The education of the complete moral person

Copyright © 1998, Marvin W. Berkowitz
Educators have long been concerned with the kind of people that leave their schools. The literature on education is filled with volumes about how to nurture the moral formation of pupils. Recently in Scotland, much has been said about this topic by such leading groups as the Scottish Office Education Department, the Scottish Consultative Council on the Curriculum, and the Gordon Cook Foundation. It is clear that there is great concern both in Scotland and elsewhere about this topic. Yet there are many problems that impede the progress toward an effective, justifiable, and practical approach to educating for morality. In this paper, it will be argued that much of the problem stems directly from the dis-integrated state of the field. Therefore, an attempt will be made to integrate the disparate approaches and suggest a model for the education of the complete moral person.

Three Problems

Humpty-Dumpty

Let us begin by reflecting on the following two questions. First, who would you nominate, living or dead, real or fictional, as a moral exemplar? Second, why did you nominate that particular individual; i.e., what about that person let you to choose him or her? Now if you think for a while, you will probably arrive at a few characteristics of the person that led you to identify your nominee as a particularly moral person. And if we were to collect all of the responses from all of the people who read this paper, we would generate a fairly long and diverse set of characteristics of the exemplary moral person. Indeed Lawrence Walker and his colleagues (Walker, Pitts, Hennig, & Matsuba, in press) have done just that. Their twelve most common descriptors, in descending order of prevalence, are: compassionate/caring, consistent, honest, self-sacrificing, open-minded, thoughtful/rational, socially active, just, courageous, virtuous, autonomous, and empathic/sensitive. Most readers will likely find that their responses are included in that list. Whereas Walker's research was conducted in Canada, I have less formally found the same basic set of responses in the USA, Scotland, Switzerland and the Netherlands.

Colby and Damon (1992) in a study of actual nominated moral exemplars find that, in addition to the moral characteristics provided for nominators, the nominees manifested a number of other attributes. These included openness to moral transformation, a balance of certainty and truth-seeking, optimism and humour, faith, and a clear union of morality with one's sense of self.

What should slowly be growing clear is that the nature of the moral person is quite complex and diverse. There are two important points to note about this. First, most authors do not even seem to recognise the problem as they focus on their favoured aspect of the moral person, a problem we will return to in a later section. Second, it is apparent that, like Humpty-Dumpty, the moral person has been shattered into bits; indeed into too many bits to be functionally whole again. Closer inspection of the cornucopia of moral characteristics reveals not only a plethora of diverse aspects, but a similarly diverse and undifferentiated set of clusters of moral characteristics; i.e., there are aspects of personality, and of behaviour, and of cognition, etc. In order to simplify this, a number of authors have offered superordinate categories of morality. Lickona (1991a) suggested that to be a moral person one must know the good, desire the good, and do the good. Walker et al. (in press) describe the domain of morality as being comprised of moral emotion, moral behaviour, moral thought, commitment, and virtue. Indeed, it is necessary to try to put the moral Humpty-Dumpty back together again. We need, in essence, an anatomy of the moral person in order to be able to design our educational endeavours so as to optimally contribute to the formation of the future citizens of our societies. After all, as Lickona (1983) reminds us in the preface to his book on parenting for morality, "A child is the only known substance from which a responsible adult can be made". To optimally educate children to be responsible adults, we need to understand more fully what it takes to be a responsible adult. Indeed we shall offer a preliminary model of the anatomy of the moral person. But before we can do so, we must address two more problems that are endemic to the literature.

The Tower of Babel

One of the less noticed and more significant impediments to any integrated model of moral education is language. There is a profound confusion of rhetoric in this field. Surely it is not a symptom unique to this specific field.

Educational discourse...is marked by a good deal of vagueness...I have encountered so little in the way of challenge to a more complete elucidation and clarification...Educational writing has for long suffered from a distressing lack of precision. (Bantock, 1965, p. 11).
Indeed, Bantock's statement is as true today as it was 30 years ago and is as true of moral education as he suggested it was of education in general.

The field that this paper concerns is currently alternatively referred to as values education, character education, moral education, personal and social education, citizenship education, civic education, religious education, moralogy, and democratic education, among other rubrics. Now, it would not be so worrisome nor so troublesome if these terms were truly interchangeable, but they are not. Values and character are not equivalent, as we shall demonstrate in a later section. Furthermore, values and character are not necessarily in the domain of morals. One can speak of moral values and non-moral values, for example.

A useful way to demonstrate this last distinction is through the work of Elliot Turiel and his colleagues. Turiel (1983) offers the distinction between three domains of social knowledge: moral, social-conventional, personal. The moral domain is characterised by issues of universal prescriptivity deriving from intrinsic harmfulness. Furthermore moral issues are unalterable in their rightness or wrongness. For example, killing is a moral issue because the ban on killing is applicable to all humans regardless of nationality, race, gender, religion, or ethnicity. It is wrong because it intrinsically robs victims of their mortal existence. And it is unalterable; i.e., no authority has the power to make killing morally right. (This is not to argue that there may not be circumstances in which killing is the lesser of two evils; only that in general, killing is morally wrong). The domain of social-conventions is a bit different. It concerns issues that are considered wrong not because of their intrinsic nature, but rather because of social agreement that they are wrong. They are therefore alterable by the appropriate social authorities and only apply to those within the purview of the social agreement. In some cultures, for example, it may be scandalous for an adult male to hold hands with an unmarried, unrelated adult female. In others it would be perfectly acceptable. The same may be said for two adult males holding hands in public as they walk casually down the street. Holding hands does no intrinsic harm, but it can be harmful if there is social agreement banning it. Without such agreement, it is perfectly acceptable. The critical point is that such social agreement does not make it morally wrong. Therefore, when schools teach social conventions, they must be prepared for moral evaluation of the validity of those conventions, no matter how widely accepted they may be. The personal domain is concerned with those issues which should not be socially regulated; i.e., matters of personal preference and taste. No one should be able to tell me what my favourite colour should be or whether I should prefer the taste of chocolate ice cream with or without fresh raspberries. Schools certainly should not teach children which flavours or colours they ought to prefer.

Now it should be clearer that values, for example, may fall into any of these domains. Does values education therefore refer to all values? One should hope not, for as we have just demonstrated, some values should not be socially regulated or indoctrinated. Does values education refer then to moral and social-conventional values, or only to one or the other? You will find little agreement about this, indeed little discussion of this, in the literature. Yet it is a very important issue. For not only is there a lack of truth in advertising when we offer values education but include different contents in the package, but we need to be concerned about the fact that moral values and social-conventional values need to be educated in different ways, at least in part because they require different justifications. Furthermore, research has demonstrated that very young children, teachers, and very religious individuals, among others, recognise that fact spontaneously in both their actions and their judgements.

We have now demonstrated that the babbling in this field is partly due to the usage of terms that are not interchangeable. Unfortunately, most educators are apparently not aware of this. For example, Lickona (1991a) titled his book Educating for Character, the first section of the book is "Educating for values and character", and, throughout, the terms "values", "character", "moral values", "moral education", "values education", "moral reasoning", etc. are used seemingly as synonyms and with little definition. Ryan and Lickona (1987), in the opening paragraph of Ryan and McClean's Character development in schools and beyond, use the following pot pourri of terminology:

Concern for the values and morals of the young is an enduring adult preoccupation. Down through recorded history, this worry about the character of the younger generation is evident. Concern, however, has never been enough to ensure that the young possess the type of character that can sustain the individual and society. Some societies have failed to transmit their values to the young. (p. 3)
Other authors equate non-equivalent terms. Carr (1995) suggests that "virtues are values". Others adopt terms merely because of their lay connotations or political effect. William Bennett (1991), former US Secretary of Education, has argued that.

Indeed, parallel with Bennett's wishes, the terminology in the United States has shifted from the former preference for "moral education" to the current favouring of "character education", whereas in Scotland the terminology has moved from "moral education" to "values education". Nevertheless in neither country is the issue settled. This is partly due to the continuing existence of factions and to the failure to define terminology. Even when a term is generally widely accepted, as the term "values" is in Scotland, there is still dissent, as exemplified by the debate between whether we should be doing "values education" or "values in education" (Barr, 1995).

Humpty-Dumpty has not only broken into pieces, but we find ourselves unable to agree on the names of the pieces or even what the task is. No wonder the field is so fractionated. There is yet another side to this problem that will be addressed before we move on to attempting the beginnings of a solution.

The Blind Men and the Elephant

The parable of the blind men and the elephant attributes the blind men's failure to accurately describe the entirety of the elephant to their physical handicap of a lack of sight. Would that the reasons for social scientists' and educators' failures to describe the entirety of the moral person were so obvious and guiltless. Rather it appears that the theoretical parochialism and arrogance of many of them are the root causes of the problem. The typical scenario is for various factions to propose diverse competing theories of moral development and moral education which are intended often as solutions to the problems of youth and society. Unfortunately, too much emphasis is placed on the "competing" and not enough on the "solutions". What results is a professional dialogue laden with antagonism, "cheap shots", and rather close-minded competition for the status as the "right answer".

William Perry (1970) demonstrated that such a search for the ultimate single right answer reflects a rather immature sense of knowledge. Such a "dualistic" epistemology should be more representative of early adolescent thinking than of high level academic discourse. Yet the field is rife with such perspectives (e.g., Bennett, 1991; Kohlberg & Mayer, 1972; Piaget, 1965; Wynne & Ryan, 1993). Asking which theory is right is simply asking the wrong question. Rarely is there simply a choice between a single right theory and a set of wrong theories. Rather, the question should be how best to explain and influence moral growth given all the available knowledge and theoretical perspectives. Such a more dialectical approach makes optimal use of knowledge from all the diverse "camps".

Certainly there are incompatible aspects of some approaches, but that simply makes the task more difficult, not impossible. The challenge is to sift through what each approach has to offer, find the commonalties and adjudicate between the discrepancies. This is a decidedly more productive course of action than the wholesale rejection of seemingly opposing theories, an approach that has unfortunately been so typical of the field.

It is important to note that there are attempts to be more integrative. For example, the single most impressive and successful program in moral education, the Child Development Project in San Ramon, California (Watson, Solomon, Battistich, Schaps, & Solomon, 1989), is a quite eclectic and wide-ranging approach to child moral development. James Rest (1985) was one of the first theorists to attempt a bridging of domains and Lickona (1983, 1991a) is the single most successful writer in the field while integrating Kohlbergian structuralism with character education and a variety of other perspectives, although his rather strident stances on specific cultural issues often precludes some readers from noticing how successfully he has integrated such perspectives (Rosenblatt, 1995).

One way out of this morass is to look for a grand theory that can encompass the major aspects of moral growth and moral education under one conceptual umbrella. Unfortunately such a grand theory is no where to be seen. Teachers tend to ignore theory and focus on the technical aspects of practice (Carr, 1986). Theorists, as already noted, tend to fall into one or another of various warring camps. To make matters even more difficult, this field actually spans a number of disciplines, most notably psychology, education, philosophy, and theology, and has its roots in quite contrasting philosophical models (Carr, 1991). Grand theory is important in part because so many of the failures in the field are due to the failure to build upon sound justifications derived from sound theory. Models instead ignore the fundamental nature of child development or rely upon an unexamined or misguided ethical theory (such as relativism), or fail to distinguish between morality and other social domains, etc.

Until we systematically reflect on the theoretical groundings of our educational work and on the ways in which we can benefit from the (even seemingly incompatible) work of others, we suffer the problems of the blind men and the elephant. Not only do we describe only a piece of the phenomenon we intend to understand in full, but we desperately refuse to listen to others who are privy to a different perspective on the broader issue, thereby impoverishing ourselves but more importantly our children and our society.

The Moral Anatomy

Just as the description of the problems with moral education began by focusing on the nature of the moral person, so must the offered solution begin. For any reasonable approach to education must be based on a clear and accurate understanding the nature of persons. Hence, an effective and justifiable approach to moral education must begin with a clear and accurate understanding of the nature of the moral person. We have already suggested that such an understanding is sorely lacking, but have pointed to some recent attempts to more adequately categorise the diverse aspects of moral personhood. Here, we shall offer a somewhat overlapping alternative scheme for categorising the components of the moral person; i.e. an "anatomy" of moral personhood. This model is not intended to represent some "truth" about the parts that comprise the morality of a person; indeed, that would be quite contradictory to the dialectical and integrative approach we are advocating. Rather this is more of a taxonomy than any sort of true anatomy. As noted earlier, some organisational scheme is required to make order and sense of the diversity of moral characteristics. This seven part model seems as good a place to start as any. At the very least it should suffice to cover most of what comprises the moral person and as a basis for generating further discussion about the nature of moral personhood. The seven parts of the moral person are as follows: moral behaviour, moral values, moral character, moral reason, moral emotion, moral identity, meta-moral characteristics. Each will be discussed in turn.

Moral behaviour

It is difficult to dispute the fact that the ultimate goal of moral education is behavioural. This is not to say that we should focus exclusively or even predominantly on behaviour control or behaviour management. It is merely to acknowledge that moral education that doesn't affect behaviour is toothless and largely irrelevant. I know of no approaches to moral education that do not consider the behaviour of pupils to be relevant and even central. The difference lies in how each approach proposes to impact on behaviour, not in whether it proposes to do so.

Two points may make this particularly evident. First, many of the opening rationales for moral education programmes focus predominantly, if not exclusively, on a litany of the behavioural ills of today's youth. The opening three pages of Lickona's (1983) "case for values education", the first three paragraphs of the preface to Damon's (1988) The moral child, the opening paragraph in Wynne (1991), the opening salvos in Ryan and Lickona's introductory chapter in Ryan and McClean's (1987) volume on character education, and the opening justification of Schulman and Mekler's (1985) Bringing up a moral child all focus on a chronicle of the modern ills of youth and society, such as high rates of substance use, suicide, delinquency, pregnancy, and sexually-transmitted diseases, as the focal justification for engaging in moral education.

Second, many of the criticisms levelled at competing models focus on their inability to influence behaviour. Now this is a particularly tricky issue which warrants some elaboration. It is tricky in part because of the lack of meaningful behavioural outcome research (or any meaningful outcome research for that matter) upon which to base valid conclusions (Laud & Berkowitz, 1995; Leming, 1993). It is further complicated by some fairly inaccurate accusations. For example, Kohlberg was criticised for educating for moral reasoning which was unrelated to behaviour. The rationale was that, even if Kohlberg and his fellow structuralists were successful in promoting the development of moral reasoning capacities, it was a wasted enterprise because such capacities had nothing to do with what people actually do. Unfortunately, this assessment was quite inaccurate, because there is a very substantial body of literature demonstrating a significant relation between one's capacity for moral reason and one's moral behaviour (Blasi, 1980; Kohlberg & Candee, 1984; Rest, 1979). This misinterpretation was due both to the "Blind men" syndrome described above and to the fact that critics mistakenly assumed an isomorphism between reason and behaviour, when, as has been suggested here, the relation between reason and behaviour is complex and indirect due to the complexity of the moral person.

We shall include here all behaviour, overt or covert, physical or verbal or mental, that affects the rights or well-being of either oneself or others. Unfortunately this definition is not fully adequate and there is strong disagreement on the criteria for moral behaviour. It is worth noting the bases for such disagreement even if we cannot resolve them here. Three such issues will therefore be identified. First, can one identify a moral behaviour by its intrinsic nature, or must one know the actor's intentions. Aristotle (1987) argued that behaviour can be classified as moral or not without reference to the actor. Kohlberg (Kohlberg & Candee, 1984) argued that intentions were critical to any definition of an act as moral or immoral. For example, if I were to try to maliciously steer you to an investment that was sure to lead to your poverty, but discover after the fact that this investment miraculously became sound and led instead to your fortune, am I benevolent? Even if many acts are fundamentally morally ambiguous without reference to the actor's intentions, many other acts are at least probabilistically classifiable on moral grounds. It would be an absurdly bizarre and improbable set of circumstances that would lead to a moral intention to torture a young child or rape a woman, for example.

Second, must the locus of acts be other humans for the act to be eligible for classification as moral or immoral? Some have argued that acts are only moral if they affect other humans whereas others suggest that self-harm is also in the moral domain (cf., Berkowitz, Kahn, Mulry, & Piette, in press). Some argue that only acts that affect human beings are in the moral domain whereas others include acts that affect animals or even the environment (Lickona, 1991a).

Third, do all behaviours that a particular tradition, culture or religion deems moral or immoral warrant that status? Some authors (e.g., Schweder, Turiel, & Much, 1981) argue that morality is culturally justified whereas others (e.g., Turiel, 1983) suggest that morality transcends social definition. It is impossible to resolve these issues here, but it is nonetheless important to be aware that such conceptual problems exist.

Moral values

"Values" is probably the most ubiquitous term in this field and certainly is so currently in Scotland. It is the term of choice for the SCCC and for the Gordon Cook Foundation. Nonetheless, it is very difficult to find any semblance of a consensus on what a value is or to agree on what types of values should be educated (not to mention the seemingly eternal problem of which values to educate). Let us begin our examination of moral values by trying to define the term. There are generally assumed to be two fundamental aspects to a value: cognition or belief, and emotion. Others include behaviour as well. Certainly a value entails cognition, for a value is centrally a belief in the desirability or lack of desirability of the focus of the value. It does not seem controversial to argue that values also entail emotion. As a verb, valuing refers to caring about something, feeling that it is important. As a noun, a value entails a feeling that something is worthwhile or not. It is less clear that values fundamentally entail behaviours other than the covert feelings and cognitions that have just been described. For instance, efforts to prevent undesirable behaviours such as unsafe sexual practices and substance use often change one's attitudes toward such behaviours without actually affecting the behaviours themselves.

Rokeach (1968, 1973) is recognised as the foremost theorist and researcher of values. He defines a value as "an enduring belief that a specific mode of conduct or end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence" (Rokeach, 1973, p. 5). He further argues that such values are organised into a value system on a "continuum of relative importance" (p. 5). Notice that Rokeach does not suggest that values are behaviours, rather that they are affectively-loaded cognitions about behaviours. Researchers in parallel fields, such as the study of attitudes (Eagly & Chaiken, 1993), make the same point; i.e., attitudes or values do relate to and predict behaviour, but are mediated by a variety of other variables so that the relation is not direct and certainly not isomorphic. This is much the same argument as offered above in defence of the relation between moral reasoning and behaviour.

Despite this definition, there is much disagreement as to what counts as a value. Rokeach (1973) felt that there are few values, that they are enduring, that they are universal, and that they are central. Others list many values. Some define values differently; e.g., the SCCC (1991) defines values as "principles or sets of principles which are consistent and which inform and direct our actions and activities" (p. 2). Carr (1995), as noted, argues that virtues are values, a sentiment that seems to be consistent with Lickona's (1991a) failure to differentiate between values and character. Despite all of these apparent disagreements, it is possible to look for common denominators among these definitions. Clearly, as already suggested, values entail beliefs and emotions. Whether character traits or behaviours are part of values is less a matter of agreement. Hence we will rely on Rokeach's (1973) definition offered above. Values are enduring tendencies to believe in the desirability or lack of desirability of a behaviour or state of being.

If we take the SCCC's use of the term "principle" to refer to such a belief, then the definitions are largely in agreement, because the SCCC suggests that values guide behaviour, not that they entail behaviour. In fact, an examination of the SCCC's five central values supports this conclusion. Those values are: appreciation of learning, respect and caring for self, respect and caring for others, a sense of belonging, and social responsibility. These are closer to beliefs than principles, in that principles usually refer to rules for decision-making.

Another interesting slant on values comes from the structuralist camp. Kohlberg's Just Community schooling approach (Power, Higgins & Kohlberg, 1989a) is based in large part on the development of norms. Norms are affectively-laden beliefs about how people or institutions ought to be or ought to act. Kohlberg has long argued that moral values are at the heart of moral education.

The school, like the government, is an institution with a basic function of maintaining and transmitting...the consensual values of society. The most fundamental are termed moral values. (Kohlberg, 1967, p. 165).
The norms or values that Kohlberg and his colleagues focused on in moral education via the Just Community approach are caring, trust, collective responsibility, and participation.

What is particularly noteworthy about the Just Community approach to norms is that they are analysed at both the individual and school level. In fact, one aspect of school norms that Kohlberg and his colleagues focused on was their development from individually-held to communally-held status. Furthermore, they study the justifications for values, recognising that the endorsement of a particular value may be for quite variable reasons. Finally, they examine the relation between moral reason and values. This once again underscores the complexity of the moral person and brings our attention to the inter-relatedness of the various components of the moral anatomy.

This still leaves us with the ongoing problem of the type of values to be included in a program of moral education. If one is to be true to the term "values education" or "values in education" then one is not limited to moral values, but is limited to values and not reasoning or character, etc. If, on the other hand, one is to use the rubric "moral values education" then it is clear that the purview is even more limited; i.e., to values and to only those values in the moral domain. "Moral education" can include values among other components of the moral person, but those values must be in the moral domain. Because this discussion is being limited to moral education, we will also limit ourselves to the study of moral values, which we will define as enduring tendencies to believe in the moral rightness or wrongness of a behaviour or state of being.

Moral character

Thus far our moral person is comprised of behaviours and beliefs about the desirability of those behaviours or states of being. What is missing? Actually quite a bit. Let us then turn to moral character. Character is clearly related to both values and behaviour. As Wynne (1991) notes, "The roots of the word character are taken from the Greek 'to mark'. It suggest a focus on observable conduct" (p. 139). There are typically two major, and closely related, ways in which the term character is used. First, there is the notion of ways of acting. Character is the way one tends to act. If one acts dishonestly, cruelly, or selfishly then one manifests bad character. If on the other hand one behaves altruistically, honestly, and lovingly, one manifests good character. Note the strong focus on behaviour. The second major use of character refers to personality. One is a person of character. Lickona (1991b) defines character as "stable dispositions to respond to situations in moral ways--manifested in observable patterns (character traits) of kindness, honesty, responsibility, and a generalized respect for others" (p. 68). This comes closest to the original Aristotelian (1987) notion of virtue, a notion that is currently reinterpreted in moral education by Peters (1981), Power, Higgins, and Kohlberg (1989b), and Carr (1991). This notion of character is also closely tied to behaviour because virtue is believed to originate in habitual behaviours and to lead to moral behaviour.

If one takes a behavioural perspective, it seems plausible to divorce character, at least potentially, from cognition. Hence one could have the character trait of honesty without valuing honesty. One might consistently tell the truth without believing in the moral goodness of truthfulness; for example, in the case of one who is afraid to be punished for lying but believes there is nothing intrinsically wrong with lying. This notion of character is far from what most character educators mean by character. In fact, this is closer to what Aristotle might call a habit, an unreflective or unjustified tendency to act in a certain way, than to a virtue. Aristotle clearly argued that reflection, as we shall see later, is central to virtue. There must be awareness of the value of the behaviour. In this sense, a value is actually embedded in a virtue. Perhaps that is what Carr (1995) means when he says that a virtue is a value. But it must be emphasised that virtues are not reducible to values, for values do not necessarily entail behaviour while character traits do. In psychological terminology, character refers to personality and value refers to cognition.

It is important to once again refer back to our ongoing distinction between the moral and other domains. Moral character obviously refers to the moral domain. We can clearly have non-moral character traits, such as endurance, passivity, or sloppiness. Just as for values, whereas schools may choose to educate for non-moral character, the justifications for such an endeavour must be different. Endurance is not a desirable end in itself; one can endure simply to be vengeful or can endure a particular tedious TV program simply out of laziness or a lack of creativity. But endurance can serve moral purposes (as well as immoral and amoral purposes). We will return to his issue in the section on meta-moral characteristics.

Moral emotion

In a sense, by introducing moral values and moral character, we have already introduced emotion. We have seen that valuing entails feeling and that character entails valuing. Lickona (1991a) considered part of the triumvirate of the moral components to be "desiring the good". Walker et al. (in press) include emotion in their five part taxonomy. It is difficult to imagine a moral person who is unfeeling. Recent books about extremely violent children (Ewing, 1990; Magid & McKelvey, 1987) describe their lack of morality or conscience rooted in a lack of feeling about the welfare of others. They simply don't care. It is not merely coincidental that the title of Colby and Damon's (1992) book about moral exemplars is Some do care.

Now the notion of emotion that has already been introduced in our discussions of values and character is really a second-order sense of specifically moral emotion. Caring about moral values is no different than caring about amoral or immoral values; i.e., returning to our definition of values, "an enduring tendency to believe in the desirability" of a moral behaviour is not different than the parallel tendency to believe in the desirability of an immoral behaviour. In other words, it is the object of the belief not the emotional charge of the belief that gives it moral focus. Indeed, we can see this lack of specifically moral specificity in the description offered by Samay (1986) in considering the role of emotion in moral development. "Besides enumerating the reasons for which people act or should act, ethics must also call attention to the energies through which people are enabled to act for those reasons" (p. 71). Moral emotion is considered to be "the general power supply of all behaviour,...the integrating force of all knowledge and valuation" (p. 73).

There is another sense in which we may discuss moral emotion, a sense in which there is a specifically moral nature to the emotion. This refers to the subset of emotions that are generally considered to be areas of morality or immorality; emotions such as guilt, shame, empathy, sympathy, and compassion. There are two types of such moral emotions. There are the aversive moral emotions that signal some form of self-censorship or negative self-evaluation in reaction to an undesirable thought or action. Here we refer to emotions such as guilt and shame. There are also emotions that are fundamentally prosocial. They tend to reflect an emotional reaction to the plight of others. Examples of this second type of moral emotion are empathy and sympathy.

It is worth noting that all theories recognise some form of emotion as central to moral personhood. Even predominantly cognitive theories, like those to be discussed in the next section on reason, recognise emotion at least in the sense that Samay (1986) does; i.e., as an energy source for moral functioning (Kohlberg, 1976; Piaget, 1981). Others coming from the same tradition go further in recognising the more central role of emotion in the development of moral reason (e.g., Hoffman, 1987). Others suggest that these models are more accurately understood as the cognitive results of prior affective sensitivity (e.g., Caputo, 1986). However, one prioritises reasoning and affect, it is unlikely that any sound theory will not include both. Let us therefore turn to a discussion of moral reason.

Moral reason

It is tempting to end our anatomy here. After all, we now have a moral person who engages in moral behaviour, believes in the desirability of moral behaviours and states of being, manifests an enduring tendency to act in moral ways, and feels moral emotions. Surely this is sufficient to send our moral person out in the world to be an effective moral agent. Unfortunately it is not enough. This perhaps becomes clearest when we consider the situation in which one either has to adjudicate between conflicting moral values or moral character traits or has to generate values or behaviours for situations that were never anticipated; i.e., for novel moral dilemmas. I for one am glad that King Solomon was endowed with moral reason along with values and character. And I certainly would want any judge or jury that sat in judgement of my destiny to have that capacity as well.

How does one decide whether to act honestly or caringly in a situation where either choice precludes the other (do you tell a timid wall-flower friend with low self-esteem what you really think of her daring "new look" when to do so would shatter her new-found self-confidence?)? And how do you decide who gets custody of the frozen embryos left from a prior failed in vitro fertilisation attempt when a couple decides to divorce (an actual case)? It is necessary not only in these extreme circumstances, but in many moral situations, to be able to reason about right and wrong, to figure out what is right or wrong in a particular situation. This is the role of moral reason.

Moral reasoning is the most studied aspect of the moral person and relies upon the strongest theoretical framework (Damon, 1977; Gilligan, 1982; Kohlberg, 1976; Piaget, 1965). Kohlberg, the most influential thinker in the field of moral reason, describes moral reason as the logic of justice; the logic that one utilises to figure out what is fair or just, to balance the claims in a given moral conflict. He describes six stages of natural development of such forms of justice logic, moving from a calculation of self-interest, to a calculation of social interest, to a reliance upon abstract universal logical principles of justice.

Kohlberg is often accused (by both foes and friends) of being monomaniacal and seeing the moral person purely as intellect (e.g.., Walker et al., in press; Wynne & Ryan, 1993). This is actually quite far from the truth. Kohlberg has clearly related moral reason to behaviour (Kohlberg & Candee, 1984), emotion (Kohlberg, 1976), character (Power, Higgins, & Kohlberg, 1989b), and to values (Power et al., 1989a). Indeed in Kohlberg's last major work on moral education, it is argued that he

is not advocating a literal teaching of the (US) Constitution. Rather, he sees the Constitution as representing the moral principle of justice and claims that it is by teaching "justice" that the schools can legitimately transmit "the consensual values of society". For justice, seen from the perspective of moral development theory, is not a given value, which can be concretely transmitted to or imposed upon children, but it is a basic valuing process that underlies each person's capacity for moral judgment. (Power et al., 1989a, p. 15)
It should be clear then that values are highly related to moral reason and that reason is applied to the interpretation and evaluation of values. The same can be said for behaviour.

Aristotle (1987) introduced the concept of practical reason, which he differentiated from scientific reason. The role of practical reason was specifically moral. It provides the individual with the rational capacity to figure out how to apply the virtues in any given situation; i.e., to figure out what is the right thing to do in a given situation. Perhaps this similarity is what led Power et al. (1989b) to attempt to integrate Aristotelian virtue theory with Kohlbergian cognitive-structuralism. Others (e.g., Carr, 1995) argue that these two approaches are fundamentally incompatible due to their differing philosophical assumptions. We cannot resolve this issue here, and it really is the domain of philosophers like Carr, but in the service of our integrationist orientation we will accept at least the plausibility of finding useful common ground between the character/virtue educator's notion of reason and the cognitive-structuralist's notion of reason.

Moral identity

Up to this point, we have endowed our moral person with five anatomical components: moral behaviour, moral values, moral character, moral emotion, and moral reason. Once again we can recognise that we have done a fine job in constructing the core of a moral person. Someone who does the right thing, believes in and cares about goodness, consistently shows the personality traits of goodness, feels regret for transgressions and concern for others, and who can effectively figure out what is good or bad is certainly someone whom I would welcome to my world. Yet our anatomy is still incomplete. The sixth component, moral identity, is a bit different, however, from the five preceding components in three ways.

First, it has only recently been taken seriously by psychologists, despite earlier statements about its importance by Erikson (1968). Erikson has long argued that the core of the adolescent's central developmental task of self-discovery is development of an ideology; and this ideology is what holds the self (the "identity") together. Second, and related, this component may be the only one that does not develop until relatively late in the formative years; i.e., it may be dependent upon, as just noted, a developmental task of adolescence. It requires, at least for its fullest manifestation, powers of self-reflection which do not develop fully and are certainly not manifested fully until adolescence at the earliest. Third, most recent mentions of moral identity come from the literature described earlier on moral exemplars (Colby & Damon, 1992; Walker, et al., in press). That is, moral identity seems to be a salient characteristic of the most highly developed moral individuals. That is not to say it is not existent in others, merely that it seems necessary for moral excellence.

Colby and Damon (1992) suggest that the notion of moral identity they have observed in moral exemplars is a sense of self as centrally a moral being. Moral exemplars, when they think of themselves, think of a self that strives to be moral and for whom being moral is critical to their sense of self. For some people beauty, or intelligence, or athletic acumen, or wealth may be critical to the way they see themselves, but to the moral exemplar being good is at the core of their sense of self. Now we can see from the data collected by Walker et al. (in press) that this characteristic of the moral person is not salient to most people. Of the 568 nominated characteristics of moral exemplars, only 6 may be representative of the concept of moral identity (5 "self-reflective", 1 "self-righteous"), and these are only tangentially related at best.

We can clearly see that one's self-concept is related to one's behaviour; indeed Aronson (1969) based his model of cognitive dissonance on the assumption that we all try to be consistent with our self-concepts and that we generally have self-concepts that suggest we are "good" people. Nisan (1985) argues that we want to think of ourselves as good people and are more willing to forgive ourselves our transgressions if we have been particularly good lately. The more specific notion of moral identity has been elaborated by Blasi (1984) and Davidson and Youniss (1991) as an independent process of personal growth based on experiences of mutual egalitarian social coordinations. Blasi argues that

Morality is more a characteristic of the agent than of either action or thinking; the ultimate source of goodness lies in good will, and good will is at the core of what a person is. (p. 130)
What is most troubling for these authors is how to relate moral identity to moral reason, a problem that we cannot hope to resolve here, but a problem that once again underscores the complexity of the relations among the components of the moral person.

Meta-moral characteristics

In a sense we have already completed our anatomy of the moral person, for this seventh and last component is not really a moral aspect of the person. By meta-moral characteristics we refer to those attributes of the person that serve moral functioning but are not moral in themselves. We have already described some characteristics of the moral person that may indeed better fit this category. The notion of moral emotion presented by Samay (1986) may actually be an example of this, for the emotions are not in themselves moral, but rather are a form of impetus for moral functioning. And Aristotle's (1987) notion of practical reason may be construed as non-moral reason in the service of moral character. All of the discovered attributes of moral exemplars reported by Colby and Damon (1992) may also be candidates for this last component. In fact their identification of faith in moral exemplars is very close to the notion of spirituality in McGettrick's (in press) article in this series, an article with a very similar theme to the current argument.

It is very important here to differentiate between non-moral characteristics that one may choose to include in a program of moral education (e.g., social-conventional values about promptness or appropriate forms of address for adults in one's culture) and those meta-moral characteristics that are intrinsically non-moral but serve to aid in moral functioning. Rest (1985) presented a broad model of moral functioning that includes processes of perception, moral judgement, moral will, and moral behaviour, but which includes many processes that are not moral in nature but that clearly influence whether one's moral judgement will lead to right action. Etzioni, Berkowitz, and Wilcox (1994), for example, suggest that self-discipline and empathy are character traits that lie outside of the moral domain yet are critical to the flowering of moral personhood. One could be self-disciplined and empathic in the service of cold-heartedly "conning" an elderly person into giving one all of their hard-earned life savings by employing self-discipline to slowly develop a relation of trust and employing empathy to better understand one's victim in order to more effectively trick her. Clearly then, self-discipline and empathy are not intrinsically moral; however, it would be difficult to imagine an exemplary moral person without self-control and who has no ability to figure out other's private emotional states.

In a similar way, Grim, Kohlberg, and White (1968) demonstrated the rather complex role that attention plays in controlling cheating behaviour. Interestingly they called this attribute "moral will" when in actuality it was not moral at all. For intelligent but low moral reasoning individuals, attention actually aided in effective cheating whereas for high moral reasoning stage individuals it aided in resistance to cheating. Wynne and Ryan (1993) suggest that moral education is "such long-honoured educational concerns as teaching character, academics, and discipline" (p. v), a rather broad and atypical definition, but one that may at least suggest the inter-relatedness of moral and meta-moral characteristics.

It is clear that we can identify fundamentally moral components of the moral person but that we must not ignore those characteristics of the moral person that are not themselves moral in nature but are essential to effective moral functioning. It takes more than morality to be an effective moral agent.

The Education Of The Entire Moral Person

In actuality, the arguments offered thus far, i.e., that our language, theory, and psychology about moral development and moral education are too fractionated and inconsistent, are the least controversial aspects of this thesis. For we are more likely to garner agreement on the moral anatomy or the list of problems with the literature than we are to reach any consensus on how to educate the moral person. This is usually where the conflicts arise. Polls generally demonstrate that the vast majority of parents are strongly in favour of public moral education; however, when such programs are proposed in schools, parents and educators suddenly find themselves quite uncomfortable. Often this is due to their misunderstanding of what moral education is or can be, and often it is because of the diverse theoretical and ideological positions that are represented in the field of moral education. Hence, any attempt to offer a comprehensive, integrated model of moral education, as we shall now attempt to do, is likely to rankle many people in many different ways.

One example of this is the fear of moral relativism. Often critics of moral education argue that approaches they reject are flawed because they embody or instil a sense that there is no means of ultimately determining right or wrong, that it is ultimately up to each individual to decide for himself or herself. Now what is most interesting about this criticism is that it has been levelled by almost every faction against almost every other faction. Kohlberg (Kohlberg & Mayer, 1972) accused both character educators and romanticists of embodying moral relativism. Character educators (e.g., Bennett, 1991; Wynne & Ryan, 1993) accuse values educators and moral reasoning educators (such as Kohlberg) of propagating an ethic of moral relativism. Clearly ethical relativism is neither a desirable philosophical grounding for education nor a lesson that one would wish to impart to pupils. But it remains quite unclear who or what is promoting relativism in moral education. The central point again is that it is difficult to avoid controversy in moral education, and therefore we shall not try. Instead we will attempt to be integrative and let the chips fall where they may.

Sources of morality

In this paper, the chosen focus has been moral education; but moral education can be understood narrowly or broadly. In the narrowest sense it refers to the explicit curriculum of the formal school. In a somewhat broader sense, it includes the implicit curriculum of the school; what is often referred to as the school ethos (SOED, 1992), the moral atmosphere (Power et al., 1989a), or the hidden curriculum (Jackson, 1968). But in the broadest sense, one is remiss in not acknowledging that the school plays only a part in the moral education of our youth. Family is clearly the strongest and earliest influence, but peers, church, society, and the media also are strong influences.

In fact much of the rhetoric about school's role in moral education centres around whether they are in fact being asked to do someone else's "job". Schools blame parents for both the need for schools to do moral education and the fact that children are often so immoral that the schools cannot ignore moral education ("if they were doing their jobs right, we could spend our time doing what we are supposed to do... teaching academics"). Teachers also often lament the fact that they are fighting "an uphill battle" because of all the countervailing forces in society providing messages contradictory to the moral messages imparted in the schools. One of the most frequently asked questions in the series of seminars for Scottish head teachers organised by the Gordon Cook Foundation in 1995 is whether moral education is futile or unfair because the pupils must go out into immoral communities and eventually an immoral adult world. My response is that it is far better to present such students with a vision of a better world than simply to train them to fit into an immoral one. Furthermore, the more immoral their context, the more they need moral education. Those pupils who live in morally idyllic families and communities are likely in little need of intensive moral education programmes. Nevertheless, despite all of these caveats, we will now turn to an attempt to sketch out a programme of moral education that will address the complexity of the moral person. School-based sources of moral education

One way to divide up the school as a moral influence is by differentiating between the informal and formal curricula (Gatherer, 1995). The formal curriculum entails that which is taught in a way that is clearly intended to be a systematic educational activity. The informal curriculum entails the influences generated from the life of the school, such as rule-making, school norms, and modelled adult behaviour. Now some (e.g., McGettrick, in press), quite rightly differentiate between the hidden curriculum and the informal curriculum. We shall not do so here because it will be argued that a central aspect of any effective moral education institution is to make the hidden curriculum an explicit aspect of the informal curriculum; that is, effective moral education entails, in part, uncovering and reflecting on the hidden curriculum, so it no longer remains hidden.

The formal curriculum can support moral education in a variety of ways. First, it can systematically incorporate lessons about morality; e.g., through the historical study of notable moral events and decisions, through the literary study of moral heroes or villains or great moral dilemmas, through the theatrical depiction of great moral narratives or conflicts, etc. Second, lessons may be created which examine moral concepts and the classical thinking about them. Third, peer interaction around challenging moral issues can be woven into many classes. Fourth, courses specifically concerned with morality can be added to the curriculum. Fifth, service learning and other forms of institutionalised moral action can be made an explicit curricular requirement.

The informal curriculum is probably even more influential however. This is quite widely recognised as can be noted from the following pair of quotes, the first from James Coleman's preface to Wynne and Ryan's (1993) quite conservative character education statement and the second from Kohlberg's quite liberal moral education statement (Power et al., 1989a). The act of educating another is a moral act...Teaching is itself a part of the moral message transmitted by a teacher to a child... That moral message may be strong and clear if the teacher is dedicated to the task of teaching, or distorted if the teacher has largely given up on the task... But the message contained in the act of teaching is only part of the values teachers and schools transmit. Teachers and schools, like parents and families, cannot avoid teaching values... Yet this teaching often goes unnoticed, unintentional and unreflective. (p. iii) Even if the values of justice were discussed in classes, if the students perceive that getting along in school runs by a quite different set of norms, they will tend to perceive the latter as the real rules of the game and the former as nice talk one engages in with teachers. (p. 21). So how does one influence the informal curriculum? There are four inter-related aspects that must be considered: (1) the explicit school ideology or mission, (2) the behaviour of adults, (3) the governance structure, (4) peer norms. We shall treat each of these in turn. School ideology should be clearly developed, articulated, and disseminated. That is, the school should have a clear statement of its purpose and values; what it is trying to achieve, why those are its goals, and how it intends to go about reaching its stated goals. Such a statement should be developed by the entire school community and the broader community of which it is a part.

The second element, adult behaviour, is one of the most contentious yet most important. David Carr (1991) has argued that the modelled behaviour of adults in schools is the single most significant influence on the development of moral character. In the field of professional ethics education (e.g., medical ethics, legal ethics, etc.), faculty modelling is seen a critical to the effective education of ethical professionals (Pellegrino, 1989). Similarly, Nicgorski (1987) has emphasised the importance of university faculty as role models in the moral development of university students. As noted in the quote from Power et al. (1989a) above, if students perceive the real actions of the school to be different than the espoused actions, then they will readily discount the latter and heed the former. Even young children learn more from what you do than from what you say. It is thus critical that adults in schools practice what they preach.

School governance refers to the ways rules are made and enforced in the school. In traditional schools these processes tend to be quite authoritarian and hierarchical. Piaget (1965) noted long ago that such a relation to children impedes their moral growth and their growth as autonomous personalities. Rather, children should be welcomed as participants in the governance of the school or classroom, at levels appropriate to their developmental capacities. The serious work of deliberate moral education entails...opening up that process and dealing with the ways everyday rules of behavior are made and enforced. for it is those rules that define the moral atmosphere--the context for moral learning--in the school. (Power et al., 1989a, p. 23) The most comprehensive study of, and most elegant statement about, the governance structure of schools as a moral agent is offered by Power et al. (1989a) in their reporting of work in establishing three Just Community schools in the USA. They describe a process of democratising schools with an explicit emphasis on promoting justice and a sense of community. Such schools promote moral growth through participation, open discussion of moral issues, exposure to differing perspectives and ideas, and living in a moral community. As Lickona (1991a) has noted, one only promotes the development of respect in children by first treating them with respect. Empowerment is a form of respect.

Finally, peer norms are also at the heart of the informal curriculum. It has long been acknowledged that what peers value and hold to be right has a great impact on how they develop and influence each other. Kohlberg has explicitly focused on the norms of the students in his Just Community schools (Power et al., 1989a), norms such as caring and participation. Peer norms both affect the moral atmosphere of the school and are affected by it. It is clear from the research on Just Community schools that such a governance structure nurtures the growth of desirable peer norms. Even more significant, from a sociological perspective, the norms shift from being individually held norms to being collective norms shared by members of the school.

Having thus identified the major influences on moral development in schools, let us turn at last to how such influences may impact on the components of the moral person. Before we do so, however, let us reiterate our caution that the field of moral education is woefully deficient in valid outcomes research. We often do not know what works because either it hasn't been studied or the studies that exist are unscientific or biased. Educating for moral behaviour

There are two ways of considering this issue: direct and indirect influences. The former generally is construed as behaviour control, and focuses on discipline, reward and punishment, and authoritarian rule-setting (Wynne & Ryan, 1993). It can effectively produce greater adherence to the desired norms of behaviour. The second type of influence, informal influence, is derivative. It assumes that education that fosters the development of moral character, values, reason, etc., will consequently lead to more responsible behaviour.

Certainly some schools and classrooms may be in more need of direct behavioural influences, but all schools require indirect behavioural influences. Therefore, schools should model moral behaviour and teach about moral behaviour as well as generally promoting the full growth of the moral person. Because indirect behavioural influence is assumed to be derivative of the influences affecting other parts of the moral person, we shall shift our attention to those other components. Educating for moral values

Values are beliefs and can be affected both by introspection and by social influences. Introspection was largely the domain of the Values Clarification movement (Raths, Harmin, & Simon, 1966). It recommended classroom activities that lead to introspection of one's values. The problem with this approach, insofar as it is applied to specifically moral values is that is offers no guidance as to what ought to stand as an acceptable moral value. Remember from our discussion of domains of social knowledge above, that moral values are universal and unalterable whereas social-conventional and personal values are alterable. Pure introspection then is inappropriate for the development of moral values, but may be an effective technique when coupled with other techniques that provide a means of judging the adequacy of moral values.

One source of moral values education is the explicit advocacy of such values by the school and school authorities. Schools should stand for something and that something should be explicit. The difficulty here is in choosing what that something should be; i.e., the eternal question of "Whose values?". There are generally two accepted solutions to this problem. One is to look for community consensus; i.e., teach those values that the community agrees are important. The second is to identify an ethically grounded set of core values. Neither are perfect solutions, because neither has ever produced a non-controversial set of values; however, philosophically, local consensus is much more susceptible to moral error than is some abstract ethical scrutiny. Many communities throughout history and today endorse some fairly reprehensible values (e.g., racial supremacy, clitorectomy, slavery, fratricide, etc.).

A second, and decidedly more controversial, approach is to make valuing a part of the school's governance agenda. That is, to let the school's democratic process examine and generate the school's moral values. There is clearly a trade-off here, with an increase in the nurturance of moral character and moral reason and moral identity at the possible expense of a justifiable moral values agenda. That is to say, participating in such a process will powerfully influence these other parts of the moral person, but the values that are endorsed may need continual examination and development. Nevertheless, values that one participates in generating or endorsing are more likely to influence one's behaviour than values that are foisted upon you by an authority in a hierarchical power structure.

Regardless of how the moral values are generated, for them to become internalised, they must be made explicit, must be endorsed by the authorities (adults in a traditional school, entire community in a Just Community school), must be modelled by the adults in the school, and must be enforced. Both the formal and informal curricula are part of this process. Educating for moral character

The primary focus of moral character education is the development of moral habits that will hopefully become character traits or virtues. The school programs' emphases are on establishing early good habits in a controlled setting so that when their young students mature, those habits will be ingrained and generalize. (Benninga, 1991, p. 129) As already noted, Carr (1991) believes that such ingrained habits or virtues are predominantly influenced by teacher modelling. Such social learning processes are indeed very powerful. Furthermore, advocacy consistent with teacher modelling is very important; i.e., as Lickona (1991a) argues, teachers must not only practice what they preach, but must preach what they practice. In a school with an explicit open focus on the moral atmosphere, such as a Just Community school, there is a check on whether teachers are being sincere or hypocritical and whether the behavioural tendencies they espouse and/or manifest are indeed justifiable. In such schools, hypocrisy or injustice is quickly rooted out and discussed.

Clearly, the promotion of unreflective habits is more developmentally appropriate in primary schools whereas the active reflection on the moral validity of habits is more appropriate in the secondary school. Lickona (1991a), for example, describes an effective primary level classroom practice in which a Primary 7 teacher regularly schedules "appreciation time", a time for students to publicly acknowledge how others have helped them. Lickona suggests that this generates the habit of affirming others, which is assumed to potentially develop into the moral character trait of respect for others. It is worth noting that this activity is simultaneously promoting self-esteem in those students being appreciated. Furthermore, it is publicly advocating the value of helping and serves to reward helping behaviour. This, once again, underscores how the components of the moral person and the processes of moral education are complex and inter-related.

At the secondary level, on the other hand, greater emphasis should be placed on reflecting on the moral adequacy of patterns of behaviour, so that habits may evolve into ethically justifiable character traits. Such reflection may come from learning about behaviour and character through the formal curriculum, or through participation in evaluation of actual behaviours and of norms in the school itself. Student participation in school judiciaries and legislatures may serve this purpose as well. Educating for moral reason

This may be where we are on our strongest empirically-grounded footing. There is a vast educational and psychological literature examining how to promote moral reason (e.g., Berkowitz, 1985; Power, Higgins, & Kohlberg, 1989a; Rest, 1979). There are two central forms of educating for moral reasoning: peer moral dilemma discussion and the Just Community.

The more practical and therefore more widely implemented technique is peer moral dilemma discussion. Appropriate moral dilemmas are selected from anthologies, current events, history, literature, or the curriculum or are created by the teacher. Classes (or other student groups) are then guided through a critical but respectful discussion of the dilemma with the explicit aim of trying to figure out what the best moral solution is. As demonstrated by Berkowitz and Gibbs (1983), discussions in which peers focus analytically on each others' reasoning promote the most development. The role of the teacher is to facilitate an active consideration of the reasoning of one's co-discussants (Berkowitz, 1985).

The Just Community is a much more ambitious project but has been successfully carried out in a variety of contexts including schools (Power et al., 1989a), residential treatment centres (Blakeney & Blakeney, 1990), prisons (Hickey & Scharf, 1980), and workplaces (Higgins & Gordon, 1985). The core of the Just Community approach is democratic self-governance tied to promotion of justice and a sense of community. Decisions are made by a direct democracy and parliamentarian procedure, but with an explicit focus on maximising justice in decisions and attempting to create and affirm a sense of the communal identity of the group. Educating for moral affect.

In some senses, moral affect is the most difficult part of the moral anatomy to account for in education. There are several reasons for this. First and foremost, most psychological accounts of moral emotions suggest they develop prior to the entrance to formal schooling (e.g., Dunn, 1987; Emde, Johnson & Easterbrooks, 1987; Hoffman, 1987; Kagan, 1981). Second, there is little research on educating for moral emotion in the schools. Third, much of the literature on moral affect concerns remediation and the lack of a typical emotional life (e.g., Magid & McKelvey, 1987). Nevertheless, we can offer some guidelines concerning educating for moral emotions.

Let us begin with the pre-school experience. It is clear that emotions, including moral emotions, emerge in the first and second years of life (Emde et al., 1987). One of the most frequently cited affective components of moral development is empathy (Emde et al., 1987; Gibbs, 1991; Hoffman, 1987). Typically empathy is understood to be natural and to have a biological base as well as to be a source of moral reason and more mature moral affect. It is also assumed by many to have origins in the early attachment bonding of infant to caretaker (Kaye, 1982). Hence children should enter school with empathic capacities. Otherwise, we are in the realm of remediation, which is beyond the scope of this discussion.

Clearly, other moral emotions should also be present when the child enters schooling; emotions such as guilt, shame, hurt feelings, pride, etc. (Emde, et al., 1987; Gilligan, 1976). What then can the school provide beyond remediation? Perhaps this is best answered by turning to the notion of moral sensibility offered by Caputo (1986) but having its roots in classical Greek ethical philosophy as well as in other prominent ethical theories such as utilitarianism. The notion is that one must "love the good" as well as simply knowing it. Affect is infused in moral knowing. Hoffman (1991) argues that moral principles are "hot cognitions" because they are bound to moral affect. The question then becomes one of guiding the moral sensibilities through education. That is, the role of the school is to direct the child to care for the good and abhor the bad; e.g., empathise with victims and despise injustice.

Unfortunately, it is quite unclear how this is done. The literature on moral education pays little attention to this issue. Some argue (along the lines of John Stuart Mill, 1979) that humans have a natural tendency to desire the good; exposure is sufficient. Others argue that such moral "tastes" must be cultivated. For example, Bennett (1991) argues for promoting "moral literacy" in our children as a means of teaching them what the good is. From this perspective schools should teach children about moral virtues, about moral heroes, about great moral acts, and should expose children to great moral literature.

There is also a need to help children identify their moral emotions. Often they do not have verbal labels for what they feel or do not know which labels apply in a given instance. Talking about moral feelings and labelling them not only reduces a child's confusion about her feelings, but also can serve to legitimise them. The child who is torn between empathy for an outcast child and a fear of ostracising by the peer group that is engaging in the bullying may feel that his empathic feelings are signs of weakness or wrong in some other way. Discussing and labelling them may serve to reduce the child's doubts about the value of his own feelings.

There is also a need in schools to allow children the opportunity to express their feelings. Schools tend to strive to avoid affective displays by children. They are messy and disruptive. A more appropriate approach is to redirect emotions in more acceptable ways. Ginott (1976) often reminded us that we should express our emotions but be sure they are verbalised in appropriate ways. Teachers of pre-school children, for instance, often tell them to "use your words, not your hands" when angry or frustrated. Educating for moral identity

Similar to educating for moral emotion, when discussing educating for moral identity, we are again in a domain about which little is known. Erikson (1968) has suggested that adolescents will naturally attempt to form an identity and will do so most effectively if (1) their prior personality development has been healthy and (2) they are given the opportunity to experiment and explore options. Davidson and Youniss (1991) suggest that moral identity is born in egalitarian co-constructions typical of peer life. It thus appears that moral identity is likely to be nurtured by introspection and experimentation with roles, models, and values, as well as by healthy peer conflict resolution and interaction. Clearly as well, providing the options and models for adolescents to consider and evaluate will also serve their formation of a moral identity. Educating for meta-moral characteristics

Because, as noted above, there is little consensus on what these characteristics are, it is difficult to describe the educational processes that will influence their growth. Self-discipline has been highlighted by both Etzioni et al. (1994) and Wynne and Ryan (1993) and the latter authors suggest that school rules form the basis of self-discipline. Those characteristics that are character traits likely are nurtured in the same way as moral character traits. But, again, until we know which characteristics are necessary for moral functioning, we are unable to design a pedagogy to support their growth.

Conclusions

What we have attempted to do thus far is to identify some structural impediments to moral education and to address them by describing the outlines of the nature of the moral person and to sketch some means of more comprehensively educating the entire moral person. Surely this project is incomplete. More needs to be done in both fleshing out the moral anatomy and, even more glaringly, in developing the comprehensive approach to moral education including explicitly linking the different educational processes to the different components of the moral person.

In doing this exercise, I have tried to be as broad-ranging and eclectic as possible while still being a critical consumer of different approaches. I believe firmly that moral education must be grounded in a sound understanding of moral psychology (indeed, isn't that exactly how Aristotle began this project over 2000 years ago?) and my moral anatomy has been an attempt to begin such a project. I also believe that moral education must be grounded in ethical theory; i.e., we must critically examine the ethical assumptions underlying different approaches to moral education. I believe that we need to clean up our use of language, so that we can more effectively communicate both within and between "camps". Finally, I believe that we must be less parochial in our approach to both the theoretical models and techniques of moral education. Surely, I have managed to violate ethical or logical soundness by being eclectic and hopefully future work will correct those errors; and possibly I will alienate everyone by willingly endorsing perspectives from their theoretical opponents. But we need to break down much of the antagonism and close-mindedness that exists in the field. After all, our goal is not to win some theoretical war, but to contribute to the moral growth of our children and the betterment of our societies.

Aristotle (1987). The Nichomachean ethics (J.E. Weldon, Trans.). Albuquerque, NM: Americn Classical College Press.

Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 4, pp. 1-34). San Diego, CA: Academic Press.

Bantock, G.H. (1965). Education and values: Essays in the theory of education. London: Faber and Faber.

Barr, I.M. (1995). Values in education: The importance of the preposition. Paper presented at the Gordon Cook Foundation Conference on Values Education, Stirling, Scotland.

Bennett, W.J. (1991). Moral literacy and the formation of character. In J.S. Benninga (Ed.), Moral, character, and civic education in the elementary school (pp. 131-138). New York: Teachers College Press.

Benninga, J.S. (1991). Moral and character education in the elementary school: An introduction. In J.S. Benninga (Ed.), Moral, character, and civic education in the elementary school (pp. 3-20). New York: Teachers College Press.

Berkowitz, M.W. (1985). The role of discussion in moral education. In M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 197-218). Hillsdale, NJ: L. Erlbaum.

Berkowitz, M.W., & Gibbs, J.C. (1983). Measuring the developmental features of moral discussion. Merrill-Palmer Quarterly, 29, 399-410.

Berkowitz, M.W., Kahn, J.P., Mulry, G., Piette, J. (In press). Psychological and philosophical considerations of prudence and morality. In M. Killen & D. Hart (Eds.), Morality in everyday life: Developmental perspectives. New York: Cambridge University Press.

Blakeney, C., & Blakeney, R. (1990). Reforming moral misbehaviour. Journal of Moral Education, 19, 101-113.

Blasi, A. (1980). Bridging moral cognition and moral action: A critical review of the literature. Psychological Bulletin, 88, 1-45.

Blasi, A. (1984). Moral identity: Its role in moral functioning. In W. Kurtines & J. Gewirtz (Eds.), Morality, moral behavior, and moral development (pp. 128-139). New York: Wiley.

Caputo, J. (1986). A phenomenology of moral sensibility. In G.F. McLean, F.E. Ellrod, D.L. Schindler, & J.A. Mann (Eds.), Act and agent: Philosophical foundations for moral education and character development (pp. 199-222). New York: University Press of America.

Carr, D. (1986). Education, professionalism and theories of teaching. Journal of Philosophy of Education, 20.

Carr, D. (1991). Educating the virtues: An essay on the philosophical psychology of moral development and education. London: Routledge.

Carr, D. (1995). After Kohlberg: Some implications of an ethics of virtue for the theory of moral education and development. Unpublished manuscript. Edinburgh: Moray House Institute.

Colby, A., & Damon, W. (1992). Some do care: Contemporary lives of moral commitment. New York: Free Press.

Damon, W. (1977). The social world of the child. San Francisco: Jossey-Bass.

Damon, W. (1988). The moral child: Nurturing children's natural moral growth. New York: The Free Press.

Davidson, P., & Youniss, J. (1991). Which comes first, morality or identity? In W.M. Kurtines & J.L. Gewirtz (Eds.), Handbook of moral behavior and development. Volume 1: Theory (pp. 105-122). Hillsdale, NJ: Lawrence Erlbaum.

Dunn, J. (1987). The beginnings of moral understanding: Development in the second year. In J. Kagan & S. Lamb (Eds.), The emergence of morality in young children (pp. 91-112). Chicago: University of Chicago Press.

Eagly, A.H., & Chaiken, S. (1993). The psychology of attitudes. New York: Harcourt Brace Jovanovich.

Emde, R., Johnson, W.F., & Easterbrooks, M.A. (1987). The Do's and Don'ts of early moral development: Psychoanalytic tradition and current research. In J. Kagan & S. Lamb (Eds.), The emergence of morality in young children (pp. 245-276). Chicago: University of Chicago Press.

Erikson, E.H. (1968). Identity: Youth and crisis (2nd edition). New York: W.W. Norton.

Etzioni, A., Berkowitz, M.W., & Wilcox, W.B. (1994). Character building for a democratic, civil society. Position paper of the Communitarian Network, Washington, D.C.

Ewing, C.P. (1990). When children kill: The dynamics of juvenile homicide. Lexington, MA: Lexington Books.

Gatherer, W.A. (1995). Values and curriculum. Unpublished manuscript. Edinburgh, Scotland.

Gibbs, J.C. (1991). Toward an integration of Kohlberg's and Hoffman's theories of morality. In W.M. Kurtines & J.L. Gewirtz (Eds.), Handbook of moral behavior and development. Volume 1: Theory (pp. 183-222).

Gillligan, C. (1982). In a different voice: Psychological theory and women's development. Cambridge, MA: Harvard University Press.

Gilligan, J. (1976). Guilt and shame. In T. Lickona (Ed.), Moral development and behavior: Theory, research and social issues. New York: Holt Rinehart & Winston.

Ginott, H.G. (1976). Between teacher and child. New York: Avon Press.

Grim, P.F., Kohlberg, L., & White, S.H. (1968). Some relationships between conscience and attentional processes. Journal of Personality and Social Psychology, 8, 239-252.

Hickey, J. & Scharf, P. (1980). Toward a just correctional system. San Francisco: Jossey-Bass.

Higgins, A., & Gordon, F. (1985). Work climate and socio-moral development in two worker-owned companies. In M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 241-268). Hillsdale, NJ: L. Erlbaum and Associates.

Hoffman, M.L. (1987). The contribution of empathy to justice and moral judgment. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development. New York: Cambridge University Press.

Hoffman, M.L. (1991). Empathy, social cognition, and moral action. In W.M. Kurtines & J.L. Gewirtz (Eds.), Handbook of moral behavior and development. Volume 1: Theory (pp. 275-301). Hillsdale, NJ: L. Erlbaum.

Jackson, P.W. (1968). Life in classrooms. New York: Holt, Rinehart & Winston.

Kagan, J. (1981). The second year. Cambridge, MA: Harvard University Press.

Kaye, K. (1982). The mental and social life of babies: How parents create persons. Chicago: University of Chicago Press.

Kohlberg, L. (1967). Moral and religious education and the public schools: A developmental view. In T. Sizer (Ed.), Religion and public education. Boston: Houghton- Mifflin.

Kohlberg, L. (1976). Moral stages and moralization: The cognitive-developmental approach. In T. Lickona (Ed.), Moral development and behavior: Theory, research and social issues (pp. 31-53). New York: Holt Rinehart and Winston.

Kohlberg, L., & Candee, D. (1984). The relation of moral judgment to moral action. In W.M. Kurtines & J.L Gewirtz (Eds.), Morality, moral behavior, and moral development. New York: Wiley.

Kohlberg, L., & Mayer, R. (1972). Development as the aim of education. Harvard Educational Review, 42, 449-496.

Laud, L., & Berkowitz, M.W. (1995). Evaluation: A critical emerging issue in character education. Unpublished manuscript. New York: Columbia University Teachers College.

Leming, J.S. (1993). In search of effective character education. Educational Leadership, 51, 63-71.

Lickona, T. (1983). Raising good children. New York: Bantam Books.

Lickona, T. (1991a). Educating for character: How our schools can teach respect and responsibility. New York: Bantam.

Lickona, T. (1991b). An integrated approach to character development. In J.S. Benninga (Ed.), Moral, character, and civic education in the elementary school (pp. 67-83). New York: Teachers College Press.

Magid, K., & McKelvey, C.A. (1987). High risk: Children without conscience. New York: Bantam Books.

McGettrick, B.J. (In press). Educating the entire moral person. Perspectives in values education.

Mill, J.S. (1979). Utilitarianism (G.Sher, Ed.). Indianapolis, IN: Hackett Publishers.

Nicgorski, W. (1987). The college experience and character. In K.Ryan & G.F. McClean (Eds.), Character development in schools and beyond (pp. 328-357). New York: Praeger.

Nisan, M. (1985). Limited morality: A concept and its educational implications. In M.W. Berkowitz & F. Oser (Eds.), Moral education: Theory and application (pp. 403-420). Hillsdale, NJ: Lawrence Erlbaum.

Pellegrino, E.D. (1989). Teaching medical ethics: Some persistent questions and some responses. Academic Medicine, 64, 701-703.

Perry, W. (1970). Forms of intellectual and ethical development in the college years: A scheme. New York: Holt, Rinehart & Winston.

Peters, R.S. (1981). Moral development and moral education. London: George Allen & Unwin.

Piaget, J. (1965). The moral judgment of the child. (C.M. Gabain, Trans.), New York: Free Press. (First published in 1932).

Piaget, J. (1981). Intelligence and affectivity: Their relationship during child development (T.A. Brown & C.E.Kaeg, Trans.). Palo Alto, CA: Annual Reviews.

Power, F.C., Higgins, A., & Kohlberg, L. (1989a). Lawrence Kohlberg's approach to moral education. New York: Columbia University Press.

Power, F.C., Higgins, A., & Kohlberg, L. (1989b). The habit of the common life: Building character through democratic community schools. In L. Nucci (Ed.), Moral development and character education: A dialogue (pp. 125-143). Berkeley, CA: McCutchan.

Raths, L.E., Harmin, M., & Simon, S.B. (1966). Values and teaching. Columbus, OH: Charles E. Merrill.

Rest, J.R. (1979). Development in judging moral issues. Minneapolis, MN: University of Minnesota Press.

Rest, J.R. (1985). An interdisciplinary approach to moral education. In M.W. Berkowitz & F. Oser (Eds.), Moral education: theory and application (pp. 9-26). Hillsdale, NJ: L. Erlbaum.

Rokeach, M. (1968). Beliefs, attitudes, and values. San Francisco: Jossey-Bass.

Rokeach, M. (1973). The nature of human values. New York: Free Press.

Rosenblatt, R. (1995). Teaching Johnny to be good. The New York Times Magazine, April 30, 36-74.

Ryan, K., & Lickona, T. (1987). Character development: The challenge and the model. In Ryan, K., & McLean, G.F. (Eds.), Character development in schools and beyond (pp. 3- 35). New York: Praeger.

Ryan, K., & McLean, G.F. (Eds.) (1987). Character development in schools and beyond. New York: Praeger.

Samay, S.A. (1986). Affectivity: The power base of moral behavior. In G.F. McLean, F.E. Ellrod, D.L. Schindler, & J.A. Mann (Eds.), Act and agent: Philosophical foundations for moral education and character development (pp. 71-114). New York: University Press of America.

Schulman, M. & Mekler, E. (1985). Bringing up a moral child: A new approach for teaching your child to be kind, just, and responsible. Reading, MA: Addison- Wesley.

Schweder, R.A., Turiel, E., & Much, N. (1981). The moral intuitions of the child. In J.H. Flavell & L. Ross (Eds.), Social cognitive development: Frontiers and possible futures (pp. 288-305). New York: Cambridge University Press.

Scottish Consultative Council on the Curriculum (1991). Values in education. Dundee, Scotland: SCCC Information and Marketing Services.

Scottish Office Education Department (1992). Using ethos indicators in secondary school self-evaluation: Taking account of the views of pupils, parents, and teachers. Edinburgh: GNP Booth.

Turiel, E. (1983). The development of social knowledge: Morality and convention. New York: Cambridge University Press.

Walker, L.J., Pitts, R.C., Hennig, K.H., & Matsuba, M.K. (In press). Reasoning about morality and real-life moral problems. In M. Killen & D. Hart (Eds.), Morality in everyday life: Developmental perspectives. New York: Cambridge University Press.

Watson, M., Solomon, D., Battistich, V., Schaps, E., & Solomon, J. (1989). The Child Development Project: Combining traditional and developmental approaches to values education. In L.Nucci (Ed.), Moral development and character education: A dialogue (pp. 51-91). Berkeley, CA: McCutchan.

Wynne, E.A. (1991). Character and academics in the elementary school. In J.S. Benninga (Ed.), Moral, character, and civic education in the elementary school (pp. 139-155). New York: Teachers College Press.

Wynne, E.A., & Ryan, K. (1993). Reclaiming our schools: A handbook on teaching character, academics, and discipline. New York: Merrill.

back to Featured Articles