When we talk together about public concerns, a whole range of phrases and concepts is likely to emerge. Imagine, for example, that the topic is a local public school: how it is doing and what should change. In talking about their own school, parents and educators may use abstract moral concepts, like fairness or freedom. They may use concepts that have clear moral significance but controversial application in the real world. For example, fairness is a good thing, by definition. It is not the only good thing, and it can conflict with other goods. But the bigger challenge is to decide which outcomes and policies actually are fair.
Other concepts are easy to recognize in the world but lack clear moral significance. We either bus students to school or we do not bus them, but whether busing is good is debatable. (In this respect, it is a very different kind of concept from fairness.) Still other concepts have great moral weight and importance, but their moral significance is unclear. You can’t use the word love seriously without making some kind of morally important point. But you need not use that word positively: sometimes love is bad, and the same is true of free and achieve.
People string such concepts together in various ways. They may make associations or correlations (“The girls are doing better than the boys in reading”). They may make causal claims (“The math and reading tests are causing us to overlook the arts.”) They may apply general concepts to particular cases. Often they will describe individual teachers, administrators, events, classes, and facilities with richly evaluative terms, such as beautiful or boring. Frequently, they will tell stories, connecting events, individuals, groups, concepts, and intentional actions over time.
All these ways of talking are legitimate in a democratic public discussion. But the heterogeneity of our talk seems problematic. So many different kinds of ideas are in play that it seems impossible to reach any principled or organized resolution. We talk for some arbitrary amount of time, and then a decision must be made by the pertinent authorities or by a popular vote. It is not clear whether the decision was correct based on the discussion that preceded it.
It seems beneficial to organize and systematize public discussion, and several kinds of experts stand ready to help:
- Social scientists propose to organize public discussions by identifying reliable causal relationships among concepts that can be empirically identified in the world. For instance, success comes to mean passing a test or graduating on time, and class size is found to influence (or not to influence) success. The hope is—if not to end the discussion—at least to focus and rationalize it.
- Managers (both actual administrators of our institutions and experts on management) hope to limit or organize public discussions by pronouncing on which strategies will work and which are permissible under the current rules and policies.
- Ideological thinkers try to simplify the discussion by putting heavy weight on certain moral concepts, which then trump others. (For example, personal liberty is a trump card for libertarians; equal welfare, for social democrats.)
- Lawyers are trained to guide public discussions by explaining which options are legal or obligatory under laws, precedents, and constitutions.
- Moral and political philosophers have less public influence than the other groups mentioned so far, but they hold the most subtle and sophisticated views of how public discussions ought to be improved. Contemporary academic philosophers are often disarmingly modest about their contributions, yet a core professional goal is to improve discussions by identifying morally clear and invariant concepts that should then influence decisions. Depending on which philosophical school one defends, those concepts might include rational autonomy, maximum utility, or virtue.
All of these forms of expert and disciplined guidance can be useful. But they often conflict, and so the very fact that they all help should tell us something. There is no methodology that can replace or discipline our public discussions or bring them to a close. This is because of the nature of moral reasoning itself.
Moral concepts are indispensable. We cannot replace them with empirical information. Even if smaller class sizes do produce better test scores, that does not tell us whether our tests measure valuable things, whether the cost of more teachers would be worth the benefits, or whether the state has a right to compel people to pay taxes for education.
But moral concepts are heterogeneous. Some have clear moral significance but controversial application in the world. (Fairness is always good, and murder is always bad.) Others have clear application but unpredictable moral significance. (Homicide is sometimes murder but sometimes it is justifiable.) Still others are morally important but are neither predictable nor easily identified. (Love is sometimes good and sometimes regrettable, and whether love exists in a particular situation can be hard to say.) A method that could bring public deliberation to closure would have to organize all these concepts so that the empirically clear ones were reliably connected to the morally clear ones.
That sometimes happens. For instance, waterboarding either happens or it does not happen. The Bush Administration’s lawyers defined it in obsessive detail: “The detainee is lying on a gurney that is inclined at an angle of 10 to 15 degrees to the horizontal. … A cloth is placed over the detainee’s face and cold water is poured on the cloth from a height of approximately 6 to 18 inches …” Waterboarding is, in my considered opinion, an example of torture. Torture is legally defined as a felony, and the reason for that rule is a moral judgment that torture is always wrong (in contrast to punishment or interrogation, which may be right). Therefore, waterboarding is wrong. This argument may be controversial, but it is clear and it carries us all the way from the concrete reality of a scene in a CIA interrogation room to a compelling moral judgment and a demand for action. The various kinds of concepts are lined up so that moral, legal, and factual ideas fit together. There is room for debate: Is waterboarding torture? Who waterboarded whom? But the debate is easily organized and should be finite.
If all our moral thinking could work like that, we might be able to bring our discussions to a close by applying the right methods–usually a combination of moral philosophy plus empirical research. But much of our thinking cannot be so organized, because we confront moral concepts that lack consistent significance. They are either good or bad, depending on the circumstances. Nevertheless, they are morally indispensable; we cannot be good human beings and think without them. Love and freedom are two examples. To say that Romeo loves Juliet–or that Romeo is free to marry Juliet–is to say something important, but we cannot tell whether it is good or bad until we know a lot about the situation. There is no way to organize our thinking so that we can bypass these concepts with more reliable definitions and principles.
A structured moral mind might look the blueprint of a house. At the bottom of the page would be broad, abstract, general principles: the foundation. An individual’s blueprint might be built on one moral principle, such as “Do unto others as you would have them do unto you.” Or it might start even lower, with a metaphysical premise, like “God exists and is good.” At the top of the picture would be concrete actions, emotions, and judgments, like “I will support Principal Jones’s position at the PTA meeting.” In between would be ideas that combine moral principles and factual information, such as, “Every child deserves an equal education,” or “Our third grade curriculum is too weak.” The arrows of implication would always flow up, from the more general to the more specific.
I think most people’s moral thinking is much more complex than this. Grand abstractions do influence concrete judgments, but the reverse happens as well. I may believe in mainstreaming special-needs children because of an abstract principle of justice, and that leads me to support Mrs. Jones at the PTA meeting. Or I may form an impression that Mrs. Jones is wise; she supports mainstreaming; and therefore I begin to construct a new theory of justice that justifies this policy. Or I may know an individual child whose welfare becomes an urgent matter for me; my views of Mrs. Jones, mainstreaming, and justice may all follow from that. For some people, abstract philosophical principles are lodestones. For others, concrete narratives have the same pervasive pull—for example, the Gospels, or one’s own rags-to-riches story, or Pride and Prejudice.
We must avoid two pitfalls. One is the assumption that a general and abstract idea is always more important than a concrete and particular one. There is no good reason for that premise. The concept of a moral “foundation” is just a metaphor; morality is not really a house, and it does not have to stand on something broad to be solid. Yet we must equally avoid thinking that we just possess lots of unconnected opinions, none intrinsically more important than another. For example, the following thoughts may all be correct, but they are not alike: “It is good to be punctual”; “Genocide is evil”; and “Mrs. Jones is a good principal.” Not only do these statements have different levels of importance, but they play different roles in our overall thinking.
I would propose switching from the metaphor of a foundation to the metaphor of a network. In any network, some of the nodes are tied to others, producing an overall web. If moral thinking is a network, the nodes are opinions or judgments, and the ties are implications or influences. For example, I may support mainstreaming because I hold a particular view of equity; then mainstreaming and equity are two nodes, and there is an arrow between them. I may also love a particular child, and that emotion is a node that connects to disability policy in schools. A strong network does not rest on a single node, like an army that is decapitated if its generalissimo is killed. Rather, a strong network is a tight web with many pathways, so that it is possible to move from one node to another by more than one route. Yet in real, functioning networks, all the nodes do not bear equal importance. On the contrary, it is common for the most important 20 percent to carry 80 percent of the traffic–whether the network happens to be the Internet, the neural structure of the brain, or the civil society of a town.
I suspect that a healthy moral mind is similar. It has no single foundation, and it is not driven only by abstract principles. Concrete motives (like love or admiration for a particular individual) may loom large. Yet the whole structure is network-like, and it is possible for many kinds of nodes to influence many other kinds. My respect for Mrs. Jones may influence how I feel about the concept of the welfare state, and not just the reverse. I need many nodes and connections, each based on experience and reflection.
I do not mean to imply that a strong network map is a fully reliable sign of good moral thinking. A fascist might have an elaborate mental map composed of many different racial and national prejudices and hatreds, each supported by stories and examples, and each buttressing the others. That would be a more complex diagram than the ones possessed by mystics who prize purity and simplicity. Purity of Heart Is to Will One Thing, wrote Sören Kierkegaard, and the old Shaker hymn advises, “‘Tis the gift to be simple, ‘tis the gift to be free, ‘Tis the gift to come down where we ought to be.” A righteous Shaker would do more good than a sophisticated fascist. But even if complexity is not a sufficient or reliable sign of goodness, a complex map is both natural and desirable. It reflects the real complexity of our moral world; it reduces the odds of becoming fanatical; it hems in self-interest; and it is resilient against radical doubt.
Four conclusions follow from this discussion.
1. We should banish a certain kind of moral skepticism which arises from thinking that moral conclusions always rest on foundations, but alas there is nothing below our biggest, most abstract ideas. For example, you may believe in the Golden Rule but be unwilling to say why it is true. You may feel that there is no answer to the “Why?” question, and therefore morality is merely prejudice or whim. Your moral house has a foundation (the Golden Rule), but the foundation is floating in air. Fortunately, our whole morality does not rest on any such rule, nor must a principle rest on something below it to be valid. The Golden Rule is part of a durable network. It gains credibility because it seems consistent with so many other things that we come to believe. If it or any other node is knocked out of the network, the traffic can route around it.
2. Moral thinking is influenced by worldly experience, by practice and by stories, and not only by abstract theories and principles. I wrote that it “is influenced” by experience; I have not shown that our thinking should be deeply experiential. But at the least, we can say that there is no reason to put abstract thinking on a pedestal, to treat is as if it were intrinsically and automatically more reliable than concrete thinking. I can be just as certain that I love my children as in the truth of the Golden Rule.
3. We can handle diversity. If individuals’ conclusions derived from the foundations of their thought, we would face a serious problem whenever we encountered people who had different from foundations from our own. It is hard to tolerate them, let alone deliberate with them. The existence of a different foundation can even provoke vertiginous skepticism in our own minds. If my worldview rests on utilitarianism, and yours depends on faith in Jesus’ resurrection, perhaps neither of us has any reason to hold our own position. But if our respective worldviews are more like networks, then they probably share many of the same nodes even though they differ in some important respects. What’s more, each person’s network must be slightly different from anyone else’s—even his twin brother’s. Thus when we categorize people into “cultures,” we are crudely generalizing. There is actually one population of diverse human beings who are capable of discussing their differences even though they may not reach agreement.
4. Expertise plays a limited role in reaching good decisions. The moral network in my mind cannot be–and should not be–radically simplified by applying any sophisticated methodology. I can learn from experts about what causes what and about how we should define various concepts and principles. But at the end of that process, I will still have my own moral network map, nourished by many sources other than the experts, and I will have to make decisions both alone and in dialog with my peers. There is no substitute for thinking together about problems and solutions.