I want to add two fairly verbose points to Peter’s post about how contemporary neuroscience very much does not close off debates about free will, no matter how many times or how condescendingly Sam Harris insists otherwise.
You don’t need to read any of that — in fact, you don’t need to know anything about the debate — to know that Harris is way too far out on a limb. Anyone who writes “there is simply no description of mental and physical causation that allows for [free will]” either hasn’t read enough or is being intellectually dishonest (see point #2 for more on this). Google Scholar shows 600,000+ articles that mention “free will” published in the last 18 months. The Wikipedia entry on the Neuroscience of Free Will has 16 sections and subsections and concludes that “there is no consensus among researchers about the significance of findings, their meaning, or what conclusions may be drawn.” Does it really sound right that all of these people just didn’t get the memo that the debate was self-evidently settled, as Harris snidely insists?
As for the content…
(1) It’s hard to escape the impression that Harris is engaging in philosophical debates the contours and stakes of which he doesn’t quite understand. He sometimes seems to confuse different technical concepts of “free will” and “rationality.” Argumentatively the confusion helps him, allowing him to dismiss one theory on the basis of convincing flaws in a related but distinct theory. But it makes it much harder to untangle what he’s claiming, why he thinks he’s so undeniably right, and how he thinks everyone else just hasn’t been paying enough attention.
Let’s separate out two different theories about why human motivation might be “opaque” to us.
The first theory is the psychoanalytic division between the conscious and the unconscious mind. We are motivated by desires and beliefs that we can’t admit to ourselves we have. Psychoanalytic structures like the id and ego obviously can’t get physically mapped on to parts of the brain, which makes psychoanalysis an incomplete theory, but it’s not a bad approximation for a lot of behavior. Psychoanalytic theory and “the talking cure” are almost totally dead in psychology departments — a June 2008 study in the Journal of the American Psychoanalytic Association described psychoanalysis as “desiccated and dead” — but practicing psychotherapists still often have Freudian sensibilities about things like reaction formations. Psychoanalytic theory also helps explain why persuasion can work even when you know how it’s working on you, as in when you still kind of tear up at awkwardly schmaltzy movies. It also accounts for the counter-intuitive finding that introspection is often counterproductive, e.g., the more people work on their cognitive biases the more biased they’re likely to objectively become, and why people are so stunningly, horrifically bad at predicting their own future behavior.
Still, psychoanalysis isn’t really the theory that Harris is talking about. He sometimes throws around pop psychoanalytic language when it rhetorically suits him but it’s not really what he’s after. But, for the sake of argument, it’s worth noting just how many different empirical and theoretical accounts of the mind converge on the idea that we don’t really have access to all our motivations. Harris isn’t totally wrong. He’s just at the beginning of a debate he thinks he’s ending (more on that in point #2 below).
In any case, the distinction that Harris relies on is a second one, the neurological division between the mind and the brain. According to Harris, the physical brain generates everything you do and then your conscious mind tries to catch up and figure out what’s going on. In the process your mind makes you think “you” are the one who sent the orders to your brain, when in fact it was the other way around. There’s an intimidating amount of empirical evidence backing this account, enough that plenty of very smart neuroscientists think it’s flat out true (although Peter’s post runs down plenty who don’t). Harris’s book is filled with expert quotes and experiments, many demonstrating that people’s bodies begin acting before those people “decide” to act.
Some of that sensibility has filtered down into folk wisdom. When lifestyle blogs advise people to smile because it will make them feel better — advice backed by experiments, by the by — that’s more or less a mind/brain issue. Very roughly, the mind sees that you’re smiling so it figures that you must be happy.
Harris’s problem isn’t the neuroscience, an area in which he’s quite literally a certified expert (although it can’t be repeated enough that he provides only half the story; see Peter’s post for the other half). His problem is how he thinks neuroscientific findings map on cleanly to debates over morality and free will. His view of what goes into making us moral or immoral is a kind of first-brush intuitive outline of a theory, developed outside of centuries-old debates over distinctions that make a difference, terms that turned out to be confounded, etc.
Human agency and morality extend beyond individual behavior, and into the relationships that we construct and the institutions that we build. Not only does some of what we produce have emergent characteristics — something that already creates problems for Harris’s simplistic “neuron fires / human acts / no one’s responsible” model — but we can even build things that later affect how we act. We can, together, in the context of collective deliberation, design things that incline us to be more or less moral. That remains true even if our individual “decisions” to participate were in a sense unwilled, and if our future “decisions” will come entirely from our brains. Even if each of us cannot be more or less moral, we can together construct more or less moral institutions.
Where this really becomes problematic for Harris is when the institution that we’re building is language itself. Here things get a little bit complicated and precision becomes kind of important. Even if Harris is right that 100 percent of behavior comes from hardwired subprograms activated in the brain, the subprograms that get activated are influenced (technically, they’re mediated) by language. To take a straightforward example: the people in Texas who felt “we” had been attacked on 9/11 were responding to hardwired impulses involving “us vs. them,” “my territory vs. not my territory,” etc. How did this happen? It can’t be that we evolved on the African Savanah to feel kinship to biologically unrelated animals beyond the horizon. Instead, what happened is that the constructed concepts of nationhood managed to trigger fairly primitive impulses in the brain. The brain may not interact with the world through language, but most of the things that hit the brain have to pass through a linguistic screen.
Our ability to tweak and design language, in other words, gives us the ability to shape how our brains shape us. We can coarsen or purify language, we can clarify or confuse concepts, and so on. We don’t need experimental evidence to show that there’s a give and take between the mind and the brain. It has to be that way, simply based on what we know about how language and meaning get in between our senses and our brains. To his credit, Harris certainly recognizes the centrality of meaning, but he doesn’t seem to follow its implications all the way downstream. If he did, he’d have a much harder time pretending that the people who treat free will as a live debate just haven’t thought about it enough.
Which brings us to…
(2) For a guy making categorical statements about the irrelevance of debates stretching back to the classical age, Harris’s style of thought on free will is striking. Some of his blogged excerpts read like the musings of a stoner trying to sound deep in an undergraduate metaphysics seminar: “Why did I use the term ‘inscrutable’ in the previous sentence? I must confess that I do not know. Was I free to do otherwise? What could such a claim possibly mean?”
Harris oscillates between these skeptical poses, meant to convey the impression of intellectual struggle, and absolutist pronouncements about the free will debate.
The entire act is an echo of Harris’s New Atheist jeremiad The End of Faith, which reads like an undergraduate trying to cut corners in an introductory epistemology seminar. Harris insisted that a society’s belief in God is exactly like any other belief, such that when the proposition “God exists” is disproved, it dissolves as opposed to getting replaced by something equally central but more sinister. Not to step on Harris’s rhetorical shtick, but there is no robust modern theory of persuasion — none — where that’s the case. Thought channeled through language simply doesn’t work that way. Bayesian neural nets sometimes do. Humans don’t. That’s one of many reasons we have trouble getting neural nets to learn natural languages like humans, and getting humans to follow through on probabilistic judgments like neural nets.
All too often Harris acts as if he’s closed an argument when all he’s done is stumble into the fundamental difficulties that set the tenor of contemporary debates. This is a veritable tic with some journalists/public intellectuals who insist that their opponents simply haven’t thought enough about an issue. To take a mundane example, liberal journalists often write as if conservatives oppose Keynesian policies because they just haven’t heard about the multiplier effect. What’s actually going on is that modern debates are precisely over the magnitude of the multiplier effect, with arguments on both sides that liberal journalists haven’t gotten to either because they haven’t been reading long enough or because they don’t read anyone outside their echo chamber.
To take an example closer to Harris, those styling themselves as “New Atheists” often write as if there are self-evident contradictions in religious traditions that should make supernatural beliefs impossible. That there are still people who have supernatural beliefs is ipso facto evidence that they haven’t been exposed to those contradictions. Christopher Hitchens is particularly grating in this regard, which is why he got epically dismantled by Ross Douthat.
So it’s not totally surprising that Harris indulges in strutting, dismissive sentences like “I don’t know of anyone who believes that these animals [chimps, dogs, and mice] have free will.” That’s true in one sense and aggravatingly, typically misleading in another. Dogs and mice and probably chimps can’t symbolize, so they don’t have free will in the same philosophical and moral sense as humans. But they certainly might engage in something approaching motivated willful action, and how that relates to the limits or our free will is very much a live debate. Slavoj Zizek certainly thinks that the question of a mice’s “will” might help unlock how humans experience everything from biomedical implants to ideology.
Zizek’s probably overstating the case, and Harris is probably closer to right than wrong, but it’s just weird to hear a critic boast how he doesn’t “know of anyone who” disagrees with him. It makes him sound like a layman dipping his toes into a debate the implications of which experts have been teasing out for centuries. It’s kind of obnoxious.