In his essay written for receiving the Erasmus Prize, “Erasmus: Sometimes a Spin Doctor is Right“, Dan Dennett argues that the idea that free will is merely an illusion—an idea promulgated by bad people like Sam Harris and me—is deleterious to society:
There is—and has always been—an arms race between persuaders and their targets or intended victims, and folklore is full of tales of innocents being taken in by the blandishments of sharp talkers. This folklore is part of the defense we pass on to our children, so that they will become adept at guarding against it. We don’t want our children to become puppets! If neuroscientists are saying that it is no use—we are all already puppets, controlled by the environment, they are making a big, and potentially harmful, mistake.
. . . the deep conviction Erasmus and I share: we both believe that the doctrine that free will is an illusion is likely to have profoundly unfortunate social consequences if not rebutted forcefully.
And on Point of Inquiry, Dec. 12, 2011, Dennett told John Shook the same thing:
“We certainly don’t want people disabling themselves with bad science. . . so I think this [claims that fee will is an illusion] is a very serious issue.
Similarly, in a piece at “Cross-Check,” his Scientific American website, John Horgan, while going after me, also argues that it’s bad for society to embrace determinism:
A recent experiment shows that belief in free will has measurable consequences. The psychologists Kathleen Vohs and Jonathan Schooler asked subjects to read a passage by Francis Crick , co-discoverer of the double helix, that casts doubt on free will. Crick wrote in The Astonishing Hypothesis (Charles Scribner’s Sons, 1993) that “although we appear to have free will, in fact, our choices have already been predetermined for us and we cannot change that.” Subjects who read this passage were more likely to cheat on a test than control subjects who read a passage about brain science that did not mention free will. Mere exposure to the idea that we are not really responsible for our actions, it seems, can make us behave badly.
This all reminds me of the famous (and possibly apocryphal) response of the wife of the Bishop of Worcester when her husband told her of Darwin’s theory that humans evolved from apes. “My dear, descended from the apes!” she said. “Let us hope it is not true, but if it is, let us pray it will not become generally known.”
And it all supports my notion that one motivation for promulgating “compatibilist” free will (i.e., the view that pure determinism of human actions is still compatible with some conceptions of free will) is that if people learned that their actions are predetermined, and that dualistic free will did not exist, they’d either behave like beasts or lapse into torpor and nihilism. (This is, of course, the same argument that religious creationists use against evolution.) Yes, I know that compatibilism preceded these psychological studies, but look at the quotes above again. An insistence on incompatibilism is, as Horgan and Dennett argue, bad for society.
Many others agree with them, often cite as support the Vohs and Schooler paper (reference and link to download below) supposedly proving the deleterious effects of incompatibilism. Here’s the abstract of that paper:
Does moral behavior draw on a belief in free will? Two experiments examined whether inducing participants to believe that human behavior is predetermined would encourage cheating. In Experiment 1, participants read either text that encouraged a belief in determinism (i.e., that portrayed behavior as the consequence of environmental and genetic factors) or neutral text. Exposure to the deterministic message increased cheating on a task in which participants could passively allow a flawed computer program to reveal answers to mathematical problems that they had been instructed to solve themselves. Moreover, increased cheating behavior was mediated by decreased belief in free will. In Experiment 2, participants who read deterministic statements cheated by overpaying themselves for performance on a cognitive task; participants who read statements endorsing free will did not. These findings suggest that the debate over free will has societal, as well as scientific and theoretical, implications.
Well, a new research team has repeated the Vohs and Schooler experiment, and couldn’t repeat their results. In the new study, reading about determinism had no effect on cheating. The work was done by Rolf Zwaan at the University of Rotterdam, as part of a university class on cognition. He describes the results on his website:
In V&S’s study, subjects in the AFW condition [those having read Crick’s view that free will is an illusion] reported weaker free will beliefs (M = 13.6, SD = 2.66) than subjects in the control condition [Crick’s “neutral text”] (M = 16.8, SD = 2.67). In contrast, we found no difference between the AFW condition (M = 25.90, SD = 5.35) and the control condition (M = 25.11, SD = 5.37), p = .37. Also, our averages are noticeably higher than V&S’s.How about the effect on cheating?V&S found that subjects in the AFW condition cheated more often (M = 14.00, SD = 4.17) than subjects in the control condition (M = 9.67, SD = 5.58), p < .01, an effect of almost one standard deviation! In contrast, we found no difference in cheating behavior between the AFW condition (M = 4.53, SD = 5.66) and the control condition (M = 5.97, SD = 6.83), p = .158. Clearly, we did not replicate the main effect. It is also important to note that the average level of cheating we observed was much lower than that in the original study.V&S reported a .53 correlation between scores on the Free Will subscale and cheating behavior. We, on the other hand, observed a nonsignificant .03 correlation.
In short, Vohs and Schooler’s result, used to buttress the value of compatibilism, is now in doubt. Do note, though, that Zwaan hasn’t yet published his result in a peer-reviewed journal, so this failure of replication must remain tentative. I’ll write Zwaan and ask if they’re going to publish this result.
In another post, Zwaan describes why his group’s result might have differed from that of Vohs and Schooler. First, there is sample size:
We ran the experiment on Mechanical Turk, using 150 subjects. This should give us awesome power because the original experiment used 30 subjects and the effect size was large (.82).
One obvious difference between our findings and those of V&S is in subject populations. Our subjects had an average age of 33 (range 18-69) and were native speakers of English residing in the US (75 males and 77 females). The distribution of education levels was as follows: high school (13%), college no-degree (33%), associate’s degree (13%), bachelor (33%), and master’s/PhD (8%).
How about the subjects in the original study? V&S used… 30 undergraduates (13 females, 17 males); that’s all it says in the paper. Kathleen Vohs informed us via email that the subjects were undergraduates at the University of Utah. Specifically, they were smart, devoted adults about half of whom were active in the Mormon Church. One would think that it is not too trivial to mention in the paper. After all, free will is not unimportant to Mormons, as is shown here and here. It is quite true that Psychological Science imposes rather stringent word limits but still…
Surprisingly, the subject population (largely Mormons) was not described in the Vohs and Schooler paper! There is simply no excuse for that; it’s sloppy writing.
Zwaaan describes other differences in the experiments that could have affected the results. For example, Zwaan’s study was online while Vohs and Schooler’s (V&S) was in the lab. V&S also employed a fake “computer glitch” that could allow for cheating, and thus there was acting involved. Zwaan says this might be telling, and makes the rather strong claim that V&S’s failure to adequately explain their protocol was “voodo experimentation”:
But there is a bigger point. If the large effect reported in the original study hinges on the acting skills of the experimenter, then there should be information on this in the paper. The article merely states that the subjects were told of the glitch. We incorporated what the students were told in our instruction. But if it is not the contents of what they were told that is responsible for the effect but rather the manner in which it is told, then there should be information on this. Did the experimenter act distraught, confused, embarrassed, or neutrally? And was this performance believable and delivered with any consistency? If the effect hinges on the acting skills of an experimenter, experimentation becomes an art and not a science. In addition to voodoo statistics, we would have voodoo experimentation. (A reader of this post pointed me to this higly relevant article on the ubiquity of voodoo practices in psychological research.)
So where does this leave us? The fact that the large (!) effect of the original study completely evaporated in our experiment cannot be due to (1) the age or education levels of the subjects, (2) subjects not reading the manipulation-inducing passages (if reading times are any indicator), and (3) subjects’ awareness of the manipulation. The original paper provides no evidence regarding these issues.
The evaporation of the effect could, however, be due to (1), the special nature of the sample of the original sample (2) the undocumented acting skills of a real-life experimenter (voodoo experimentation), or of course (3) the large effect being a false positive. I am leaning towards the third option, although I would not find a small effect implausible (in fact, that is what I was initially expecting to find).
He also observes that another researcher is replicating the V&S experiment.
I emphasize again that Zwaan’s study hasn’t been peer-reviewed or published, and until it has we shouldn’t conclude that the V&S paper is deeply questionable. However, I am disturbed at the lack of replication, which may be a common feature of such small-scale psychological experiments. These studies often use college students, and it’s not clear how representative they are of people as a group (or even Westerners as a group). Small sample size is also an issue.
In an ideal world, every experiment would be replicated, and in molecular or evolutionary genetics such replication is often inherent in later studies, which frequently build on earlier ones. But in psychology—and much of evolutionary biology and ecology—there’s no professional payoff in repeating what someone else has done. Because such experiments often involve special populations or locations, I’ve often mused that replicating most evolutionary or ecological experiments using different materials and/or locales (or even the same ones!) would probably give the original result only about 30% of the time. (That’s just pure speculation on my part.) Remember that when you use conventional statistical criteria, you’ll get false positives at least 5% of the time. What I’m saying here is that we should be wary of single studies with flashy results that nobody tries to repeat.
In this case, we should be wary of using the V&S experiment to argue that belief in determinism has inimical effects for society. But even if it did, I can’t countenance hiding our belief in determinism—a belief shared by Dennett and many compatibilists—from the general public on the grounds that it’s bad for society. That’s simply condescending. The truth is the truth, and shouldn’t be suppressed. And, of course, there are some potentially positive effects of accepting determinism and the idea that free will is an “illusion” (i.e., not the dualistic behavior it seems to many people). Those positive effects include salubrious reforms of the criminal justice system and the demolition of an important cornerstone of religion.
This is all independent of whether one can philosophically forge a compatibilist notion of free will. To me that endeavor is futile and incoherent, though others disagree. Nevertheless, such an endeavor should derive from philosophical motivations alone rather from some supposed effect of incompatibilism on society.