:: Article

Questioning willusionism

Eddy Nahmias interviewed by Richard Marshall.

eddynahmias

 
Eddy Nahmias is cool because he doesn’t know it. He’s an anti-willussionist. He stands amongst the smoke of his burnt out armchair on the same land as Josh Knobe’s but has another one nearby. He’s writing a book Rediscovering Free Will. He co-edited another one Moral Psychology. He won a superior honors teaching award from the Florida State Honors program, which shows he’s pedagogically groovy.
 
3:AM: When did you start being interested in philosophy? Were you always interested in brooding on things like freewill and the deep questions or was it something that just kind of happened?
 
Eddy Nahmias: I have a bad memory, so my “narrative self” is partly confabulation. But the way I like to tell the story is that I was always interested in big questions, including why humans are the way they are (and think the way they think). For a while in my teens I thought natural selection could explain just about everything about these questions and I was always coming up with ‘just so’ stories I also loved good literature. I ended up majoring in English at Emory (writing a thesis on T.S. Eliot and on Toni Morrison). I thought that it would be the subject that allowed me the widest latitude for synthesizing from every field, but I kept finding the philosophical questions in literature the most interesting, and I came to believe (mostly correctly) that it’s philosophy that allows the most breadth and synthesis. So, I migrated to philosophy and also began studying sciences relevant to understanding the mind, mainly psychology, neuroscience, and also primatology (I still like to think about how human minds evolved in comparison to our nearest relatives). It’s funny, but I can’t remember what exactly got me started on the free will problem, but I suspect I was worried about what I think most folks worry most about when it comes to free will – that science might explain everything in such a way that my conscious thinking – my self – would somehow be illusory or unimportant, and we would not be morally responsible for what we do.  
 
3:AM: You’re thinking about free will and you argue that we need to be careful about what we think free will is and what it entails. To some, determinism is the opposite of free will, and it seems to be a bad thing. Determinism seems to imply the end of responsibility and stops us from being able to make our own choices. But you think that folk don’t always think determinism is a bad thing. You say they make a distinction between determinism and reductionism, epiphenomenalism and/or fatalism, which people think is threatening, and determinism that doesn’t imply these things. So can you say what your evidence is for saying that people don’t always think determinism is a bad thing?
 
EN: As you say, determinism is often presented as the opposite of free will (if that’s what ‘determinism’ meant, it’d be silly to debate whether it is compatible with free will). But people understand ‘determinism’ in many ways, and it’s not always clear how it is meant to threaten free will. In my dissertation I used a metaphor of a many-headed monster – if we can distinguish, and take on, the various heads one by one, we can see more clearly what the threats are supposed to be and how they might each be confronted (hopefully, it is not a hydra that will grow back two heads for each we cut off). We also learn more about free will and responsibility by seeing more clearly what exactly it contrasts with- what we are free from (hint: it does not really make sense to say we are free from determinism).
 
Determinism is sometimes presented to mean that the past and laws control us or that our actions are pre-determined, as if someone planned them. But it should not be anthropomorphized in these ways. The Big Bang did not plan our lives, and it didn’t really cause what we do in any useful sense of ‘cause’. Determinism should also not be confused with fatalism, the idea that certain things (like your actions, or Oedipus’ sleeping with his mom) will happen no matter what – that is, no matter what you want or try to do (or no matter what Oedipus tries to do to avoid his fate). Quite the opposite – determinism suggests that what happens in the future depends on what happens in the past and what we do in the present. Finally, determinism should not be confused with what I call bypassing – the idea that our conscious mental activity is not causally relevant to our decisions and actions. Determinism does not mean that our minds don’t matter.
 
So, what does determinism mean? In the philosophical debates, it is a specific thesis about the relations between events or states in the universe, as governed by laws of nature. It says that, holding fixed the laws of nature and the state of the universe at one time (e.g., the distant past), there is only one possible state of the universe at any other time (e.g., the future). As such, determinism does not mean that reductionism is true – i.e., that everything can be explained in terms of low-level physical events (such as interactions between quarks or neurons) – nor does it mean that epiphenomenalism is true – i.e., that our (conscious) mental states play no causal role in what we do. Some people worry that reductionism entails epiphenomenalism, and most people (rightly) worry that epiphenomenalism would threaten free will. But these are different heads of the monster, distinct from the (now-shrinking) head of determinism.  
 
If determinism, properly construed, threatens free will, it is because it either rules out a specific type of ability to do otherwise (holding fixed everything, including our mental states) that some philosophers think is required for free will, or it is because it means that there are ultimately conditions (e.g., in the distant past), over which we have no control, that are sufficient for our choices and actions. Responding to these two threats is serious business, and compatibilists (i.e., those who think determinism is irrelevant to free will) work hard to show that the sort of choice or control we need to be free and responsible does not require an unconditional ability to do otherwise (holding fixed everything) nor an ability to make choices that are somehow untethered from prior events.  
 
I can’t rehearse these arguments here, but here’s just one quick thing to consider. Do we think that the truth of determinism would make it false to say, “That leaf [that just landed there] could have landed somewhere else”? I don’t think so. I think we think the leaf could have landed elsewhere, and we probably think that if it had, it would have been because something had been slightly different (e.g., strength of wind, time it broke off of tree, etc.). And those earlier differences were possible too. Determinism is entirely compatible with this analysis of ‘could have happened otherwise.’ It doesn’t make everything inevitable. Now, do we mean something different in kind (metaphysically different) when we say, “That student could have chosen a different paper topic”? Do we commit ourselves to the universe being indeterministic when we say or think such things about human choices? I doubt it. We do think that our choices involve lots of complex factors and capacities that tree leaves (and dogs and babies) don’t have. I’m doing some experimental philosophy studies to try to show that ordinary people aren’t committed to an unconditional sense of choice, nor to anything as metaphysically fancy as the libertarian theory of agent causation. If they aren’t, that doesn’t prove libertarians are wrong about free will – but it means they have to explain why free will demands more than our ordinary thinking, intuitions, or practices suggest.
 
I think my previous x-phi work already suggests that the traditional threat posed by determinism is not the one that ordinary people are worried about. And the threats from determinism to our ability to do otherwise or to be ‘ultimate sources’ are also not ones that are bolstered by the scientific discoveries that some have taken to threaten free will (see below). What people worry about are the other heads of the monster – some of which are suggested by – though not established by – these scientific discoveries.
 
3:AM You argue that people think that determinism doesn’t imply the end of free will and you agree with them. You argue that this intuition of the folk tracks reality and that the evidence for this is found in the study of the mind. This is the compatibilist view, that determinism is compatible with free will, or vice versa. In fact you have some pretty groovy experiments that suggest that what people think about determinism and free will, and whether people will hold people responsible or not for their actions, depends on what scenarios the issue is part of. Can you tell us about these findings and what conclusions we should draw from them?
 
EN: If incompatibilism is right, then our having free will makes certain demands on the universe – it’d have to be indeterministic and, given what most incompatibilists say, it’d also have to be such that the indeterministic “gaps” happen at the right time and place in the brain and perhaps also that we have particular causal powers that can “fill these gaps” to cause one intention rather than another (e.g., “agent causation”). The more you demand from free will, the more possibilities there are for us to lack it. And that’s fine: I don’t want to preserve free will only by watering it down. But these demanding conditions need to be motivated, especially if we assume, as I think we should, that free will marks off the control conditions to be morally responsible – to deserve praise and blame, reward and punishment for actions. What might motivate such conditions? One possibility often suggested by incompatibilists is that they are motivated by our ordinary understanding of free will and responsibility. For instance, Bob Kane, who develops one of the most elegant libertarian theories of free will, says, “If compatibilism is to be taken seriously by ordinary persons, they have to be talked out of this natural belief in the incompatibility of free will and determinism by means of philosophical arguments.” Peter van Inwagen, the bane of modern compatibilists, asserts that “It is almost impossible to get beginning students of philosophy to take seriously the idea that there could be such a thing as free will in a deterministic universe.” And Galen Strawson, who develops a powerful argument for skepticism about free will and responsibility, targets conceptions that he thinks are demanded by ordinary thinking: “In our nature to take determinism to pose a serious problem for our notions of responsibility and freedom.”
 
Some former students and I have tested whether these armchair assertions about ordinary intuitions are true and also what factors might drive people’s intuitions about free will. So, there are two different descriptive projects here: (1) uncovering the patterns of folk judgments about free will, responsibility, blame, control, etc.; and (2) exploring the psychological sources of such judgments. This information is useful for the normative project of developing a plausible theory of free will and responsibility. For instance, information about (1) can help establish when a theory is consistent with our pre-philosophical intuitions about free will and practices of praise and blame, or when a theory revises our ordinary concepts and practices. And information about (2) can help us discover if people’s intuitions provide reliable or unreliable information about free will and responsibility. Finally, this descriptive information, in combination with the normative project, can inform the prescriptive project of how we might need to revise people’s beliefs about free will or our practices of assigning responsibility and whether and when people deserve credit or blame. (My collaborators, who deserve credit for the good aspects of this research and who do not always share the normative and prescriptive conclusions I draw from it, have included Thomas Nadelhoffer, Jason Turner, Steve Morris, Trevor Kvaran, Justin Coates, and Dylan Murray. Shaun Nichols deserves credit for the names of these three projects, and Josh Knobe for helping us start our experimental philosophy projects way back in 2002, though he and Shaun have a different view than I do about the descriptive facts.)
 
So, here’s one scenario we’ve used to present determinism:

Imagine there is a universe that is re-created over and over again, starting from the exact same initial conditions and with all the same laws of nature. In this universe the same initial conditions and the same laws of nature cause the exact same events for the entire history of the universe, so that every single time the universe is re-created, everything must happen the exact same way.  For instance, in this universe a person named Jill decides to steal a necklace at a particular time (time T) and then steals it, and every time the universe is re-created, Jill decides to steal the necklace at time T and then steals it.

Among the participants that correctly answer comprehension questions about the scenario, 82% judge that Jill has free will, is fully morally responsible for her act, and deserves to be blamed for it. We get similar results when we describe determinism is different ways (e.g., a supercomputer can predict, before people are born, what they will do at a particular time) and when we ask about other types of actions (e.g., saving a child or going jogging). On the face of it, these results suggest that the armchair speculations of incompatibilists are mistaken.
 
Now, what they might mean (despite the way they put it in the quotations above) is that you can get people to see the problems posed by determinism with some explanation or argument. Perhaps. But I suspect that either those explanations or arguments are presenting determinism in the guise of more threatening heads of the monster (e.g., fatalism or manipulation), which happens a lot in introductions to the “problem of determinism,” or the arguments are not so commonsensical after all. For instance, van Inwagen’s Consequence Argument and Strawson’s Basic Argument both include a crucial principle that says, essentially: in order to have a choice about X, or to be responsible for X, one has to have a choice about or be responsible for the conditions that entail (or bring about) X. That somewhat complicated principle may seem plausible enough when applied to specific cases, but it is controversial as a universal truth. And its truth strongly suggests (given other plausible premises) that no one can have a choice or can be responsible for anything at all. So, weighed against that counterintuitive conclusion, perhaps the principle is not so plausible after all.  
 
Three more quick conclusions I think these x-phi studies suggest (but let me emphasize it is very early in the game). First, across a variety of descriptions of determinism, almost all of those participants (the minority) who say that agents lack free will and responsibility also respond that the agents’ mental states have no effect on what they end up doing. That is, they read the deterministic scenario to imply that our beliefs, desires, and decisions are bypassed such that they play no role in what we do. That is a mistake. Determinism entails no such thing. Dylan Murray and I suggest that this offers an error theory for incompatibilist intuitions. These intuitions arise from unreliable sources. Among participants who recognize that determinism does not mean that our mental states don’t matter, the vast majority also say that agents can be free and responsible in deterministic worlds.
 
Second, these are participants who have also understood that the scenario does entail that “given everything that happened before time T, Jill has to decide to steal the necklace at time T.” Yet, most of them do not seem committed to the unconditional sense of choice required by some forms of incompatibilism: 75% also judge that she has a choice about what she does. As I suggested earlier, it looks like people do not typically assume that a person’s having a choice or having been able to do otherwise is metaphysically different than a leaf’s having been able to land elsewhere or a dog’s having been able to catch a ball that he dropped. Of course, human choice is very different than these simpler events, but it is different primarily because of the complexity of the processes involved: human choices involve conscious deliberation and reasoning.
 
This leads to the third point: in these and other studies it looks like people are most concerned about a reductionistic or mechanistic picture of human decision-making that undermines the complexity and causal relevance of conscious deliberation and reasoning. In studies I did with Trevor Kvaran and Justin Coates, we found that determinism, when described as ‘working through’ “thoughts, plans, and desires in the agent’s mind,” is not threatening, but determinism, when described as ‘working through’ “chemical reactions and neural processes in the agent’s brain” led people to say agents could not be free or responsible. Again, the worry is not that there are sufficient conditions in the past for all of our actions (i.e., determinism), but that there are conditions that bring about our actions while bypassing what we identify with our selves (our mental, including conscious, activity). If people can see how their minds matter, even if their minds are part of the physical (perhaps deterministic) world, then they are unlikely to think free will is an illusion. So, I predict that a good naturalistic theory of the mind, including the conscious mind, will largely dissolve the free will problem… that is, as long as that theory shows that the conscious mind plays the right sort of role in our choices and actions.
 
3:AM: You discuss the consequences of misinterpreting results in modern mind sciences as implying that free will is really an illusion, (‘willusionism’). So psychologists like Jonathan Bargh say that, “The phenomenological feeling of free will is very real [...] but this strong feeling is an illusion,” and John-Dylan Haynes says, “[t]here’s not very much space for operation of free will. The outcome of a decision is shaped very strongly by brain activity much earlier than the point in time when you feel to be making a decision.” Josh Greene and Jonathan Cohen say, “The net effect of this influx of scientific information will be a rejection of free will as it is ordinarily conceived with important ramifications for the law”. You don’t just think that the idea that free will is an illusion is not right, but you also argue that if you tell people that free will is an illusion bad things happen and that doing so is unjustified. Can you explain this? And if it turned out to be an illusion after all, would you still think it a bad idea to tell people? (Not that you would have any choice about that!) So this question is partly about whether you think science research should take account of possible ethical consequences resulting from discoveries?
 
EN: Since I went on too long on the last question, I’ll try to answer briefly and just point interested readers towards my New York Times article on this topic and your interview with Al Mele, which also covers this territory. Two preliminary points: (1) I am a big advocate of considering scientific research relevant to free will, including research that threatens it (e.g., I think research in social psychology on unrecognized factors that influence our thinking and behavior suggests that we have less free will than we think and suggests we should tone down how much we hold people responsible for their situation in life and their actions). I am a “cagey compatibilist” because unlike other compatibilists, I think that showing that determinism is not a problem for free will does not guarantee that we humans have free will; we still need to carefully consider other (distinct) challenges – the monster has more than one head. And (2) I am not an advocate of hiding the facts from people, even if knowledge of the facts might have some bad results. Despite what one willusionist once told me, I do not “have an agenda” (though, on his view, my unconscious processes might have an agenda and my conscious self can do nothing about it!).  
 
Free will matters. Our views about free will influence our self-conception and our moral and legal practices. (If someone means to say that something they call ‘free will’ is an illusion but it’s not anything that matters in these ways, then all I can say is: Who cares? But also, Stop it, since people will interpret your claim in terms of their understanding of free will, which does matter). So, we need to make sure that scientists who are the most prominent, and well-publicized, advocates of the position that free will is an illusion (“willusionists”) are justified in their claims.  
 
Whether they are justified depends on three things: whether their target conception of free will is a reasonable one to use, whether their target conception matches what people think free will is, and which, if any, of these conceptions of free will the scientific evidence plausibly challenges. Most of them use unreasonable definitions of free will, ones that require supernatural or magical powers. For instance, willusionist Jerry Coyne says, “Free will is, I believe, an illusion that we have that we can somehow affect the workings of our brain and free them from the laws of physics.” If that’s how you define free will, then we don’t need science to show us that it’s implausible.  
 
Why would scientists use such non-scientific definitions of free will? Because, from the armchair, that’s what they think ordinary people think. I think they are wrong, but I also think we can try to sort this dispute out empirically (scientists should appreciate that). And I think my evidence (plus some that Mele discussed with you) suggests that most people do not think free will is ruled out by plausible forms of naturalism.
 
If we start with a supernatural conception of free will, then scientific evidence that challenges the existence of a supernatural soul or a non-physical mind would indeed challenge free will, and the body of scientific evidence certainly offers inductive evidence for such naturalism (that’s one reason I’m a naturalist). But in that case, the specific evidence discussed in these contexts, such as Benjamin Libet’s EEG study or John-Dylan Hayne’s fMRI research, is not particularly useful. Rather, this sort of evidence is relevant if the goal is to show that the mental processes involved in conscious reasoning and decision-making play no role in action – they are bypassed, for instance, by non-conscious processes. If we allow that the conscious mental processes are themselves instantiated in neural processes (i.e., naturalism), then the question is whether those neural processes are bypassed. To make a very long story short, I do not think the existing evidence shows that those neural processes that instantiate conscious mental states are bypassed. They might play a less significant role than we think, especially in the milliseconds before action, but they have not been shown to be irrelevant when, for instance, we deliberate long and hard about tough decisions like what job to take or we make complicated plans for future action like how to accomplish six tasks before lunch.  
 
So, either (a) the willusionists define free will poorly, in which case their claims might lead people to think they lack what they think free will is – and that may have bad results – plus the evidence discussed would not be particularly relevant, or (b) on a more plausible view of free will, the evidence does not show we lack free will entirely, though again, it might show we have less free will – and may be less praiseworthy and blameworthy – than we tend to think.
 

moralpsychology

3:AM: You do have kind of armchair moments of philosophy too. So one of the subtle arguments you put forward to defend a compatibilist viewpoint is the argument that considers two conditions that seem to be required in cases of free will: the deliberation and control condition on the one hand and the alternative possibilities condition on the other. This is an argument that seems to get to the heart of many of the conundrums of free will and determinism. Could you say how you basically run the argument and how thinking about these conditions helps us to see how we might yet be compatibilists?
 
EN: Wow, it sounds like you are one of the very few people who have read my paper, ‘Close Calls and the Confident Agent’! I certainly have my armchair moments – armchairs should not be avoided – but most of mine include some attempt to consider how our ordinary experiences and intuitions map onto philosophical debates. In this case, I was trying to diagnose why libertarians think the unconditional ability to do otherwise (holding fixed everything) is an important component of free will, and to try to put pressure on this claim. As many have noted, it is not attractive to have the ability to do otherwise, holding fixed your reasons and desires, if those reasons and desires clearly point towards one course of action – if you are sure you want to choose job A, why would you want the ability not to choose job A (i.e., the possibility in those exact circumstances of not choosing A)? The unconditional ability only seems attractive or useful in “close calls,” such that, at the moment of choice, you feel conflicted about which of two actions would be best (or least bad).  
 
So, I suggest we imagine a Confident Agent, one who feels conflicted about what to do when considering future courses of action, just as we do, but not after she has deliberated about it. At the moment of choice, she always feels confident about what to do. There’s more to the argument, but the basic idea is to say (1) this thought experiment is coherent, (2) there are no good, non-question-begging reasons to say that the Confident Agent lacks free will, and more so, (3) she seems to lack a certain constraint on freedom that we sometimes feel – that is, not knowing what would be best to choose when we need to make a choice. So, these points suggest that the ability to do otherwise is not necessary for free will. Perhaps the libertarian wants us to have a power to do otherwise even when we have no reason to want to exercise that power and even if we never in fact exercised it, but that’s a strange power to want. More plausibly, we want to have the power to recognize various future options for action (determinism does not rule that out), to consider reasons for each of them, to come to recognize which reasons are best (while the process of deliberation is active, we don’t seem to control which reasons we have or which come to mind), to choose on the basis of this reasoning, and to control our actions in light of these choices (e.g., to exercise self-control). These capacities are maximally realized in the Confident Agent, who does not need the unconditional ability to do otherwise. We have these capacities too, but are sometimes in close call situations where they do not lead us to an obvious choice. But those close calls do not seem to make us more free, in control, or responsible for what we do. The capacities for reasoned deliberation and self-control, on the other hand, do seem to make us more free, in control, and responsible for what we do.  
 
I won’t be converting any incompatibilists with these arguments (especially those who care about “sourcehood” rather than alternatives), but I hope to help neutral parties (including non-philosophers) recognize that close calls are not so attractive, and that without them, the unconditional ability to do otherwise is unattractive, while the compatibilist capacities for deliberation and rational self-control are attractive, and crucially, neither determinism nor naturalism challenges our possession of these capacities or our opportunity to exercise them in action.
 
3:AM: Of course, you’re a leading member of the extra cool experimental philosophy clan of which the extra cool Josh Knobe is another member who we’ve interviewed here at 3:AM. You’ve written about why it’s important to have the kind of data you draw on for philosophical arguments. Can you say why you think philosophical arguments and ideas are helped by this data, and perhaps equally importantly, can you explain what would be a wrong or bad way of using the information. Linked to this, you’ve also written about how experimental philosophy is developing and changing as it matures. So perhaps you could also say what experimental philosophy is doing better now than it was, what mistakes you feel might have been made and where you see it all going.
 
EN: Well, I’m either a bad marketer or just not “extra cool.” You see, when x-phi was beginning in the early 2000s there was some discussion about what to call it. My memory is that Josh came up with the name “experimental philosophy” and I was against it. I worried the “experimental” moniker sounded like we were trying out a new drug or just doing experiments, not philosophy. I have also been reluctant to embrace the hip ‘x-phi’ label or image of burning armchairs (tongue-in-cheek or not). Luckily, hotter heads prevailed and the name and images have taken off, and I’ve gotten to go along for the ride. If experimental philosophers – led by the example of Josh – were not some of the most inclusive, cooperative, helpful, and nice philosophers around, then I’d be more worried about the way the movement has presented itself. But if one knows anything about it or (most of) the people in it, then some of the (sometimes scathing) attacks on it sound a bit silly.
 
I don’t believe in burning armchairs. I think we philosophers should be doing most of our work in the armchair (or better, a circle of armchairs) – and the best scientists do much of their work there too, thinking up theories and hypotheses to test, and considering paradigms to test them. I am not against the methodology of appeal to intuitions (I’m not in the ‘negative’ camp of x-phi that says the evidence suggests that all intuitions are unreliable), not even appeal to one’s own intuitions, nor the use of such appeals in one’s arguments. But I can’t imagine how attempts to get more systematic information about people’s intuitions or the psychological sources of them (the descriptive project) does any harm to philosophy.  
 
How does it help? Well, at a minimum it allows us to test certain philosophical claims about what is commonsensical or widely believed or intuitive to ordinary people – and such claims are common in philosophy (when someone writes something like, “Obviously, we believe X,” I tell my students to pounce). It also allows us to gain insight into the sources of our intuitions and judgments, and that is important both to teach us about how our minds work – a quintessential philosophical question – and to give us some evidence about the reliability of our intuitions. For instance, if Knobe is right that our moral judgments can precede and influence our judgments about intentional action, causation, and knowledge, that is important for both our psychological theories and our philosophical theories. Certainly, such information helpfully supplements philosophical theorizing. I do not think it replaces it.
 
X-phi is only a decade old, and it’s already developed quite a bit in terms of the methods it employs (including more complicated statistics, reaction time studies, cross-cultural studies, etc.), the collaborations philosophers have made with practicing scientists, and the attention it has paid to its meta-philosophical implications. Two problems experimental philosophers (especially the students attracted to it) should avoid are: (1) failing to understand the philosophical literature fully enough to design the right materials to test hypotheses and to interpret the relevance of results to philosophical debates (that’s why trained philosophers need to be involved in x-phi projects rather than turning it all over to scientists), and (2) failing to explain the relevance of the results to the philosophical debates.
 
3:AM: Josh Knobe is both a professor of philosophy and cognitive science, and you are also working in a department of both philosophy and neuroscience. Does this suggest that philosophy as you conceive it is actually just a branch of the natural sciences after all? Does this mean that philosophy as a humanities discipline is over from your perspective? Given the difficulties of getting grants outside of science these days in many universities, experimental philosophy might be thought of as another sign that culture is increasingly being dominated by a single research programme (i.e. natural sciences) and that the pluralism of culture is being eroded. How do you respond to this sort of worry? Should we be concerned, or should we be pleased that philosophy has finally learned to get serious?
 
EN: Philosophy examines just about everything, including ourselves, to find the deepest and broadest understanding of just about everything. With content like that, it had better let many flowers bloom when it comes to methodology. The natural sciences provide many methods and answers to get such understanding. So do the social sciences. And so do the humanities (literature and film provide a rich understanding of human life that I have not seen on offer from other methods). Philosophers have the luxury of surveying all these flowers and trying to synthesize the fruits of their labors (nothing like a mixed metaphor).  
 
I don’t think x-phi has helped philosophy “get serious”! It has helped inform some philosophical debates in ways I’ve already suggested. And it has helped some people (including me) get grants they might not otherwise get. I’m very glad my career does not depend on getting grants, and I really feel for the scientists who live and die by them, especially now that the money is flowing less freely. Given what I already said, you can see that I want the humanities flourish as well as the sciences, though there is sub-par work in subfields of each. I should admit that I think philosophy classes (including critical thinking or logic) should be mandatory for undergraduates – I think that (typically) we teach students to think, read, and write better than most other disciplines. Perhaps I’m biased…
 
3:AM: I guess an important criticism of experimental philosophy is one that links with the previous question to some extent. It is that you take for granted the conditions for enquiry as given by research into the natural science and therefore don’t ask the meta-question about what is required for that line of enquiry to be possible in the first place. So this line of challenge isn’t one that’s just skeptical about the outcomes of some of the experiments and the evidence being used, but is one that wants to know why you aren’t asking the philosophical question: what are the philosophical presuppositions required for this programme? This might be the kind of thing that someone like John McDowell might be asking: is experimental philosophy closing down on another strand to the naturalistic approach to philosophy – a tradition that might be characterized as one that runs from Kant through the Hegelians to Wittgenstein, Anscombe, Rorty, Davidson and the later Putnam, –  by staking everything on the other naturalism that might be loosely defined as the one running from Kant to the positivists, Dewey, Quine and Rosenberg, by burning the armchair? Or is burning the armchair burning both these lines of naturalism and opening a new vista?  
 
EN: There’s a lot in this question. My brief response is that one of my favorite things about experimental philosophy, and one of the things I’ve tried to do with it, is to force more consideration of these meta-questions about philosophical methodology. What is the role of intuition in philosophy? Whose intuitions matter? How do we find out what our intuitions are (my own and others’)? How reliable are they? What are we doing when we do conceptual analysis? How do our minds work such that we make the judgments we make about philosophical (including moral) issues? I’m a metaphysical naturalist (all that exists supervenes on what, in principle, can be studied by empirical sciences), but I’m a “wait-and-see epistemological naturalist” (I’m not sure whether the natural sciences will ever develop to the point where we can understand everything in the vocabularies they deploy—I doubt it). But I better leave it at that for now.
 
3:AM: What I find very engaging about the experimental philosophy approach is its dialogic nature. You wrote a great paper on how you have tried to change the teaching of philosophy classes which captures the spirit of your whole approach I think. I think many people will recognize your description of how the Socratic method disappears in a typical philosophy classroom. You write: “people often point to Socrates’ dialogues as the quintessential method of philosophical thinking and teaching. How ironic that these dialogues too often depart from what we actually think of as the ‘Socratic method’, since Plato’s Socrates (especially after the early dialogues) usually takes over discussions to offer monologues interrupted only by the interlocutors’ assents of ‘By all means,’ ‘Precisely so,’ and ‘Of course.’”’ Your response to this, and a diagnosis of the problem, is very interesting: “If we want to use the Socratic method, then we should guard against falling into a Socratic lecture about what the right answers or the right questions are.” Can you explain what you mean by this and how you think teaching could be improved?
 
EN: Thomas Nadelhoffer and I wrote a paper, ‘Polling as Pedagogy’, where we argue that philosophy courses, especially intro, could be enhanced with more systematic use of what instructors tend to do anyway – poll their students on the thought experiments in the reading. If you give students the scenarios to read ahead of time and let them think a bit about their intuitions and write down their responses and some reasons for them, it can help in several ways. It can help ensure that they engage with the thought experiments. It can ensure that the proportions of responses are accurate (especially if they are taken up online or with ‘clickers’), sometimes surprising the teacher. And it can help inspire debate when students see how many of their peers have different intuitions. It’s helpful to have the minority respondents try to convert the majority (and you can cold call on people who don’t talk enough since they had time to think about why they responded as they did). There are tons of good survey scenarios to use from the literature, from some recent books on thought experiments, and from work in x-phi. We can also get our students to think about what thought experiments and intuitions are supposed to do in philosophical debates.  
 
To me experimental philosophy is a way of doing philosophy. It forces me to think hard about how to present philosophical ideas in a way that is understandable to non-philosophers. If a survey (or argument) uses a scenario that people can’t understand, then the results are useless. At the same time, trying to discover how people are interpreting a scenario helps me to see how they are thinking. And it helps me think through various approaches to a problem as I design experiments in response to previous results. This process might be seen as Socratic dialogue in slow motion.  
 
3:AM: Your new book in progress is called Rediscovering Free Will. Do you find writing something that helps you to shape your ideas or is it just a matter of reporting what you already think? Have you ever changed your mind in the process of writing something? In fact it seems pretty rare to find philosophers who ever admit to changing their minds much at all. Yet the Socratic spirit you want to encourage would suggest that this is a bad thing if philosophers don’t do much changing in the light of other arguments. What would make you change your mind about what you think about free will, for example? And what do you think about this general issue of the truculence of philosophers and their own beliefs?
 
EN: In theory philosophers should be more willing to change their minds in the face of evidence and arguments than just about anyone. But in practice, they may build the walls of their castle so strong they can’t break them down. There’s psychological evidence that people believe things more strongly the more they argue for them. Since philosophers argue for positions more than anyone, we likely come to believe those positions strongly. We also try to construct coherent sets of beliefs, which means that we see how giving up just one belief might suggest we have to give up lots of other beliefs, and to use Neurath’s image, it may be difficult to give up most of the planks in one’s boat unless one already sees how to replace them. On the other hand, philosophers are very good at playing devil’s advocate for their own positions, and one risk of playing devil’s advocate is that one comes to believe what your client believes. In any case, I don’t think I’ve changed my mind enough about much. I’ll keep working on it.  
 
3:AM: And where next are you going to be researching once you’ve had the last word on free will?
 
EN: Well, I’m sure I won’t have the last word on free will. And I’ll probably keep thinking about it for a long time. I came to free will in part because of my interest in consciousness, so I’d like to go back to solving that “easy problem” – actually, as I’ve said, I think that once we solve the hard problem of explaining how conscious processes exist and work in a naturalist framework, the problem of free will should largely dissolve. I’m interested in many areas of moral psychology, including moral intuitions and moral cognition – I’ve done some “trolleyology” experiments and would like to try more x-phi work on moral judgments. I’d like to understand how we understand our own and others’ mental states (when we do) and how much our moral judgments are influenced by our epistemic judgments (and vice versa). And I haven’t lost my teenage obsession of figuring out how human nature evolved to be the way it is.
 
3:AM: When you’re not brooding or burning armchairs, what kind of stuff do you do that helps you with your philosophy? Are there significant books or films or art shows or tv that have been particularly illuminating to you?
 
EN: (Well, as we’ve established, I need my armchair too much to burn it.) When I’m not working, I’m spending most of my time with my family (my wife, Cheryl, and I have three kids, ages 10, 7, and 4). I tell my students that philosophy is about the questions that children ask, “Why?” and that teenagers ask, “Why do I have to?” So, in addition to helping to keep me grounded, my kids remind me what philosophy is all about. Sometimes they give me examples for blog posts at Flickers of Freedom – for instance, recently I got honked at for holding up traffic, and my son said, “You deserved that!” I used that to launch a discussion of what we mean by deserving something and how this case of low-level desert compares with a murderer deserving to be punished.
 
Instead of ending on a high-brow note with a list of all the cool films I watch (have you seen how sophisticated kids’ movies are these days?) or the uncountable books I read outside philosophy, I’ll offer a poem that I wrote. It’s inspired by one of the books I’ve been reading (to my kids): Shel Silverstein’s ‘Runny Babbit’ (all you need to know is that he cleverly transposes the initial sounds of nearby words to write punny foems).

Runny Babbit Pheads Rilosophy
 
Runny Babbit chose, he thought,
to read about wee frill.
He’d heard that philosophy
could be thrite a quill.
He tried Veter pan Inwagen,
​but was unable to trender it rue.
He then read Frarry Hankfurt,
without really wanting to want to.
Wusan Solf drove him crazy.
​Is JoJo mad or bad or lust jazy?
Next up was Reedom and Fresentment
by elder Streter Pawson.
Runny lorgave him for its fength,
since he thought it pretty awesome.
He then read the son Galen,
who says it’s all just a latter of muck.
Runny decided silosophy phucks,
and most silosophers phuck.

For those who have not “submerged” themselves in the free will literature, perhaps this poem will inspire you (ha!) to take a look at the wonderful work of van Inwagen, Frankfurt, Wolf, and the Strawsons. And while you’re at it, I encourage you to read some others, against whom I did  did not commit poeticide (but do feel free to Sel Shilverstein their names), such as John Fischer, Bob Kane, Al Mele, and Derk Pereboom, not to mention ‘youngsters’ like Dana Nelkin, Manuel Vargas, Neil Levy, and Tamler Sommers. Runny Babbit’s conclusion is all wrong. It’s a very exciting time for philosophical and scientific discussions about free will and responsibility, and I am lucky to be a part of it. Tanks for thalking with me, Richard!



richardmarshall

ABOUT THE AUTHOR
Richard Marshall is still biding his time.

First published in 3:AM Magazine: Friday, May 25th, 2012.