I have not tired to argue that the notion of the freedom of the will in its interesting, that is, metaphysical, form is a lost cause. There will never be proof that a person, or any other creature, has acted freely because there can be no such proof. If it existed, it would require a demonstration that the person’s (or the creature’s or the rock’s) behavior was neither determined by causes nor truly random. No one even knows what that might mean beyond the mere phraseology of the behavior being freely willed. When people claim, usually with regard to their own behavior (but also with regard to other people whose behavior they find objectionable), that they acted out of free will, they mean that they can imagine having acted differently. Of course they can imagine different actions. Fantasy is cheap and rangy (though not itself freely willed). It is easy to imagine the sun rising in the West and traveling to the East (or to think about the Earth rotating in such a way to make it look that way). The ease of imagining anything has no evidential value whatever. It is easy to imagine Santa Claus sliding down the chimney or going on a date with Jennifer Lawrence. Yet, ideation does not make reality. Whoever said it did? Did I miss the memo?
Perhaps I'm being unfair by using examples of events we consider impossible (Earth’s alternative rotation, the existence of S. Claus). Free willers will happily restrict themselves to possible behaviors. Speaker of the House, Boehner (pronounced "Bayner") said the word “ass” in public, but we can imagine an alternate reality in which he restrained himself; and presumably so can he. I counter that if he could have restrained himself, he would have, but he didn’t. So there. Anti-counterargument: Krueger is just tautologically claiming that whatever happened had to happen (or was random). Sure. I can live with this tautology because the alternative is worse. Being able to imagine Mr. Boehner using politic speech says nothing about his freedom to actually do so. It’s all in your head.
Now power. Interpersonal power is a wonderful topic. Bertrand Russell thought it was so important and pervasive that social science should be built around it. Susan Fiske once said (ca. 1995) that power is a dirty word among liberal social psychologist, so dirty in fact that few would dare study it. Times have changed. Led by my Messrs. Galinsky (Northwestern) and Keltner (Berkeley), among others, social psychologists have rediscovered the dark and fascinating world of interpersonal power. Their research shows that individuals with a high-power mindset feel good, entitled, and optimistic. They go out and get stuff (done). Power (The Will to Power, to use Nietzsche’s term) is a powerful motive. Who would prefer having less power to having more power? C’m on! I know you like this rhetorical question.
It is difficult to define “power” to everyone’s satisfaction. According to one definition, power resides in the control over someone’s else’s outcomes. Suppose Ali gets to decide whether Yusuf gets a job. Ali is powerful; Yusuf is not. We think of bosses as more powerful than the rest of us. Bosses hire, fire, set salaries, grant and withhold assignments and bonuses; and we just have to live with it. In the labs of behavioral economists, the Dictator Game is the paradigm for power. There is a pot of money (e.g., 100 dirham) and Ali gets to divide it between himself and Yusuf. Yusuf’s job is to accept the result. As an aside, the Dictator Game yields a confounded picture of how much people care about others because their caring about others is perfectly (and negatively) confounded with caring about themselves. In a modified Dictator Game, Ali gets 100 dirham no matter what and gets to decide how many dirham (0 – 100) Yusuf will receive.
According to another definition, power lies in the ability to influence others’ behavior. This boils down to the ability to get people to do things they’d rather not do (buy more insurance, get married, turn in a lab report on time). As interpersonal behavior is, well, interpersonal and highly dynamic, human behavior involving some aspect of behavior control is more common than the pure kind of distributional arrangement discussed in the context of the first definition. It is perhaps for this reason that this second definition is popular. Galinsky, though, pointed out a problem. Defining power as influence confounds the power of the agent with the weakness of the target. If Ali gets Yusuf to pass the oasis without having a drink, this episode tells us as much about Ali’s power as about Yusuf’s weakness – until we gather more information.
Let me introduce a third definition: Power is unpredictability. He (she, if you like) who manages to behave unpredictably has power. More specifically, power is relative. If Ali manages to behave less predictably than Yusuf, he has more power. Power is to keep the other guessing – and thus off balance. If Ali’s behavior is predictable, be it nasty or nice, Yusuf can act accordingly, make plans, exploit Ali, or at least protect himself from being exploited. In the history of psychology, thousands of experimenters have subjected their lab animals to their power of giving or withholding rewards. If the rewards come reliably, the behavior of the non-human animal is closely aligned with the experimenter’s behavior. At the limit, there is no logical reason to see who is conditioning whom. As rewards are offered on a probabilistic schedule (in the advanced stages of learning), the experimenters assert their power over the pigeon. Still, the pigeon has the power “to make” the experimenter release fewer food pellets by pecking less frequently or less consistently.
Pigeons are one thing; god is another. Scripture (and the pope emeritus) says that god is omnipotent. How would one know that? Well, one source is assertion, with reference to a text presumably authored by the agent whose omnipotence is being asserted (and they accuse me of tautology). Another source is experience. God’s responsiveness to human suffering is probabilistic, so much so that the null hypothesis of chance variation has not been rejected by those who have bothered to look (ask Job). In other words, god’s perceived power comes at the price of his benevolence not being taken all that seriously. The miracle of faith is the ability to assert the existence of a higher being whose actions are so random as to maximize attributions of power while minimizing attributions of existence. If god had her wits about them, she would act in ways that are partially predictable. That strategy would optimize the balance between power and (the lack of) doubt.
Humans who want to maximize their power over others without being ignored or abandoned face a challenge. By reducing the predictability of their actions, they can increase their power, while risking the collapse of their relationships. Arguably, the mark of social intelligence is to reduce predictability as far as possible without going over the cliff.
What does this have to do with free will? If, in their quest for power, people strive to make their actions unpredictable to others, they face a problem. They want to accomplish (partial) unpredictability without resorting to chance. They could throw dice to guide their behavior, but this strategy will undercut their sense of power. They want behavior that looks partially random from the outside but willfully chosen from the inside. The solution is to conjure the conviction that behavior is freely willed. If Ali believes that he can freely choose actions so that they appear partially random to Yusuf, he’s got it made. He has the power, but he paid for it with an illusion.
As Benedetto Sedicesimo reverts to Joseph Ratzinger, I feel the sting of respect. J. R. did something extraordinary and thus unpredicted. He committed his greatest act of power by giving up power.
E' un miracolo!