennings, W. S., Kilkenny, R., & Kohlberg, L (1983). Moral-development theory and practice for youthful and adult offenders. In W. S. Laufer & J. M. Day (Eds.), Personality theory, moral development and criminal behavior (pp. 281-351). Lexington, MA: Lexington. Jesness, C. (1966). The Jesness Inventory. Palo Alto, CA: Consulting Psychologists Press. Jesness, C. (1971). The Jesness Behavior Checklist. Psychologists Press.  Palo Alto, CA: Consulting  Jurkovic, 0. (1980). The juvenile delinquent as a moral philosopher: A structuraldevelopmental perspective. Psychological Bulletin, 88, 709-727. Jurkovic, 0., & Prentice, N. (1974). Dimensions of moral interaction and moral judgment in delinquent and nondelinquent families. Journal of Consulting and Clinical Psychology, 42, 256-262. Jurkovic, 0., & Prentice, N. (1977). Relation of moral and cognitive development to dimensions of juvenile delinquency. Journal of Abnormal Psychology, 86, 414420. Kahn, T. (1990). Pathways: A guided workbookfor youth beginning treatment. Orwell, VT: Safer Society. Kasen, S., Johnson, 3., & Cohen, P. (1990). The impact of school emotional climate on student psychopathology. Journal ofAbnormal Child Psychology, 18, 165-177. Keasey, C. B. (1971). Social participation as a factor in the moral development of preadolescents. Developmental Psychology, 5, 216-220.  Kegan, R. (1984). The child behind the mask: Psychopathy as a developmental delay. In W. H. Reid, D. Dorr, J. I. Walker, & J. W. Bonner (Eds.), Unmasking the psychopath: Antisocial personality and related syndromes (pp. 45-77). New York: Norton. Kerr, M. E., & Bowen, M. (1988). Family evaluation: An approach based on Bowen theory. New York: Norton. Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research (pp. 347-398). Chicago: Rand-McNally. Kohlberg, L. (1981). Essays on moral development: development (Vol. 1). New York: Harper & Row.  The philosophy of moral  Kohlberg, L. (1983). The moral atmosphere of the school. In H. Giroux & D. Purpel (Eds.), The hidden curriculum and moral education: Deception or discovery? (pp. 64-81). Berkeley, CA: McCutchan. Kohlberg, L. (1984). Essays on moral development: development (Vol. 2). New York: Harper & Row.  The psychology of moral  105. Kohlberg, L. (1986). The just community approach to corrections. Correctional Education, 37, 54-58.  Journal of  Kohlberg, L., Boyd, D. R., & Levine, C. (1990). The return of Stage 6: Its principle and moral point of view. In T. Wren (Ed.), The moral domain: Essays in the ongoing discussion between philosophy and the social sciences (pp. 151-181). Cambridge, MA: MIT Press. Kohlberg, L., & Candee, D. (1984a). The relation of moral judgment to moral action. In W. M. Kurtines & J. L Gewirtz (Eds.), Morality, moral behavior, and moral development (pp. 52-73). New York: Wiley. Kohlberg, L., & Candee, D. (1984b). The relationship of moral judgment to moral action. In L. Kohlberg, Essays on moral development: Vol. 2. The psychology of moral development (pp. 498-581). San Francisco: Harper & Row. Kohlberg, L., & Higgins, A. (1987). School democracy and social interaction. In W. M. Kurtines & J. L. Gewirtz (Eds.), Moral development through social interaction (pp. 102-128). New York: Wiley. Kohlberg, L., Kauffman, K., Scharf, P., & Hickey, J. (1974). The just community approach to corrections: A manua4 Part II. Moral Education Research Foundation, Harvard University. Kohlberg, L., Yaeger, J., & Hjertholm, E. (1968). Private speech: Four studies and a review of theories. Child Development, 39, 691-736. Krebs, D., Vermeulen, S., Carpendale, J., & Denton, K. (1991). Structural and situational influences on moral judgment: The interaction between stage and dilemma. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Research (Vol. 2, pp. 139-169). Hillsdale, NJ: Eribaum. Kruger, A. (1989, April). The effect ofpeer and adult-child transactive discussions on moral reasoning. Paper presented at the meeting of the Society for Research in Child Development, Kansas City. Kruger, A., & Tomasello, M. (1986). Transactive discussions with peers and adults. Developmental Psychology, 22, 681-685. Kupfersmid, J., & Wonderly, D. (1982). Disequilibration as a hypothetical construct in Kohlbergian moral development. Child Study Journal, 12, 171-185. Larose, C., Dionne, J., & Larivee, S. (1992, November). The effects of group discussions about social and moral norms on misbehaviors, on moral atmosphere and on individual moraljudgment development inside a Just Community Program for male delinquent adolescents. Paper presented at the meeting of the Association for Moral Education, Toronto.  Leming, J. S. (1973). Adolescent moral judgment and deliberation on classical and practical moral dilemmas. Unpublished doctoral dissertation, University of Wisconsin. Leming, J. S. (1976). An exploratory inquiry into a multi-factor theory of moral behavior. Journal ofMoral Education, 5, 179-188.  106. Lickona, T. (1991). Educating for character: How schools can teach respect and responsibility. New York: Bantam. Light, P., & Perret-Clermont, A. (1989). Social context effects in learning and testing. In A. Gellatly, D. Rogers, & J. Sloboda (Eds.), Cognition and social worlds (pp. 99-112). Oxford: Clarendon. Lockwood, A. L. (1978). The effects of values clarification and moral development curricula on school-age subjects: A critical review of recent research. Review of EducationalResearch, 48, 325-364. Lyman, D. R., & Selman, R. (1985). Peer conflict in pair therapy: Clinical and developmental analyses. In M. Berkowitz (Ed.), New directions for child development: Peer conflict and psychological growth (No. 29, pp. 85-103). San Francisco: Jossey-Bass. Maag, J. (1989). Moral discussion group interventions: Promising technique or wishful thinking? Behavior Disorders, 14, 99-106. Maitland, K., & Goldman, 3. R. (1974). Moral judgment as a function of peer group interaction. Journal ofPersonality and Social Psychology, 30, 699-704. Malarek, V. (1984). Hey, Malarek!: Toronto: MacMillan.  The true story of a street kid who made it.  Martinson, R. (1974). What works? Questions and answers about prison reform. Public Interest, 35, 22-54. Martinson, R. (1979). New findings, new views: A note of caution regarding sentencing reform. Hofstra Law Review, 7, 243-258. McCann, D. C., & Prentice, N. M. (1981). Promoting moral judgment of elementary The influence of direct reinforcement and cognitive school children: disequilibration. Journal of Genetic Psychology, 139, 27-34. Meichenbaum, D. (1977). Cognitive-behavior modification: An integrative approach. New York: Plenum. Miller, S., & Brownell, C. (1975). Peers, persuasion, and Piaget: Dyadic interaction between conservers and non-conservers. Child Development, 46, 992-997. Moos, R. (1975). Evaluating correctional and community settings. New York: Wiley. Moos, R., & Moos, B. (1986). Family environment scale manual. Consulting Psychologists Press.  Palo Alto, CA:  Moran, T. (1988). Moral development and moral action: A study ofyouthful offenders. Unpublished doctoral dissertation, University of British Columbia, Vancouver. Mosher, R. L. (1980). Moral education: development. New York: Praeger.  A first generation of research and  Murray, F. (1982). Teaching through social conflict. Psychology, 7, 257-271.  Contemporary Educational  107. Murray, F. (1983). Equilibration as cognitive conflict. DevelopmentalReview, 3, 54-61. Murray, H. (1938). Explorations in personality. New York: Oxford University Press. Nelson, J. R., Smith, D. J., & Dodd, J. (1990). The moral reasoning of juvenile delinquents: A meta-analysis. Journal of Abnormal Child Psychology, 18, 231239. Nettler, G. (1978). Explaining crime (2nd ed.). New York: McGraw-Hill. Niles, W. (1986). Effects of moral development discussion group on delinquent and predelinquent boys. Journal of Consulting Psychology, 33, 45-51. Nucci, L. (1982). Conceptual development in the moral and conventional domains: Implications for values education. Review ofEducationalResearch, 52, 93-122. Nucci, L., & Weber, E. (1991). The domain approach to values education: From theory to practice. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Application (Vol. 3, pp. 251-266). Hilisdale, NJ: Eribaum. Piaget, J. (1965). The moral judgment of the child. New York: Free Press (original work published 1932). Piaget, J. (1980a). Adaptation and intelligence. Chicago: University of Chicago Press (original work published 1974). Piaget, J. (1980b). Experiments in contradiction. Chicago: University of Chicago Press (original work published 1974).  Polsky, H. W. (1962). Cottage Six: The social system of delinquent boys in residential treatment. New York: Russell Sage Foundation. Polsky, H. W., & Claster, D. S. (1970). The structure and function of adult-youth systems. In H. W. Poisky & D. S. Claster (Eds.), Social system perspectives in residential institutions (pp. 683-699). East Lansing: Michigan State University Press. Power, C. (1980). Evaluating Just Communities: Toward a method of assessing the moral atmosphere of the school. In R. Mosher (Ed.), Moral education: A first generation of research and development (pp. 223-265). New York: Praeger. Power, C., Higgins, A., & Kohlberg, L. (1989). Lawrence Kohlberg’s approach to moral education. New York: Columbia University Press. Power, C., & Reimer, J. (1978). Moral atmosphere: An educational bridge between moral judgment and action. In W. Damon (Eds.), New directions for child development: Moral Development (No. 2, pp. 105-116). San Francisco: Jossey Bass.  Powers, S. I. (1983). Family interaction and parental moral development as a context for adolescent development. Unpublished doctoral dissertation, Harvard University.  108. Rest, 3. R. (1973). The hierarchical nature of moral judgment: A study of patterns of comprehension and preference of moral stages. Journal ofPersonality, 33, 45-51. Rorte, W. (February, 1994). Do we need ethical principles? Paper presented at the Conference for Applied Ethics, University of British Columbia, Vancouver. Rosenkotter, L. I., Landman, S., & Mazak, S. G. (1980). Use of moral discussion as an intervention with delinquents. Psychological Reports, 46, 91-94. Ross, R., & Fabiano, E. (1981). Time to think: Cognition and crime: remediation. Department of Criminology, University of Ottawa. Ross, R., & Gendreau, P. Butterworths.  (1980).  Effective correctional treatment.  Link and Toronto:  Ross, R., & Gendreau, P. (1984). Offender change: An effective alternative to incarceration. Programs Branch User Report, Ministry of the Solicitor General of Canada, Ottawa. Ruback, R. B., & Innes, C. A. (1988). The relevance and irrelevance of psychological research: The example of prison crowding. American Psyclthiogist, 43, 683-693. Sabornie, E. J., & Kauffman, J. M. (1985). Regular classroom sociometric status of behaviorally disordered adolescents. Behavior Disorders, 11, 268-274. Samenow, S. (1984). Inside the criminal mind. New York: Times Books. Sanvitale, D., Saltzstein, H. D., & Fish, M. C. (1989). Moral judgments by normal and conduct-disordered preadolescent and adolescent boys. Merrill-Palmer Quarterly, 35, 463-481. Scharf, P. (1973). Moral atmosphere and intervention in the prison. doctoral dissertation, Harvard University.  Unpublished  Schonert, K., & Cantor, G. (1991). Moral reasoning in behaviorally disordered adolescents from alternative and traditional high schools. Behavioral Disorders, 17, 23-35. Scott, G. D., & Trent, B. (1982). Inmate: The casebook revelations of a Canadian penitentiary psychiatrist. Toronto: Optimum. Selman, R. (1980). The growth of interpersonal understanding: Developmental and clinical analyses. New York: Academic Press. Selman, R., & Demorest, A. (1984). Observing troubled children’s interpersonal negotiation strategies: Implications of and for a developmental model. Child Development, 55, 288-304. Smetana, J., & Killen, M. (1991). Children’s reasoning about interpersonal and moral conflicts. Child Development, 62, 1000-1010. Snarey, J. (1985). Cross-cultural universality of social-moral development: A critical review of Kohibergian research. PsychologicalBulletin, 97, 202-232.  109.  Stern, G. (1970). People in context: Measuring person-environment congruence in education and industry. New York: Wiley. Stierlin, H. (1973). Family perspective on adolescent runaways. Archives of General Psychiatry, 29, 56-62. Sullivan, E. V., & Beck, C. (1975). Moral education in a Canadian setting. Phi Delta Kappan, 56, 697-701. Taguiri, R. (1968). The concept of organizational climate. In R. Taguiri & G. Litwin (Eds.), Organizational climate: Exploration of a concept. Boston: Harvard University, Graduate School of Business Administration. Tannehill, R. L. (1987). Employing a modified positive peer culture treatment approach in a state youth center. Journal of Offender Counselling, Services, & Rehabilitation, 12, 113-129. Thornton, D., & Reid, R. L (1982). Moral reasoning and type of criminal offense. British Journal ofSocial Psychology, 21, 231-238. Trevethan, S. D., & Walker, L. J. (1989). Hypothetical versus real-life moral reasoning among psychopathic and delinquent youth. Development and Psychopathology, 1, 91-103. Turiel, E. (1966). An experimental test of the sequentiality of developmental stages in the chilcVs moral judgments. Journal of Personality and Social Psychology, 6, 611-618. Turiel, E. (1974). Conflict and transition in adolescent moral development. Development, 45, 14-29.  Child  Turiel, E. (1977). Conflict and transition in adolescent moral development, II: The Child resolution of disequilibrium through structural reorganization. Development, 48, 634-637. Turiel, E. (1983). The development of social knowledge. University Press.  Cambridge: Cambridge  Turiel, E., Smetana, J., & Killen, M. (1991). Social contexts in social cognitive development. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development: Research (Vol. 2, pp. 307-332). Hillsdale, NJ: Erlbaum. Villenave-Cremer, S., & Eckensberger, L. (1985). The role of affective processes in moral judgment performance. In M. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 175-196). Hillsdale, NJ: Erlbaum.  Vorrath, H., & Bendtro, L. (1985). Positive peer culture. New York: Aldine. Vygotsky, L. (1986). Thought and language. Cambridge, MA: MIT Press (originally published in 1934). Walker, L. J. (1980). Cognitive and perspective-taking prerequisites for moral development. Child Development, 51, 131-139.  110. Walker, L. J. (1982). The sequentiality of Kohlberg’s stages of moral development. Child Development, 53, 1330-1336. Walker, L. J. (1983). Sources of cognitive conflict for stage transition in moral development. Developmental Psychology, 19, 103-110. Walker, L. J. (1986). Cognitive processes in moral development. In G. L Sapp (Ed.), Handbook of moral development Models, processes, techniques, and research (pp. 109-145). Birmingham, AL: Religious Education Press. Walker, L. J. (1988). The development of moral reasoning. Development, 5, 33-78.  Annals of Child  Walker, L. J., de Vries, B., & Bichard, S. L (1984). The hierarchical nature of stages of moral development. Developmental Psychology, 20, 960.966. Walker, L. 3., de Vries, B., & Trevethan, S. (1987). Moral stages and moral orientations in real-life and hypothetical dilemmas. Child Development, 58, 842-858. Walker, L. 3., & Richards, B. S. (1979). Stimulating transitions in moral development as a function of stage of cognitive development. Developmental Psychology, 15, 95103. Walker, L. J., & Taylor, J. H. (1991). Family interactions and the development of moral development. Child Development, 62, 264-283. Wertsch, 3. V. (1984). Conceptual issues in the zone of proximal development. In B. Rogoff & J. V. Wertsch (Eds.), New directions for child development: Children’s learning in the “zone of proximal development” (No. 23, pp. 7-18). San Francisco: Jossey-Bass.  Wertsch, J. V. (1985). Vygotsky and the social formation of mind. Cambridge, MA: Harvard University Press. Whitehead, J. T., & Lab, S. P. (1989). A meta-analysis of juvenile correctional treatment. Journal ofResearch in Crime and Delinquency, 26, 276-295. Yochelson, S., & Samenow, S. (1976). The criminal personality: A profile for change (Vol. 1). New York: Jason Aronson.  Youniss, 3. (1980). Parents and peers in social development: perspective. Chicago: University of Chicago Press.  A Sullivan-Piaget  Youniss, J., & Damon, W. (1992). Social construction in Piaget’s theory. In H. Beilin & P. B. Pufall (Eds.), Piaget’s theoiy: Prospects and pitfalls (pp. 267-286). Hillsdale, NJ: Eribaum.  111.  APPENDIX A Moral Judgment Interview (from Colby & Kohlberg, 198Th) Form A In Europe, a woman was near death from a rare kind of cancer. There was one drug that the doctors thought might save her. It had recently been discovered by a druggist in the same town. The drug was expensive to make, but the druggist was charging ten times what the drug cost him to make. He paid $400 for the ingredients but charged $4000 for a small dose of the drug. The sick woman’s husband, Heinz, went to everyone he knew to borrow the money, and tried every legal means, but he could only get together about $2000, which is half of what it costs. He told the druggist that his wife was dying and asked him to sell it cheaper or let him pay him later. But the druggist said, “No, I discovered the drug and I’m going to make money from it.” So having tried every legal means, Heinz gets desperate and considers breaking into the man’s store to steal the drug for his wife. Should Heinz steal the drug? Why or why not? Is it actually right or wrong for him to steal the drug? Why is it right or wrong? (If subject favors stealing) If Heinz doesn’t love his wife, should he steal the drug fOr her? (If subject favors not stealing) Does it make a difference whether or not he loves his wife? Suppose the person dying is not his wife but a stranger. Should Heinz steal the drug for a stranger? (If the subject favors stealing the drug for a stranger) Suppose it’s a pet animal he loves. Should Heinz steal to save the pet animal? Is important for people to do everything they can to save another’s life? Why or why not? It is against the law for Heinz to steal. Does that make it morally wrong? Why or why not? In general, should people do everything they can to obey the law? Why or why not? How does that apply to what Heinz should do? Heinz did break into the store. He stole the drug and gave it to his wife. In the newspapers the next day, there was an account of the robbery. Mr. Brown, a police officer who knew Heinz, read the account. He remembered seeing Heinz running away from the store and realized that it was Heinz who stole the drug. Mr. Brown wonders whether he should report that Heinz was the robber.  112. Should Officer Brown report Heinz for stealing? Why or why not? Suppose Officer Brown were a close friend of Heinz. Should he then report him? Why or why not? Officer Brown finds Heinz and arrests him. Heinz is brought to court and the jury finds him guilty. It is up to the judge to determine the sentence. Should the judge give Heinz some punishment or should he suspend the sentence and let Heinz go free? Why? Thinking in terms of society, should people who break the law be punished? Why or why not? How does this apply to how the judge should decide? Heinz was doing what his conscience told him when he stole the drug. lawbreaker be punished if he is acting out of conscience? Why or why not?  Should a  Thinking back over the dilemma, what would you say is the most responsible thing for thejudgetodo? Why? Form B There was a woman who had very bad cancer, and there was no treatment known to medicine thatwould save her. Her doctor, Dr. Jefferson, knew that she had only about six months to live. She was in terrible pain, but she was so weak that an overdose of a pain-killer like morphine would make her die sooner. She was delirious and was almost crazy with pain, and in her calm periods she would ask Dr. Jefferson to give enough of the drug to kill her. She said she couldn’t stand the pain and was going to die in a few months anyway. Although he knows that mercy-killing is against the law, the doctor thinks about granting her request. Should Dr. Jefferson give her the drug that would make her die? Why or why not? Is it actually right or wrong for him to give the woman the drug that would make her die? Why is it right or wrong? Should the woman have the right to make the final decision? Why or why not? The woman is married. Should her husband have anything to do with the decision? Why or why not? What should a good husband do in this situation? Why? Is there any way a person has a duty or obligation to live when he or she does not want to, when the person wants to commit suicide? Why or why not? When a pet animal is badly wounded and will die, it is killed to put it out of its pain. Does the same thing apply here? It is against the law for the doctor to give the woman the drug. Does that make it morally wrong? Why or why not? In general, should people try to do everything they can to obey the law? Why or why not? How does this apply to what Dr. Jefferson should do?  113. Dr. the Dr. Dr.  Jefferson did perform the mercy-killing by giving the woman the drug. Passing by at time was another doctor, Dr. Rogers, who knew the situation Dr. Jefferson was in. Rogers thought of trying to stop Dr. Jefferson, but the drug was already administered. Rogers wonders whether he should report Dr. Jefferson.  Should Dr. Rogers report Dr. Jefferson? Why or why not? The doctor does report Dr. Jefferson. Dr. Jefferson is brought to court and a jury finds Dr. Jefferson guilty. It is up to the judge to determine the sentence. Should the judge give Dr. Jefferson some punishment or should he suspend the sentence and let Dr. Jefferson go free? Why? Thinking in terms of society, should people who break the law be punished? Why or why not? How does this apply to how the judge should decide? The jury found Dr. Jefferson legally guilty of murder. Would it be wrong or right for the judge top give him the death sentence (a legal possible punishment)? Why? Is it ever right to give the death sentence? Why or why not? What are the conditions when the death sentence should be given (if ever) in your opinion? Why are these conditions important? Dr. Jefferson was doing what his conscience told him when he gave the woman the drug. Should a lawbreaker be punished if he is acting out of conscience? Why or why not? Thinking back over the dilemma, what would you say is the most responsible thing for the judge to do? Why?  114.  APPENDiX B Moral Climate Interview (from Power, Higgins, & Kohlberg, 1989) Stealing Norm I -  One day Marcel got called out of his unit to the nurse’s office. He left in a rush, leaving some of his tapes in the games room. Tom notices the tapes, particularly two of them, the better, more popular ones. Tom is thinking of taking the two tapes. 1.  What do you think Tom should do in this situation? Why?  2.  Should Marcel have been trusting like that in this situation, or should he have been more careful?  3.  What would you do if this situation occurred in your unit?  4.  What would most guys do? Why?  5.  Would there be a general feeling or expectation in your unit to not steal?  6.  Is there an general agreement or an unwritten understanding among the kids in your unit not tO steal each other’s stuff? Why or why not?  7.  If not, do you think there should be an understanding about not stealing? Would you bring it up (Would you talk about it/propose it?) Why?  8.  Would you be disappointed if someone in your unit stole from another? Would most others in your unit be disappointed?  9.  Would you say something if someone stole in your unit? residents in your unit?  10.  Finally, would you report someone if they stole in your unit? Would most others?  Would most other  Stealing Norm II -  Bob brought back a new ghetto blaster from a T.R. a few weeks ago. Staff advised him to keep it in his personals, but he decided to keep it in his room instead, and frequently let other residents in his unit use it. Bob had become quite confident that he didn’t need to worry about his ghetto blaster being stolen and often left it “lying around” throughout the unit. One weekend, Bob left the ghetto blaster in the games room while he was on his weekend T.R. One of the other residents is thinking of taking it. 1.  Should the other boy take the radio? Why or why not?  2.  Should Bob have been so trusting; should he have kept the radio in his personals? Why or why not?  115. 3.  What would you do in the other boy’s situation? Why?  4.  What would most of the other guys on your unit do in the other boy’s situation? Why?  5.  Is there a general agreement in your unit about taking others’ possessions? Do you think there should be?  6.  Would you be disappointed if someone in your unit stole from another? Would most others in your unit be disappointed?  7.  Would you say something if someone stole in your unit? residents in your unit?  8.  Finally, would you report someone if they stole in your unit? Would most others?  Would most other  Helping Norm I -  Sam has several months of custody left. He doesn’t really care what happens to him and he hates being here. He has told Frank and some other residents that he might take off or “do himself in.” Now Sam is not a very popular kid on the unit; he talks too much and generally bugs the kids. Now, there are a few things you could do to help; you could try to talk him out of it, or spend some time with him, or mention it to staff who could tell him how hurting himself or running wouldn’t be a good idea. 1.  What should Frank do in this situation? Why?  2.  What would you do if you knew about Sam’s situation? Why?  3.  Would most guys on your unit help Sam, or would most guys on your unit think it would be a good thing if someone helped Sam? Why or why not?  4.  Is there an understanding or an unwritten agreement in your unit about helping another?  5.  Do you think there should be an understanding or agreement to help other residents? Why?  6.  Would you be disappointed if someone on your unit did not help a guy like Sam? Would most people on this unit be disappointed?  7.  Would you express your disapproval or say anything if someone did not help? Why or why not? Would most kids on this unit? (Why or why not?)  Helping Norm II -  Rodger is a boy with a lot of personal problems. He is not good at school work, and his mind wanders a great deal. He causes some difficulty for the other kids on the unit; he is frequently late or slow in getting his chores done (or his shower), and this causes delays in program activities and meals. Sometimes Rodger gets into fights with the other residents because he feels that they are deliberately trying to make it tough for him. It is true that the other kids laugh at him and the strange things that he does, and sometimes  116. they call him names like “freak” and “retard.’ Tony, who lives on the same unit with Rodger, sometimes thinks that Rodger needs a lot of help. 1.  Should Frank try to help Rodger? Why or why not?  2.  Should Frank tell the others to stop laughing at or bugging Rodger? Why or why not?  3.  Would you help Rodger? Why or why not?  4.  Would most other guys on your unit help Rodger? Why or why not?  5.  Do you think there should be an agreement (or an understanding) in your unit about helping a guy like Rodger? Why or why not?  6.  Do most of the other guys think there should be an understanding? Why or why not?  7.  Would you be disappointed if no one on your unit helped a guy like Rodger? Why or why not?  8.  Would you say anything to someone on your unit who was bugging or laughing at a guy like Rodger? Why or why not?  9.  Would you report it, or bring it up at a group meeting, if someone was bugging or laughing at a guy like Rodger? Why or why not?  Contraband Norm I -  Upon arriving at CYOC each resident becomes aware of the expectation of not having or using contraband, for example cigarettes or drugs. It is made clear that if contraband is found either the individual responsible or the group as a whole will be consequenced. Due to a recent incident involving contraband on your unit, resulting in a group consequence, each resident made a commitment not to use or bring contraband onto the unit. Then while at school a kid from another unit offers Bob a cigarette and a piece of hash. 1.  Should Bob accept it and bring it back to your unit? Why or why not?  2.  What would you do? Why?  3.  What would most guys on your unit do? Why?  4.  Is there an understanding or unwritten rule in your unit about not bringing contraband onto the unit? Why or why not?  5.  Do you think there should be? Why or why not?  6.  Would you be disappointed if someone brought contraband onto the unit after committing not to? Why or why not?  117.  Contraband Norm II -  Frank’s roommate Bob went on a weekend T.R. and upon returning, he brought back a lighter and some marijuana. He told Frank not to say anything to anyone about this. 1.  What should Frank do in this situation? Why?  2.  What would you do? Why?  3.  What would most guys on your unit do? Why?  4.  Is there an understanding or unwritten rule in your about not bringing contraband onto the unit? Why or why not?  5.  Do you thing there should be? Why or why not?  6.  Would you be disappointed if someone brought contraband onto the unit after committing not to? Why or why not?  118.  APPENDIX C Scoring Criteria for Stage of Community  (adapted from Power, Higgins, & Kohlberg, 1989)  Stage 1 (Rating 1)  Everyone looks out for themselves, believes in “doing their own time,” insisting on not being involved. The Unit is simply random aggregate of persons, exclusively selfinterested, committed to avoiding punishment or hassles from each other.  Stage 1/2 (Rating 2)  Still no sense of community but there is at least a bond by common need to avoid punishment, group consequence, or “heat on the unit.” There is a simplistic reliance on the rule that one does not “rat.”  Stage 2 (Rating 3)  There is still not a clear sense of community apart from exchanges among group members. Community denotes a collection of individuals who do favors for each other, or a series of cliques that do favors and rely on each other for protection. The group is valued in so far as it meets the concrete needs of its members.  Stage 2/3 (Rating 4)  There is a beginning sense of community that involves valuing others. At this stage there is discussion of “helping each other out” and doing things together, but it is unclear as to whether this is for instrumental purposes (doing favors) or because of mutual appreciation and trust.  Stage 3 (Rating 5)  The sense of community refers to a set of relationships and sharing among group members. The group is valued for its friendliness, its willingness to help each other, to work together, and to trust each other simply because they prefer it that way.  119.  APPENDIX D Correctional Institutions Environment Scale Subscale Descriptions (from Moos, 1975)  Involvement:  Support:  measures how active and energetic residents are in the day-to-day functioning of the program (i.e., interacting socially with other residents, doing things on their own initiative, and developing pride and group spirit in the program)  measures the extent to which residents are encouraged to be helpful and supportive toward other residents, and how supportive the staff is towards residents  Expressiveness: measures the extent to which the program encourages the open expression of feelings (including angry feelings) by residents and staff utonomy: assesses the extent to which residents are encouraged to take initiative in planning activities and take leadership in the unit Practical Orientation: assesses the extent to which the resident’s environment orients him toward preparing himself for release from the program: training for new kinds of jobs, looking to the future, and setting and working toward goals are among the factors considered Personal Problem Orientation: measures the extent to which residents are encouraged to be concerned with their personal problems and feelings and to seek understanding of them Order and Organization: measures how important order and organization are in the program, in terms of residents (how they look), staff (what they do to encourage order), and the facility itself (how well it is kept) Clarity: measures the extent to which the resident knows what to expect in the day-to day routine of his program and how explicit the program rules and procedures are Staff Control: assess the extent to which the staff use regulations to keep residents under necessary controls.  120.  •  APPENDIX E  Behavior Frequency Rating  Unit  Start Date  Resident  Frequency Behavior 1.  swearing at another resident  2.  uncommunicative  3.  failing to comply with routine  4.  fighting  5.  damaging property  6.  provoking another  7.  defying staff  8.  bullying  9.  running away  10.  room confinement  11.  Q.R. (special confinement)  Day 1  Day 2  Day 3  Day 4  Day 5  Day 6  Day 7  121. Legend 1.  swearing at another resident: This refers to any incident of posturing, intimidation or defensiveness, or aggressive behavior that does not involve pushing or striking.  2.  uncommunicative: This behavior involves any unusually quiet behavior, that is, less talkative or active than usual, withdrawal, spending more time than usual in his or her room, or apart from other residents. If this behavior happens for a period on one shift, or for the entire shift, it should be scored once, for each shift.  3.  failing to comply with routine/disobedience: This involves any failure to do a routine chore, or to do a chore to the expected standard, or any failure (more passive than active) to do what has been laid out in the unit rules or orientation manual.  4.  fighting: This involves any altercation, involving hitting, kicking, etc. between two or more residents. However, do not score this category if the behavior more appropriately fits into the bullying category. “Fighting 11 usually is evidence of a less adaptive way of resolving conflicts, disagreements and other interpersonal problems.  5.  damaging property: Any incident of damage to Center or others property, including defacing walls, breaking equipment, pictures, tapes, dishes, etc.  6.  provoking another: This refers to only those aggressive acts which involve “egging another resident on,” or name calling, putting another resident down, or making statements that serve to provoke another resident into similar behavior or loud complaints or yelling, or aggressive behavior.  7.  defying staffidisobedience: This involves any incident where the request of a staff member is verbally challenged or refused, or where there has been a deliberate violation of a Center rule.  8.  bullying: This involves striking or pushing another resident, whether on a 1-to-i, or as a participant of a group assault. However, do not score as fighting any act which, in your view, is clearly a defensive act, that is, that a resident is fighting in order to defend or protect him or herself against others who are acting in genuinely assaultive manner.  9.  running away: Any AWOL incident or attempt, whether it is official or not.  10.  room confinement: Any incident which leads to the resident being sent to his bedroom for purposes of “time out.”  11.  Q.R. (special confinement): Any confinement (time out) in the secure quiet room, regardless of length of such confinement.  122.  APPENDiX F Dilemmas Used for Dyadic Discussions  Day 1 Al. Bernie is a kid with a lot of problems. He is immature, most other kids think he looks funny, he bugs others a lot, and, as a result, he is very unpopular. Recently, as a gift, his mother gave him a new “Walkman,” which Bernie says he wants to give to Frank in order to make friends with him. Frank, like the other people on the unit, does not like Bernie. Your instructions are to put yourself in Frank’s/Bernie’s shoes. Think about this situation as if you were Frank/Bernie, answer the questions below and carry on a debate with your partner. 1.  Should Bernie give his Walkman away? Why or why not?  2.  If he does try to give it away, should Frank take Bernie’s Walkman? Why? or why not?  3.  Should anything be done about the fact that Bernie wants to give away a new Walkman? Why or why not?  4.  How would Bernie’s mother feel if he gave it away? Why?  5.  In general should a person always try to be helpful to another? Why or why not?  6.  What if Frank were a friend of Bernie. Should that make a difference? Why or why not?  7.  Does Bernie have the right to do what he wants with his property? Is it morally right for Frankie to accept the Walkman?  8.  In general, when dealing with problems someone who is less able than yourself what is the most important thing to consider? -  -  123. A2. Joe is a 14-year-old boy who wanted to go to camp very much. His father promised him he could go if he saved up the money for it himself. So Joe worked hard at his paper route and saved the $100 it cost to go to camp, and a little bit more besides. But just before camp was about to start, his father changed his mind. Some of his friends decided to go on a special fishing trip, and Joe’s father was short of the money it would cost. So he told Joe to give him the money he saved from the paper route. Joe didn’t want to give up going to camp, so he thinks of refusing to give his father the money. Your instructions are to put yourself in the place of the father/Joe. Think about this situation as if you were the father/Joe, and carry on a debate with your opponent, answering the following questions. 1.  Should Joe refuse to give his father the money? Why or why not?  2.  Does the father have the right to tell Joe to give him the money? Why or why not?  3.  The father promised Joe he could go to camp if he earned the money. Is the fact that the father promised the most important thing in this situation? Why?  4.  In general, why should a promise be kept?  5.  Is it important to keep a promise to someone you won’t see again? Why?  6.  In general, what should be the authority of a father over his son? Why?  7.  What do you think is the most important thing a son should be concerned about in his relationship to his father? Why is that the most important thing?  124. Day 2 Bi. One day Bob was passing by the entrance to the gym and heard some noise. When he checked, he saw two bigger kids beating on Carlos, Bob’s friend. When Bob came up to them, the other two kids (who were known in the center for being tough), left the area but not before threatening Bob and Carlos telling them they had better “keep their mouths shut.” Carlos is not sure about whether to report this incident to staff because the other two boys know something about him that could get him into trouble. -  Your instructions are to think about this situation from the point of view of Bob/the other two boys. Carry on a debate with your opponent as if you were Bob/the other boys. 1.  Should Bob report the bigger boys? Why or why not?  2.  Should Carlos report the bigger boys? Why or why not?  3.  In thinking about whether to report the incident to the staff, Bob thinks about the fact that, although they are bigger, the other two boys are not that well liked on the unit. Should that be an important consideration for Bob?  4.  Is it important that people not be afraid of being physically hurt? Why?  5.  What if Bob is not Carlos’ friend; does that make any difference in what should be done in this situation? Why or why not?  6.  What if Bob didn’t even know Carlos; does that make any difference as to what should be done in this situation? Why or why not?  7.  In general, should boys like the two here be punished? Why or why not?  8.  What is the most important thing to consider in being a friend? important?  Why is it  125. B2. In a country in Europe, a poor man named Valjean could find no work, nor could his sister and brother. Without money, he stole food and medicine that they needed. He was captured and sentenced to prison for 6 years. After a couple of years, he escaped from the prison and went to live in another part of the country under a new name. He saved money and slowly built up a big factory. He gave his workers the highest wages and used most of his profits to build a hospital for people who couldn’t afford good medical care. Twenty years had passed when a tailor recognized the factory owner as being Valjean, the escaped convict whom the police were looking for back in his home town. Your instructions are to think of this situation from the point of view of Valjean/the tailor. Carry on your debate with your opponent, thinking of Valjeanlthe tailor. 1.  Should the tailor report Valjean to the police? Why or why not?  2.  Does a person have a duty or obligation to report an escaped prisoner? Why or why not?  3.  Suppose Vaijean were a close friend of the tailor? Should he report Vaijean? Why or why not?  4.  If Vaijean were reported and brought back before the judge, should the judge send him back to jail or let him go free? Why?  5.  In general should people who break the law be punished? Why or why not?  126. Day 3 Cl. In one of the units in this Center, a sum of money had been stolen from the Chaplain’s office. Mario knows which resident stole the money, bUt he is reluctant to say anything to the staff, because he might be called a “rat.” However, the entire unit has been cut passes and T.R.’s until the person who took the money admits it. The person who stole the money is very popular. Mario is thinking of confessing to taking the money himself, claiming that he had already spent it. Your instructions are to resolve this situation thinking about it from the point of view of Mario/the resident who stole the money. 1.  Should Mario “confess” to taking the money? Why or why not?  2.  Should Mario report the resident who stole the money? Why or why not?  3.  What should the staff do in this situation? Why?  4.  What if the other residents didn’t like the resident who stole the money. Should that make any difference in terms of what they should do? Why?  5.  In general, what authority should a staff member have over a resident? Why?  6.  In general, what is important for a resident to consider in dealing with other residents? Why is this important?  127. C2. In Korea, a company of Marines was way outnumbered and was retreating from the enemy. The company had crossed a bridge over a river, but the enemy were mostly still on the other side. If someone went back to the bridge and blew it up, with the head start the rest of the men in the company would have, they could probably then escape. But the man who stayed back to blow up the bridge would not be able to escape alive. The captain himself is the man who best knows how to lead the retreat. He asks for volunteers, but no one will volunteer. If he goes himself, the men will probably not get back safely, as he is the only one who knows how to lead the retreat. Your instructions are to take the position of the captain/one of the men. Talk about the problem with your opponent with the captain/one of the men in mind. 1.  Should the captain order a man to go on the mission or should he go himself? Why or why not?  2.  Should the captain send a man (or even use a lottery) when it means sending him to his death? Why?  3.  Should the captain go himself when it means that the men will probably not make it back safely? Why or why not?  4.  Does the captain have a right to order a man if he thinks it is best? Why?  5.  What’s so important about human life that makes it important to save or protect? Why is that important?  128.  APPENDIX 0 Sample Transcript of Dyadic Discussion  Day 1 I: Should Bernie give his Walkinan away? Why or why not? S: I don’t think he should. Well simple, if everyone on the Unit doesn’t like me then what’s the use of even trying. I: Well, yah, you should give it to me, cause if you give it to me then I will make others like you and then I’ll like you and it will just be wonderful, you know. S: Yah, but how do I know that everybody will leave me alone? I: Cause I said they would. S: I give you some of my property and all you are giving me is your word. How do I know 9 I: I don’t know. You don’t know, you are just going to have to trust me, Bernie. S: Well you’re one of them doesn’t like me anyways so... (But you want me to like you and...) I: If he should try to give it away, should Frank take Bernie’s Walkman? Why or why not? Sure. I think I should take the Walkman cause if he’s going to give it to me I think I should take it right. S: I disagree with that one. If he’s going to give it away just for friendship it’s bogus, why, when he can listen to it like everybody else. I: Your trying to tell me if somebody wanted to give you a Walkman you wouldn’t take it. S: For friends no. Personally, without this I would. I: So your saying okay then. (I guess so) Why? Because I like the person.  5: Say someone offered you a T.A. for two days or so would you take it just to like them? (No I wouldn’t) Should anything be done about the fact that Bernie wants to give away his Walkman? I: Unless there’s something wrong with this kids head he wouldn’t give away his Walkman. It would make his mom feel all freaked out you know, he must not care about his family or something you know.  129. S: Yah but it’s my life, yah my parents, mother gave it to me but it’s also mine, I can do whatever I want to do with it. I: Still you must have a problem, you must feel pretty bad about yourself if some guy can buy somebody’s friendship. Right. (Yah I guess so.) Then you need help. S: Well nobody will give me help. (Well maybe you didn’t ask) I: How would Bernie’s mother feel and why? I think his mother would feel really bad about Bernie you know. She gives Bernie a present you know and Bernie gives it away. S: (Why would you choose to give the Walkman away even though your mothers going to feel like that?) Who cares? (Why don’t you care) Cause she hasn’t cared in the past. I: I don’t know,. maybe you should start caring, you know. In my mind she must care to give you it you know. Be nice you know, maybe you should start caring. S: She could be doing the same thing that I’m doing with you. Trying to buy me back. I: At least she’s trying. S: In general should a person always try to be helpful to others? Yah, well if somebody’s in trouble or something they can’t really deal with it by themselves. They need somebody to support them and what they do. I: Whenever you try to help somebody all it does is cause problems you know. Try to help somebody and the next time they have problems they ask you again, and like you try to do something else and they ask you again and then like everybody’s asking you cause they always see you helping somebody. Then people start taking advantage of you cause they know your a nice person you know. S: Well some people don’t take advantage. How do you know if one of these times you say, “No I don’t want to help you”, and they really do need that help. They can accuse you of not needing anything and just coming to you for the help because I: Well then if you don’t ask for nothing, you don’t give nothing, no-one expects anything. What if Frank is a friend of Bernie’s should that make any difference why? Sure it would make a difference man, cause if Frank is his friend then Frank probably wouldn’t take the Walkman because the whole situation would be different. (In what way?) Well because Bernie wouldn’t be trying to buy Frank with gifts you know then, you know like if they were friends he could just use it when he wants to. You don’t buy your friends, especially when he’s already your friend you know. Especially if you knew he got it from his mother man. S: I think I wouldn’t do it. None of my friends ever do that for me. I got whatever it was from my mom. They’d ask to borrow it for a couple of hours or something and I’d give it to them.  130. I: Does Bernie have the right to do what he wants with his property? Is it morally right for a friend to accept the Walkman? The first one is yes (Why?) Cause well it’s his. S: I can do whatever I want, like see these are my jeans. I could give them to Tex, which I never will do but... (Well why don’t you give them to Tex?) Why? it’s nobody else’s business. I: Well it’s still wrong, I mean like sure he has the right like legally he has the right but not like in his head man, cause you know it’s a gift from his mom, man. He’s not giving it for any reason, he just wants to buy a friend man. He’s not going to feel good about that anyway. He’s just going to feel bad cause you know he knows he just bought somebody, you know. You might feel good for a little bit, but as soon as he thinks, he’s going to feel bad you know. So it’s not really right, legally it is. In his mind, it isn’t, man. S: But I don’t care. I: It’s still wrong; it doesn’t matter. Just because the law says something’s right doesn’t make it right, man. S: Yep, I agree. Cause if I want to give him something then if I want him to have it then he should like say, okay well whatever. Just like for Christmas I gave a present. I: Yah but morally it’s not right because Frank knows that this kid is just trying to buy his friendship, you know, and if he accepts the Walkman he’s still not going to be this kid’s friend. He might be nice to him for a week, but he’s still not going to be this kid’s friend, you know. And so this Frank kid shouldn’t take the Walkman if he’s not going to be nice to this kid. You know it’s just not right to let somebody buy you. I guess he doesn’t have much respect for himself if somebody can just buy him for a Walkman. In general in dealing with a person who has problems, someone who is less able than yourself. What is the most important thing to consider? Well the most important thing is his feelings, pretty well, what’s better for him. If your going to help somebody who has like low self esteem what’s going to help him you know. Maybe it’s not the nicest thing cause if you rejected his Walkman that maybe make him feel just as bad cause you know he’d feel like rejected about another thing. 5: Well I don’t know. I think maybe he’d feel a little better about himself. (If they didn’t take it) Yah. Cause like he’s trying to buy Frank’s friendship right, and Frank said, “No I don’t want to”, then that reassured that, that tells Bernie that he doesn’t want the Walkman cause he ain’t going to be nicer. He’s being nice to him. I: Yah, but he also might feel that, you know, the kids not taking the Walkman, cause he really hates this kid and he doesn’t even want to try to be nice to him. I think the most important thing to consider is, like not his feelings but what’s going to help him.  I: Should Joe refuse to give his father the money. Why or why not?  131. S: Hey Dad, I’m not going to give you my money. It’s my money. You told me if I worked all summer on my paper route I can go to camp and then when it comes time to go and you say give me your money man. It’s my money I made it. What, do you say I got to give it to you for? I: Cause I changed my mind and I’m your father and.I am older than you and you have to obey me. S: It doesn’t matter man. IVs my money. You can’t take my money. (So) So what. I: So, I changed my mini S: Okay, so I’m not going to camp, I’m still not giving you my money. I: Well I’ve got to go fishing so give me your money. S: I don’t have to man. It’s my money okay. Did I make that money? Did I make that money? (Yah) Then why shouldn’t I spend it man. I: Cause I’m telling you can’t. S: Then why did you like to me man, you said I could go to camp. I: I changed my mmd. S: What do you mean change your mind. It’s a lie, just a lie. I: Well I changed my mind. You can’t go. S: What do you mean I can’t go. (You can’t go) You gonna stop me. (Yah) What do you think you are man (Your Dad) Should I give you my money do you think? (Yes) No, I don’t think I should. It’s my money, I made my money. (So) So you don’t give me your money that you make. I: I give you an allowance (No you don’t) 5: It’s my money, I made it, I don’t have to give you my money. I: You have to pay me rent, your going to have to pay me rent pretty soon, your almost getting of age. I pay for the rent of the house so you have a roof over your head, I pay for food, I pay for your clothes, your school. S: I know man, your my dad. Your supposed to, that’s your job. I: That’s not my job, I can say flick you. 5: Then kick me out.... Does the father have the right to tell Joe to give him the money? You don’t have the right to tell me that man. What makes you think you have the right to. Cause you told me one thing man and then you changed your mind. You don’t have the right to do that.  132. I: Yes I do, I’m your father, I can do whatever. S: I also respect my elders man, but this is getting pretty sick you know. You don’t have that right man, it’s my money like you can’t just tell me to do that you know. I: I want to go fishing. (I want to go to Camp) Well you can’t go to camp. (Why not) Because I said your not going to go. (Well I want to go) Your not going to. S: You don’t have that right to tell me to give you the money. You don’t have that right. (I do) Who says you do man. (I’m your father) I’m your son. (So I’m older than you) So I’m younger than you. I: I’m the one that gave you life. The father promised Joe he could go to camp if he earned the money, is the fact that the father promised the most important thing in this situation? No. Cause it’s my choice I can do whatever I want. S: It is important man. Your trying to tell me that your word is nothing. You have no honor or what? (Sometimes) What do you mean “sometimes” man? You don’t have no honor if you do that. Who the fuck do you think you are? It’s my money and I’m and you promised me man, that’s the thing man, you promised me. I work and save the money man and then I think oh my dad promised me and I can go to camp you know and then it comes time, oh I changed my mind. You didn’t change your mind, you lied, you went back on your word. I: I changed my mind cause I don’t want you to go. I want to go fishing. S: It doesn’t matter, you still changed your mind. You lied man, that’s the most important thing. You can’t lie. (Well I just did) You can’t lie to your son. In general why should a promise be kept? S: A promise should be kept because that’s the most important thing you have man. You know you can take away anything man, but if you promise something you have to do it, that’s the most important thing to a person. I: Okay, so I promise these shoes cause I think I’m getting out tomorrow. I don’t get out, I don’t give you the shoes cause I don’t get out tomorrow. S: If you promised that, then you have to do that. I: If something changes the promise is then fucked. S: It’s still a promise. It doesn’t matter. So somebody promised me. You still promised me that though. Look man you can take away anything man but you still have a promise. What else can you do man. It doesn’t matter when you promised something like that man, you got to do it. I: No, not necessarily. (Well I say you do) Is it important to keep a promise to someone who you won’t see again?  133. I: No. Cause if your not going to see the person again then why the flick make a promise. S: Well it doesn’t matter man, it’s still your word man. What else do you have, they can take away your car, they can take away your boat whatever man but they can’t take away what you promise you know. Your word is the only thing you have that they can’t take away. A promise is the most important thing there is. I: Say I promise you my boat okay when you come back. You never come back. Fuck you. Your beat. S: I promised you that so I still got to do that. I: So you can still break it. You promise me that your not going to do crime, if you get out tomorrow which is bulishit but and I come back in and you’re in here again. I said, hey you promised me that you wouldn’t you do crime and you said oh I flicked up, sorry. -  5: Then your promise isn’t worth anything. If that’s what you say and then you lie about it. You still broke your promise man and that’s something you can’t do. Next time you promise somebody something their going to know, oh yah sure. If somebody promises you something, you expect them to do it right. But how can you expect somebody to do what they say if you don’t do what you say. I: But how are they going to know that you fucked them around. S: They won’t know but you will. You will (So) Your not a good person man cause you lied. I: So it’s my choice if I want to do it. In general what should be the authority of a father over his son? S: I think a father should have authority over him to a point. But if a father tells you something he can’t lie about it and go back on his word. I: Well if something comes up then it’s a different story. S: What makes it better for you to go fishing than me go to camp. Because I’m getting educated and your just going with some punk-assed punks. You still don’t have the authority to tell me what to do man you know. I made the money, I can spend it, I shouldn’t have to give it to you. I: Your only 14. (Am I) Yes. S: It doesn’t matter man, I can spend money man. $100 isn’t much. If I knew I would have spent it as soon as I got it if I wasn’t going to camp. I: Your not going to camp. (Well it’s my money) No man you can’t go to camp man. (Yah I can) No you can’t. S: You promised me that man. I: So I broke my promise.  134. S: It’s wrong you know. You don’t have that authority, you can’t tell me. (I’m your father) What do you think is the most important thing a son should be concerned about in his relationship with his father? Why is that the most important thing? I: Cause well, a son should obey his father. Like if something came up then he could do it another time. He could do it next year. S: His living man, his loving each other you know like. Kind of what they say about each other you know. The most important thing man is understanding each other. Even if you disagree man an understanding of each other you know. (Who cares) What do you mean who cares?  I: I think you should obey me seeing that I’m your father and I am the one who brought you into this world. I could take you out. S: I’m not going to do totally stupid things just cause you tell me you know. What’s this shit man, you can’t tell me what to do. IVs understanding each other that’s what the most important thing is. I: I can tell you to do a chore right now, and you have to go do it cause I’m your father and you should obey your father. S: I should but I still don’t think that’s the most important thing.  135. Day 2 I: Should Bob report the bigger boys? No. I don’t think so cause if he rides somebody bigger than himself then it’s his turn to get pounded on. S: Yah, but people shouldn’t be allowed to do that. I: But still that is a fact of life. (They shouldn’t be able to do that.) They shouldn’t be able to do what? I think people should be able to beat up on whoever they want cause that’s just the way life goes. You know what I mean. (Yah) It don’t matter. (What don’t matter?) S: Yah it matters, it fucking hurts. I: Yah it does but still you get into a fight, your in pain for a couple of days and then you (Too much noise from P.A. announcements) Carlos report the boys S: Yah, I think both of them, they should both be reported, they shouldn’t be allowed to do that you know, it’s taking advantage of somebody that’s smaller and using them. I: Fuck that. I don’t think he should report it cause if he does report it he’s just going to get it worse next time. S: Well not if they, maybe teach those kids to fucking stop doing that. I: I don’t know but I don’t think S: I think your wrong. I: I know I’m right. I’m right sorry. In thinking about whether to report the incident to the Staff Bob thinks about the fact that although the two boys are not that well like on the Unit, should that be an important consideration for Bob? I: I don’t think so. Cause well, the kids are bigger flick. I don’t think Bob should put through a report S: Well it doesn’t matter I just asked you if you should consider that other people don’t like him on the Unit? I: So you know, It’s not his problem. Take into consideration man, no-one else likes him. Who cares man, their sort of geeks. It’s the other people’s problem, it’s not his problem. 5: Yah but other people don’t go saying we want him off the Unit cause we donft like him. It’s not other people’s problem.  136. I: But no-one likes him so if they fucking tell the Staff on this guy (then Carlos) it won’t make any difference man. cause everybody fucking doesn’t like him. No-one cares you know. If somebody ever likes him then everybody’s going to be mad at you, but no-one likes him anyways and everybody’s not going to fucking care or anything. S: Yah but the people that do like him, their going to be after that person that ratted them off. (Right) There’s a lot more people on their side than people who don’t like them. I: I still disagree. S: Why is it important that people not be afraid of being physically hurt? I think it is important, well it is impossible not to be afraid to some extent. You have to be able to feel safe and if you can’t feel physically safe I’m sure you can’t feel morally safe, and then you turn into a mental case and be paranoid with everybody. I: But you have to, you gotta, there’s times, there’s things in life that you just can’t, you won’t trust or there’s things in life that you really want to do and you don’t do it. S: Well how should I put this, well basically everyday someone get’s into a fight like if somebody gets into a fight, like if you walk down the street and this guy doesn’t like you, he’s gonna pound the shit out of you cause he don’t like you. You can’t be safe all the time. (Well why not?) Cause well Jesus Christ cause it’s just a fact that you can’t be safe all the time and you always have to be expecting something. I: Is it important though? You said it’s not important that people feel safe or physically safe, but why isn’t it important? Why isn’t it important? (Cause) Cause why? (I’d trying to think) I said it’s important because if you can’t feel safe physically you can’t feel safe emotionally and then your scared of everything and paranoid if that’s the case. You can’t trust anybody if you think their going to punch you out. S: That’s when you start working on your own, like that’s why you work by yourself like be selfish and do everything by yourself for your own. I: You can’t do everything by yourself man, you need people to help you. (Not all the time) All the time yah. S: No. I don’t think so, cause there’s some things that you have to do on your own and some things that If Bob is not Carlos’s friend, does that make any difference as to what should be done in this situation? I: Yah, I think it does. Because if Bob wasn’t Carlos’s friend then Bob would keep his nose out of it and not say nothing at all. He shouldn’t say anything you know if Bob wasn’t his friend. If Carlos isn’t a friend you don’t get involved with him. You don’t get involved with things that you don’t have anything to do with. S: But there’s some things you want to get your nose into just for the sake of feeling important. I: Where do you get these ideas. What it says is would it make a difference if Bob wasn’t Carlo&s friend. If Bob was Carlos’s friend yah, but if he’s not your friend man then get out of it. Don’t get involved. I would think so.  137. S: Yah but if you see someone getting the shit kicked out of them on the Street like an old lady or something like that you’d go and help her wouldn’t you. I would. (It’s none of your business) So. You just don’t go seeing this and just look at them and go well okay and walk away. You just can’t do that.  I: I’d do that. What if you jumped in man and maybe got beat up you know. S: So that’s why you have a friend or just stay out of it. Well is this cause you gotta help somebody, you gotta help. I: It’s easier to stay out man. Stay out of it and there’s no problem eh. I don’t why your risking your neck for some person you don’t even know. They’ve got friends. It’s not worth it man cause why are you risking your neck for somebody you don’t even know. S: Cause then you could probably become friends with them and then so forth. I: Who wants to become friends with somebody who’s getting beat up. 5: A lot of people. Everybody gets beat up. Everybody. But there’s some things that people help like, I don’t mean like jump in and start kicking the piss out of the other guy that’s beating this one guy up, I mean like breaking the fight up. I: How do you plan doing that? S: Well you might as well break it up before the cops get there anyway cause then there’s going to be charges laid. I: Why don’t you just leave. You don’t even know these people. Let them have charges laid, who cares. What if Bob didn’t even know Carlos does that make a difference as to what should be done in the situation? 5: That was the last one we did. We lumped the two together. I: Yah I know but what happens if you just know him but he’s not your friend? I: Then you’d help him out. 5: What has he done for you man? He’s just some guy you know. He hasn’t done nothing for you, why should you do anything for him man. I: Cause then you could return the favor some time. 5: Yah but then you’d get the shit kicked out of you. He could be bigger man. So you could just break it up like say hey, you don’t have to kick the piss out of him, you don’t have to do that, just break it up and tell him hey hold on like. I: Then they’d grab you and beat you up. 5: Well I just totally disagree with you on that one. In general should the two boys be punished? Why or why not?  138. S: No. (Why not?) Cause they probably have a reason. Cause they did what they did and they wanted to so it’s their choice. I: Oh, so flick I want to go kill, so I shouldn’t get punished for that. You gotta have some fucking law and order. S: So. Like who cares. I: Everybody cares or the whole fucking world would be chaos. People do it cause they want to sure, but do it man but you can expect to get punished for what you do. You can’t just have people walking around killing each other. S: No, I disagree with that. I don’t.think there should be a law system or whatever. I feel they should be able to do what they fucking want to do. I: Face reality. Your crazy. (I’m supposed to disagree with you) Well not that much. If they don’t get punished man their just going to think well I can get away with it, I got away with this and next time they’ll do something else you know. Every time they will just get away with it. Everything’s just going to get out of hand man. Stop they now, it will be worse later. So you think they should be just let off. (Yah) Why? S: Because, well because they, they only beat the fuck out of them cause they had a reason. I: There are other ways you can deal with it. (No) Yes (No) S: Well I just don’t think so. I: They have to be punished though. S: No they don’t, they can just wait until a week or so down the road until they do something more serious like maybe kill somebody. I: Oh, that’s a good argument. Why if they don’t kill nobody S: Cause it’s a minor crime, it’s minor, it’s nothing important. What is the most important thing to consider in being a friend? I: The most important thing for being a friend is to stand up for each other man. S: No, but if they don’t do anything for you then (their not your friend man) You can hang out with them and whatever but they still don’t have to (their not friends then) I: What is the most important thing with friends. S: To be there for you like when you get in shit. I: What do you mean when you get in shit? S: Like standing up for you, like say Carlos was getting the piss kicked out of him by these two older guys right. A friend would come up to the guys and say hey leave him alone or stuff like that.  139. I: Why don’t they jump in there and fight with him man? Only two people there is, is friends for each other.  Back each other up man.  Should the Tailor report Vaijean to the police? S: I don’t think so. (Why not?) Cause like if he’s got his own business and all this then I don’t think he should report him because he’s making money, he’s got a good job he’s not causing any trouble. I: Well he’s still a criminal man. He’s still wanted by the law. (So) So. S: Well if he got like, if he’s making good wages or whatever then why should he be turned in. I: Because he committed a crime man. You gotta pay for what you do. S: If you escape then I think and ten or twenty years later you get busted, you have your own home and country and all this shit, you wouldn’t want to get like busted for I: Well probably not, but even if I didn’t want to, it’s my fault for doing it, nobody else’s man. 5: So then I don’t think he should turn him in. (Well why not?) Because he’s making his own way, he’s not doing any more crime. I: You don’t think people should pay for what they do. S No, well to a certain extent yah. I: Well he was a criminal man. He hasn’t done what they gave him for time man. S: Well if you get out or you escape it’s like oh wow. If your not getting into shit and all this then I think it’s okay. I: But don’t you think you should have to pay for what you did? S: But he already did a couple of years. I: But he got six years man. The government says if you gotta do six years cause you stole and he only did two years man so he owes the government four years to make up for what he did. S: But this is twenty years later. I: It doesn’t matter. Look at war criminals man, that was forty years ago, fifty years ago. They still go to jail man because their criminals. If that happens man, everybody’s going to escape and just fucking get away, and that way.... (P.A. making too much noise. S: Our security has nothing to do with it man. I: People do finish their time you know. You gotta take responsibility for what you do. I do something, I don’t care, you know if they catch me then they catch me you know.  140. Does a person have a duty or an obligation to report an escaped convict? I: Yah they do man, cause their a convict man. People want unity you know. Maybe this Vaijean guy isn’t doing something bad man but maybe somebody else who escaped and has been out for twenty years is man. S: Yah, but that’s two different people. I: That doesn’t matter man. It’s the same thing man. Their escaped convicts. Escaped convicts, you gotta turn them in man cause you never know. S: Well if he managed to stay out for twenty years, have his own company or factory, making his own money, then he’s obviously doing something right. I: You can’t judge one person and then another man. You have to look at everybody the same. Cause even judging probably wouldn’t make no difference cause he’s a fucking bum he committed a crime. It’s the same thing man, both of them do the same time. S: Not necessarily cause there’s rich people that have the money to buy things off. I: Everybody’s the same man. It’s your duty man to uphold the law to keep everything running smooth in your country. You can’t have people getting away with crime or everybody would be doing crime and getting away with crime man. Keep it orderly. S: But it’s also the tailor’s decision. I: It’s his decision, but should he report the person morally. (No) Suppose Vaijean were a close friend of the tailor’s. Should he report Vaijean? S: No, cause he’s a friend. A friend would help him, not get away from the cops but he wouldn’t rat him off. I’d say like I know where your from and all this stuff. I don’t think he should tell the cops where he is. I: That’s what you said before. What’s the difference if he’s your friend or if he’s not your friend. 5: Well cause if he’s your friend you just don’t go around telling the cops where your friends are and all this. I: But he’s still a criminal man. (So) 5: If he’s a friend he would help him, he would say like help him out, like tell him that the cops are looking for him and well give him a place to stay or whatever. I: The cops aren’t looking for him or anything man. It’s just this guy knows who he is. If Vaijean were reported should the Judge send him back to jail or let him go free? 5: I think he should let him go free man. He’s a good person now you know, he made a mistake and he has turned it around. I: He’s got four years to do left, so why doesn’t he just do it.  141. S: Because he’s turned his life around man. That’s what jail is supposed to do, rehabilitate you and stuff, and it looks like he did a pretty good job on himself. I: But he’s also got time left. S: Well make him do time, make him pay money or restitution or something like that but you can’t take away his life after he’s done so much with it. Helped everybody you know, there’s more to it than that. I: He’s still a criminal. He did crime, he’s got to pay for it. S: Look how much he’s accomplished man. His own factory and built a hospital for poor people. You got to take into consideration what’s being done with his life, how much he’s turned it around. I: But if other criminals take off, they have their own house, they might not own the company where they work but they’ll still get the time. S: Twenty years later, I doubt it. Twenty years later man, your not going to send somebody back to jail man. In general should people who break the law be punished? I: Yes, cause it teaches them a lesson. S: If they don’t get caught they shouldn’t be punished man. Let him get away with it you know. It’s not his fault if society is being so stupid and don’t get him. I: But if he does get caught then he should be punished because then it will teach him not to do it again. Or else not to be caught. S: Everybody, well not everybody, but the majority of people do crime and they have to be taught a lesson, like cause if you do something wrong at home you get grounded or whatever. Same thing. I: Your saying that everybody that commits a crime should have to go to jail. (Yah) So all those thousand of people that break crime every day throw in jail. (Yah) The guy who rips off $5 from petty cash to buy coffee should go to jail? (Yah, it’s a theft) Only if you get caught should you go to jail. Do you know how full the jails would be if everybody who got caught was thrown in jail? Any idea what would happen to all those people man if everybody who committed a crime went to jail? S: That’s why you make a big jail.  142. Day 3 Should Mario confess to taking the money? Why or why not? I: Now he didn’t do it actually. Based on morals. S: Well then why should he say he robbed him. I: He doesn’t have to say he robbed him he just has to say no S: Yah but nobody gets T.R.’s I: Correct. Who cares. 5: A lot of people care. People who get T.R.’s. I: Well he doesn’t get any so he doesn’t care and he shouldn’t. I think he’s best say he didn’t take it. 5: He shouldn’t confess. (no) I: No-one said he’s supposed to man. Why confess to something you didn’t do? S: Just to get it out of the way. I: What do you mean get it out of the way. 5: So that people will quit harassing other people to start telling somebody where it is. I: Do you know what’s going to happen, then he’ll get all his T.A.’s and shit cut. S: So, if he wants to he will. I: Why doesn’t the person who took it confess? 5: Cause he doesn’t want to get busted. (What do you mean get busted?) He doesn’t want his T.A.’s to be cut. I: Why doesn’t he tell the guy that took it to give the money back? 5: They wouldn’t cause then they’d be a rat or something like that. I: No they won’t. Tell the person to give the money back. You know what I mean. I: Yah but still.... Should Mario report whoever stole the money? S: No, cause if he does then he’ll be labelled a rat. I: Well yah, he could tell them cause everybody’s losing T.A.’s and I don’t really agree with you on that. I think he should tell him to give the money back and if he doesn’t and they lose their T.A.’s then he should be punished cause he stole the money man.  143. What should the Staff do in this situation? S: They should leave it up to the kids to settle. I: Yah, but what should the Staff do about the money though like not everybody’s T.A. like. Do you think they should cut everybody’s... S: Yah, or else take it out of their account. I: But they don’t know which one it is. S: So, take it out of everybody’s. Same amount out of everybody’s until it’s all paid for. I: Well Mario said he would take the blame and pay for it. S: Is that what he said? I think everybody should pay. Well it has to be paid back, everything has to be paid back. I: I stole a lot of stuff and I never paid nothing back. S: That’s true but your doing time to pay for it. I: What percent though? I’m not sure. What if the other kids didn’t like the guy who stole? Should that make any difference what they should do? S: Well yah. Cause if they don’t like somebody then they’ll obviously set him up so they won’t get in shit. I: I know but should it make a difference if they tell or not? (no) Should they tell on him or what? S: Yes and no. In some cases like they should. (In this case) Well yah (They don’t like him man) So yah they should rat him off. I: But should they? (Yah) They should. (Yah) Why? S: Cause well nobody likes him. Nobody will care if he gets in shit or not. I: I know but isn’t it like being a rat on somebody you like the same as being a rat on somebody you don’t like. (No) Being a rat on somebody you don’t like is the same as being a rat on somebody you do like. S: No. (Why not?) Because somebody you don’t like you can set them up and they’ll get the shit. I: It’s the same thing man to rat people off. Well you shouldn’t tell on anybody you don’t like man because it’s the same thing man, it makes you look as bad man. You could make them give it back man, you could make them give back the money. Well you shouldn’t rat on somebody no matter what. S: It’s my opinion.  144. In general what authority should a Staff member have over a resident and why? I: The Staff has ultimate control over a resident, man. I don’t know why. It’s because you broke the law, man, and you’re in a controlled environment, you know. That’s the way it is. S: Yah, but they have to help you in some ways too. I: You have to make your decisions about what you do after. But when your in here they do, man. They have to, man. It doesn’t matter what you say, you know? You don’t have much say about anything. S: You have to live your life in here as well. In general is it important for... S: It’s important to think about with that person what the consequences are both in what you see and what you do. I: What if the person in here deserves it? S: Deserves what? Well it doesn’t say that in the situation in general. Yah but he gets in shit for something. He deserves to get in shit. He deserves to get in shit if he did it that’s all. He deserves to get in shit if he did it. I: Well say it again. Keep talking. S: If you get in shit then you got to pay the price. I don’t know.  Should the Captain order somebody else to go or should he go himself? I: He should go himself. (Why?) Yah but he should go himself cause he can always tell somebody where to go, how to retreat but he can’t send somebody to their death because it’s the same thing as a death order. He should do it himself if he wants to save the others. S: Well then who’s going to be in charge of the troops then? I: Well the next higher ranking officer. Just delegate him to be in charge. S: But still he’s the only one who knows and then he can’t describe how it’s supposed to be cause he knows. I: Yah, but he could just tell the guy what to do. S: He can’t describe it as well. I: Well it’s a risk you have to take, man. Either one man or all men and you know you can’t send somebody to death so he’s going to have to do it himself if he wants it done. Should the Captain send a man, or he can use a lottery when it means sending him to his death and why?  145. I: The Captain should use a lottery. I don’t know if the Captain should use a lottery to send someone on the mission or should he go himself. No, I’m right you can’t send a man. If you use a lottery it means sending him to his death. I don’t think he should send a man to his death, man. Like it’s not right. S: Well if nobody would volunteer then he has to order somebody. I: Why doesn’t he go himself man. (He’s the Captain) It doesn’t matter man he can’t be God man. S: He’s the higher rank, he would be able to pick somebody. I: Do you think they’d actually send somebody to his death man. They’d actually do it. S: Yes. If you were ordered to, you would. I think yes he would. Should the Captain go himself knowing that he’s the men get back safely? S: I think he should have other guys do it for him so that he can make sure that their going to get back safely. I: No. I think the Captain should go himself cause no-one else is going to go unless he orders them and it’s his obligation as an officer. 5: The Captain is needed. I: You can’t order somebody to death man. You can’t. (In the army you can) You can’t. It’s a suicide mission. He should do it himself. (The Jap’s did it.) The Jap’s were fucking nuts. You can’t send somebody to their death. (He has to do it, it’s his rank...) But don’t you think it’s morally wrong to order somebody to death. Wouldn’t you feel guilt. S: No. In the army, yah, you would. You’d feel guilt man but you still have to do it. (It’s wrong) Does the Captain have the right to order a man to his death if he think’s it is best? 5: He does cause he’s the higher rank and he can basically tell them what to do and... I: I know about that man, but death is different man, you can’t kill somebody. 5: The Captain figures right that he should seeing that (seeing what?) seeing that he’s the higher rank and that. What’s so important about a human life that make’s it important to save Why is that? I: Life man. There’s nothing else other than life man. There’s nothing else important. S: if you can. I: You say you can, like maybe you can but I mean S: Still it’s like, you have to do as ordered.  146. I: I wouldn’t would you. (Yah) If you were told to go back and kill yourself you would. S: If it was an order yah, but if he just said would you go back and kill yourself wouldn’t, but if it was an order. I I: That doesn’t make it right man. It’s human You got nothing else. There’s nothing else. life and what else is there but human life. precious thing you got man. You could have Once your dead, your dead. It’s the most but if you don’t have life then your just fuckall the money and everything you want man ing man so how can you take it away and tell them dead eh. Then you don’t have nothing to give it up. S: Well it’s not just for me it’s for the rest of the guys too. I: Well why doesn’t somebody else go and do it. S: Because you were ordered. I: And why doesn’t he go then. S: Because he’s needed to lead the troops.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items