A Critical Role for Intuitions in Moral Theory

James T. Anderson

Moral intuitions, while ubiquitous in moral reasoning, have been the cause of considerable controversy in philosophy. My purpose here is to describe the most reasonable role for intuitions in moral theory, in order to look at some problems that arise, particularly for theories of justice, when intuitions are presumed to have this role.

I will first specify what I mean by "moral intuition". Intuitions are not a separate moral "sense" that directly perceives the moral quality of situations; nor are they mere emotional reflexes. Moral intuitions are judgments, that is, beliefs for which reasons can be given if we are called upon to reflect on them. As judgments, what is peculiar about intuitions is just that they are made without reflection. We routinely judge a situation as just or unjust, or an action as right or wrong, or a person as admirable or despicable, without first having to deliberate as to the evidence, or the principle under which the evidence is to be subsumed. One sees an act of unprovoked aggression, and one immediately judges it to be wrong. But the fact that this judgment is made without reflection does not mean that one cannot, if necessary, reflect on, and justify, the judgment.

As nonreflective judgments, intuitions have two main functions in our moral lives. First, they determine most of our everyday decisions about how to act. Most of our behavior as good or bad people takes place routinely, without reflection; it is only the occasional moral dilemma or very novel circumstance that requires us to stop and reflect. The second function is a theoretical one. We use moral intuitions in the evaluation of moral theories: for example, we are inclined to reject a moral theory when it entails that the right action in a particular situation is one that we intuitively know is wrong.

This second function is the one I want to focus on. The justificatory role of intuitions has given rise to considerable controversy in moral theory. The most notorious attempts to give intuitions a prominent role in moral theory come under the heading of intuitionism. In this school of thought, intuitions are made the epistemic foundation of moral theory. Prichard, for example, claims that all moral judgments are intuitions, so that there is no need for theory at all. Ross, on the other hand, argues that only general moral principles are known intuitively, and that particular judgments can be inferred from them with varying degrees of certainty. In both cases, it is claimed that the process of justification ends with intuitions. Moral intuitions are simply known to be true, and no reasons can be given to justify them. According to Ross, simply knowing the meaning of the terms in a moral principle should be sufficient to know that the principle is true.

The problem with this view of intuitions is obvious. There is no evidence for the claim that our moral intuitions are infallible, and there is abundant evidence to the contrary. Yet if moral intuitions are not infallible, and if no justification is possible for them, then there is no way to know whether they are correct, and they are therefore useless for moral reasoning.

At the opposite extreme from the intuitionists are those who claim that intuitions have no role whatsoever to play in moral reasoning, that relying on intuitions amounts to a kind of subjectivism, and that because of their irrational origins we should eliminate intuitions entirely and reason exclusively from the correct theory. But it seems difficult, if not impossible, to justify moral principles without any reference to our most stubborn nonreflective moral judgments. Most often such a view attempts to show that a moral principle can be derived simply from an analysis of the concept of morality. Critics of this approach charge that, aside from the requirement of universalizability, little can be done through analysis alone, leaving us with an empty formalism.

An alternative to these two foundationalist approaches that has gained increasing acceptance over the past few decades is the modified coherence theory known as reflective equilibrium. This model of moral theory has the virtue of granting both intuitions and theory some degree of authority, while regarding neither as self-sufficient or infallible. Rather than having principles justified entirely by intuitions, or judgments justified entirely by principles, the idea is to have reciprocal justification of each by the other.

Rawls is the best-known adherent of this approach in moral theory.1 As he describes it, reflective equilibrium works roughly like this. In developing a moral theory, we start with our considered judgments—our already-existing beliefs about right and wrong, justice and injustice. We then try to come up with general principles that systematize these judgments. The test of the accuracy of the principles is whether the judgments we routinely make conform to them. That is, if we were to reason from these general principles before undertaking any action, would the conclusions we come to regarding the right action coincide exactly with the judgments we make intuitively? If not, then we need to revise the theory to bring the principles more in line with our intuitions.

Thus far the process of reflective equilibrium makes firmly-held moral intuitions formed in the proper circumstances the standard by which moral theories must be justified. But we cannot regard even these well-formed intuitions as infallible. We need to acknowledge the possibility that some of them will fit poorly if at all with the best set of principles. In that case, some moral intuitions that conflict with these principles may on further examination turn out to be vestiges of an earlier period of moral underdevelopment, or ascribable to self-interest. We may decide to abandon those conflicting intuitions of which we are least certain. Then we move back to the theory, tweak it to bring it into line with our intuitions, and so on, back and forth, until, ideally, intuitions and principles stand in equilibrium.

In moving toward reflective equilibrium we develop general principles that systematize our intuitive judgments. But this is not simply a process of inductive generalization. Instead, we aspire to elucidate and articulate the fundamental concepts and principles that underlie our judgments. Thus the process of reflective equilibrium is one that permits us to discover our deepest values and helps us to understand why we make the judgments we do. Rawls uses an analogy with linguistics to illustrate this point. The principles of grammar are not given to us ahead of time when we learn our first language, not are they apparent to us once we've mastered the language: instead, the principles are embedded in our daily practice, and only a great deal of careful analysis is able to bring these principles to light.

The notion that by systematizing our moral intuitions we can articulate the deep underlying structure of morality has a problematic implication. Rawls thinks that if you can reach reflective equilibrium, then as far as moral truth is concerned, you've done the best you can. It is conceivable that two people, each with a significant number of moral intuitions which conflict with many of the other's, will come up with different sets of principles to which those intuitions conform. They could both arrive at reflective equilibrium within their own systems of moral thought, without internal contradiction. In that case, we would have something akin to subjectivism, albeit a rigorous version requiring internal consistency. We cannot argue with each other on the basis of fundamental principles that only one of us accepts. So reflective equilibrium could conceivably bring moral discourse to a dead end. We would, in effect, be back where subjectivism and emotivism leave us, in that any disagreements other than factual ones are impossible to resolve.

Rawls does mention this possibility, and even says that it would be good to get to that point, so that we could understand clearly just what sort of challenge confronts us as moral theorists. But a dead end is not a challenge. There simply is no way to reconcile two conflicting theories if both can be brought into reflective equilibrium with the moral intuitions of those who hold them.

It seems that for reflective equilibrium to work as a model of moral theory construction, for it to avoid subjectivism or relativism, it must be assumed that the framework underlying our most basic moral intuitions is held in common. That is, the process of reflective equilibrium, at least as Rawls describes it, implies that there is a universal schematism that underlies our considered moral judgments.

Proving that there is such a universal moral schematism would be an enormous task, and demonstrating what it must look like would be even greater. But it might help to enhance its plausibility if we return to the analogy that Rawls suggests with linguistics, as the activity of discovering and articulating the principles that, unseen, structure our everyday activity.

Rawls actually refers to Noam Chomsky here, and an analogy between moral theory and Chomsky's idea of linguistics may be useful. Chomsky is famous for, among other things, arguing that human beings necessarily possess an innate language acquisition device that limits what kinds of human language are possible. He claims that analysis of a single language and the prerequisites for competence in it shows that if a child were born a linguistic tabula rasa, it would be impossible for the child to do what children do in mastering a language. The child's capacity to learn sophisticated rules of transformational grammar must therefore be innate, a universal characteristic of the human species. And since a necessary evolutionary condition for the existence of any human language is that it be learnable by children, it follows that the possibilities for any human language must be circumscribed by this innate capacity.2

The important claim that Chomsky makes is that we are born with a universal set of parameters that get narrowed down through our exposure to our particular language. We are born capable of learning any language, not because we are born capable of learning everything, but because every language is an instantiation of the universal grammar with which we are born. Having an innate universal grammar does not mean that we are born already knowing how to use language, any more than we are born already knowing how to walk: instead, we have an innate capacity and predisposition to learn the rules and principles that govern the activity of our (or any) verbal community. It takes environmental stimulus of a particular kind to employ this capacity, to actually become competent.

This analogy suggests that the process of reflective equilibrium may discover the structure and limitations of a universal capacity for understanding and making moral judgments. As Rawls says, there is reason to believe that the fundamental principles brought to light through this process are complex and require considerable intellectual effort to articulate, just as with the principles governing language use.

Furthermore, the evidence for an innate moral capacity is the same as for an innate language capacity, namely, the ability of children to learn the complex rules, implicit knowledge of which is presupposed by competence. As language learners, children quickly develop the ability to generate an infinite variety of well-formed sentences, as well as to recognize instantly when a sentence is or is not well-formed, despite never having seen the sentence before, or knowing nothing about the subject matter. In the same way, children quickly develop the ability to recognize when a transaction is fair or unfair, despite never having seen the transaction before, and never having analyzed the concept "fair". Just the investigation of the basic concepts of right and wrong has occupied philosophers for centuries and is still ongoing, yet we do not need to explain to children what "right" means or what "wrong" means (assuming we could), in order for them to use these concepts correctly. Children understand the difference between a right act and a wrong act, although they might be unable to explain this difference.

I here only suggest the idea of an innate capacity for making and recognizing moral judgments. In order to get the discussion back to moral intuitions, I would like to carry the analogy with language just one step further, and then note a problem that is generated by it.

Chomsky's notion of universal grammar is a claim about the condition in which we begin life, a condition which is altered by our experience of language. Supposedly in universal grammar there are parameters for possible languages, a set of switches, and as one is exposed to one's native language, the switches are thrown in one direction or another. Native competence presupposes this setting of switches, and one's switches being already set is why learning foreign languages can be so difficult.

By analogy, an innate moral capacity might work this way. We have an innate capacity to learn how to make moral judgments and how to assent to or withhold our assent from moral judgments, without having to reflect or infer. But we are not born with a set of innate moral intuitions, as intuitionists believe. Instead, our possession of moral intuitions, that is, our ability to make moral judgments without reflection, requires environmental stimulus, in the form of our experience of relationships with others and the ways that others evaluate those relationships. The intuitions that are available to us in everyday practice will be determined in part by the social structure of the relationships in which we live. This means that the intuitions to which we have ready access could vary according to culture, or to one's role in a social structure. Contrary to cultural relativism, however, moral intuitions are not merely the product of culture or upbringing, because of the universal innate capacity that governs and limits the range of possible intuitions. Nor does the possession of this capacity guarantee the correctness of whatever moral intuitions we do learn to make, any more than the possession of a universal grammar guarantees the production of well-formed sentences.

This way of understanding moral intuitions and their origin may permit us to understand differences in moral judgments that correspond to cultural and social-role differences, yet at the same time leaves open the possibility of moral theory articulating the conceptual basis of all human morality.

The analogy with language acquisition suggests that we are born with the capacity to learn to employ the whole range of moral intuitions, but that our subsequent experience necessarily limits the range of those that we find self-evidently true. Conceivably one could, with effort, learn to adopt the moral intuitions that come naturally to others, somewhat in the way that one learns a foreign language.

We thus can think of the feminist critique of liberal moral theory as being based partly on the absence of certain intuitions in liberal theory, an absence that can be attributed to a difference in the way that men and women traditionally experience others. This difference in the way others are experienced is in turn a function of the social structure of relationships. One of the main claims of feminist moral theory of the past thirty years has been that the socialization and experience of women, primarily in the domestic sphere, has enabled them to bring a different moral perspective to situations, a moral perspective that entails or is informed by different moral intuitions from the ones that males typically produce or recognize as obviously true.

I suggest that the feminist critique of traditional liberal moral philosophy is, in part, a call to understand and grant equal status to what traditionally have been feminine moral intuitions. The idea is that, as the history of moral theory has been almost exclusively the province of males, the intuitions to which men traditionally have access have become enshrined in theory as the self-evident values of moral theory: autonomy, obedience to law, freedom, individuality, and individual well-being. Meanwhile, the traditionally feminine intuitions having to do with the importance of care, and of maintaining relationships, are like a foreign language to moral philosophers steeped in the tradition. Annette Baier's advocacy for a moral theory that harmonizes justice and care can be seen as an appeal for moral bilingualism, however much effort may be required for one to learn the language of the other.3

From this example, it appears that fallibility is not the only problem with moral intuitions. Moral intuitions are not infallible, but those that have been tested reflectively against principles may be trusted. Nevertheless, even a correct set of intuitions can be incomplete, setting up a conflict with a different set of equally correct but incomplete intuitions. To the degree that the intuitive differences arising from differences in the experience of others produce theoretical conflicts, any common understanding, not to speak of agreement, will be more difficult to achieve.

There is one more example of a problem caused by the absence of moral intuitions that I would like to discuss. One of the most important ways that we experience others is in relationships of dependence. To depend on someone is to be unable to realize one's goals or aspirations, to be unable to exercise one's will, without the cooperation of the other. For most of the moral tradition, dependence appears as a weakness, the absence of independence and autonomy. But in fact the necessity of dependence pervades our everyday lives. This is most obvious in the case of small children, the elderly, or the disabled, but on reflection it can be seen that we constantly depend on others for the very existence of the world we live in. The everyday world consists of what we stand on, sit on, shelter beneath, drive, wear, eat, and everything else that is in any way necessary for us to live, much less accomplish anything. All of this must be created and put here by other people.

Of course, these observations are true in a trivial way. But there is something revealing in that very triviality. Throughout almost all of history, until relatively recently, this dependence has been transparent: that is, the common human experience has included an awareness of the roles that concrete particular others—community members and ancestors—play in the creation and maintenance of the prerequisites of life.

We can try to get a sense of what this experience might have been like by imagining life on a remote island with a few other people, where each makes an indispensable contribution to the survival of the group. In this scenario, in contrast to our normal experience, one is immediately aware of this dependence: one knows the people on whom one depends. The residents of this island, I suggest, would have immediate access to a moral intuition that is remote from most of us. Each would know that she is not merely an individual but part of something greater, a common striving with which she identifies. Consequently, each would know without reflection that it is right to sacrifice for the common good, that an injury to one is an injury to all, and other apparently outdated moral truths. For lack of a better term I will call this the moral intuition of solidarity.

This unrealistic scenario is only an idealized form of what was once normal human existence. Our current situation, in which material goods seem to occupy their own independent realm apart from the social, is an anomaly, a relatively recent occurrence in history. We enjoy a rich social life, filled with friends, family, co-workers, and fellow members of teams, clubs, associations, and political parties. But whereas social life once was congruent with the common effort to advance the flourishing of the whole, today the two have become completely divorced.

Most of us work, and we receive money in exchange for our work. In order for this exchange to occur, our work must be marketable, which is to say that what we do must somehow be needed by others. Thus the dependence of others on us comes into our experience as the money we receive for our work. And our dependence on others for the things around us is experienced as the effortless exchangeability of that money for the goods they have produced. Personal interdependence now appears as an interaction of the products of others' labor with the money received for our labor. The things we need reside in stores, awaiting not us but our money, with which they speak a common language and which they recognize as their equal.

It is odd to speak of goods and money interacting in the place of ourselves and those with whom we are interdependent, but this way of speaking is not without precedent. Early in the book Capital, Marx has a mysterious and obscure discussion of something he calls the fetishism of commodities, a term which refers not to an obsession with consumer goods, but to the propensity of commodities to take on characteristics of the living. There he speaks of "a definite social relation between persons, that assumes, in their eyes, the fantastic form of a relation between things." I believe that Marx had in mind the phenomenon I am suggesting: the relations of interdependence that for thousands of years coincided with social relations and were therefore directly present in experience, have undergone a metamorphosis and are no longer experienced as such.

This change in the way we experience our mutual interdependence may be due, as Marx apparently thought, to the capitalist mode of production, or it may merely be due to production taking place on a global scale. In any case, it is a problem for moral theory, insofar as it deprives us of a moral intuition, solidarity, that has played a role in all of human life until recently. I will briefly give an example of how this problem emerges in current political discourse.

Libertarianism is enjoying unprecedented popularity in the U.S., and one could argue that it is the dominant philosophical orientation among those who rule the country. This is the theory based on the notion that individual liberty and the right to property are absolute, that is, can never be overridden for any other moral reason. This entails, among other things, minimal government, and no taxation except as necessary to protect liberty and property rights.

As with any other moral theory, libertarianism is often attacked for the counterintuitive results that it endorses: for example, that gross inequalities, including abject poverty in the midst of great wealth, are just as long as they result from a series of voluntary transfers from an initial just situation. Some who call themselves libertarians try to get around this objection simply by denying that such results obtain. When government is minimized, it is claimed, the invisible hand of the free market will make everyone prosperous. Even if it were plausible, this response is not available to the true libertarian, because it is at bottom a utilitarian response, and libertarianism is a deontological theory. True libertarians will bite the bullet and claim that, as long as it is the result of voluntary transactions, poverty is preferable to the injustice of redistribution of wealth. That is, they deny that the result is counterintuitive at all.

When moral debate gets to the point where people agree on what a theory entails, but disagree about whether that entailment is counterintuitive, we have to wonder whether the moral intuitions with which we are operating are shared at all. I want to suggest that libertarianism is plausible only to those who do not find its consequences counterintuitive, and that this is because they are lacking the moral intuition of solidarity, most likely for the reasons given above. If this is the case, it would seem to be incumbent on the libertarian to understand and, if possible, learn this intuition, in order for dialogue to continue.

It would appear that the way in which a social structure forms our experience of others and thus determines which moral intuitions will be accessible to us, ought to be a concern of the theory of justice. That is, if moral dialogue is hindered or even rendered impossible because a social system structures or obscures relations so as to deprive people of an important moral intuition, then that ought to be a factor in evaluating the justice of that social structure.

But if that is the case, then the theory of justice faces a kind of paradox. The theory of justice is supposed to be a consistent set of principles used to evaluate the justice of a social structure. Assuming that reflective equilibrium is the most reasonable way to construct a moral theory, the theory will be justified according to its ability to produce judgments that conform to our moral intuitions. But if a theory is to be evaluated in accordance with the intuitions that we actually have, and if the theory is to evaluate the social structure in which we live, on what basis can the theory criticize the social structure for intuitions that are absent? Isn't the theory doomed to a kind of inherent conservatism, doing its best to systematize our status quo intuitions?

I am not certain of how serious this problem is, or if there is an answer to it. But if there really is a universal, innate capacity for learning to make moral judgments, then perhaps the ultimate task of an ideal theory of justice will be to reveal the deep structure of the possibility of all coherent moral judgments, and thus to provide grounds for criticizing a social structure that hinders the formation of any of them.

1. John Rawls, A Theory of Justice (Cambridge: Harvard University Press, 1971), pp. 17-22, 46-53.

2. Vivian Cook and Mark Newson, Chomsky's Universal Grammar, Second Edition (Oxford: Blackwell Publishers, 1996), pp. 75-133. The analogy with Chomsky is introduced by Jeff McMahan in "Moral Intuition", The Blackwell Guide to Ethical Theory, Hugh LaFollette, ed. (Oxford: Blackwell Publishers, 2000).

3. Annette Baier, "The Need for More Than Justice", Moral Prejudices (Cambridge: Harvard University Press, 1994)