Friday, July 22, 2005

Quantum Agents versus Mechanistic Handpuppets

'Is there anything more boring than using Heisenberg's uncertainty principle
to try to prove free will?'

-Walker Percy, Lost in the Cosmos

Since my sophomore year in college, I haven't put much store in using contemporary physics theories to establish the freedom of the will. For one, I haven't the advanced knowledge necessary to speak with any authority on quantum physics. I've read Hawking's Brief History of Time, a few other articles written for laymen, and I've had a few basic physics classes. What's worse, it seems to be a major talking point for crappy new age inspirational works, one example being, I am told, the movie "What the Bleep do We Know?" So on the principle that better to remain silent and be thought a fool than to speak and remove all doubt, I simply passed it by. Besides, I hadn't come across any explication of how quantum mechanics could be brought to bear on the question.

That is, until I came across QUANTUM PHYSICS IN NEUROSCIENCE AND PSYCHOLOGY: A NEUROPHYSICAL MODEL OF MIND/BRAIN INTERACTION.

By necessity, it's an interdisciplinary work written by a Berkeley physicist, a neurologist, and a neuropsychiatrist. The paper has been published in the Philosophical Transactions of the Royal Society, which hopefully means it has passed through a crucible of quality peer review.

I sure can't critique many of the arguments invoking the big names of twentieth century physics. However, my concern about this article is that it seems to believe that human influences on descriptions of the world actually shape the world itself. Perhaps this isn't a confusion, but a foundational discovery or assumption of the new physics, but when the authors write "Thus the choice made by the observer about how he or she will act at a macroscopic level has, at the practical level, a profound effect on the physical system being acting upon" I suspect they are blowing something out of all proportion. I can grant that some observations change the observed thing, but I haven't a clue how that works on a cosmic level. If somebody wants to explain to me how telescopic measurement of light a star put out 50,000 years ago affects the star today, please do.

The paper jibes with some recent findings on how mental habits can change even the physical structure of the brain. I believe those studies concentrated on Buddhist monks, and oddly enough the Buddhist virtue of mindfulness pops up in this paper. As soon as Buddhist writings were cited, I feared I had stumbled onto some crappy new age pseudophysics, but the Buddhist remarks were relevant and not dwelled upon.

Anyway, the authors claim that, in contradistinction to classical Newtonian physics, wherein the observer was simply one more part(or sum of parts) in a mechanical universe, contemporary physics incorporates the observer's "inner space:"

The core idea of classical physics was to describe the “world out there,” with no reference to “our thoughts in here.” But the core idea of quantum mechanics is to describe both our activities as knowledge-seeking and knowledge-acquiring agents, and also the knowledge that we thereby acquire. Thus quantum theory involves, basically, what is “in here,” not just what is “out there.”

This philosophical shift arises from the explicit recognition by quantum physicists that science is about what we can know. It is fine to have a beautiful and elegant mathematical theory about a really existing physical world out there that meets various intellectually satisfying criteria. But the essential demand of science is that the theoretical constructs be
tied to the experiences of the human scientists who devise ways of testing the theory, and of the human engineers and technicians who both participate in these tests, and eventually put the theory to work. Thus the structure of a proper physical theory must involve not only the part describing the behavior of the not-directly-experienced theoretically postulated entities, expressed in some appropriate symbolic language, but also a part describing the human experiences that are pertinent to these tests and applications, expressed in the language that we actually use to describe such experiences to ourselves and to each other. And the theory must specify the connection between these two differently described and differently conceived parts of scientific practice.


Happily, the writers do not follow the habit of so many armchair quantum mystics and leave "free will" a vague and cheery phrase. For the purposes of their paper, free choices are free "in the sense that they are not specified by the currently known laws of physics." This is because for the Copenhagen school of quantum physics, "the human experimenter is considered to stand outside the system to which the quantum laws are applied." (In this there is an echo of another Dane, Kierkegaard, who pointed out that the self is always a leftover of one's theory.) I suspect here the difficulties are passed over, which might lead a reader even more casual than I to believe their argument is more solid than it is. In one sense, this can be caricatured as the statement "our system presupposes it, therefore it must be true."

The paper also, alas, invokes the Heisenberg Uncertainty Principle to support free will. as I read them, they hold that nobody will ever be able to completely capture the activity of the calcium ions bouncing between nerve endings, and so determinism is unprovable by scientific means. Obviously, the impossibility of proof for determinism isn't a positive proof for free will. I hope I have misread them here, but it seems they are going about proving human agency, and not human freedom. However, in their defense I believe that they aren't so much arguing for free will as a belief requiring full assent to the belief "this is real," as for a more pragmatic belief "this works." As the psychological section declares, "Quantum physics works better in neuropsychology than its classical approximation because it inserts knowable choices made by human agents into the dynamics in place of unknowable-in-principle microscopic variables." In addition, the paper makes a few predictions about "attention density," a concept I'm not going to touch, the measurements of which they claim should vindicate their position.

Supposedly, quantum mechanics doesn't apply to the atomic level at which the brain works. For what it's worth, the authors of this paper argue against it, claiming neurobiology can be described by "almost classical physics," but not classical physics simpliciter. I'll leave that evaluation to my scientific betters. Also of note, the article touches upon the Libet Experiment, which various popularizers presented as hard evidence for determinism.

Part of their conclusion is worth recording here:

Materialist ontology draws no support from contemporary physics, and is in fact contradicted by it. The notion that all physical behavior is explainable in principle solely in terms of a local mechanical process is a holdover from physical theories of an earlier era. It was rejected by the founders of quantum mechanics, who introduced crucially into the basic dynamical equations choices that are not determined by local mechanical processes, but are attributed rather to human agents.


Their last line reads: "A shift to this pragmatic approach that incorporates agent-based choices as primary empirical input variables may be as important to progress in neuroscience and psychology as it was to progress in atomic physics." In my eyes, an "agent" is not quite a person, but the concept is certainly better off than the marionette theory of the self.

No comments: