There is actually a nice little book, Quantum Mechanics and Experience, that discusses the difficulties in quantum mechanics related to measurement. It gets to the gist of the issues without being very mathematically demanding of the reader. It was recommended by Richard Muller to his class Physics for Future Presidents.

Moreover, the author of the book is a (gasp) philosopher, David Z. Albert. Apparently Albert had a lucid moment, and setting aside the penchant of some philosophers for obfuscation, wrote a fairly clear little book. Neither "noumena" nor its modern father, Kant, seem to play a role in the book. Albert seems to have maintained contact with science and with reality.

It turns out that the Bohmian approach is coming under fire in this regard. You must first recognize that the Bohmian approach has no one specific version, and has not really been worked out. It is more of an idea than an exact theory. A number of people on working on it, but it is not exactly mainstream at this point.

However, some of the main elements of the program can be tested. For example, see:

A first experimental test of de Broglie-Bohm theory against standard quantum mechanics (2002)

This excludes many Bohmian type theories. There have been other papers as well. Naturally, the Bohmians reject these experiments. But do far, EVERY proposed theory which gives "almost" identical results to Standard Quantum Mechanics (SQM) has been experimentally rejected. So only Bohmian theories which give 100% identical results to QM - and there may not be any of these possible - can currently claim to be feasible.

The point is: when you look at the details of a full Bohmian theory, it is always different than SQM in some minor respect.

__________________
The map is not the territory.-Korzybski

If a theory is formulated so as to agree with standard QM 100% of the time, is not in fact standard QM ? It seems to me if it provides the same predictions then it is the same theory, even if you paint it chartreuse.

Yes, this is my objection too-- it's a fine thing to strip a theory of all unnecessary elements and still have it make the same predictions, but it's something quite different to add a whole bunch of extraneous bells and whistles just to make it sound like classical mechanics, but make no different predictions for all the effort. So if you take a version that changes none of the predictions, it is pretty pointless, and if you take a version that does change the predictions, it fails experimentally. Not a lot to recommend it, on the whole.

I respectfully disagree:

The more stripped down version of two empirically equivalent theories is not necessarily preferable. Consider a grand master-level chess-playing computer. Your job is to predict its next move. One can treat it as a black box and predict its moves based on chess strategy, or one can take it apart and attempt to reverse engineer its circuitry and decompile the software. The preferable approach depends on one's purposes. But taking the former approach does not prove that there is no rich inner life taking place inside the box.

There is an analogous situation in evolutionary theory: gene centered selection theories are basically empirically equivalent to selection theories that emphasize hierarchical selection involving individual organisms and groups of organisms. But the gene selection theories treat organisms and groups as black boxes and are (rightly, in my view) derided as mere "bookkeeping" models that don't capture the full richness of what is actually going on in nature.

Similarly, before much progress in neurology was possible, behaviorism of the B.F. Skinner variety was a popular approach to the study of the human consciousness. Behaviorism treats consciousness as the product a black box and cares not at all how particular behaviors are produced; behaviorism was merely concerned with predicting future behavior. Needless to say, treating the human skull as a black box is not very philosophically satisfying.

The Copenhagen interpretation of quantum mechanics treats the Heisenberg limited regime as a black box; that is, it is a "behavioristic" theory of physics--it's mere bookkeeping in other words. Unfortunately, because of the HUP, there is no prospect of directly looking inside as is possible with modern neurological techniques.

So the question is, does treating the quantum realm as if it were a black box imply that the quantum realm really is a black box? Empirically, the answer doesn't matter. Technologically, the answer doesn't matter. But philosophically, the answer is very important.

Indeed, what does it really mean to say that the quantum realm is a black box and that nothing more can be said? For one thing, such a position is self-defeating and fundamentally mysterian and therefore anathema to the proper spirit of scientific enquiry.

For another, it implies that the quantum realm is unlike anything else in experience: it implies that the quantum realm is naturally indeterministic and apparently partless. Indeterminism and partlessness maybe mathematically simple to model, but they are ontologically huge bells and whistles--it's not at all clear to me at least such an explanation is the more parsimonious.

The Bohmian approach multiplies entities, to be sure, but these at least are not ontologically distinct from ordinary tables and chairs. The "hidden variables" are reassuringly deterministic and imply that there are parts at work, even though the parts are beyond our ability to resolve.

Warren wrote:
I will agree that parsimony is a tool, not an absolute rule. But I must ask why different realms of experience must be assumed to follow the same rules, that the rules are scale-invariant (one that comes to mind is symmetry-breaking in the fundamental forces as a function of temperature/energy).

My own wonder-switch is flipped by flipping a coin a thousand or two times and watching the accumulated runs of heads and tails approach the binomial distribution. No matter how you try, as long as you don't cheat the result is the same, a reproducible pattern arising from individually indeterminate outcomes. I think the most I've done is about two thousand.

Now, if you were to argue that the individual flip is in principle predictable if you had a perfect understanding of the masses and forces involved, I suppose we would be back to the origin of the discussion. Or I could use my double-headed quarter.

__________________
If they can get you asking the wrong questions, they don't have to worry about the answers.

Of course you disagree. That is completely predictable, and therefore demonstrably not dependent on quantum mechanics.

If there is a rich inner life then presumably that rich inner life is reflected in the strategies and specific responses to board positions that result. Hence a prediction predicated on an understanding of the inner workings would be more accurate than a "black box" prediction. In that case the two models not produce exactly the same predictions and they are in fact different models/different theories.

It is generally an understanding of the "rich inner life" that permits one to construct a more precise theory than one constructed in absence of that knowledge. In practice one cannot usually find the transfer function for the black box without knowing quite a bit about what goes on inside, enough to construct a detailed model of it.

That is why thermodynamics identifies state variables, why the explanation of classical thermodynamics by statistical physics is important, and why Kalman's state space approach to control theory was such a major advancement over the older transfer function approach of servomechanism theory. Science recognizes the advantage of understanding what happens inside a black box and most certainly is not satisfied with black boxes.

Your characterization of physicista as being satisfied with black box approaches is quite far off the mark.

Evolutionary theories are notoriously non-quantitative, but gene centered theories offer real hope of producing truly quantitative models. I simply cannot believe that black box bookkeeping models can be relied upon to produce 100% agreement with theories based on fundamental science.

Here you have clearly departed from any sort of quantitative models at all. If you think this example is relevant, then please provide and explanation of how these models are in the 100% agreement that was assumed in the post with which you disagree.


What is the "Heisenberg limited regime" ? And what in the hell does that have to do with "modern neurological techniques ?

Quantum mechanics is no more treated as a black box than is gravity in either Newtonian theory or general relativity. Science has never tried to explain WHY things work the way they do, only to provide an explanation with predictive power as to HOW they work.

Nearly every scientist would be beside themselves with glee if a deterministic theory could be devised to replace quantum mechanics. Unfortunately the best available experimental data suggests that it is impossible to construct such a model. This has nothing to do with parsimony. It has everything to do with accuracy.

The quantum real does indeed seem to be "unlike anything else in experience". Nobody particularly likes it. But that is what the data says.

"There was a time when the newspapers said that only twelve men understood the theory of relativity. I do not believe that there ever was such a time. There might have been a time when only one man did, because he was the only guy who caught on, before he wrote his paper. But after people read the paper, a lot of people understood the theory of relativity in some way or other, certainly more than twelve. On the other hand, I can safely say that nobody understands quantum mechanics." – Richard P. Feynman in The Character of Physical Law

No one has said that nothing more can be said about quantum theory, or any other physical theory. That is why there continues to be research in physics.

I am sure that the models are not clear to you. You seem to specialize in the unclear in your approach to philosopohy. Scientists have the opposite charter -- to be quantitatively accurate and as clear as possible.

If and when hidden variables can be shown to provide any additional insight in the form of new and verifiable predictions or even simplifications of existing models I am sure that they will be accepted. However, I believe that it has been shown that hidden variables will not do the job -- I would love for someone to prove this statement wrong by producing a superior predictive model for quantum behavior.

To that I would add, the search for a "rich inner life" is quintessentially nonscientific, because in science, a rich inner life is something we see if we see it, but we never search for it. The search is for just the opposite-- a replacement of the rich inner life with a simpler set of rules that produce the same objective behavior. The latter is what science is really all about, even in regard to life itself. It is all we can hope to add to the situation-- we already have the rich inner life. We don't do science to be "reassured" that everything makes sense, we do it to find what we can make sense of, but we have to try hard not to beguile ourselves into simply retrofitting preconceived notions to achieve a sense of "reassurance"-- that practice comes under a different name.

Sure about that?

Actually, I agree. The Bohmian approach is only one sign of dissatisfaction with the Copenhagen, "black box" theory. Michio Kaku himself said on the radio the other night that the philosophically abhorrent "Many-Worlds" interpretation is fast becoming the consensus among physicists these days. Of course, I don't buy into that idea; but it does point to systemic dissatisfaction with mere bookkeeping.

The hope for a truly quantitative model of evolutionary biology is as forlorn as the hope for a truly quantitative model of the helium atom.

The Heisenberg limited regime is that physical regime where our perception is fundamentally limited by the Heisenberg Uncertainty Principle. The disanalogy with neurology is that the human seat of consciousness has become more open because of modern neurology techniques.



You have a point here.But here you falsely generalize an idiosyncrasy of physics to the rest of science.

Now you are overreaching. Bohmian theories are empirically equivalent. The real question is whether the Copenhagen interpretation is in fact more parsimonious; my point is that it is not.

Not a very reassuring view, I'm sure you'll agree.

HAHA! Great quote sir! But it only reinforces my point that there is no understanding a black box. It might be possible to predict the behavior of a black box; but the black box as such cannot be understood.

Yes, the kettle is black.

That statement cannot at this time be proved wrong; on the other hand, it has not been proved correct either--not by a long shot. If it were, the Many Worlds view would never get traction.

This view has been a disaster for who have had to suffer live vivisection by humans who denied their rich inner lives. If the a priori, Bayesian, discounting of rich inner lives in biology has proved to be a disaster, why should we expect that the physical sciences should be any different.

That in fact is exactly what we do. We do science in order to reassure ourselves that our faith that the world is in fact orderly, predictable--and therefore controllable--is justified.

For sure, that's the question. My point is pragmatic--the tried and true categories that have served us so well in the macroscopic regime are not to be given up lightly. If they can be retained, they should--even if that entails considerable mathematical cost.

There is nothing else for the scientist to study but behavior.

Ken brings his umbrella when it is cloudy because he too often got caught in the rain. That is an observation of behavioral relationships readily visible in the environment. A study of the neuronal fireworks that go on in Ken is simply a study of more behavior--in this case the behavioral relationships among the neurons and the environment. (Note too that without understanding the "outer" environmental behavior, the "inner" neuronal behavior would be quite meaningless.)

There is nothing special about the inner behavior such that the easily observable behavior degrades to "mere bookkeeping."

By your own account, if you discover any sub-quantum behavior, it too will be just more black boxes. You will always be philosophically dissatisfied if you always discount what you do observe and use to your advantage in favor of what may be hidden beneath the surface boundary of what you happen to be observing.

Doesn't quantum mechanics allow us to make the most precise predictions of any science? "Indeterminism" doesn't seem to be an apt description.

Quite true. Thanks for the reminder.

What we're really talking about is reducibility. If one successfully "reduces" outer behavior, does that imply that we can eliminate talk of outer behavior? Not necessarily. One can explain the behavior of a bomb as a whole by describing the behavior of its parts. Yet the capacity to explode is a property of the whole that none of the parts possess. Nothing "mere" about that! My point is that we shouldn't be content with surface behavior, especially if there is a research program that promises to uncover the inner behavior.

Really, the question boils down to this: since Bohmian approaches cannot be proved wrong, should we then defund Bohmian physics?


Very good point, again, Joe. I certainly don't want to say that it's turtles all the way down. The question is whether the fundamental limit just happens to coincide with our human limitations or whether there are still perhaps one or two levels yet to go.

Indeterminisms can be quite precise, as Mike's coin flip example above demonstrates.

Warren wrote:
I would only say that this is where we may be using the same words to mean slightly different things. If we accept (just for the moment, no gotchas intended) that a coin flip is truly indeterminate, that is, unpredictable in advance, then the outcome of the flip is only precise in that it can result in one of only two states (heads or tails). The distribution of a large number of such flips is predictable, to any desired level of accuracy, depending on the total number of flips performed. The solution to the binomial distribution, predicated on equal outcomes of two states, offers the solution as well. (My sense of wonder at how this always works out is an emotional response. The rational side of my head understands this perfectly well, the kid in me still goes "Oooooh!")

The experimental verification of this is so strong that any significant deviation from the pattern in a large number of tests is a priori taken as evidence of experimental bias. And I do something like this many times every day.

Chromatography is based on the partitioning of a substance between two immiscible phases. As one phase passes over the other, individual molecules are distributed between the two (a dynamic equilibrium), based on the differential solubility bewteen the two phases. Even for the smallest sample the total number of molecules is on the order of 10^10 to 10^13. The shape of the chromatographic peak is Gaussian (the number of individual trials is so large as to move from the discrete distribution to the continuous), the result of ten to a thousand billion 'flips', each one completely unpredictable, results in a pattern so exact that any deviation form the peak shape is evidence of extra molecular interactions or a chromatographic column going bad.

I mention this because it is a more homely example of other indeterminate processes, such as nuclear decay, where the breakdown of a single nucleus is also unpredictable, but with millions or billions of them each second the decay rate is extremely predictable. It is so predictable that any deviations point to either experimental bias or new science.

__________________
If they can get you asking the wrong questions, they don't have to worry about the answers.

Yes, you clearly get it. The Law of Large Numbers and the Central Limit Theorem work quite well, don't they ?

I am amazed at the number of people who think that because a 7-sigma event is theoretically possible that the fact they seem to have encountered one is no big deal, that the data is reflective of what was expected, and the event is just one of those normal fluctuations of the cosmos.

It is true that if you put a moneky at a typewriter long enough he will produce the complete works of Shakespeare, and in fact infinitely many copies. But if you assume a reasonable typing speed, it will take quite a bit longer than the current age of the universe for the first one to appear.

Your bomb example here describes a reduction to components. In what sense is Bohm's physics that sort of reduction of quantum physics? It seems to me that these are two different recipes for calculation and any notion that one recipe describes the inner mechanism of the other is mistaking theory for mechanism.

There are two types of reduction: (1) theoretical reduction, as when it is said that Newton's mechanics can be reduced to Einstein's special relativity; and (2) part-whole reductionism whereby the emergent properties of a whole (those properties not shared by the parts) are explained in terms of the properties of the parts. So yes, it's my hope that Bohmian physics offers a potential reduction of the latter kind.

Physics is unique--and not in a good way in my view--in that at least one school of physics would like it if it were possible to reduce all things to the equations of mathematics--literally. That way, all metaphysics--i.e., all physical ontologies--can be dispensed with: there is nothing more to say about things other than that they are permanent possibilities of sensation (PSS's). This program of "mathematical assent" has only been carried out so far for the lone hydrogen atom: supposedly, the behavior of a lone hydrogen atom can be completely described by the equations of quantum mechanics. Ideally, all talk of things like atoms and molecules would be eliminated entirely and replaced by mathematical formulas with no loss of content.

Bohmian physics as I understand it, on the other hand, holds out for the possibility that there might still be mechanisms analogous to a camshaft that cannot be seen by the application, however precise, of clumsy photons. Imagine trying to explain the behavior of an internal combustion engine if you weren't allowed to take the thing entirely apart. You would come up with a mathematical description that would contain variables that would describe the hidden mechanism of the camshaft. Hence, you would have a theory of "hidden variables"; and it wouldn't be a category mistake to characterize that theory as an attempt at part-whole reductionism: you can't directly access the hidden mechanism, but you take it on faith that it is ontologically similar to the things you can access--that is, that the hidden mechanism is mechanical in nature.

If you heard anything in my comments about "denying" a rich inner life, then it was you who put it there. My words included no such concept.

Sorry Ken--I didn't mean to imply that you endorse cruelty to animals.

You said it best earlier in this thread: "The preferable approach depends on one's purposes." Leibniz thought the major flaw of Newton's theory was that it didn't explain the cause or mechanism of gravity. Newton thought his theory was stronger because it worked regardless of any cause. Newton puts the universe in reach of the capabilities of high school students. Einstein doesn't. Just because a theory may be simpler doesn't make it "clumsy." As you suggested, it is the context provided by one's purposes that defines a theory's value.

This is where I part company with everyone else here. I don't think a theory is a replica of the phenomena it is about. A mathematical description of an engine's behavior is not itself a kind of engine. A hidden variable is not a species of camshaft. To borrow from something Ken said earlier, we already have the engine. We need to use the engine. Mathematics is a process of organizing our activities around the engine. (The ink markings of the equations otherwise are just that: scribbles on paper.) Hidden variables, as I understand them, allow us to fine-tune our actions, to be responsive to the subtler behaviors of the engine.

That's why I don't think quantum mechanics (or Newton, or Einstein, or any other physical theory) can be faulted for missing an underlying mechanism. A theory is not a mechanism itself. A theory puts us in contact with the aspects of phenomena that are most relevant to us at the time. Our purposes will define a theory's value. Bohm will become the better theory when we can do with it more than we can do with quantum mechanics.

You're forgetting that emitting the thermal radiation cooled the water, so you're not measuring it's temperature, you're actually measuring its cooling down.

__________________
‘To those who regard “crime fiction” as some sacred icon which must follow a rigid formula, I will always be the man who writes 18-syllable haiku.’Trying to make sense of computers, The Error Log.

What a pleasant way to spend Christmas Day, reading debate on the philosophy of physics. My two cents, rushing where angels fear, is that the HUP was wrongly interpreted as a refutation of Laplace's Demon, the claim Heisenberg was the first to show that this problem is in principle insoluble for science because of the observer effect. However, it is entirely wrong to say that because we cannot know the position and direction of a particle, that the particle does not actually have a unique position and momentum. It is just that we finite creatures cannot know it. People often equate truth with knowledge, whereas Warren's point here, as I read him, is that truth is a noumenal reality independent of human knowledge. Heisenberg did not show that the universe is indeterministic, only that it is indeterministic for science.

Heisenberg is pulled over by a policeman whilst driving down a motorway, the policeman gets out of his car, walks towards Heisenberg's window and motions with his hand for Heisenberg to wind the window down, which he does. The policeman then says ‘Do you know what speed you were driving at sir?', to which Heisenberg responds ‘No, but I knew exactly where I was.

Dr Rocket said:

I don't quite understand. Doesn't trying to explain why things work imply understanding what the things actually are? How do mathematical predictive models differ from a black box approach? Neither know what things really are.

I don't think so. What does it mean to understand "what things actually are" ? That sounds rather mystical.

Some mathematical models are black boxes. But when they are they tend to be purely phenomenological with an attendant lack of understanding of the applicability.

Better mathematical models such as Newtonian mechanics, classical electrodynamics, quantum mechanics, and relativity, provide more than just input-output relations but also provide some insight into the dynamics that produces those relations.

I have no idea what you mean by "what things really are". When you start talking about quantum mechanics I have less than no idea.

Why does it sound mystical? When I say "what things really are", I mean what is the reality of the microworld? QM has its limitations, does not include gravity, is incompatible with GR at extreme conditions, like black holes or the Planck length.

String theory, which is totally unproven to date, postulates vibrating 1 dimensional strings for the elementary particles. M theory brings n dimensional branes into play.

These are all different mathematical models of the microworld.

But what is actually present down there? (I know we will probably never be able to know...)

QM is is incompatible with GR under all conditions -- one theory is stochastic and one is deterministic. Fortunately each provides good approximations until you have the extreme conditions in which quantum effects and gravitational effects are important simultaneously.

String theory is not just totally unproven, it has been superseded, by M theory. M theory is supposed to be, based on a 1995 talk and paper by Witten, a unification of what were 5 distinct competing string theories, but so far as I know the "dictionary" providing the M theory translation among those theories remains conjectural. So one of the remaining questions of M theory is "what is it ?".

But the question of "what is it?", has not been answered at much higher levels. What is an electron ? There is a very good theory for describing the behavior of phenomena involving electrons -- quantum electrodynamics. But it has some pecularities, and the description of the electron itself is a bit fuzzy -- as reflected for instance in the predicted infinite self-energy of the electron.

This statement is not refutable, because we can no more easily say a particle does not have a position than we can say it does. But the argument lacks plausibility, in my view, because it basically says that we created a certain notion called position, which worked so well we thought is was absolute, and then we discovered it did not work as universally and uniformly well as we first thought. Is the plausible reaction to that that the concept transcends its own usefulness to us, and has some separate existence that is not limited the way its usefulness is? Wasn't the whole purpose of the idea its usefulness? It just seems to lose track of why we generate concepts like position in the first place. We should never take our efforts at understanding reality too seriously, there's simply no reason for us to think our concepts transcend their demonstrated limitations.

Again, this statement is quite true, but is not determinism a scientific concept in the first place? So why should we imagine the concept transcends science? We don't know we can't, but there wouldn't seem to be any reason to think we can.

Why can’t we just take the HUP at face value? It seems to me that humans have a hard time accepting that perhaps we can not know all factors which contribute to the evolution of our physical universe. Leaving aside the HUP; even on a macroscopic scale the universe is evolving and becoming more complex, introducing new emergent properties all the time.

And under such circumstances I find it a bit much we can claim the universe is Deterministic. All we can do is better and better approximations with the “promise” of exactitude in predictions if given infinite approximations over an infinite amount of time. It seems to me that the bigger message coming from nature is that it is impossible for us to know all factors or conditions necessary to predict the future with certainty. Perhaps it is the emotional and practical need for certainty which requires humans to idealise a Deterministic universe.

Actually I have no problem with calling the universe deterministic, as long it's realized that a system capable of emulating the universe precisely is at least the size of the universe.

What physics is about is making approximations that are good enough for normal work.

__________________
‘To those who regard “crime fiction” as some sacred icon which must follow a rigid formula, I will always be the man who writes 18-syllable haiku.’Trying to make sense of computers, The Error Log.

Agreed.

However, what has always fascinated me is the question of "what is really down at the micro level". Just pure curiosity. Sure, our science is wonderful, but it does not tackle this question. Many say, "if the predictive systems work well, who cares what is really down there...". OK, but I still find it one of the most fascinating questions there is.