Feb
13
Determinism is Dead
February 13, 2007 | 18 Comments
Stephen Barr is probably my favorite science writer these days: he is both an authority in theoretical particle physics at the Bartol Research Institute at the University of Delaware, and an orthodox Roman Catholic who believes, as do I, that science and religion are not only compatible in the sense that they do not contradict each other, but they are, in the words of John Paul the Great, the two wings of human experience that together lift us to knowledge of the Good and the True. So I was particularly excited when the March number of First Things appeared in my mailbox today sporting a tidy little essay by Professor Barr entitled “Faith and Quantum Theory.”
The article presents an argument that I find quite congenial, having made a version of it myself over ten years ago (see Robert Brandon and Scott Carson, “The Indeterministic Character of Evolutionary Theory: No ‘No Hidden Variables’ Proof, But No Room for Determinism Either,” in Philosophy of Science 63 [1996], pp. 315-337. The article is available online if you have access to JSTOR. Otherwise, you’ll just have to wait until the movie comes out). The upshot of both his and my articles: Determinism is dead. Not just dead as in, we can’t figure out how to make it work. Dead as in, it’s metaphysically not possible unless you’re willing to commit yourself to a metaphysics that is, in a word, insane.
Sometimes, when I teach the history of Presocratic philosophy, I will have students who ask questions about the metaphysical schemes of the Pythagoreans, questions about the viability of such schemes and about the very motivation behind seeing the world in the way that they did. If you’re not familiar with the Pythagoreans or their schemes, one aspect of their thought that has made them rather notorious is their proposal that the observable world is, in some sense, reducible to numbers or, as Philolaus proposed, known through numbers. To understand this proposal fully would require a rather detailed knowledge of ancient metaphysics and philosophy of mathematics, so I won’t go into more detail here. Suffice it to say that, when I get these questions from my students, I’m often tempted to point out that it is not all that different from what contemporary physicists are prone to do. We, too, represent the world numerically, particularly in the form of scientific equations but also in the form of predictions and explanations that are expressed in terms of probabilities.
Now, if you are a determinist, you view these probabilities as epistemic in nature. That is, if someone tells you that there is a 30 percent chance of rain tomorrow, if you are a determinist you think that what is meant by that is that, if only we knew everything there is to know about a particular weather system, down to the disposition of every last material particle, then we could say with 100 percent certainty whether it will rain tomorrow or not. But we cannot have that kind of complete knowledge of such a vast and complex physical system as the entire atmosphere of the earth, so our prediction reflects a certain amount of uncertainty on our part that is due to our lack of complete information.
There is another way to interpret probabilities, however, and it is much more congenial to both quantum mechanics and evolutionary theory. We may call this the metaphysical interpretation, because it posits that probabilities reflect, not limitations in our own knowledge of the physical universe, but the stochastic properties of a non-deterministic universe. That is, on the metaphysical interpretation of probabilities, we say that there is a certain chance that something will happen not because we do not know for sure what all of the variable are that will lead to one outcome rather than another, but because it is genuinely undetermined, at the time of our prediction, which outcome is going to take place.
Although he does not say much about this in his article, Professor Barr notes that this view of probability is congenial not only for quantum theory, but also for those who endorse the idea that the human will is free. As I mentioned in one of my earlier posts in this series, if the universe is a purely deterministic place, where materialism is the only correct metaphysics, then our wills cannot be free. So if you happen to believe that your will is free, it seems that you have to abandon either materialism or determinism. Some will want to abandon both, of course: the Christian must, in my view. But even if you are wedded to your materialism it might be time to dump that dinosaur, determinism.
We can’t, of course, take it for granted that our wills are free. Part of the point of these explorations of mine is to enquire into whether it is, in fact, possible to make any sense out of human freedom. It is worth emphasizing, however, that unless global determinism is false, we cannot be free. That is a controversial view (it was denied, for example, by Hume, and continues to be denied by the more dogmatic compatibilists), but I will not argue for it, since I doubt that many of my readers will find it all that problematic. In these matters, much depends upon one’s starting points.
Comments
18 Comments so far
My copy of FT just arrived–and Barr’s was the first article I read. Have you read his book, Modern Science and Ancient Faith? Superb.
“As I mentioned in one of my earlier posts in this series, if the universe is a purely deterministic place, where materialism is the only correct metaphysics, then our wills cannot be free.”
I can’t see why. Could you please give a reference to your earlier post so I can see your argument?
John–
Haven’t read that one yet, but I’ll have to have a look, I keep hearing great things about it.
Martin–
I can do better than that: I’ll limb the argument for you right here.
If materialism is true, then we as human beings are, body and soul and everything, reducible to material components. That means that all properties and processes involved in our experience are reducible to properties and processes of matter.
If determinism is true, then all material properties and processes are determined by the sum total of all properties and processes of matter combined with all the physical laws governing the state of the system under consideration.
So, if materialism is true, then all of our mental states–thoughts, beliefs, choices, preferences, emotions, etc–are nothing more than the interactions of (material) neurotransmitters and (material) neurons, and if determinism is true all of these interactions will be determined by the physical properties, processes, and laws that govern them. We will have the illusion of freedom, of course, because naturally when we make a choice it feels as though we could have chosen otherwise, but this is just an illusion, an artifact of our capacity to reason and imagine things being otherwise. When I go to the kitchen for breakfast, I might choose Weetabix over Corn Flakes and feel certain that I could have chosen Corn Flakes instead if I had really wanted to, but this feeling is nothing more than a feeling. Indeed, the fact that I wanted one over the other was itself determined by things beyond my control, and my deciding to choose what I wanted rather than something I didn’t want was also determined by things beyond my control.
The Christian, then, cannot be both a materialist and a determinist, and I have maintained in some of my other posts that the Christian cannot be a materialist. That leaves only determinism, but I think the Christian has reason to abandon that, as well, though the reasons are principally philosophical rather than theological.
I’m a researcher in the physical sciences and a philosophy neophyte. No- thats actually not fair to neophytes. Case in point: I’m even having a hard time plowing through a very introductory book by Ralph McInerny on Aquinas which I bought as a prelude to reading Barr’s “Modern Science, Ancient Faith.”
So with that background, can you clarify something for me that wikipedia could not:
I don’t get the distinction between materialism and determinism. I mean, if you are a materialist, then you should believe (I think) that given the proper equation, and initial conditions (the state of the universe) at time t, you should be able to determine what the state will be at t dt. But that sounds like determinism.
Neither seem in anyway compatible with Christianity.
“I mean, if you are a materialist, then you should believe (I think) that given the proper equation, and initial conditions (the state of the universe) at time t, you should be able to determine what the state will be at t dt. But that sounds like determinism.”
That would assume that the relationship between the material (which is all that exists) is deterministic. But there are plenty of models of material interaction that are non-deterministic, quantum mechanics being the most glaring example.
God bless you for this post, Dr. Carson. What you say here needs to be repeated early and often. Incidentally, have you read Anthony Rizzi’s The Science before Science? I consider it an excellent primer in how to think straight about reality, and I was wondering if you had an opinion about it (or a parallel recommendation).
Fletch–
I think you’re right, actually, that the Christian ought not to accept either. Certainly materialism is out, and determinism may be as well, but it’s not as clear.
Materialism is, fundamentally, a metaphysical stance, that is, it is an attitude regarding what sorts of things exist. Briefly, it asserts that nothing exists apart from matter (and energy, in the modern version of the view). There is no argument in support of either materialism or anti-materialism–they are both a priori commitments (though the adherents of materialism usually either do not realize this or are unwilling to admit it). Materialism is attractive to empiricists, because on their view the only things we can have knowledge of are those things that we can test empirically, and only matter can be so tested, since only matter can serve as the proper object of sense perception. Since a Christian is committed to the existence of God, a non-material entity, he cannot be a materialist; since the Christian believes that it is possible to know that God exists, he cannot be an empiricist, either (though Bas Van Fraassen is one rather famous example of an empiricist who is also a Christian).
Determinism is not a metaphysical stance, but a methodological one. That is, it is a view about how to understand and test causal processes. Your description of it is not a bad one. The Christian, I think, can be what we call a local determinist: he can regard certain closed systems (such as all purely physical systems above the level of quantum particles) as deterministic. Global determinism, however, will have to be regarded as false.
A Christian can also adopt a kind of local materialism: he can regard science as a kind of language game that assumes materialism and then he can work in a scientific field using both local materialism and local determinism as purely hermeneutic tools for making limited claims within the domain of his science. As long as he does not adopt a global version of either view as a definitive view of all of reality, he will be OK.
I hope that clarifies things at least a little bit.
I think you meant “limn” rather than “limb” but I don’t know American very well so I may be wrong.
I see that you think what I thought you were thinking. This is interesting: thank you.
Martin
I did mean “limn”. I’d like to be able to blame Americanisms or even my automatic spelling corrector, but I’m afraid I just hit the wrong key! I’m probably the world’s worst typist. I can at least say that in American the two words sound the same, as I assume they do in proper English as well.
Jonathan
Thanks for the comment, and the reference. I haven’t seen Rizzi’s book, but you’re not the first person to recommend it, so I think I’d better have a look. I’d like to be able to recommend something in return, but most of what I have to read in my line of work is stuff that I can’t really endorse. As I think I’ve mentioned, I do very much like Stephen Barr’s work, so I can recommend that, but I’m not sure whether it’s on the sort of scale that you have in mind.
Thank you Jonathan and Scarson for the clear and patient answers. I realize now that my view is clouded by the fact that I work in Newtonian continuum mechanics and am just the type of “local materialist” scarson described. While I appreciate that materialism is metaphysical and determinism is methodological, it sounds as though, the distinction between the two is still due in no small part to the probabilistic nature of quantum physics. Were it not for this indeterminancy, I can’t imagine how a purely material universe would not by definition also be deterministic. Again, being a Newtonian guy, I’m not familiar with models of material interaction (other than quantum mechanics) that are not deterministic.
I realize it was kind of selfish to use such a thoughtful post to clarify a 101 level question. I’ll go back to lurking in admiration now!
Fletch
You’re right that if we restrict the domain to a purely physicalist system, then determinism would appear to be a necessary feature of that system, locally.
There is an interesting book by R. I . G. Hughes, The Structure and Interpretation of Quantum Mechanics (Harvard University Press, 1989) that deals with the question of whether quantum indeterminacy can “percolate up” into the macroscopic level of physicalist systems. One of the arguments of the publication of mine that I mention in the main post is that these quantum level indeterminacies may have an effect at the level of biological populations. If this is true then determinism turns out not to be a necessary feature of all purely materialist systems. You might want to check out both the book and my article (I think they’re both fairly accessible).
All questions are welcome here, by the way!
I hate it when I type homophones by accident, so it was unfair of me to rub your face in it!
I thought that the evolution of the state of a quantum system was always deterministic; and that this applied even in the case of measurement, as long as you recognised that the system whose state was evolving was {apparatus experiment} and not {experiment}. Is this wrong?
Bohmian models of quantum theory are AFAIK wholly deterministic. They may be unfashionable or even repugnant but they are not actually inconsistent with the theory: how, therefore, can the theory be said to entail indeterminism if it has deterministic models?
Using computers, it is possible to simulate all sorts of systems with all sorts of laws. Many of these simulations are deterministic: for instance, the merging of two galaxies of 10^9 stars can be simulated by calculating the gravitational interactions of all possible pairs of stars. It is, of course, possible to use laws that relate to our own laws of nature or to use any other laws you like. Depending on the problem domain and the chosen laws, it is possible for “creatures” to emerge (galaxies, in the gravitational case; something more biological if you had more biological laws). Are you saying that, even with unlimited computing capacity, no set of laws that we chose to simulate could ever lead to the emergence of what could be considered (in the simulated world) to be “creatures” complex enough to have intelligence? I think that you have to say this, and it is an interesting and strong thing to be saying.
Martin
The work of John Bell in the mid- to late-sixties showed quite conclusively that one of three things must go: (a) quantum mechanics itself; (b) determinism; (c) locality. Most physicists are unwilling to give up either (a) or (c), though there are some who would rather give up (c) than (b). My own view is that determinism is merely an a priori assumption in the first place, so why not give it up?
The Bohmian model is discussed by Barr in his piece in First Things, as is the so-called “many worlds” interpretation. Barr favors a modified version of the Copenhagen interpretation.
Jonathan,
Can you say something about Rizzi’s view on time travel?
I’m not sure whether it’s on the sort of scale that you have in mind.
I suspect it will be. At present, I am looking for even a handful of books to refute this annoyingly persistent “metaphysics is dead” idea that seems to permeate society. I’d like to have at least a handful of good books useful for clearing the cognitive cobwebs about just how much metaphysics must be taken for granted simply to be able to say what they say so glibly (and foolishly).
Can you say something about Rizzi’s view on time travel?
I think he might be one step shy of the level of abstraction the subject requires (I have a similar notion for his idea of rest frames). Given that angelic existence allows an atemporal-yet-causal sort of existence, it seems needlessly rigid to say that something like retroactive causation cannot take place. He might be right, but I am unconvinced that he is necessarily so. He seems to have glossed over some distinctions between spiritual and material substances, and he seems to assume matter as the principle of individuation, which was probably the weakest area of St. Thomas’s metaphysics. Given the complications of matters like aionic time (the time metric for angels, saints in Heaven, etc.), I think it might be too reductionist to restrict what can and can’t happen in time to material sorts of change.
Jonathan
Doesn’t final causality count as a kind of backwards causation, though not, strictly speaking, temporally retroactive?
Doesn’t final causality count as a kind of backwards causation, though not, strictly speaking, temporally retroactive?
Yes, that’s a good example. When you start looking at how different orders of causation interact at the intersection of the natural and supernatural, it doesn’t seem to follow quite the same relational rules regarding the transformation of act into potency. There’s a similar degree of confusion with predestination; if you take the analogy of literal motion from act to potency too far, then you can end up with an erroneous sort of Augustinianism. Likewise for prophecy; there is clearly some sort of definiteness about the future that it not in itself causally determinative in the sense of motion. For that reason, I don’t think that time travel would be impossible on metaphysical grounds simply because it might involve some appearance of circular causation. Otherwise, it would seem that God’s miraculous action in time would be metaphysically impossible, and we know that is not the case.