Did Evolution Give Us Free Will?
If you pick up a book up about free will by a materialist neuroscientist, you are generally safe to assume that the point of it will be to explain that free will is merely an illusion — that we are actually at the whim of the blind forces of Nature, and are therefore not responsible for our actions. So it’s surprising and somewhat refreshing to see a self-proclaimed naturalist defend free will. That’s what Trinity College Dublin neurobiologist Kevin Mitchell sets out to do in Free Agents: How Evolution Gave Us Free Will.
As Denyse O’Leary has noted about the book, the scientific debate over free will seems to be reviving a bit, with another book by another prominent scientist arguing the opposite position released the same year (Determined by Robert Sapolsky). So after reading Mitchell’s book, I thought it would be worth digging into the details of his argument a bit for Evolutions News readers.
Does the book succeed? By my assessment, no and yes and no. There are really several different questions at play here: Do we have a will? Is it free? Did evolution give it to us? And if so, how? Each of these subjects has its own set of scientific and philosophical difficulties, and the book is not equally persuasive on every point. To keep the various strands of the argument straight, let’s go in order, following the subtitle. We’ll start with “how evolution gave us”…
“How Evolution Gave Us…”
Anyone expecting a defense of the claim that Darwinian processes can or did create complex neurological systems will be disappointed. That’s not the point of the book. With a very few exceptions1, Dr. Mitchell works from the tacit assumption that (a) there is no real limit to what Darwinian processes can achieve, and (b) that anything that exists in biology must have arisen through Darwinian processes. That means the book is largely concerned with describing what exists in nature, with “evolved” acting as a synonym for “is.”
Thus, phrases like “mechanisms evolved” prevail throughout the book. Complex systems are simply “built” or “invented” or even “designed,” without much concern given to the concrete details or the relevant engineering problems. The following passage is typical:
More complex creatures emerged, colonizing and creating new niches, with expanded repertoires of possible actions. A system was then required to coordinate the movement of all the organism’s constituent parts and select among actions. Muscles evolved, along with neurons to coordinate them, initially distributed in simple nerve nets. As evolution proceeded, the nervous system became more complex, linking sensory structures to muscles via intervening layers of interneurons. The meaning of signals became disconnected from immediate action, giving rise to internal representations…
In all fairness, Mitchell presumably did not set out to defend Darwinian evolution against other possible explanations. The heart of Free Agents is not really in explaining how we evolved to be what we are, but rather in simply describingwhat we are, according to the cutting edge of neurobiology. That’s where the book shines.
“Free”
One view of free will, called “compatibilism,” maintains that materialistic determinism and free will are really compatible. This position is apparently quite popular in philosophy of mind circles, and has been argued by Daniel Dennett and other famous philosophers. The argument says, first, that it doesn’t matter if an organism “could have done otherwise” — what matters is that the organism is the source of the action. That is, we can reasonably be said to have free will if we are able to do what we want, even if we are not able to want what we want. Second, compatibilists point out that organisms and their environments are so complex that there is no way, even theoretically, to predict what an organism will do in a future situation. So for all practical purposes, we are free.
Mitchell finds these arguments unconvincing. They seem to be saying that if we just change our perspective, or our definitions, the problem will go away. “But I cannot escape feeling that some sleight of hand is part of this line of argument,” he writes. “It feels as if some (presumably unwitting) misdirection is going on — as if the primary problem has been circumvented or even denied, rather than confronted.” Instead, there ought to be some genuine indeterminacy in the system, or else “no matter how complex, the agent will be pushed around deterministically by its own components.”
I think the “sleight of hand” Mitchell senses is the confusing of epistemology with ontology: confusing what can be known with what is. Regardless — Mitchell argues that the fuss is unnecessary. There is really no reason that free will needs to be compatible with strict determinism, he says, because physics, as it turns out, is not strictly deterministic. That requirement is a relic from a bygone era, when everything seemed to move inexorably according to simple Newtonian laws. Most modern quantum physicists, in contrast, agree that particles seem to actually have a degree of freedom or true randomness to their movement. So, Mitchell says, “there is nothing in the laws physics that rules out the possibility of agency or free will, a priori.”
In fact, various studies seem to show organisms acting in a non-deterministic way. In one fascinating experiment, an electrical probe was attached directly to a leech’s central nervous system, allowing the experimenters to bypass the complexities of environment altogether and administer the exact same stimulus, repeatedly. Even under such perfectly controlled conditions, there seemed to be no way to predict how a leech (like the one pictured above) would respond to the stimulus each time.
This apparent indeterminacy scales all the way up to more complex behaviors and situations, resulting in what is known as the Harvard Law of Animal Behavior: “Under carefully controlled experimental circumstances, an animal will behave as it damn well pleases.”
So Far, So Good
But what about the experiments that seem to show the opposite, that free will is a mere illusion?
There are quite a few famous experiments of this kind, but in Mitchell’s professional opinion, they show nothing of the sort.
For example, Benjamin Libet’s now-famous 1983 experiment showed a signal called a “readiness potential” in the brain a fraction of a second before the subject was conscious of choosing to move his hand. Many have taken this to be definitive proof that free will is only an illusion: at the moment we think we are freely choosing, the brain has actually decided beforehand.
Mitchell writes that this interpretation is “to put it mildly, a drastic overinterpretation”:
That is because the design of the experiment makes it effectively irrelevant for the question of free will. The participants made an active and deliberate decision when they agreed to take part in the study and to follow the instructions of the researchers. Those instructions explicitly told them to act on a whim: “to let the urge to act appear on its own at any time without any preplanning or concentration on when to act.” They had no reason to want to move their hand more at one point than another because nothing was at stake. And so, it seems they did indeed act on a whim: they (decided to) let subconscious processes in their brains decide, by drawing on inherent random fluctuations in neural activity.
This is what a different group of neuroscientists, led by Aaron Schurger, concluded from analyzing the data from the original experiment — that the test subjects had (instinctively, of course) set a certain potential level of neuronal activity, deciding that when random fluctuations in the brain reach that level, they would take the proscribed action.
So now you have two plausible interpretations of the data.
But Which One Is True?
Another experiment, led by Uri Maoz and Liad Mudrik, sought to distinguish between the two possibilities. The researchers gave half the test subjects a decision with no serious consequences, and half a decision with consequences that they cared about. Sure enough, when the subjects were given inconsequential decision, a readiness potential preceded the decision, as in Libet’s experiments. But when the decision mattered, no readiness potential was detected.
“Overall then,” Mitchell writes, “Libet’s experiments have very little relevance for the question of free will. They do not relate to deliberative decisions at all, where readiness potential is not observed. Instead, they confirm, first, that neural activity in the brain is not completely deterministic and, second, that organisms can choose to harness the inherent randomness to make arbitrary decisions in a timely fashion.”
So much for “free.” We’ll examine what Mitchell has to say about “will” tomorrow.
Notes
1.E.g., Mitchell mentions that the now-classic view that symbiosis might have been necessary to make the switch from prokaryotic to eukaryotic life.
No comments:
Post a Comment