These are early days in the philosophical study of character. We know very little about what most peoples’ character looks like. Important virtues are surprisingly neglected. There are almost no strategies advanced by philosophers today for improving character. We have a long way to go.
This is bound to sound exaggerated. After all, the topic of character is one of the oldest in both Western and Eastern philosophy, and has enjoyed a renaissance in analytic philosophy since at least the 1970s with the revival of virtue ethics. To say that we are mainly in the dark about these matters sounds like I haven’t been paying attention to my history lessons. Maybe that’s true, but first let me explain what I mean.
Let’s begin at the start of the 2000s. The philosophical discussion of virtue ethics had already been flourishing thanks to the work of Elizabeth Anscombe, Alasdair MacIntyre, Rosalind Hursthouse, Christine Swanton, Philippa Foot, and Michael Slote, among others. However, very little had been done up to that point to connect concepts like character and virtue with the vast amount of empirical research in psychology on moral behaviour and motivation.
Enter Gilbert Harman and John Doris. In a series of articles, and in Doris’s 2002 book Lack of Character, they sparked a very lively discussion about what the empirical literature in psychology does and does not show. In their minds, a number of studies – like the Milgram shock experiments, the Darley and Batson Good Samaritan helping experiment, and the Isen and Levin phone book helping experiment – collectively give us good reason to think that most people do not have the traditional moral virtues such as compassion. Doris did, to his credit, admit that there might be a few exceptions, but he held that by and large the psychological results do not paint a very flattering picture for the rest of us.
Harman and Doris did not stop there. They took their argument one step further by drawing implications for ethical theory. Given the rarity of the traditional virtues, they held that any approach in ethics which makes significant use of those traits needs to be called into question. Aristotelian approaches to virtue ethics were the most obvious and natural target of their attack, but not the only one.
Aristotelians pushed back. The ensuing years saw the publication of over 50 articles and at least a dozen books and edited volumes devoted to exploring the empirical adequacy of the virtues and what philosophical implications the psychological research is supposed to really have.
I find all this very exciting. Philosophical work in ethics has expanded in a major way into territory it has hardly ever occupied.
But I do not plan on reviewing this literature here, nor will I weigh in on whether Aristotelians have been able to respond adequately to this so-called “situationist challenge” or not. My interest in this essay is less in what has already been said, and more in what has not yet been said at the intersection of the philosophy and psychology of character. Specifically, let’s look closely at three areas of comparative neglect.
1. What about the Other Virtues? When Harman and Doris jumpstarted the contemporary discussion of the empirical credentials of virtue, they focused their attention on experiments in psychology which had to do with helping and harming others. Finding a dime led to much greater helping. As did walking past Mrs. Fields cookies. Or coming out of the bathroom. On the flip side, pressure from an authority figure led people to administer what they thought were electric shocks, eventually seeming to cause another person’s death. Being in a hurry to complete a task led to a man’s suffering being ignored. And helping was rare when a nearby stranger did nothing to help, even if there was a clear emergency in the next room. Not exactly the patterns of behaviour you’d expect to find if most of us are compassionate people!
This narrow focus on helping and harming was a wise move strategically. Rather than introduce a few studies with respect to each of the virtues, which would hardly warrant any interesting empirical inferences, Doris in particular chose to go into detail with respect to one area of morality. As a result, his conclusion strictly speaking couldn’t be about the lack of virtue in general, but rather just about the lack of compassion.
Leaving aside whether he is right to draw that conclusion, the natural question that arises is how the other virtues – temperance, humility, courage, fortitude, forgiveness, gratitude, generosity, and so forth – fare on empirical grounds. Are we as susceptible in these areas to morally irrelevant considerations in the environment, such as smells and going to the bathroom, as we appear to be when helping others? What about pressures from authority figures or the influence of group inactivity?
This is a place where philosophers and others who engage in research on normative (as opposed to empirical) topics can make a real contribution. For we need to have criteria to use to decide what counts as honest motivation or forgiving behaviour, for instance. We can then take those criteria and bring them to bear on the experimental results. Empirical findings themselves cannot give us such normative criteria.
Unfortunately, philosophers don’t have any idea what is going on empirically with the other virtues (and I’m not sure anyone else does either, for that matter!). Part of the reason why is that for some moral domains there just isn’t the wealth of existing studies to analyse, in the way that there is for helping behaviour. Stealing is a good example – as you might imagine, it is hard to do experimental studies testing what factors influence actual (rather than self-reported) theft.
But part of the reason is also that examining the empirical credentials of individual virtues just hasn’t been on our philosophical radar screens. Our attention has been elsewhere.
Take the virtue of honesty, for instance. If any virtue is on most people’s top five list, it is that one. Yet it has had no traction at all in the philosophical literature. I can quantify what “no traction” means – there has not been a single paper in a mainstream philosophy journal on the moral virtue of honesty in over fifty years.
So clearly conceptual work needs to be done on honesty (for what it’s worth, that’s my current book project). But our interest here is empirical, and there have been numerous studies of both lying and especially cheating which could be examined. To take just one example, Lisa Shu at the London Business School and her colleagues in a 2011 study found that a control group averaged 8 correct answers on a 20 problem test. They had no opportunity to cheat and were paid 50 cents per correct answer. Test-takers in another group knew they would get paid 50 cents too, but they also knew that they could cheat if they wanted to and get away with it completely undetected. The result? 13.22 problems answered “correctly.” This is a startling result which raises a host of interesting questions (such as, if you are going to cheat, why not cheat more?). Yet philosophers have been almost completely silent about what studies like this one have to teach us about how honest (or not) we tend to be.
Or take the virtue of generosity. Again, it is sadly neglected by philosophers – just three papers in mainstream journals in the past forty years (by way of comparison, there are over two dozen papers on the virtue of modesty – who would have expected that to happen?). And again, there is plenty of relevant empirical literature to examine to see whether people tend to possess this virtue. For instance, there are hundreds of studies looking at what happens in variations of the dictator game, in which a participant is given a good, such as money, and can choose how much (if any) to share with another person. This research can serve as a useful jumping off point for probing our character when it comes to matters of generosity.
So the upshot is that compassion may indeed be a rare virtue. But at this point it is not at all clear how we are doing with the rest of them.
2. If Not Virtue, What? I admit, though, that it is hard to accept that most people are virtuous (or even have a few moral virtues). The news, politics, and Hollywood give us plenty of examples to suggest otherwise. And from my own examination of the empirical literature pertaining to cheating, donating, and the like, I don’t see much evidence of widespread virtue possession there either.
At this point, I think most philosophers who work in this area are comfortable with the claim that the moral virtues are rare. That was Aristotle’s claim as well thousands of years ago.
But if not virtue, then what? What is the correct story supposed to be about how our character actually works a story which can successfully explain the empirical data such as it is? Here too contemporary philosophy doesn’t seem to have much of an idea.
One could just dispose of character traits all together, whether of the traditional kind or any other. The idea would be that other factors are enough to explain moral behaviour. But no philosopher working on these issues has gone this far, for good reason. Harman came close (he once wrote a paper called “The Nonexistence of Character Traits”), but even he ended up accepting some role for traits (more on this in a moment).
In developing a story which appeals to character traits, we should consider these choices:
(i) Most people have either “global” or “local” character traits. Global character traits function across a wide variety of relevant situations. So if Jones is honest, then he reliably tells the truth when appropriate in the courtroom, and in the office, and in the bar (among other places). On the other hand, local character traits function only in a narrow range of situations. So Smith might have honesty-just-in-the-courtroom, which is compatible with his not having honesty-in-the-bar.
(ii) Most people have character traits that tend to give rise either to virtuous motivation, or to vicious motivation, or to some other kind of motivation.
(iii) Most people have character traits that tend to give rise either to virtuous behaviour, or to vicious behaviour, or to some other kind of behaviour.
If we want to say that most people have morally excellent characters, then we would go with character traits which are (i) global, (ii) tend to give rise to virtuous motivation, and (iii) tend to give rise to virtuous behaviour. In other words, we would say that most people have the traditional virtues. But as noted, that option is off the table on empirical grounds.
Another option is to go with widespread vice instead of widespread virtue. The idea is that instead of honesty and compassion, there exists rampant dishonesty and callousness. But while a few philosophers have recently shown sympathy for this view, it has yet to be worked out in detail as far as I am aware.
Doris himself floated the idea of local virtues as opposed to global ones in his 2002 book, and Harman seems to have endorsed it as well. These are traits like courage-on-playgrounds or temperance-on-vacation. The claim is that such local traits are widespread, although each person’s constellation presumably differs to some extent. Global traits like courage or temperance (period), on the other hand, are rare. Unfortunately, Doris only mentioned this possibility in a few paragraphs in his book, and it has never been developed subsequently.
Let me mention just one more option. One could say, on the basis of the experimental evidence, that most people do have global character traits, but they issue in motivation and behaviour which is neither virtuous nor vicious. Rather it is very much a “mixed bag”. In some situations it is admirable. In others, it is not. And the only difference between the two situations might be the presence of a moral reminder or a pleasant fragrance. These traits can be called “mixed traits” since they are neither good enough to qualify as virtues nor bad enough to qualify as vices. They occupy a middle space between the opposites of virtue and vice. This is the view which I think has the best story to tell about how most of us are put together, a story which I have developed in detail in two recent books. But so far I seem to be the only one who thinks this way.
The upshot here as well is that philosophers have paid little attention to developing a positive story about what is going on with our characters. When they have, there is no clear sense of what that story should be.
3. Virtue Development. So long as there is consensus about our general lack of virtue (which there is), and so long as there is consensus about the importance of being a virtuous person (which there also is), something should be said about how to address this state of affairs. Yet here too is another area of neglect on the part of philosophers.
To put the point differently, it seems apparent that there is a sizable character gap:
A Virtuous Character
The Character Gap
Our Actual Character
There are moral exemplars, people like Abraham Lincoln and Sojourner Truth, whose character is morally virtuous in many respects. Examining their lives ends up reflecting badly on most of us, myself included, since it illustrates in vivid terms just how much of a character gap there really is.
To try to at least reduce this gap, it would be helpful to have some strategies which can, if followed properly, enable individuals to make slow and gradual progress in the right direction. Philosophers needn’t be the only ones who can come up with these strategies, but advocates of virtue ethics in particular should have something to say that is practically relevant, empirically informed, and actually efficacious if carried out properly.
By and large, they haven’t had much to say. But there are signs that this is beginning to change. In fact, the development of character improvement strategies strikes me as an area that is primed to take off in the coming decade, and many good and innovative dissertations are there for interested graduate students to tackle.
Nancy Snow, Julia Annas, and Jonathan Webber are among the leading philosophers who are starting to contribute in this area. Here, by way of illustration and simplifying greatly, is one approach that Snow has discussed in several places. She notes that we often have goals, such as the goal to drink coffee, whose functioning can become automatic and habitual. This can also describe more virtuously relevant goals, like being a good parent. For goals like that, people can come to routinely perform virtuous actions which contribute to being a good parent, such as being more patient when playing games with their children. Given such a virtue relevant goal, Snow articulates how a parent can come to see patience as instrumentally valuable, and eventually as intrinsically valuable. The parent as a result might come to not just do patient acts, but do them from the actual virtue of patience. If Snow’s approach is promising, then it will be important to explore how people come to have virtue relevant goals, and what can be done to encourage them to recognise the intrinsic value of their habitual virtuous actions.
My hope is that this groundswell of interest in how to cultivate the virtues will continue to expand in the coming years. First, though, it would be nice to get a much clearer picture of what our actual character looks like, as well as address the conceptual parameters and empirical adequacy of neglected virtues like honesty and generosity. As I said, these are indeed early days in the philosophical study of character. And exciting days too, full of so many research possibilities.