Home

VIRTUE ETHICS AND EDUCATION

Dr. Frank M. Flanagan Education Department Mary Immaculate College University of Limerick Republic of Ireland.
Tel: +353 61 204354 Fax: +353 61 313632 frank.flanagan@mic.ul.ie

Introduction

John Dewey’s description of the relationship between education and philosophy provides a good starting point for any discussion on the moral aims of education.

Education is the laboratory in which philosophical distinctions become concrete and are tested. ... If we are willing to conceive of education as the process of forming fundamental dispositions, intellectual and emotional, toward nature and fellow men, philosophy may even be defined as the general theory of education.[1]

This is more true of ethics than of any other branch of philosophy. If our moral theories cannot help people to become virtuous then they are of little value. For the purpose of the study of morality in ethics is ultimately to learn how to live well. And in relation to the purpose of education it can be asserted that the fundamental question of educational philosophy is “Can we teach our children to be good?”

In what follows I will attempt to identify some of the inadequacies of an excessively cognitivist approach to moral formation and to suggest that a return to an Aristotelian concern with formation through habituation might be a more reliable and productive way to conduct the moral education of our children.

Universalism

Socrates held that in relation to morality or moral education we could not judge a particular case until we had clarified the universal criteria that determine morality. In response to the question “Can virtue be taught?” in Meno he responds by challenging Meno to define “virtue” so that they can then begin the process of examining its teachability.

And so of the virtues, however many and different they may be, they have all a common nature which makes them virtues; and on this he who would answer the question, "What is virtue?" would do well to have his eye fixed.

However the initial attempts are discouraging

Ever and anon we are landed in particulars, but this is not what I want; ... Had I the command of you as well as of myself, Meno, I would not have enquired whether virtue is given by instruction or not, until we had first ascertained "what it is."

And the conclusion is disappointing for two reasons: the definition of virtue which Socrates sought, of trans-personal, trans-cultural, transcendant validity, has not been found and, in any case according to Socrates virtue is a gratuitous dispensation (it cannot apparently be taught):

Then, Meno, the conclusion is that virtue comes to the virtuous by the gift of God. But we shall never know the certain truth until, before asking how virtue is given, we enquire into the actual nature of virtue.

Socrates identified virtuous conduct with knowledge: it was sufficient to know the good to want to do it. This view was exacerbated by later theories which separated mind and body, cognition and sense experience, reason  and experience. For as the link between thought and behaviour became more tenuous it became increasingly difficult to explain moral motivation. Even when we know the good, why should we do it? Eventually Kant attempted a purely rational grounding for moral thought which marginalised natural inclination entirely. Duty, not inclination, became the primary feature of moral life. Kant’s Categorical Imperative was an attempt to meet the kinds of universalist criteria which Socrates required. Kant defined a moral act as that which it would be right for anyone, anywhere, at any time, to do.

The Deontological approach

Kantian, or deontological ethics, presents duty as the most fundamental moral motivation. A moral action is good to the extent that it involves a person's doing, or trying to do, his or her duty. On this account a virtuous character is someone who has a disposition to do his or her duty.

We ought always act for the sake of duty, that is, because it is the right thing to do. How do we know our duty? Our duty is to act in such a way that the maxim or principle behind our act can be willed as a universal law, it must be universalizable; this is the Categorical Imperative (categorical because unconditional, an imperative because obligatory).

In Kant’s moral philosophy duty is always supposed to prevail over inclination: what one ought to do is superior to what one wants to do. This priority dichotomises duty and inclination, reason and emotion. Kant urges us to follow the path of duty and reason even where this is contrary to our inclinations (especially where this is contrary to our inclinations).

In the Kantian scheme the basis for evaluative judgements of the moral worth of actions is provided by the formal character of the moral judgement, the Categorical Imperative. As a result the moral agent is rather abstract and psychologically insipid: an impersonal reason rather than a flesh and blood human sets the moral agenda.  The net result is that

The emphasis is not on being a 'good person' but rather a just and fair administrator (being a good person is presumably the same). The model, thinly disguised by the evasive logic of 'universalizability' is the bureaucrat, who treats everyone the same and has no relevant personality of his or her own.[2]

Lawrence Kohlberg brought this Kantian, rationalist perspective into the mainstream of education. His is perhaps the most influential theory of moral development and moral education of the last 40 years.

Kohlberg proposed six stages in the development of moral thinking at three different levels. The first and second stages comprise what he calls the  pre-moral level. At this level rules are obeyed to avoid punishment (stage I) or in order to obtain rewards (stage II).

Level 2 involves the morality of conventional role-conformity. At stage III, the child is concerned with maintaining good relations: she conforms to avoid disapproval. This conformity develops at stage IV into an authority maintaining morality: the child conforms to avoid censure by authorities and to avoid the resulting sense of guilt.

At level 3, the level of autonomy, the prevailing morality becomes a morality of self-accepted principles. Stage V is characterised by the morality of contract; duty is defined in terms of contract with others which entails a corresponding general avoidance of violation of the rights of others.

Stage VI is the highest stage in Kohlberg’s anatomy of the development of moral thinking. At this stage the individual thinks in terms of the morality of individual principles of conscience. Conformity is no longer due to external factors like punishment, reward, authority, law or contract. Now the individual conforms to avoid self condemnation. This is the equivalent of Kantian moral reasoning, self-defined moral principles which are held to be universally binding on all.

Kant's purpose was to formulate a moral theory; Kohlberg's was to formulate a test of moral reasoning. However adequate in its own terms neither, on its own, is of much practical assistance in dealing with the moral complexities of life or in providing for an adequate and practicable moral education.

Ethics is an intensely practical discipline: morality is, primarily, about action not theory. “If we were to ask of a person,” Hare declares, “ ‘What are his moral principles?’, the way in which we could be most sure of a true answer would be by studying what he did”.[3]

According to Kohlberg one discovers another’s principles not by studying his actual behaviour but by “analyzing and interpreting his verbal responses to a hypothetical dilemma”.[4] The hypothetical dilemmas which Kohlberg provides for the purpose of exercising moral thought lack the urgency of actual experience (or even the richness of serious literary invention). Moral conflict is construed exclusively in terms of conflict between competing moral principles. (Whether to do X or Y in situation Z). Kohlberg’s approach cannot answer questions about the origin or validity of the principles invoked. Nor does the Kohlbergian scheme explain how we develop the capacity to perceive morally significant situations, that is, how we learn to identify certain situations as being morally significant, how we select the appropriate moral principle(s) which should be applied, or how we acquire competence in the application of these principles.

Part of the problem is that Kohlberg’s account of moral development is based on experimental situations. The reality of the ‘morality’ of such situations is questionable and there is no clear distinction between verbal justifications of moral judgement and motives for moral action. As Straughan says: “We do not always want to do what we believe we ought to do. Reasons for action of a justifying kind, therefore, do not always provide us with reasons for action of a motivating kind.”[5] The Kantian paradigm which undergirds the highest level of the Kohlbergian scheme 'essentially renders the self a disengaged observer of moral actions'.[6]

In short, the Kantian approach fails to explain why moral agents who know exactly what they ought to do morally, fail to do it. (Straughan “And remain a bastard!”). It also fails to explain how someone who would otherwise have given in to weakness holds out under the influence of training. For if such a person acts from habit rather than from reason she is not, in the Kantian view, acting as a genuine moral agent at all.

Aristotle: virtue ethics, character

There is a danger that, under the influences of the cognitivism of Kant and the developmental theory of Kohlberg, moral education will be considered a technical matter based upon the findings of experimental psychology.[7] Such a technicized view of moral reason divorced from the complexities and contingencies of ordinary life situations has been rejected by a return to consideration of Aristotelian-inspired virtue (or virtues) as central to moral development. Increasingly it is being asserted that pre-theoretical human moral experience is crucial to our understanding not only of what it is right to do, but also to our understanding of how we come to know what it is right to do and how we learn to do it.

The Kantian approach to the issues of determining the right or wrong course of action and the explanations people might give for their morally relevant judgements and decisions[8]  is silent on the ways in which people are motivated to do what is right and to refrain from doing wrong. There appears to be a presumption that people will do what is right once they have a rational grasp of what is the right or wrong thing to do, thus invoking a Socratic identity between knowledge and virtue. Virtue ethics, on the other hand, sees moral behaviour as springing from an individual's character and ‘not just from the formal determination that an act is right or wrong'.[9]

In Aristotle’s ethics the central question is "What kind of person should I be?" The focus is on character rather than on the nature of actions, on virtues and vices, rather than on rules and calculations. Aristotle’s virtues are character traits which promote eudaimonia, human flourishing, his vices are weaknesses of character which impede human flourishing.

Aristotle, contrary to Plato and many later, post-Cartesian philosophers, proposes a non-dualist philosophical anthropology: man is not made up of two separable parts. For Aristotle the human experience is holistic: the cognitive capacities function within a wider context of human experience which cannot be described or explained independently of a complex of social, practical, historical, cultural, and affective features. Aristotle describes morality as a matter of deliberation within frameworks of value which are received rather than constructed by the individual moral agent: we don’t make up our maxims, we inherit them. (How we inherit them is a matter of ‘reconstruction’ rather than ab initio invention.)

The distinctive mode of moral thinking, according to Aristotle, is practical intelligence (phrónêsis)[10]. Phrónêsis (as the guide of human action) is the capacity to comprehend the necessary interplay between the true nature of the individual and the welfare of the sustaining community. In order to act rightly, i.e. morally, individual desires must be properly co-ordinated to correct goals or ends. Phrónêsis excludes a detached perspective with regard to moral values. Moral life is not invented, it is already a going concern into which the individual is thrust as a member of a moral community.

Aristotle's view of the nature of the state,[11] as the supreme form of human association, involves three fundamental claims. The first is that human beings have natural ends, the realization of which constitutes an important part of the human good life. The second stresses the collaborative nature of human experience: human beings cannot fully realise their ends without living in a polis, that is, in association with others. The third claim is that the polis develops naturally and exists for the sake of the human good life, the two are inseparable. So, in the Aristotelian account, the natural ends of human life are achievable only in association with others in the sustaining context of a community.[12] The state exists by nature and man is by nature a political animal. 'Anyone who by his nature and not simply by ill-luck has no state is either too bad or too good, either subhuman or superhuman ...'[13] Furthermore, the state has a natural priority: it is both natural and prior to the individual.[14] This does not mean that somehow the state pre-dates the individual historically (which would be absurd) but that we are born into, and die out of, a community. The community transcends the lives of its individual members.

Reason aids the assimilation of the prevailing values. We do not create values as an ab initio procedure. The prevailing values themselves provide the context and the limit of the moral experience and the concomitant responsibility. In Sandel’s formulation

Open-ended though it be, the story of my life is always embedded in the story of those communities from which I derive my identity - whether family or city, people or nation, party or cause. On the communitarian view, these stories make a moral difference, not only a psychological one. They situate us in the world, and give our lives their moral particularity.[15]

Aristotle’s approach to moral reflection and action was much less formally stringent than Socrates’ requirement of a universal definition. In the first place we cannot expect certainty or a universally compelling definition of virtue or right action. In the Nicomachean Ethics he tells us that

We must be content, then, in speaking of such subjects and with such premises to indicate the truth roughly and in outline, and in speaking about things which are only for the most part true and with premises of the same kind to reach conclusions that are no better. In the same spirit, therefore, should each type of statement be received; for it is the mark of an educated man to look for precision in each class of things just so far as the nature of the subject admits; it is evidently equally foolish to accept probable reasoning from a mathematician and to demand from a rhetorician scientific proofs.[16]

In contrast to Socrates Aristotle advocates an understanding of moral life which is irremediably rooted in particularities; his concern is how we should live. Morality is, in the first instance, what people do, not the rules, principles, or procedures they follow.

Moral action is rooted in character. Character education has been devalued by association with an older, heroic tradition; it has also been contaminated by cold showers, brutality, inhumanity in the treatment of children, an ideal of ‘manly’ behaviour or a robust Christianity, which frequently meant that one should not complain about abuse, bullying, neglect. Too often, if not invariably, such ‘character formation’ led to a coarsening rather than to a refinement of moral discrimination.

Character is formed throughout a lifetime. It begins as habituation. We learn to do good by doing good under the guidance of an adult who is well versed in the culture of our moral community. By degrees we learn to act well on our own initiative and gradually we learn why certain classes of actions are good.[17]

In virtue ethics it is virtues, rather than rules, which constitute the primary area of interest. The real moral world is not an ideal arena where moral dilemmas can be worked out rationally. Injustice, misfortune, moral luck, mutual misunderstanding abound. Rationality alone cannot describe, much less explain, significant features of the moral landscape, trust or betrayal, pain, love, or loss. These are not things that we can experience or learn purely on the basis of reading about them (although reading about them may improve our understanding) but only by living a life in which they are experienced as morally significant features of human experience.[18] Kantian ethics obscures the wholeness of the individual human agent rooted in a particular community. The virtue based approach restores the social character of the moral agent to a central position.

Moral perception is a crucial feature of the virtuous agent. The social context in which people engage with real moral dilemmas is complex, and the ethically salient features are not always easily discerned. In order to identify the morally relevant features of a situation the moral agent must have an appropriate and adequate moral vision. Possession of principles is not sufficient; the immediate relevance and specific application of principles always depends on the discretion of the agent. Moral principles themselves do not contain the prescription for their own application, they require a learned 'tacit knowledge'.[19] The moral agent must become 'enmeshed in the particulars of the situation rather than attempt to apply a principled solution of universal effectiveness'.[20] The general moral question to be addressed is not about the procedure by which the moral decision will be (or was) arrived at in the first instance. The substantive moral question relates to the acuity of the moral vision: was the moral situation perceived correctly in the first place?

Virtue ethics also acknowledges emotion as an integral component of the agent's moral vision and response. The affective response of the moral agent is neither an irrelevant side effect or an avoidable contaminant of the agent's moral deliberations. Emotions serve a dual role: they serve as guides in the practical reasoning process (does it feel right) and are themselves 'a mark of virtue' (the moral agent feels good about the right things). Frequently it is our initial feelings which prompt moral action rather than deliberative application of abstract principles to situations. This is not to say that emotions on their own are infallibly reliable as moral pointers: the practical reasoning process comprises both cognitive and affective factors.

Selfhood and morality are interwoven constructs. A practical morality must be rooted in some form of personal identity; personal identity requires a moral core. Virtue ethics places morality at the core of selfhood, what Dewey calls 'the essential unity of the self and its acts ... The unity of self and action underlies all judgement that is distinctively moral in character'.[21] This is the sense in which moral actions are self-revealing: they demonstrate the type of person the agent has become, the kind of person who acts in this manner, given these circumstances. But in terms of individual development there is also a progressive move towards the moral in the self-definitions of individuals. It is the moral acts themselves which give us our character as moral individuals: our acts make us, we are our acts. Moral decision making does not just reflect the character of the agent it also constitutes this character.[22]

One's moral identity must be understood as a cumulative process over time. It is formed by way of a long series of "present moment" actions. Moral actions determine the kind of person the agent is in the process of becoming. One does not just perform a compassionate action but in the process one becomes (more) compassionate. Therefore, from a virtue ethics perspective, moral life is a process of self-making, of becoming. This perspective follows MacIntyre in seeing the virtuous self as 'an unfolding narrative of moral continuity. ... Narratives are constructed (or, better yet, lived) which establish a thread of continuity between one's past actions or beliefs and the future self one is intent on becoming'.[23]

As Aristotle made clear the virtuous self is also a relational, interdependent self: it requires a community of other individuals which provides the context within which moral lives are played out. The 'cognitive, affective and behavioural qualities which are constitutive of the virtuous self cannot be formed and maintained in isolation'.[24] Others continually clarify and expand the agent's moral vision.

How do we acquire virtues?

Aristotle dismissed the Socratic notion that knowing the good invariably leads to doing it. He recognised that we often fail to behave well even when we are reasonably certain of the right thing to do. As Roger Straughan writes: “In making a real life moral decision, my motives, feelings, wants and emotions may run counter to my hypothetical reasoning and judgements, which will often need to be modified if I actually find myself in such a situation”.[25] Goodness in ordinary life is not the subject of a Philosophy examination.

According to the Aristotelian view dispositions to act in morally appropriate ways are acquired by habituation. Neither demonstrative knowledge of the sort employed in science nor aesthetic judgement of the sort applied in the arts are relevant to morality. But the Aristotelian explanation has the appearance of circularity: to have virtues one must be virtuous, but to be virtuous, one must have virtues.

How does it work? A young child chooses to do what seems to be pleasant or to her advantage. If she is being reared well parents and teachers will ensure that what she perceives as pleasant or advantageous will be acts in accordance with virtue. They will provide extrinsic incentives for her to engage in such acts even when they are contrary to her inclination. Over time the acts involved will become more attractive because of the reward and approbation she receives. (‘Good girl!’) Simultaneously, she will be helped (through adult example, exemplary stories, exhortation, etc.) to realise that there is more to such actions than the extrinsic rewards which have been linked to them. They mean more, not alone in themselves, but to the community into which the child is growing: becoming a full member of that community entails acknowledging the value of such actions, doing them, and becoming the kind of person who does them because of their intrinsic worth.

Parents and teachers will try to ensure that what the child sees as pleasant and useful will be acts which accord with virtue (as defined by the sustaining community). When the child is required to act against her initial desires, parents and teachers will provide incentives other than the intrinsic value of the acts themselves. As she becomes habituated to such acts they will appear less painful/objectionable. Soon she will be guided to the realisation (including perception and feeling) that the relevant actions have an intrinsic value which is quite distinct from the extrinsic reward that has been arbitrarily attached to them.

In summary, then we acquire these specific moral dispositions because

1. They are valued in our society;

2. Their practice is, as a consequence, encouraged, praised and celebrated;

3. They are embodied in the stories we tell our children as characteristics of admirable role models;

4. In some cases they are embodied in law so that, if we cannot make their practice mandatory we can at least punish failure to implement them (the educative value of law and punishment);

5. We learn them as we learn language - as part of the natural order of things; only later can we (in terms of capacity rather than permission) question them.

Objections

It can be objected that there is a serious conflict here between the way in which virtue is acquired and exercised, and claims to moral freedom. As outlined the acquisition of virtue itself appears to constitute a limitation of one's freedom because it requires a directive upbringing. The Aristotelian habituation takes place without the fully free and rational consent of the individual concerned, and it imposes a particular conception of the good life. So the acquisition and present possession of virtue was and continues to be an involuntary act. For if the virtues as settled dispositions result from one’s upbringing then one cannot claim responsibility for one’s virtue since neither its acquisition nor its exercise is fully rational nor free.

In addition acts which result from a settled virtuous disposition are not necessarily expressions of rationality and freedom, since they may be performed simply as matters of habit. It would seem then that the exercise of rationality and freedom is not necessary for possessing virtues nor acting virtuously. Then it appears to follow that the activity of virtue does not fully exploit our definitively human capacities.[26]

Such objections are based on a rather extreme conception of freedom: that it is an either/or, all or nothing affair. But surely there are degrees of freedom between total lack of constraint and no freedom at all. There is a presumption that it is possible to constrain human freedom totally (without destroying the humanity of the individual).[27] We would not want to say, for example, that the acquisition of language constrains all that an individual can communicate. Or that we have no responsibility for our utterances because we acquired the language naturally. In fact submitting to the rules of language, the constraints of vocabulary, grammar and syntax is a necessary pre-requisite to the possibility of effective communication: it is not a loss of freedom (pre-linguistic babbling is not language nor does it constitute communication) but a necessary structure and constraint (a discipline). So we should see the acquisition of dispositions to virtue as an empowering discipline rather than as a radical constraint on moral freedom.

In any case human beings are not just rational and autonomous beings. There are human capacities, other than rationality and choice, which are equally valuable in themselves. Privileging rationality and choice appears to cast other human capacities (physical, emotional, psychological, and mental) in a secondary and subservient role.[28] Failure to develop and nurture all of these capacities leads to serious harm. (Who will say, for example, that the capacity to love is inferior, or subordinate to, autonomy? Or that if the capacity for self expression is unrealised the individual is not seriously harmed?) The good of the individual is promoted by 'promoting all aspects of his or her well-being, subject only to a reasoned recognition of what that well-being currently constitutes'.[29]

Clearly, one's upbringing strongly influences the dispositions one acquires. There are two ways of considering this: on the one hand it seems to limit one’s freedom, rationality, and consequent responsibility. On the other hand moral habituation can be seen to make rationality and freedom more effective. The conflict here illustrates the difference between the liberal view that the individual’s freedom and rationality are necessary for the ‘self-making’ which is the focus of the liberal agenda and the communitarian view that our capacity to become more fully who we can become is both limited and strengthened as a consequence of our cultural and historic existential location.

But even if we acknowledge the centrality of rationality and choice the very basis of human agency is socially constructed.[30] Without someone’s physical care and an appropriate socialization children could never become competent agents. Crucially we recognise children's developmental needs for care, and education of a certain standard, and have established legally supported social practices which recognise the rights of children to have those needs met.[31] Without such consistent, systematic, and socially sustained intervention children would have no hope whatever of becoming the kinds of purposive, autonomous agents envisaged in any moral universe.

Early habituation to virtue does not preclude the exercise of freedom or rationality. The acquisition of reason is gradualist and  developmental. As Locke pointed out: 'age and reason as they grow up loosen (the bonds of the parental subjection) till at length they drop quite off, and leave a man at his own free disposal'.[32] This would lead us to expect that as reason advances the need for habituation diminishes. But, no more than freedom, reason is not an all or nothing affair. Locke advocates reasoning with children, for instance, in the following terms[33]

(W)hen I talk of reasoning, I do not intend any other but such as is suited to the child's capacity and apprehension. No body can think a boy of three or seven years old should be argued with as a grown man. Long discourses and philosophical reasonings, at best, amaze and confound, but do not instruct children. When I say, therefore, that they must be treated as rational creatures, I mean, that you should make them sensible, by the mildness of your carriage, and the composure even in your correction of them, that what you do is reasonable in you, and useful and necessary for them; and that it is not out of caprichio, passion or fancy, that you command or forbid them anything.[34]

Let me return to the analogy of language acquisition. Cognition and language are closely linked. Children are born with a capacity for language and for non-verbal thought. This does not mean that they are born with a language but that, in the normal course of events, they are sufficiently equipped, physiologically and cognitively, to acquire a language. From their earliest age infants manifest this capacity in their vocal behaviour. Left to their own devices, however, children would not progress to the mastery of speech in any language, for human language is a social phenomenon and can be mastered only in a linguistic environment. In an analogous way children are born with the capacity to reason. It is only through the social interaction of thought and language, and the internalisation (or appropriation) of public, interpersonal communication that the child becomes capable of structured thinking.[35]

So even the small child can reason (even if the reasoning is faulty) in the same way as a young child can speak (even if the utterance is grammatically flawed, syntactically incorrect, or factually untrue). Human development involves learning the appropriate application of the power of reason as it involves learning the appropriate application of the power of speech. Habituation to virtues increases the possibility of eventual freedom by empowering the individual to acquire and deliberately sustain a specific character. This character has been described as “the true fulfilment of her potential as a rational being, with the firm and stable disposition of directing herself in accordance with her accurate perception of reality”.[36]

Can virtue be taught?

Philosophers in the rationalist tradition often present an inadequate characterisation of childhood. Children are seen as simply lacking 'the capacity to act rationally in pursuit of their own self-chosen goals'.[37] To say that children are immature is a tautology and it is indisputable that they require greater protection than most adults. However, the portrayal of children as totally lacking in reason until a particular age 'leaves no room in our thinking about children for things like thinking and learning, or development in understanding and character and all those other concepts which refer to processes and not to states'.[38] The philosophers cited by Hughes present 'a pretty unconvincing picture' of children and childhood. Hughes is particularly concerned with the way in which philosophers concentrate on rationality. She could also have criticised them for presenting a static picture of children, when in fact the experience of childhood is dynamic and developmental.

Any adequate understanding of our ethical systems and moral life must begin with children: any worthwhile moral theory must take account of the ubiquity of children in human life (at least outside of Academia) and of the reality that all of our moral experience began in individually experienced childhoods which formed our moral perception, our moral judgement, and our motivating commitment to the values we internalised from our sustaining community - family, education, society. For morality requires more than a knowledge of the good: although undoubtedly one must have knowledge of what is moral. One must also have a formed disposition to will the moral action; whatever else it is, morality is not accidental, haphazard, occasional, or arbitrary: one must act out of a consistent character. (This is what gives sense to a concept like ‘moral outrage’. Certain observed behaviour is so contrary to one’s settled dispositions that it elicits a strong emotive response. There is no room in Kant’s ethics for such feelings.)

Moral education via specific "character development" programmes is not a solution. The thought that such training is necessary is similar to the thought that children normally have to be taught their first language. Although children do not have to be taught their first language in the sense that they require explicit instruction, it is clear that they will not learn their first, or any, language if they are not members of the language community in which this language is the norm of communication. Analogously, children may not need to be explicitly taught the principles of moral behaviour (or the right thing to do) but they do need the example, response and correction which is a necessary condition to their learning.

Should schools in a democracy promote moral education? Virtually every public institution in any society (schools, law courts, legislature, media, medical and financial establishments, government agencies) actively promotes moral education, a vision of the good life, in various ways. The question for formal public education is whether the state - however defined - has a right, or a duty, to control the mooral education of its citizens. The tradition of educational discourse is emphatically political: it is a prolonged discussion as to how the future society shall be, and it tends to reinforce the belief that freedom cannot be sustained outside of a secure political context. Individual freedom must be restrained so that individual freedom can be exercised. In the democratic state there is a perennial tension between the freedom of the individual and the constraint of the law: ungovernable individuals are restrained by the law, unjust laws are challenged by belief in individual freedom.

Knowledge of the (moral) law is insufficient, nor is it sufficient to know how to apply the rules correctly. What is important is that the knowledge is consistently (and appropriately) implemented in action. The task of teaching children to be good is that of inculcating an intelligent (and reflective) understanding of the good in such a way that it is translated into habits of action.

How much can the schools do?

Virtue theory places special emphasis on moral education since virtuous character traits are inculcated in the young. But there is a very definite constraint on the capacity of schools to deliver meaningful character training.

Following Gilbert Ryle[39], we should understand that "moral education" has both task and achievement senses. Even if there is factual agreement on moral ideals or virtues (such as honesty, justice, compassion, loyalty) such factual agreement may be without much practical consequence. Even though we all might agree on the virtues, and even agree that we want students to be honest, just, compassionate, loyal, there is the practical problem. We can undertake every reasonable means to instil the virtues and be quite successful in the task sense without any guarantee of achievement.[40] Even excellent teachers and exemplars cannot ensure virtuous pupils.[41]

But even if we succeed in teaching knowledge of the virtues, the skill (if such it is) of moral perception, analysis, application we still may fail on the hurdle of disposition. For unfortunately, pace Plato, to know the right is not necessarily to love it. Failure to act is a failure of disposition and is the pivotal, though typically ignored, dilemma for teachers as moral educators. Moral education in school has little prospect of success if, on the one hand we aim for character development and simultaneously treat the acquisition of knowledge and understanding (lessons about morals) separately; if, that is, we continue to dichotomise theory and practice in our pedagogical arrangements. As free agents, students are always capable of accepting or rejecting what is taught; hence, there will always remain the very real possibility that a moral decision will not be acted upon.

For Aristotle the moral response develops in stages. At first, the child performs actions which are virtuous for reasons which have nothing to do with the actions or virtue themselves (for reward, to please an adult, etc.) As the child becomes accustomed to performing such actions she learns to appreciate what is good and feel aversion to what is not. So the child does not just perform just actions but comes to realise their goodness and performs them because they are good.  Aristotle realises that virtue requires appropriate training of perception, emotion, and desire as well as of rational capacities. But eventually habituation is not sufficient: character must be prepared to be interrogated by reason.

While argument and teaching, we may suspect, are not powerful with all men, but the soul of the student must first have been cultivated by means of habits for noble joy and noble hatred, like earth which is to nourish the seed. For he who lives as passion directs will not hear argument that dissuades him, nor understand it if he does; and how can we persuade one in such a state to change his ways? And in general passion seems to yield not to argument but to force. The character, then, must somehow be there already with a kinship to virtue, loving what is noble and hating what is base.[42]

So there must be moral character as a pre-requisite to moral reason.

Embracing a virtue-theoretical understanding of moral life and education does not require the acquisition of any new-fangled technical expertise, but only our return to a familiar and readily accessible pre-theoretical experience of human moral triumph and failure, as a basis upon which to build an educated appreciation - with the help of that vast repository of human wisdom which has and will always continue to inform such appreciation - of the rich nuances of ordinary human motivation and association.[43]