What caused a pair of bombs to go off at the Boston Marathon on April 15? One correct answer could be a fuse, but that would be unsatisfying. We want to know who is responsible for the atrocity. Much effort is now being made to apportion blame. But what if the answer is no one? What if no one is ever responsible for anything?
Neuroscientist and atheist author Sam Harris is convinced that free will does not exist, and he’s gone to considerable trouble to convince you, as well. He’s written a book to make his case:
"Free will is an illusion. Our wills are simply not of our own making. Thoughts and intentions emerge from background causes of which we are unaware and over which we exert no conscious control." (Free Will, New York: Free Press, 2012, p. 5)
Such a stance is not without consequences. Harris himself suggests that criminals cannot be responsible for their acts, since their causes lie out of sight and out of mind. “How can we … hold people accountable for their choices, given the unconscious origins of our conscious minds?” (p. 5) Later, he writes, “Once we recognize that even the most terrifying predators are, in a very real sense, unlucky to be who they are, the logic of hating (as opposed to fearing) them begins to unravel.” (p. 53)
I intend to refute his case—and to do so with more than armchair reasoning. I will offer some experiments you can do on your own—the poolside leap test, the dissolving word demonstration and the tic tac toe wheeze (among others). I need these, because Harris claims more than just theoretical justification. He says he has science on his side.
But first: what the hell are we talking about here? Sam’s a bit slippery about just what it is he denies. (Although, like the Groucho Marx character Rufus T. Firefly, whatever it is, he’s against it!). “If determinism is true, the future is set—and this includes all our future states of mind and our subsequent behavior. And to the extent that the law of cause and effect is subject to indeterminism—quantum or otherwise—we can take no credit for what happens. There is no combination of these truths that seems compatible with the popular notion of free will.” (pp. 29-30, emphasis added.)
Harrumph. One of Harris’s claims is that fMRI (a magnetic brain scan) can read subjects’ intentions as much as 10 seconds before they are aware of them—with 80 percent accuracy. With that in mind (so to speak), let me offer a working definition of free will: the ability to consciously decide one’s actions so that they can, in principle, be unpredictable.
That last clause is, by no coincidence, an analog to the Turing Test. (In 1950 computer pioneer Alan Turing proposed that if an AI program could fool an human interrogator—via teletype, or these days, instant messaging—into believing that it was human, then the program would truly be intelligent and likely conscious.) By extension, if self-declared conscious acts can be made unpredictable on demand, that should be held prima facie evidence of free will.1
Here’s the line of demarcation: even if we assume that there is an unbroken chain of causation from the quantum to the quandam, the line of predictability may be severed at some point. Where it is, something more than particle bumping into particle, like so many billiard balls on a table, is going on.
That something is emergence. This is the idea that as simple things interact, new properties and relations emerge. They can only be understood at a higher level of interpretation because the simpler units give no clue about the more complex. For example, there is nothing about a molecule of H2O that tells you about its ability to exist in three distinct phases: ice, water or steam. Examine a water molecule in any one of those phases and you will see no change. The difference between ice, water and steam is an entirely emergent property. Nothing spooky about that—it’s just a fact of nature.
One of the more interesting emergences in the history of the Earth is the tendency of certain molecules to replicate themselves. This requires the ability to store knowledge. The world champion at this is DNA. It stores whole libraries of how-to manuals for building various kinds of life that harbor replicators.
One set of libraries tells proteins how to build brains. Some brains are mere “if-this, then-that” devices (see ants), but some have another emergent property: consciousness. It’s the Fifth Avenue Penthouse of pricey emergent properties. In humans, the brain soaks up a fifth of the energy budget to maintain it. Does consciousness have some causal power over our actions?
Harris thinks to accept that is to buy into magic, and so he argues that the appearance of higher-level causes is an illusion. He tells us it can be dispelled, ironically enough, by causing ourselves to pay attention to what’s really going on inside. “We are mistaken about our experience. Not only are we not as free as we think we are—we do not feel as free as we think we do. Our sense of our own freedom results from our not paying close attention to what it is like to be us. The moment we pay attention, it is possible to see that free will is nowhere to be found…” (p. 64). Might it not be more probable that while introspecting, Harris wills himself not to see?
It doesn’t matter. Harris is fundamentally wrong. You don’t have to consider the entire chain of causation to explain an event. As the physicist David Deutsch points out, “cause” is itself an abstract concept, not a fundamental property of nature. We interpret certain sequential relationships as cause and effect.2
Not arbitrarily, however. We can debunk invalid claims, such as the assertion that “lifestyle choice” rather than HIV causes AIDS.3 But even with a valid claim, where we draw the lines is often a matter of interpretation. Did a gun kill JFK? Did Lee Harvey Oswald? Or was it a particular arrangement of all the atomic nuclei in the universe shortly after the Big Bang? All could be valid causal descriptions, but a lot rides on which you choose.4
Harris simply does not see how emergent causes can be considered independent of their foundations in lower-level causes. This leads him to view the self as a helpless wanderer in a labyrinth. “My choices,” he tells us, “matter—and there are paths toward making wiser ones—but I cannot choose what I choose. And if it ever appears that I do—for instance, after going back and forth between two options—I do not choose to choose what I choose. There is a regress here that always ends in darkness.” (p. 39)
Nonsense. First of all, as the interpreter of cause, you can (and should) ignore the regression the moment it becomes irrelevant to what you are attempting to understand. Let’s say lightning strikes a healthy tree in your backyard and it topples onto your roof. Although the tree is clearly the proximate cause of the damage, it would be unreasonable to stop there: lightning was responsible. But suppose your insurance company shows that the lightning was caused by a high-velocity cosmic ray, and that the ray was emitted by a supernova, and by the way your policy doesn’t cover astronomical events? Can it get out of paying the claim? Certainly not. Only relevant causes count.
Still, Harris has a second line of attack on free will. He argues that thoughts have no causal power—they just pop into mind, followed by another unforeseeable volley of thoughts, and another after that. If indeed our thoughts are just a fireworks show high above the neuronal cannons that form brain states, then indeed there seems to be no way for them to affect anything. But there is something wrong with this argument.
To spot it, just substitute “consciousness” for “thoughts” and consider the implication. If there’s nothing consequential about consciousness, we can subtract it from the list and just regress from the most recent brain state to the Big Bang, or vice versa. We may then claim to have a complete account of everything real in that particular chain of history. After all, a thing that has no effect whatsoever is just another Invisible Dragon in the Garage.5
Here we come to the heart of Harris’s problem with free will: it’s really about the mystery of consciousness. How can consciousness play a causal role in anything? To know that definitively, we’d have to know just what consciousness is, but we don’t. We’re confident it is a natural phenomenon—it is certainly subject to material influences (viz., a shot of vodka), and there is every reason to believe that it is subject to the laws of physics.
Lacking a full understanding of consciousness does not grant us a license to deny its existence. Indeed, if there is one thing in all of existence that it is preposterous to deny, it is the existence of the mysterious “I” to which Harris repeatedly refers. As DesCartes rightly observed centuries ago, all other things may be illusory, but how could anyone be fooled into believing they exist when they don’t? (Possibly, I exist in some wildly different form—data scattered across a holograph, or whatever—but if I think I exist, I damn well do!)
Within the bounds of what we know, reasonable conjectures about how “I” have a causal influence over my actions are in reach. You don’t have to deny determinism or the laws of physics or to acknowledge multiple levels of causation, including causation by minds—virtual agents that knit data into knowledge and use it to make decisions that affect the world.
How can this be in a (mostly) deterministic universe? Through top-down causation. Sometimes, emergent phenomena can affect the lower-level phenomena that are their causes. Simple example—one snowflake cannot alter the Earth’s temperature—after all, its formation was caused, in part, by a certain temperature. But if enough of them fall, their reflectivity can cause the Earth’s temperature to plunge, leading to more snow, and so on until the entire Earth is frozen over. This appears to have happened in geological history in a period known as Snowball Earth.
That’s trivial compared with how brains can (and have) altered the Earth. Snowflakes are passive. They just happen. Brains are active. They gather information, compare it with memory and draw on various mental tools to model alternative futures.
But brains are just physical stuff, the Determined Determinist says. True enough. But minds are virtual agents—abstract objects that run on the wetware of the brain and may someday run on hardware of some other kind. One of their remarkable features is that minds can externalize many of their operations—memories can be written down or recorded, ideas can be voiced, mental images can be sculpted, built or drawn. Minds can even co-opt indeterminacy! Some free-will searchers have tried to find a brain mechanism that can do the job (a quantum tubule, for example), but it’s a futile and unnecessary search.
First, as noted, we can co-opt randomness in our environment (more on this in Part Two.) But more to the point, determinism has nothing to do with free will, at least not of the kind meaningful to us. If you want to deny free will on a cosmic scale, then determinism becomes relevant—and you run smack into new challenges posed by quantum theory. Let’s leave those aside for now. If, on the other hand, you want to defend the position Harris takes—that in the realm of human affairs what seem to be consequential decisions are mere illusions of choice—then determinism is irrelevant.
That’s right. Who cares how 10^80 tiny particles are predestined to dance from moment to moment when we have emerged at the highest level of complexity with the freedom to make meaningful choices? So what if those choices are embedded in the particle dance? No one—not even with the aid of the most powerful computer imaginable—can look at the arrangement of particles and predict your next move on that basis—not if you want to be unpredictable. You have that virtual yet genuine freedom.
Let’s wind up with a whimsical example. From March until November 1980, much of the world was gripped by a question. It was this: “Who shot JR?” The cliffhanger ending of the ’79–’80 season of the TV show “Dallas” left viewers wondering which of the dastardly Ewing family members had tried to knock off its sneering patriarch. Betting parlors set odds, and huge stakes were laid. The actor playing JR was offered a six-figure bribe to spill the beans. (At $100,000 an episode, Larry Hagman didn’t need and didn’t take it.)
But if determinism dictates, the answer was in plain sight all along. All a person would need do is study the inbound path of each electron in the cathode tube of their TV as it struck each phosphor cell, and then predict the outbound path of each photon that it would emit, and the next and the next, leading inevitably to the on-screen revelation nine months in the future that JR’s scheming mistress, Kristin, done did it.
So where’s the problem? Well, we haven’t included all the relevant particles. What about the photons from the broadcast tower that prompt the electron beams in the cathode tube to form a particular pattern? What about the nerve impulses that caused the writer’s fingers to hit certain keystrokes? Really, to be sure of getting to the right conclusion, we need to take a snapshot of all the particles in the universe at the season finale and go from there.
Only we cannot. It’s not just a practical challenge. Our universe has built-in limits on the speed and precision with which data can be gathered. The absolute limit on the transmission of information is the speed of light. We cannot see every particle simultaneously. The absolute limit on precision is the Heisenberg Uncertainty principle. We cannot gain perfect resolution on both location and momentum for any particle. Together, these guarantee that no one can ever take a forensic snapshot of particles to predict what will happen next.
Unless you resort to an omniscient God (something Harris would never do), determinism as the playbook of fate is mere fantasy. When it comes to human choice, it is as irrelevant as the phosphor screen dots are to who shot JR. And when it comes to deciding who is responsible for the Boston Marathon bombings, the relevant causes have nothing to do with particles and everything to do with virtual minds that influence others and even themselves through those mighty symbols, words.
Words create minds, and minds create words. From that recursive relationship, immense power grows. The power, for example, to make reasoned decisions about the future, decisions that can change the world. With the power to make decisions comes responsibility. We can choose to fire off nuclear missiles or to send a man to the moon, to run a marathon race or to set off bombs at the finish line.
Harris does not see it that way, of course, and he claims to have scientific evidence that would falsify my position. To secure the claims I have just made, therefore, we will have to consider his scientific evidence and present some experimental evidence of our own. By all means, respond to this column, but don’t miss Part Two, where the great poolside leap test and other DYI scientific tests of free will will be unfurled.
3. In 1987 molecular biologist Peter Duesberg first hypothesized this and has not changed his mind, but massive, worldwide evidence that HIV is responsible for AIDS (and that disabling HIV prevents AIDS) has convinced virtually all other researchers in the field that he is wrong. (See en.wikipedia.org/wiki/ Peter_Duesberg.)
4. By remarkable coincidence, the philosopher Daniel Dennett uses the example of the JFK assassination to make a similar though deeper point in his book “Freedom Evolves” (p. 84). In much the way that Schubert might have wished he hadn’t heard Beethoven’s symphonies before composing his own, I postponed reading “Freedom Evolves” until after I had gotten my ideas down on paper. For anyone who wants to really understand this subject, I can do no better than to recommend Dennett’s book.
5. In his book “The Demon-Haunted World,” Carl Sagan offers the example of an invisible dragon that cannot be detected by any means whatsoever as something that may be legitimately considered unreal. (See rationalwiki.org/wiki/The_Dragon_in_My_Garage)