Today I interview philosopher John Doris. Among other things, we discuss:
- If animals don’t “deserve” praise or blame for their actions, then why should humans?
- Does empirical research support the notion of “moral character”?
Download CPBD episode 063 with John Doris. Total time is 38:36.
- John Doris at Washington University in St. Louis
- John Doris at the Moral Psychology Research Group
- Doris, Lack of Character
- Doris, The Moral Psychology Handbook
Links for things we discussed:
- Strawson, “Freedom and Resentment“
- Manuel Vargas and Gary Watson
- Shaun Nichols
- Oliner, The Altruistic Personality: Rescuers of Jews in Nazi Europe
- Oskar Schindler
- Is-Ought Problem
- Jesse Prinz, Steve Stich, Joshua Knobe
Luke Muehlhauser: Dr. John Doris teaches philosophy at Washington University in St. Louis. He is the author of the 2002 book, “Lack of Character” as well as numerous scholarly articles. John, welcome to the show.
John Doris: Thank you. Thank you.
Luke: John, one of your papers draws attention to Strawson’s 1962 paper, “Freedom and Resentment.” How did Strawson try to change the agenda for moral philosophy and what is some of the work that has been done on those fronts?
John: Strawson’s paper is probably one of the more influential philosophy papers of any kind, but particularly on the subject of what people call free will for the past half-century or so. One of the reasons, if you’re me, for its enduring influences — although it’s kind of written in a disarming style and presented in this very commonsensical way — it’s actually an incredibly difficult paper, and it’s not entirely clear what Strawson was up to, despite there are some very good scholars like Manuel Vargas and Gary Watson who have tried to sort it out.
But the kernels of whatever exactly Strawson’s views were – the kernels of the main ideas were really influential.
The first thing in Strawson’s memorable phrase is that the whole worry about free will and determinism was based on what he called philosophers’ “panicky metaphysics.”
The panicky metaphysics are something like this: plausible to suppose that every event has a cause. Or if you like, is governed by deterministic causal law. Well, the events in your head that make you do stuff are events, so they too must be covered by deterministic causal laws. And therefore it’s hard to see how you could be free.
Another way that this is sometimes put is: you can’t change the past. That’s the fixity of the past. You can’t change the laws. That’s the fixity of the laws. Since the past is connected to the future in law-like ways, your behavior is fixed, as it were, or governed by the laws in such a way that it makes no more sense to talk about you being free than it does to talk about, say, your automobile engine being free, so there’s something deeply and mechanistic.
Now what the problem here, it sure seems like we make decisions. I can’t do this on the radio. But, what I usually say is, “I’m going to raise my right hand to make a philosophical point.” So instead, I’ll just say, “I chose to say I’m going to raise my right hand [laughter] to make a philosophical point”.
OK. There’s no question that we have feelings of agency, although it’s easy to see that doesn’t exhaust the issues. Because you know that sometimes you feel like you choose, and you know you really didn’t so on and so forth. There’s also a lot of empirical evidence to this effect.
Strawson’s way out of this mess, this conflict between this sort of plausible metaphysical picture and the appearances, was to basically say we need to attend to the practices. Others have tried — the so-called libertarians — to say human behavior is sometimes exempt from the causal laws. Maybe some actions are self-caused or some decisions are self-caused or something like that.
People found that either spooky or unintelligible. Not everybody, because there are libertarian philosophers. But Strawson proposed something much simpler than trying to figure out what an uncaused decision would look like, or a self-caused decision would look like.
He said, look, there is no question that we respond to one another with a range of emotions and attitudes: anger, admiration, and so on and so forth that seem to presuppose that we are responsible agents or — in the vocabulary that I don’t like — that we are free.
Right? Makes sense for you to be angry, if I carelessly… It makes sense for you – this is crucial – to be angry at me if I carelessly break your valuable antique. On the other hand, if you’re valuable antique is damaged in a windstorm, you might be angry, but it would be odd to subject the wind to the reactive attitude of blame. [laughter]
Similarly, if I maul you, I’m evil. I’m culpable. If a bear mauls you, it’s just doing what a bear does. Right? So Strawson supposed that we subject one another to these participant reactive attitudes.
He contrasted to that to what we call the objective reactive attitude, what he called the objective reactive attitudes, which are the attitudes we have toward children and the mentally ill.
There, we are not in a reciprocal relationship. Children and the seriously mentally ill we think of as objects of treatment and control. But we think of each other, normal healthy adults, as participants in certain kinds of relationships. How does this get us out of the loggerhead?
Strawson said, “Look, you can’t help but do this. You can’t help but do this. You are doomed to play the game, as it were. You can’t purge yourself of these attitudes.” What philosophers, for the past 50 years, many of us have thought was right, is the way to think about responsibility was to wonder whether someone is appropriately subject to the reactive attitudes.
So I’m wondering. Take a woman who kills her child in a post-partum depression psychosis or something like that. One way to think about “is she responsible for what she did” is what reactive attitude is appropriate. My instinct says something like pity. That makes me think that she was a victim and not an agent. Or a patient and not an agent.
When you’re thinking about whether you should be resentful towards your spouse for leaving her coffee cup in the living room for the 500,000 time, you may ask yourself, “What else does she have going on? Is she stressed out? Is she getting bullied by the boss at work?”
In thinking about which reactive attitude is appropriate, it seems to draw our attention to factors that are implicated in our actions that are relevant to how we feel about one another.
Now just to quickly end in not-too-long a story. Strawson seemed to think that that was the end of the story. Many people like me think it’s not just enough to say “we have the reactive attitude, deal with it.” We need a story about whether that practice makes sense. So Strawson is sometimes right — this is where the essay gets obscure. It’s not wanting to ask those further questions, being what philosophers call a “quietist.”
But whatever you think about that — his insight that what we think about is not whether things were caused or not, but how they were caused, in thinking about responsibility — is the way to get past this kind of block in the road presented by causal determinism.
Other philosophers have tried to do this. He sort of gave us the tangible and concrete way to think about how agents living in a world of causes could still be responsible.
Luke: I think a very intuitive way to view this situation is that people respond to our emotional reactions. So if we respond with disgust at a particular option, then a healthy adult may be less likely to perform that action in the future. Whereas it may not have such an effect on a young child or on someone who is seriously mentally ill. Or a grizzly bear.
John: Excellent. So two things about that. One thing that isn’t noticed often enough in talking about Strawson is how deeply social this process is, that it takes two or more to play. And so that suggests that the agency of an individual — if there is such a thing — is deeply social.
It also suggests ways in which we help each other to be agents, because we can engage in this kind of discourse. So your disgust when you see me sneaking a smoke after I said I quit, your response to me imposes a cost on smoking that helps me do what I want to do. Which is, if I’m like 80% of smokers, I want to quit.
So that’s the kind of story is, not only does a Strawsonian story suggest that we treat each other as agents, but it also suggests a story about how we help each other to become agents. Or this what Shaun Nicholls and I call “collaborativism” about agency and rationality.
So I think that is a very important bit. And we don’t, in fact, enter into collaborative relationships with grizzly bears, which is one of the things of course that makes them wonderful animals and different than dogs and cats. But it does suggest that they’re at least not human agents, because they don’t enter into the kind of co-regulative discourse that people do.
Luke: Now does this kind of realization about the effects of our reactions to each other give any legitimacy to some of our reactions? For example, it doesn’t make sense to respond to a windstorm with resentment. But maybe, because people will react to us, it makes sense for us to respond to them with resentment, even though they’re just as caused as the windstorm.
John: Yes. Notice you could go two ways here, right? You could say it makes sense, but only for forward-looking reasons. So it makes sense for me to be angry at you, because then you won’t do it again. But someone could say, “That’s true, ” and that’s what utilitarians have famously said about punishment. But someone could say, “Look, nobody’s a responsible agent, so I don’t deserve your anger, but it’s useful.” So you could be kind of Strawsonian, but the only purpose of the reactive attitudes was to regulate future behavior.
I actually think that our reactive attitudes are in some cases appropriately grounded in desert. So you yell at your dog, or do whatever is — if you’re fortunate enough to have a dog that does anything you say. You do this top regulate his future behavior. You don’t think it deserves being yelled at. It’s just doing its doggy thing. It doesn’t have a moral failing.
So you could think the reactive attitudes were more important, but not think that anyone has free will. I think the reactive attitudes are important, and thinking in the right way about that importance enables us to see how human beings can in fact function as agents.
Luke: Now why is it that the dog is just doing its doggy thing and it doesn’t deserve your scorn though it might be useful to the dog to regulate its behavior, but you wouldn’t apparently say exactly the same thing about me if I hurt you. You wouldn’t just say “Oh, that’s Luke. He’s just doing his human-y thing and he doesn’t deserve my scorn.”
John: The short answer is maybe there’s not a difference. And then the long answer is the subject of many books. Why think that the natural system that are human beings are appropriately subject to these attitudes, and the natural system that are dogs are not? And there might be no good answer. I think there’s a good answer, because I think that human beings direct the course of their lives in ways that animals do not. And so it makes sense to think of humans authoring their lives in a way that it doesn’t make sense to think of a dog doing it.
So just take a simple, simple way of sort of getting at this linguistically. We say something like “I admire him. He made something of himself.” Now it would be a little bit odd to say “I’ve got a great dog. He made something of himself.”
Fewer of us who admit to being analytic philosophers who think that words solve things, or reflecting on words solve deep problems. But I think in that case, the grammar as it were, looks pretty deep. That human beings are to a certain extent — although it’s something that’s necessarily done as inhabitants of cultures and groups — human beings can be self-formative in ways that other animals cannot. Although of course we don’t really know in full detail what goes on in animals’ heads.
But it looks like human beings can be self-formative in a way that other animals cannot. And some of the conduct and judgment that is expressions of those formations is quite appropriately praise- or blameworthy, and not just because it will affect future behavior that we can’t, but because the person so-formed deserves credit or blame.
Luke: Well, we humans are extraordinarily self-aware and we do think about the possible things that we could do, and deliberate about them, and end up at a decision to do one thing rather than the other. Those may be the kinds of differences between us and a dog that you’re considering?
John: I don’t want to go there, because there’s this kind of large philosophical literature — which is what I’m working on in my current stuff. I’m writing a book called “The Natural History of the Self,” which is just about these issues we’ve been talking about, how to think about agency. And I think there’s lots of times we do things that we don’t — I think we don’t deliberate and reflect very much. It’s obviously conspicuous. But I also think there’s lots of things we do that aren’t deliberative and reflective that are appropriately the subject of responsibility attributions.
So take any kind of case of habit. Take athletic achievement. It’s natural to credit LeBron James for what he does. Now you might not think that, but we do do that. It doesn’t seem crazy. And a lot of the things that he does that are so great are precisely non-deliberative.
Indeed, the difference between ordinary athletes like me and great athletes like LeBron James, one of the things is probably that he deliberates less and in fact he is more effortlessly intuitive.
Luke: So where would you locate the difference in the way that human beings act versus the way that dogs act?
John: The key thing about human beings in their self-direction, is that human beings’ self-direction is sourced in their values. OK, and now good question: what is a value? But at the least let’s say it’s either a special type of desire or a relative of desire or a relative of an attitude. So I take it my kitty cat does what she desires, And sometimes I do what I value, despite what I desire. Despite some of my desires. So for me that’s supposed to be the difference, that human beings — not always, but often — have behavior that is sourced in evaluative structures in ways that, so far as I can tell… At least about animals, and you know insofar as people I believe or most animals… Interesting question about what to think about fancy, big-brained animals like elephants and dolphins and gorillas.
But leaving that side, that behavior can be evaluatively structured for human animals seems to me an important difference. And it’s a difference that’s natural enough to locate on the dimension of agency.
Compare another kind of case — not with two different kinds of human animal. Take someone who has obsessive-compulsive disorder or something like that. So they have to go back and check and see if their toaster is off 57 times or something like that.
What’s natural to suppose with various types of psychopathologies or mental illnesses is that their behavior is not self-directed. Indeed, that’s one of the things you might think is, intuitively at least, bad about being mentally ill.
In what you might suppose about the addict at least, or the obsessive compulsive, is the forces that decisively move them to act — maybe these are motives or cravings or maybe there’s some deeper level brain story we should be telling — are counter to their values.
So that the decision to smoke, as it were, doesn’t express who they are but undermines who they are.
Luke: Right. So things like shame or resentment or praise, there might be some people who deserve those things, or a lot of people who deserve those things, because they are self-directed in this way, where they might be able to choose certain values over their immediate desires.
John: Right. Yes. It’s definitely too long a story to talk about here, but that’s at least where I’d look.
Luke: Sure. Your book, “Lack of Character,” argues that social psychology has raised some tough problems for the way that we typically think about moral personality and character. What do you think are the common assumptions that are called into question by this research?
John: There’s cultural and individual variation, but there’s pretty good evidence in the so-called “attribution theory” literature to the effect that especially in – let’s just say – American culture, we expect that people have robust characters that generate consistent behaviors. So that we think some people are honest and behave honestly, whether it’s easy to do so or not. A bit more fancifully, honest people behave honestly, whether the situation they’re in is conducive to honesty or not.
In fact, it turns out that people are wildly sensitive to situational variation in their behavior. And this seems to call into question the notion that someone’s character is sturdy or robust or something like that, in the way that the story I just told about honesty seems to assume.
Luke: What are some of the examples of studies that have shown that we are extremely sensitive to the situation?
John: Yeah. People might be ruder when they read rude words. They might be more likely to avow collectivist or communal values if they read plural pronouns than if they read singular pronouns. They may be more likely to help another person if they found a dime, or if they are in front of a good-smelling coffee shop. Or maybe even they may be more likely to help if they are holding a warm cup of coffee, as opposed to a cool iced tea. They might be more likely to behave badly in the dark than in the light, and so on and so forth.
So, many times, peoples’ behavior seems to be caused by factors that they would not reference as a justification. You might say something like this, “Well, why did you take your coat off?”
“Because I got too hot.” And there, what caused you to behave also you can point to as a justification of your action.
But, if you say, “Why didn’t you help?” and I say, “There was a lawnmower running loudly nearby,” it looks like it’s a causal story. At least the experiment suggests it is. But, if I were mad at you for not helping, it wouldn’t do to say, “I was holding cold iced tea and not hot coffee”.
Luke: [laughs] Right. Some of these swings in behavior in the research are pretty extreme. In one case, 80% of the people will do the helping thing. Then, you change the situation and 80% will not help, or something like that. Right?
John: Sometimes, there are big swings. Sometimes there are significant swings that aren’t as big. It varies with the experiment. It’s psychology, so it’s always “for the most part.” But, yeah. Some of the effect sizes seem to be quite large. You don’t need fancy statistics to see that something is going on. Other times, they are smaller, but they’re there. Sometimes things that you would expect to matter in situations don’t matter so much.
So it’s very messy and very imperfectly understood, but it sure seems like surprisingly small differences in the situation can at least sometimes make a big difference in how people behave. And if that’s true, it looks like it would be kind of a mystery on the supposition that people have robust characters.
Luke: Now, do you think that maybe some of the reason we like to think that people have robust characters is because it’s easier for us? It lessens the calculation. We can say, “Oh, John, he’s honest. I can trust him to say that”.
John: Yeah. I think often enough that works pretty well, actually. I think you are right. You might think of these kinds of personality attributions as what psychologists call “heuristics.” So, “He’s a good guy” is shorthand for, “I expect these behaviors.” And often enough, it works. So why would that work? Because you might say, “Well, if you are as mixed up as you say we are, Doris, then why would that work so well?”
Take an example. You go down to the corner bar, and maybe the bartender isn’t friendly every time or grumpy every time, depending on the type of persona he’s cultivating to maximize his income or minimize his hassles.
But you know what? You know what to expect from him. But notice it’s not just the person that is constant in that case. It’s the situation. You probably don’t see him in a lot of other situations.
One possibility is that the heuristic works because we see people in a limited number of situations. And because the heuristics work, the belief that it’s the person and not the situation is able to persist.
Luke: Yeah. Now there are some researchers who might say that you’ve gone too far in dismissing the robustness of character.
John: Many. Many researchers. Yeah.
Luke: Many researchers. Some people point to the study of Samuel and Pearl Oliner, about rescuers during the Holocaust. What was that research? What did they have to say?
John: Well, they set out to see if there were individual differences that could help to explain why some people served as — or behaved as, since they weren’t working as an institution usually — as rescuers during the Second World War. Which has a particular meaning in this context, which is they helped to shelter Jews from the Nazis not for the sake of personal gain, like they were paid or something, but just because… At least not for the sake of personal gain. Presumably because they thought it was the right thing to do. So if you took these kinds of risks with no thought of personal gain, you end up counting as a rescuer.
Now, a tiny, tiny, tiny fraction of the population in the occupied territories did this. So since it’s such a small number of people, it must have to do somewhat with the people, their upbringing. Sometimes it’s suggested that they have a more inclusive worldview and so on and so forth.
I think there may well be personal factors at play in these extreme behaviors, but it’s also true that there are situational factors in play. Some nations who have less of a history of anti-Semitism — Denmark, I believe, comes to mind — did more to shelter the Jews than nations like Poland, where they had a longer history of more virulent anti-Semitism.
Now, I’m not an expert on the history, but you could imagine how those kinds of factors might be at play: sort of broader, social, contextual factors and not just individual differences. Some situations are conducive to helping and some not.
And I think what the literature would show is that even for these great heroes, there’s some of each. Which is probably true for most any behavior, right? Everybody is supposed to be an interactionist now, both circumstance and individual differences count, and I think that too.
So I don’t think it serves as a counterexample. One, because it’s a tiny, tiny fragment of a population. And I’m not denying that no one is consistent. But also because when you look closely at that literature, if you look at that and the other books on the rescuers, you not only find that situational factors made a difference for whether people helped or not, you find amongst the helping individuals who helped quite morally-variable behavior – characters that appeared not unified on a moral dimension, but fragmented.
Of course, Oskar Schindler being the famous example. One of the most heroic and effective of the rescuers, but yet in other ways maybe not the most morally upstanding character.
Luke: John, let’s say you’re right about your descriptive claims about human psychology and that we’re greatly influenced….
John: I’m happy to accept that stipulation.
Luke: [laughs] Let’s just assume, shall we? And now what would these descriptive claims about human psychology have to say about prescriptive ethics or normative ethics, do you think?
John: Yes. As you know, this was arguably the sort of animating insight of ethics in philosophical ethics in the 20th century, that there seems to be a difference between prescriptive and descriptive statements, or “ought” statements and “is” statements. So it’s not clear that anything follows from the fact that drinking and smoking are implicated in cancer, undeniable descriptive claims, and what any individual should do. Perhaps you’re here for a good time and not for a long time. In which case, the dangers of smoking and alcohol shouldn’t move you — assuming you find drinking and smoking fun, as many of us do.
So there looks to be this kind of inferential gap that, as you know, Hume is at least alleged to have thought that “is” does not imply “ought.” At the same time, one of the things that became clear during the wane of the 20th century is that descriptive statements and prescriptive statements are not easily disentangled.
So supposing I say, “You’ve got a healthy baby.” Presumably I’m making a kind of an evaluation that the baby is in good shape. But presumably you can’t make that evaluation for just any baby. And so a sick baby or a dead baby, for all that, you don’t have a healthy baby. So it looks like there’s often a lot of leakage between descriptive and prescriptive statements.
Another nice place to look at this is in sort of thinking about psychopathology. We say someone is depressed, we’re saying we can expect these kinds of behaviors, but most of us also understand that as saying they’re in a condition we should want to avoid.
Now there might be some people who like being in a clinical depression — although if you’ve ever been in a clinical depression you might doubt that. So it’s pretty tempting to say that facts may nudge evaluations in one direction or another. And of course evaluations nudge our conceptions of the facts.
So, for example, we have standards of evidence that are normative, you should believe something when there’s empirical evidence for it, say. And that affects the way you describe the world, certain kinds of prescriptive commitments. OK, so it looks like it’s a two-way street.
And then what I suppose is it’s natural in thinking about character, you’re saying people who are described this and that way, as honest or courageous, should be responded to this and that way, with admiration. So if it turned out that no one had these character traits, you might wonder about certain practices of admiration or approbation associated with character traits.
And if it turned out further that it was inordinately difficult to inculcate these character traits — maybe you can, but only with sort of like sending everyone to boot camp — you might wonder about what kind of practices of moral education you should have. What is it we should be trying to do in moral education?
So it seems like there’s lots of “shoulds” in the area of normative ethics — what we should aspire to, how we should respond to one another — that are disciplined by descriptive claims. And that you know disciplined is a willfully vague notion. The notion is not straightforwardly inferential. The process is probably more along the lines of what Rawls imagined with reflective equilibrium or something like that.
But the fact that “is” does not imply “ought,” does not – without further premises anyway – imply that evaluative inquiry can go on in complete isolation from normative inquiry. And in fact, unsurprisingly, it never has. Everywhere you look in the moral philosophy literature, people make descriptive claims.
So my modest suggestion — along with my colleagues like Jesse Prinz and Steve Stich and Joshua Knobe and Shaun Nichols. Our modest suggestion is let’s get the facts right. If we’re going to make factual claims when doing ethics, and guess what, everybody does, let’s get the factual claims right. And many of the interesting factual claims are about human psychology.
Luke: Well, do you think that these findings in human psychology about the situational effects upon our behavior, do they make certain normative theories less plausible than they may have been if the more traditional view of robust characters was upheld by the data?
John: Well, supposing there’s empirical worries. Now we know that here’s a very popular approach to normative ethical theory called virtue ethics, which at least plausibly is thought to presuppose the problematic nature of character. Of course, this is contentious, not very virtue ethicist thinks that it does. But supposing it does, one thing you might have thought is that one of the kinds of things that was cool about virtue ethics is that it was psychologically realistic. It was sort of right about what the psychologies that we admire looks like.
Now if it turns out not to be right about what actual human psychologies look like, it’s lost a competitive advantage.
Luke: Right. And do you think there are particular normative theories that the descriptive data fits with best?
John: That’s a further question where the devil is in the details. So I don’t advocate a particular normative theory. But if you think of something like a contractualist theory that emphasizes human beings co-planning and negotiating and muddling along together to figure out what’s right and figure out how to act according to it, that kind of model seems to me to fit better the kind of animals we are then the kind of moral theory which emphasize individual character or sovereign artifices or so on and so forth. And this all said without tagging particular thinkers with particular views. This is not the place, I take it, to play “pin the tail on the philosopher.” Since you can be sure that if I’m saying that something is a mistake, no one will have said they said it.
John: I’m no different, right? “No, that’s not quite what I mean,” right?
John: “Is that false? Then I disagree.”
Luke: [laughs] “That’s not what I really meant.”
John: Yes, exactly.
Luke: Well, I can certainly agree with you that more research needs to be done, and I look forward to what you find next. John, it’s been a pleasure speaking with you. Thanks for coming on the show.
John: Thanks for having me.
Previous post: 6 Questions with which to Stump Conservative Christians
Next post: Sean Carroll on Stephen Hawking’s New Book