Chance and Quantum Mechanics

[I wrote this in February, 1998 to someone who claimed that the failure of quantum physics to specify complete deterministic explanations for events was no more than an argument from ignorance for the reality of irreducible chance events. The current revision (July 2001) incorporates the fruits of discussion with another friend, especially about Bohm-type interpretations.]

The case for thinking there are real, irreducible, chance events is not just an argument from ignorance. I told you where to go to find the evidence and argument on the subject – talk to the physicists. Read a good popularization. Heinz Pagels’ The Cosmic Code is good as is John Gribbin’s In Search of Schrödinger’s Cat (both written, by the way, by people who themselves have doctorates in physics). If you’re up to something more advanced, look at Feynman’s Lectures on Physics.

However, I’ll say a little more than just telling you where to go for information. If there were an ironclad proof that there are no chance events, then it would be reasonable to say that ignorance of causes does not in any way support the claim that there are chance events. But there is no such proof and since there is not, the fact that we cannot find a causal account in a certain area is certainly relevant evidence as to whether there is a (completely deterministic) causal account. Quantum physics generates predictions of probability distributions over outcomes – probability distributions that do not reduce to underlying deterministic relations.

There are deep reasons for this in quantum theory; it’s not just that we haven’t found (or, more hopefully expressed, haven’t yet found) the underlying causal mechanism. In fact, there’s a proof (von Neumann’s) that it’s impossible for there to be underlying, deterministic causal mechanisms unless you deny locality – the assumption that causal influences do not propagate faster than light. (Von Neumann thought he was proving something stronger, that no underlying determinism at all was consistent with quantum mechanics, but he overlooked the fact that he was assuming locality.)

The upshot is that if you want a deterministic interpretation of quantum mechanics, it’s going to have to be non-local. Now, maybe non-locality isn’t a problem by itself – we may have to admit some non-locality anyhow to deal with the experimental results that show violations of Bell’s inequality (the altenative may be some kind of non-realism about quantum phenomena). But what’s on offer if we go looking for some kind of non-local determinism? In essence, only one thing: some version of Bohm’s interpretation.

In the 1950s, David Bohm noticed the locality loophole in von Neumann’s proof, and worked out an interpretation that exploited the loophole to generate a deterministic but non-local reading of quantum physics. Bohm’s version, as he presented it, yields the same experimental predictions as other interpretations (as do some minor variations upon it). Somewhat wilder variations upon his ideas have been worked out that diverge in what they predict. Unfortunately, those variations, whenever they have been tested, have turned out to be decisively in conflict with experimental results. So, for non-local, deterministic readings of quantum physics, that leaves only Bohm’s original plan, or some minor variant, in the field.

Is that good enough? Perhaps. It suffers from some technical problems, such as making apparently arbitrary distinctions between baryons and other particles. But to my mind, there’s a more fundamental problem. Remember that Bohm’s interpretation makes exactly the same predictions as other interpretations. That includes yielding probability-distributions for predicted events. What it does not give is any kind of experimentally accessible causal account. According to Bohm’s interpretation, there is a non-local, deterministic explanation for any quantum phenomenon you care to mention. But because it is non-local, the alleged determining factors are forever and in principle inaccessible to experimental investigation. (They cannot – ever – be brought within a local field of investigation because, whatever could be brought within any local field of investigation would be subject to non-local influences.) So, my question is: Why accept this picture? For anything that can be empirically investigated, it tells us nothing that more standard interpretations do not. Its only warrant seems to be an insistence that somehow the truth must be deterministic. It would only be reasonable to buy Bohm’s interpretation if you already had some strong argument for determinism. In its absence, this looks not much better than an act of faith.

Is there anything left? Very little. Leaving aside Bohmian interpretations, the only way to consistently deny that there are real chance events is to hold that quantum physics is fundamentally mistaken.

But diagnosing fundamental defects in quantum physics doesn’t seem a very attractive option either, since quantum physics is, quite simply, the best tested and most successful physical theory ever. It has been tested over enormous ranges of scale, from the submicroscopic to the size of stars by very bright people who would love to find something wrong with it – since finding something wrong with it is Nobel-caliber work – and so far, over 70-plus years, it has passed every test with flying colors. This includes tests of things that people initially thought were crazy and sure to bring down the theory. But over and over, experiment after experiment, what quantum theory predicts is what gets observed when it’s put to the test.

There’s really not much maneuvering room if you want to deny there are real chance events. That quantum mechanics is fundamentally wrong is a very bad bet – a bit like betting that the flat-earthers are right after all. That Bohm’s interpretation is right is still a pretty bad bet, though not completely beyond the pale. That there are really chance events is a pretty good bet. If there were some proof that the third option was false, that might justify picking one of the others. In its absence, it just isn’t reasonable, assuming you’re aware of the evidence and argument, to resist the conclusion that probably there are real chance events.

Rob
_____
Rob Bass
rhbass@gmail.com
http://oocities.com/amosapient