Your Fancy-Pants Brain

James Gleick opens his essay from our January 18 issue with a bang: “Nobody was holding a gun to your head when you started reading this. You made a choice.” From this playful premise, he examines where philosophical questions of free will converge with science, making the case for human agency by teasing out which physical processes give rise to volition. He quotes the neuroscientist Kevin Mitchell: “The universe doesn’t have purpose, but life does.”

Gleick is the author of seven books, including biographies of Isaac Newton and Richard Feynman, an introduction to chaos theory, and a history of information, which won the 2012 PEN/E. O. Wilson Literary Science Writing Award. Since 2008 he has written twenty articles for the Review, on subjects ranging from dictionaries to death. He was also one of the inspirations for the character Ian Malcolm in Michael Crichton’s Jurassic Park.

I wrote Gleick last week to ask him about chaos theory, how to test for free will, and whether we’ll know if AI has gained consciousness before it’s too late.

Daniel Drake: Early in your essay, you note that Kevin Mitchell employs a “complex-systems perspective” in his thinking about biological organisms and free will, as opposed to a “reductionist” approach, like that taken by physicists or neuroscientists who focus on smaller or more fundamental parts and processes. Could you explain a little what is meant by “complex systems”? I have an intuitive sense of how it contrasts with reductionism, and of what it would mean to try to model highly contingent, multiply determined processes, but is there a more concrete scientific definition?

James Gleick: Complex-systems theory arose alongside chaos theory as a discipline unto itself in the 1970s and 1980s. The climate is a canonical complex system, as are the brain, the global economy, immune systems, food webs, epidemics, social networks—entities made up of many different parts, interacting and interdependent. Complex systems can behave in chaotic and unpredictable ways, yet they reveal patterns and structures that carry over from one discipline to another.

The more traditional reductionist program in science looks for explanations at the deepest or smallest levels, where fundamental particles might reveal basic laws. The great Steven Weinberg, a proud reductionist, felt the rise of complexity theory was a rebuke to his specialty, particle physics. “There is a low-intensity culture war going on,” he wrote in this magazine two decades ago, “between scientists who specialize in free-floating theories of this sort and those (mostly particle physicists) who pursue the old reductionist dream of finding laws of nature that are not explained by anything else, but that lie at the roots of all chains of explanation.”

What vindicates the complex-systems people, as I see it, is the recognition of “emergent behaviors,” unexpected phenomena that arise from seemingly simple interactions. Organized strategies for foraging and defense emerge in ant colonies, demonstrating a collective intelligence that can’t be found in individual ants. Emergent behaviors are seen in traffic patterns, computer networks, and, of course, brains. All the fun stuff, in other words. Mitchell’s rebuke to Weinberg would be, you can’t reduce biology to physics. A complexity theorist says that even when you learn everything there is to know about neurons and synapses, you’ll still fail to capture what’s really going on when we think, feel, or act. You can write down all the equations of fluid dynamics and still be surprised by a tornado.

You convincingly cast doubt on the Libet experiments, which used electrodes attached to subjects’ brains to discover neuronal activity in advance of any awareness of a decision. Do you know of an experimental design, or even a thought experiment, that might more convincingly tell the difference between free will and determinism?

Here’s an experiment that proves we don’t have free will—no electrodes required. Arrange the subject in a seated position with legs dangling freely. Place one hand behind the knee, and with the other hand use a little rubber mallet to give a sharp whack just below the kneecap. Expected result: the subject’s leg will kick forward, willy-nilly. This is the automatic reflex that goes by the name “knee-jerk reaction.” Your quadriceps muscle feels the jolt and signals some motor neurons, by way of the spinal cord. Your fancy-pants brain doesn’t enter into it.

Most of us have our patellar reflex tested this way as part of a regular check-up. To me it always feels a bit uncanny, as if I’m somebody’s marionette. My knee kicks out and I think, Whoa! Who did that? The free-will deniers say we’re all somebody’s marionette. The problem with designing an experiment to prove otherwise is that we’re already running that experiment, more or less continuously. For example, try the following: look at your watch, and in fifteen seconds raise your hand—either your right or your left. I’ll wait…. Did you raise your hand? Or did you decide not to bother? Either way, you just exercised free will.

Now, a physical determinist might say that whether or not you raised your hand was merely a matter of elementary particles moving this way and that, through your brain and your muscles, obeying the laws of physics, all part of a grand machine that has been running since the Big Bang. If my typing these words led, some days later, to your reading them, followed by your hand going up, there was still an unbroken chain of particle interactions all along the way. If the universe keeps surprising us, the determinist says it’s because we haven’t done the math.

You may have noticed a flaw in my so-called “experiment”: the only person who knows for sure that you freely chose whether to raise your hand is you, looking inward, examining your own thoughts. An observer has to take your word for it. Scientists rightly mistrust evidence from introspection. When we look into our own minds, the view is notoriously cloudy. For that matter, an AI-powered robot could certainly be programmed to raise its hand, or not, on command.

How might we be able to tell if or when an AI device or system has started striving or planning?

They might tell us themselves, when the time comes.

In our newest issue, Meghan O’Gieblyn writes about William Egginton’s book The Rigor of Angels: Borges, Heisenberg, Kant, and the Ultimate Nature of Reality, and a tendency Egginton calls “metaphysical overreach”: “the problems and paradoxes that emerge when we try to extend our ordinary way of seeing the world beyond the human scale.” There are a number of interesting convergences between your essay and O’Gieblyn’s—in particular time, as the thrust of her essay concerns continuity and discontinuity of knowledge and the self over time. Could you expand a little on the comments you make in your essay about how free will and time are linked?

Long before physicists started telling us that free will was an illusion, theologians wondered how our feeling of freedom could be reconciled with divine foreknowledge. Augustine addressed God this way: “In the sublimity of an eternity which is always in the present, you are before all things past and transcend all things future, because they are still to come.” A God outside of time, omniscient, seeing past and present in a tapestry already woven, doesn’t leave us much room for choice.

We mortals can aspire to that godlike perspective—“that purely transcendent, objective vantage outside the totality of things,” as O’Gieblyn puts it—but she eloquently reminds us that we can’t have it. We live in time, and when we forget that, we fall into metaphysical traps. “Problems and paradoxes” indeed. “Weird things happen,” she says, “when humans try to cast off the shackles of linear time and assume the vantage of eternity.” That’s what certain physicists do when they imagine the universe as a four-dimensional space-time continuum, in which past and future are interchangeable. They replace the all-seeing God with the laws of physics, eternal and immutable.

I read O’Gieblyn’s essay as a welcome plea for humility. She quotes Heisenberg: “We have to remember that what we observe is not nature in itself but nature exposed to our method of questioning.” We can construct powerful mathematical models and use them to great effect, but the models are not the universe (as Borges loved reminding us). Her paraphrase of the sixth-century philosopher Boethius strikes me as particularly apt: “We are free because any knowledge that exists outside of time is irrelevant to the very nature of choice, which must necessarily take place in time.” As long as the future is still to come, our choices might matter.

In the last five years alone, you’ve written for us about the moon, conspiracies and misinformation, Stephen Hawking, clocks, Buckminster Fuller, invisibility, and, now, free will. How do you choose—or will, as it were—what subjects to write about?

That sounds like quite a hodgepodge, when you list them that way. My books have also been fairly wide-ranging, but I’m aware of certain themes connecting them. Vaguely aware, anyway. I don’t think I’m a reliable analyst (see above re: introspection). Clearly, I’m drawn to frontiers of science, though I’m no scientist myself. I’m a sucker for technology, and I mistrust it. I have a liking for paradox and for unexpected connections between things.