My favorite empirical research programs try to help something good work in the world. For instance, scholars who study Positive Youth Development assess initiatives that give young people opportunities to contribute to their communities. Scholars of Common Pool Resources study how communities manage common property, such as fisheries and forests. Scholars of Deliberative Democracy investigate the impacts on citizens, communities, and policies when people talk in structured settings.

These are empirical research programs, committed to facts and truth. They do not seek to celebrate, but to critically evaluate, their research subjects. However, an obvious goal is to make the practical work succeed by identifying and demonstrating positive impacts and by helping to sort out the effective strategies from the ineffective ones. Underlying these intellectual efforts is some kind of hope that the practical programs, when done well, succeed.

As a philosopher, I am especially interested in that hope and why scholars have it. I like to ask what motivates these research projects. The motives are largely hidden, because positivist social science cannot handle value-commitments on the part of researchers; it treats them as biases to be minimized and disclosed only if they prove impossible to eliminate. Often the search for motives is critical and suspicious: one tries to show that a given research project is biased by some value-judgment, cultural assumption, or self-interest on the scholars’ part. But I look for motives in an appreciative spirit, believing that an empirical research program in the social sciences can only be as good as its core values.

Note that it is not at all obvious why we should hope that Positive Youth Development, Common Property Resource Management, and Deliberative Democracy work. These are expensive and tricky strategies. For instance, the core empirical hypothesis of Positive Youth Development is that you will get better outcomes for youth if you help them contribute than if you use surveillance and remediation. But it would be cheaper and more reliable if we could cut crime with metal detectors in every school instead of elaborate service-learning programs. So why should we hope that Positive Youth Development is right?

Likewise, it would be easier to turn all resources into private or state property than to encourage communities to manage resources as common property. And it would be easier for professionals to make city plans and budgets than to turn those decisions over to citizens. So why do scholars evidently hope that good common property regimes produce more sustainable and efficient economic outcomes than expert management, and that deliberations generate more legitimate and fair policies than governments do?

I think part of the reason is simply that things are not going very well in the world, and scholars seek alternatives that may be uncontroversially better: more efficient or sustainable, less corrupt and wasteful. That’s part of the reason, but it doesn’t fully explain the focus of these research projects. If you’re worried about violence in American high schools, you should look for something new that works. But why should that new approach include service and leadership programs, instead of better metal detectors and video cameras?

Ultimately, all three of my examples are anchored in commitments that I would describe as “Kantian.” The individual is a sovereign moral agent and our responsibility to others is always to help develop their capacities for autonomy and voluntary cooperation. Real Kantianism is dismissive of utilitarian outcomes (such as efficient public services) and is willing to defend autonomy even if the consequences for health and welfare turn out to be bad. But real Kantianism just doesn’t fly. It doesn’t influence power and it doesn’t satisfy most people’s intuitions. So I think the research projects I have mentioned here are motivated by a kind of soft or strategic Kantianism. The best initiatives, on this view, are the ones that achieve efficient and reliable improvements in tangible human welfare by enhancing people’s autonomy. Strategies like Positive Youth Development and common property regimes stand out as worthy of study because of their Kantian values. But they deserve critical scrutiny on utilitarian grounds. If they fail to deliver the promised practical outcomes, they should be improved before they are abandoned. The same attention should not be given to surveillance systems or top-down managerial structures. In theory, those solutions might work just as well, but helping them to succeed would not enhance autonomy.

I realize that it is a risky strategy in our culture for scholars to admit their core moral commitments. The smartest move is to pretend that a research program is simply scientific and all the outcomes of interest are utilitarian. But those assumptions have the disadvantage of being wrong. They distort research in various subtle but damaging ways. Even though it is idealistic, I think we should take on positivism directly and not accept the presumption that values are simply biases.

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.

John Rawls was the most influential recent academic political philosopher in the English-speaking world, or at least the most influential academic who defended liberal views. If you take him at face value, he is a very abstract kind of thinker. In fact, he says in section 3 of A Theory of Justice:

My aim is to present a conception of justice which generalizes and carries to a higher level of abstraction the familiar theory of the social contract as found, say, in Locke, Rousseau, and Kant. In order to do this we are not to think of the original social contract as one to enter a particular society or to set up a particular form of government. Rather, the guiding idea is that the principles of justice for the basic structure of society are the object of the original agreement.

In a famous methodological move, he defines the “original position” as one in which persons are ignorant of all morally irrelevant facts so that each cannot “tailor principles to the circumstances of [his or her] own case.” By making us ignorant of most empirical facts about ourselves, Rawls makes his theory seem more abstract than even Kant’s.

As Rawls works out the actual framework of justice, it turns out that the government should do certain things and not others. Parties to the original contract would want there to be “roughly equal prospects of culture and achievement for everyone similarly motivated and endowed. The expectations of those with the same abilities and aspirations should be not be affected by their social class.” To achieve this outcome, the government should fund education and channel educational resources to the least advantaged. I presume it should also regulate employment contracts to prevent discrimination, thus enacting the principle of “careers open to talents.” But the government should not be in charge of child-rearing, even though families affect people’s capacities and motivations. (“Even the willingness to make an effort, to try, and so to be deserving in the ordinary sense is itself dependent upon happy family and social circumstances.”) The state should compensate people from unhappy families, but should not take over the family’s traditional function.

Why not? One answer might be that Rawls was insufficiently radical and consistent. He arbitrarily excluded the family from his program of reform because of prejudice. I have a different view than this–more favorable to Rawls’ conclusions but less supportive of his methods.

I don’t believe that his reasoning was nearly as abstract as he claimed. Instead, I think he was a reader of newspapers and an observer of life in America, ca. 1945-1975. He observed that the actual government did a pretty good job of providing universal education but could still improve the equality of educational opportunity. The government policed employment contracts increasingly well to prevent racial and gender discrimination, albeit with room for improvement. But the government didn’t do child-rearing well. (The foster care system was only an emergency response that, in any case, relied on private volunteers.) Rawls derived from the immediate past and present some principles for further reform.

That interpretation makes Rawls a good thinker, sensible and helpful, but not quite the kind of thinker he believed himself to be. In my view, he was less like Kant (elucidating the universal Kingdom of Ends from the perspective of pure reason) and more like Franklin Roosevelt, defending the course of the New Deal and Great Society in relatively general and idealistic terms. Or he was like John Dewey, critically observing reality from an immanent perspective. The reason this distinction matters is methodological. As we go forward from Rawls, I think we need more social experimentation and reflection on it, not better abstract reasoning about the social contract.

Yesterday, for fun, I posted a clip of the philosopher Jonathan Dancy on the Late Late Show. His interview raises an interesting and serious question. Asked whether philosophers should dispense moral advice, Dancy says: No. I would agree with that, for reasons stated below. But Dancy goes further and suggests that philosophers shouldn’t address substantive moral issues at all. He implies that people’s ethical judgments are already in pretty good shape. A philosopher’s job is to understand what kind of thing an ethical judgment is. In other words, moral philosophy is meta-ethics.

That is a controversial claim. John Rawls, Peter Singer, Robert Nozick, Judith Jarvis Thomson, and many other modern philosophers have advanced and defended challenging theses about morality. Since the great renaissance of ethics in the English-speaking world (1965-1975), its ambitions have diminished, I think, and a distinction has arisen between ethics (which is very “meta”) and applied ethics (which is mostly about a given topic area, and not very philosophical). This split seems a harmful development, because the best moral philosophy is methodologically innovative and challenging and also addresses real issues.

Why shouldn’t philosophers dispense advice? Because what one needs to advise people well is not only correct general views (which, in any case, many laypeople hold), but also good motivations, reliability and attention, fine interpretative skills, knowledge of the topic, judgment born of experience, and communication ability (meaning not only clarity but also tact). There is no reason to think that members of your local philosophy department are above average on all these dimensions.

But correct general views are valuable, and philosophers offer proposals that enrich other people’s moral thinking. You wouldn’t ask John Rawls to run a governmental program or even to advise on specific policies, but your thinking about policies may be better because you have read Rawls. It so happens that he held some interesting ideas about meta-ethics, but those were merely complementary to his core views, which were substantive.

I’m afraid I detect a general withdrawal from offering and defending moral positions in the academy. Humanists like to “problematize” instead of proposing answers. Social scientists are heavily positivist, regarding facts as given and values as arbitrary and subjective (thus not part of their work). If moral philosophers begin to consider the offering of moral positions as beyond their professional competence, there’s virtually no one left to do it.

I’m an adherent of a very small and obscure philosophical school called “particularism.” (Of course, because I’m an academic, I have to have my own special flavor of it.) The best known particularist is Jonathan Dancy, whom I only met once but who nicely reviewed a book manuscript of mine. And his work has had a big influence on me, even though I come at things from a different angle. Anyway, unbelievable as it may seem, here he is explaining particularism on Craig Furgeson’s “Late Late Show” on CBS:

I’ve never seen his show, but this Furgeson guy strikes me as pretty smart. And Dancy does a credible job in a terrifying situation. It turns out he’s the actress Claire Danes’ father-in-law. That relationship–rather than the arguments in “Are Basic Moral Facts both Contingent and A Priori?” (2008)–may be the reason for Dancy’s new TV career. Whatever the reason, long may it prosper.