Interpretations of Quantum Mechanics

[Grey]

Over here (http://www.bautforum.com/showthread.php?p=704392#post704392), Spherical suggested it might be interesting to discuss the various interpretations of quantum mechanics. This (http://en.wikipedia.org/wiki/Interpretation_of_quantum_mechanics) is a pretty good introduction to the most prevalent, with some explanation of the similarities and differences. So, what's your opinion of the various choices? Is there one that's "obviously" the most reasonable? Should we hold out on making a judgment until someone comes up with a way to distinguish these cases observationally, or is it worthwhile to talk about it even we can't distinguish them?

And what about things like Feynman's sum over histories method? This is really a mathematical technique, but it treats particles as taking all possible paths from a given starting point to a given destination, including paths that take a side trip to another galaxy on the way. Should we take that as a serious representation of what's really going on? If reality doesn't actually behave anything like that, why does the math work out so darn well?

Or is this all just philosophical nonsense, and we should just shut up and calculate (http://www.aip.org/pt/vol-57/iss-5/p10.html)? What do you think?

[Daffy]

Wish I understood the math. John Von Neumman's speculations (particularly the "chain") are especially fascinating to me. Does consciousness exist because of matter? Or vice versa? Or both?

[Tensor]

Over here (http://www.bautforum.com/showthread.php?p=704392#post704392), Spherical suggested it might be interesting to discuss the various interpretations of quantum mechanics.

snip...


What do you think?

I'm of the "shut up and calculate" school. Trying to figure out the different interpretations is, to me, philosophy, not science. But, I'm sure there are others with different interpretations.;)

[JohnD]

All,
I'm not capable of these heights.
But I would use the example of the discipline of pharmakokinetics, the modelling of drug movement into and out of the body. There, it may be necessary to assume that the drug dissolves in a volume that is several times the size of the human (in this case) body. Nonsense? No, it works, it allows you to predict the behaviour of drugs, to decide rationally on dosage and to control infusions.
That it demands an impossible condition doesn't matter.
In other words, shut up and calculate!

John

[hhEb09'1]

I'm of the "shut up and calculate" school. Trying to figure out the different interpretations is, to me, philosophy, not science. But, I'm sure there are others with different interpretations.;)At this point, I agree. It appears that they all come up with the same conclusions--but they also seem to be strikingly different.

I think the hope is, as far as each of their proponents is concerned, that by pushing the philosophical arguments far enough, we'll eventually attain a quantifiable difference that would allow us to test each version.

But another goal has already been served. There were all sorts of philosophical conclusions drawn from the randomness and uncertainty principles of quantum mechanics. The other interpretations show that those sort of conclusions are not necessarily forced by the physics. In other words, ethics and teleology are still not on as firm of a foundation as molecular biology or pharmacology. But we probably guessed that.

[gzhpcu]

What is the goal here? To understand how the universe works or to find out what corresponds to reality? IMHO, as long as we do know (if ever) the fundamental answers to what is matter, energy, space and time, we will have myriads of mathematical models and myriads of interpretations. They can not all be right. There is only one answer, but will we ever know what it is?

[Grey]

But another goal has already been served. There were all sorts of philosophical conclusions drawn from the randomness and uncertainty principles of quantum mechanics. The other interpretations show that those sort of conclusions are not necessarily forced by the physics.That's a good point. So if you really think the universe is deterministic rather than random, you can find an interpretation that allows that, though you'll end with some weirdness somewhere else...

[Tensor]

Ok, here's my interpretation:

The universe is weird. How's that?

[Disinfo Agent]

Ok, here's my interpretation:

The universe is weird. How's that?Weird. :p


I'm of the "shut up and calculate" school.I knew I had mentioned that quote a while ago in another interesting discussion (http://www.bautforum.com/showthread.php?p=619846&highlight=calculate#post619846). :)

[Nereid]

If philosophising can help produce some new theories which, upon testing, turn out to be better than anything we've got today - specifically, an extension to QM - then let the philosophising begin!

History, that ever unreliable guide, counsels against putting too much hope into such an endeavour however.

[Van Rijn]

Well, there may be ways to distinguish between interpretations. From here, a column by physicist John Cramer:

http://www.analogsf.com/0410/altview2.shtml

Many (including me) have declared, with almost the certainty of a mathematical theorem, that it is impossible to distinguish between quantum interpretations with experimental tests. Reason: all interpretations describe the same mathematical formalism, and it is the formalism that makes the experimentally testable predictions. As it turns out, while this "theorem" is not wrong, it does contain a significant loophole. If an interpretation is not completely consistent with the mathematical formalism, it can be tested and indeed falsified. As we will see, that appears to be the situation with the Copenhagen and Many-Worlds Interpretations, among many others, while my own Transactional Interpretation easily survives the experimental test.

I'm not sure about more recent work on this, though.

[Grey]

What is the goal here? To understand how the universe works or to find out what corresponds to reality? IMHO, as long as we do know (if ever) the fundamental answers to what is matter, energy, space and time, we will have myriads of mathematical models and myriads of interpretations. They can not all be right. There is only one answer, but will we ever know what it is?Actually, that can be the subject of debate, too. If you have two models of reality with completely different underlying frameworks, but that can be shown to make identical predictions in all cases, do you really have two models of reality after all? I'd probably say yes, but the point is arguable, I think.


I'm not sure about more recent work on this, though.There haven't been any published journal articles about this yet, but it's created a lot of buzz. Most physicists think Afshar is mistaken, and doesn't understand a few important points. For example, from here (http://motls.blogspot.com/2004/11/violation-of-complementarity.html):


In the complementarity principle, we first determine which set of photons we consider, and then we calculate both V as well as K from this set of photons. The contrast is computed from the pictures created by all these photons, and we also want to determine the "which way" information of all of them if we want to claim that K=1.

If you read the section 3 of his PDF file, Afshar seems to compute the contrast of his interference picture from a very small subset of his photons that he uses for the calculation of K - only from the photons that interact with the wire grid.

If true, that's of course silly. We can always arrange an experiment with 2 million photons - the first million will be used to create a perfectly sharp interference picture, and for the second million we will be exactly able to determine the pinhole. But this does not mean that we have K=1, V=1. We must consider the same set of photons if we want to determine K, V.Still, your general point is valid, that we might be able to come up with some other experiment which would let us distinguish between the various interpretations.

[Peter Wilson]

What is the goal here? We have myriads of mathematical models and myriads of interpretations. They can not all be right. There is only one answer, but will we ever know what it is?

Only one answer? What about the wave-particle duality of QM? Neither interpretation is "right;" neither is "wrong." If getting the "interpretation of QM" right means holding mutually contradictory ideas, then I guess we have to accept mutually contradictory ideas (i.e. more than one).

Am I not both right and wrong?

[Disinfo Agent]

Am I not both right and wrong?There's a 50% chance you are.

[gzhpcu]

Only one answer? What about the wave-particle duality of QM? Neither interpretation is "right;" neither is "wrong." If getting the "interpretation of QM" right means holding mutually contradictory ideas, then I guess we have to accept mutually contradictory ideas (i.e. more than one).

Am I not both right and wrong?

Not so sure. It is wave-like when not observed, and particle-like when observed. Since it is not both at the same time, I see no contradiction.

[snarkophilus]

And what about things like Feynman's sum over histories method? This is really a mathematical technique, but it treats particles as taking all possible paths from a given starting point to a given destination, including paths that take a side trip to another galaxy on the way. Should we take that as a serious representation of what's really going on? If reality doesn't actually behave anything like that, why does the math work out so darn well?

It does work well. Really well. Really, really, really well. It shocked me when I first learned how well.


Or is this all just philosophical nonsense, and we should just shut up and calculate (http://www.aip.org/pt/vol-57/iss-5/p10.html)?

Yes and no. How's that for a quantum answer?

There are good reasons for finding a deeper physical basis for this. First of all, curiosity. Second, because a good physical hypothesis will suggest new experiments that can be done. Third, I'm as sick of the EPR paradox as I am of Riemann's hypothesis. Somebody solve the stupid things already. :)

That was a joke, by the way. But it's one of the reasons to shut up and calculate. Maybe we're nowhere near knowing enough to solve those problems yet. Maybe our time would be better spent learning about different facets of the world, and if we go back to those problems fifty years from now, we'll have the tools to investigate more efficiently. A lot of it really is philosophical nonsense, but some of it may have value if we can figure out how to use it. The problem may be that we don't yet know enough to intelligently decide what is useful -- maybe we're really just in the alchemy stage of physics.

For most people, I really lean toward "no." If it was, for instance, Neils Bohr telling me that he was looking into a deeper meaning of it, then who am I to argue? (As an aside, I went to see that play the other day, but it was sold out :( ) But if it's some guy who took an English class in university and decides that because of that he can lecture to me about how the universe really has eleven dimensions and we are not responsible for our actions because of Heisenberg's principle, I feel an obligation to tell him to shut up (and calculate, I suppose, though really I just wanted him to shut up ;) ).

[Nereid]

This (http://en.wikipedia.org/wiki/Afshar_experiment) is the wikipedia entry; in part:
Papers by Afshar on the experiment have been published in the American Institute of Physics and SPIE conference proceedings; however, as of May 4, 2006, neither a description of the experiment, nor any discussion of its theoretical interpretation, has been published in a refereed physics journal.Does anyone know what's happened since early May?

In particular:
a) has anyone replicated the experiment?
b) done a similar one, with only one photon in the apparatus at a time?

The Afshar Q&A blog (ref 15 in the wikipedia article) seems to be unmoderated, so is (completely?) full of spam (and so useless).

[gzhpcu]

Only one answer? What about the wave-particle duality of QM? Neither interpretation is "right;" neither is "wrong." If getting the "interpretation of QM" right means holding mutually contradictory ideas, then I guess we have to accept mutually contradictory ideas (i.e. more than one).

Am I not both right and wrong?

What I meant was the different interpretations of QM: Everett's many worlds, Copenhagen interpretation, etc.

[Ken G]

If you have two models of reality with completely different underlying frameworks, but that can be shown to make identical predictions in all cases, do you really have two models of reality after all? I'd probably say yes, but the point is arguable, I think.

I'll take the position that the connection of theories to reality is defined entirely by their testable predictions. Thus two such theories you describe would be different theories, but not in any physically significant way-- the difference is purely, yes, pedagogical. Now I've used that word quite a bit around here, because I find a persistent confusion between what is a theory (just shut up and calculate) and what is a pedagogy (how can I picture what I just calculated, to help me do the calculation right, and get a sense of meaning behind the calculation?). Once one understands the crucial difference between these, and how one is the objective testable component of science and the other is the subjective human component, a lot of these seemingly tricky questions just go away. So my answer is, certainly do not shut up and calculate, rather, talk a lot and calculate. But never confuse the talk for the calculation. Ultimately, the test of the calculation is its accuracy and reliability, and the test of the talk is its track record for leading to proper calculations and the all-important sense of comprehension that it conveys. Note the latter may be different from individual to individual, but the former should not.

[gzhpcu]

I'll take the position that the connection of theories to reality is defined entirely by their testable predictions. Thus two such theories you describe would be different theories, but not in any physically significant way-- the difference is purely, yes, pedagogical. Now I've used that word quite a bit around here, because I find a persistent confusion between what is a theory (just shut up and calculate) and what is a pedagogy (how can I picture what I just calculated, to help me do the calculation right, and get a sense of meaning behind the calculation?). Once one understands the crucial difference between these, and how one is the objective testable component of science and the other is the subjective human component, a lot of these seemingly tricky questions just go away. So my answer is, certainly do not shut up and calculate, rather, talk a lot and calculate. But never confuse the talk for the calculation. Ultimately, the test of the calculation is its accuracy and reliability, and the test of the talk is its track record for leading to proper calculations and the all-important sense of comprehension that it conveys. Note the latter may be different from individual to individual, but the former should not.


I would think that if you have two different theories delivering the same testable predictions, then IMHO, we are not covering all aspects and making all possible measurements. Not sure if we can ever get a pedalogical view of physics (much too complex). Increase the scope of the testable environment and one of the two will hit its limits.

We are talking about 0 dimensional points, about vibrating 1-dimensional strings, about 10 spatial dimensions, but this is just math. We can't really picture it. Sure we can talk about "curled" up dimensions, but it is really not imagineable from our experience.

[Ken G]

I would think that if you have two different theories delivering the same testable predictions, then IMHO, we are not covering all aspects and making all possible measurements.

I think another key distinction between mathematics and pedagogy is that mathematics is so self-contained that if two different mathematical processes lead to the same answer in all situations, they may be said to be equivalent and for all intents and purposes they are identical. For example, 2+2=4 (the "length" approach) can also be solved by 2=1+1 and 1+1+1+1=4 (the "counting" approach). No one would say these are two different ways of doing math, but they really are pedagogically different. The equivalence is more obvious on the mathematics side than the pedagogy side, as the first can be seen in a more formal way.

An example that is more relevant to this thread is the equations of wave mechanics. One way to find what a wave will do is to solve a second-order differential equation, the "wave equation", but one can also use a mechanistic approach-- Huygens principle (each point on the wave crest is a new wave-crest source, that idea). The latter is really just a technique to solve the former, and the former is a way to describe the latter in terms of equations, but pedagogically they sound quite different. So it's not always obvious when two pedagogies are equivalent.

But the point is well taken that two different pedagogies might diverge in a realm that has so far been untested (example, Newton's laws and relativistic mechanics make all the same predictions at very low speeds, so seem like equivalent pedagogies but only over a restricted input domain.) When this is true, the mathematics are not equivalent either, but we don't know that until we consider the new domain. Thus we have another reason to favor multiple pedagogies-- not only do they allow us, as individuals, to choose the one we "prefer", it also gives us an opportunity to discover new physics in domains where the diverging pedagogies lead to different mathematical results.

[gzhpcu]

Thus we have another reason to favor multiple pedagogies-- not only do they allow us, as individuals, to choose the one we "prefer", it also gives us an opportunity to discover new physics in domains where the diverging pedagogies lead to different mathematical results.

I like the second part of the statement, but the first part, less. I don't like choosing what I "prefer", I'd like to know what corresponds to the truth.

I suspect that no current theory is correct for all possible cases, and that consequently, it is treading on thin ice when one attempts to correlate the underlying math in a theory with a physical picture/image.

[Ken G]

I like the second part of the statement, but the first part, less. I don't like choosing what I "prefer", I'd like to know what corresponds to the truth.

I suspect that no current theory is correct for all possible cases, and that consequently, it is treading on thin ice when one attempts to correlate the underlying math in a theory with a physical picture/image.

Yes, but we are just intelligent apes, when will our pedagogies ever be "the truth"? And how would we know? We never get beyond the inside of our head, yet we can interface with nature, and its great consistencies, and so we progress in practical terms. But as a human exercise, science does inevitably involve personal preference ("Occam's razor" being the classic example). Of course, that message must be made very carefully to avoid being heard wrong-- the testable aspects are not personal preference, and they are the anchor of the usefulness of science as something more than an intellectual pastime. Still, you cannot test a pedagogy, only its predictions.

[hhEb09'1]

But as a human exercise, science does inevitably involve personal preference ("Occam's razor" being the classic example). I've said this before, and I'm only saying it again because I think it is gzhpcu's point as well, that Occam's Razor is not a scientific principle. It's more of a personal preference, as you say.

[Ken G]

And I will briefly reiterate that scientific principles are defined by the practice of science. Part of that practice is to value understanding, which involves simplification, which is where Occam's razor comes in, right in the middle of the foundations of the practice of science. (And note, the fact that you and I don't seem to agree on much of anything lately, even though we use all the same mathematics, is proof positive of the important role of pedagogy in scientific understanding.)

[gzhpcu]

Yes, but we are just intelligent apes, when will our pedagogies ever be "the truth"? And how would we know? We never get beyond the inside of our head, yet we can interface with nature, and its great consistencies, and so we progress in practical terms. But as a human exercise, science does inevitably involve personal preference ("Occam's razor" being the classic example). Of course, that message must be made very carefully to avoid being heard wrong-- the testable aspects are not personal preference, and they are the anchor of the usefulness of science as something more than an intellectual pastime. Still, you cannot test a pedagogy, only its predictions.

Correct me if I am misunderstanding you here. Are saying essentially "we are limited intellectually and incapable of understanding the truth because it is so complex and foreign to our way of thinking?".

[Ken G]

Yes, that seems quite likely.

[gzhpcu]

Then would it be better to have theories based on abstract math, like Matrix mechanics of QM, which do not lend themselves to imagery?

[Ken G]

I'm not sure I'd agree that abstract math doesn't lend itself to imagery. Indeed, the more abstract, the greater the need for imagery to guide one's thinking. When's the last time, when someone said 2+2=4, you actually imagined two things and two more things coming together?

[gzhpcu]

Yet, there is a reality out there... the search may very well futile, but fascinating regardless.... and it would be interesting to be aware of as much of it (as accurately as possible) as our limitations permit us to be...

[Disinfo Agent]

I don't like choosing what I "prefer", I'd like to know what corresponds to the truth.Wouldn't we all? But, in a way, that's not what science is about. Knowing the truth, I mean.

[Peter Wilson]

What I meant was the different interpretations of QM: Everett's many worlds, Copenhagen interpretation, etc.

Seems the “interpretation” issue boils down to determinism vs. indeterminism: is the collapse of the wave-function deterministic, or not? What does that math really say? Or does it? And the reason there is so much fuss about “the collapse” is because the old question of free-will hinges on it.

In the Many Worlds interpretation, when I order pizza, if the guy asks if I want garlic bread with it, the universe splits in two, because I cannot decide, and in one universe I say, Sure, and in another I say, No thanks.

All I know is, free will—as it is popularly imagined—is only possible if somewhere in the mechanism of the organism there is a state-change that is not entirely deterministic. Scientists have mapped out many deterministic bio-mechanisms, and doubtless we are automons to a large degree. Creatures of habit, to say the least. But if the sensation of choice is real, that can only be possible if somewhere in the bio-mechanism real choice is possible, and the collapse of the wave function is the only place I can imagine an indeterminate choice can be made.

If there are “hidden variables,” as Einstein imagined, and the collapse of the wave function is deterministic, then the age-old debate is pretty much settled: no such thing as free will is physically possible; it must be an illusion. However, the experimental tests seem to show the collapse is indeterminate, so the possibility lives on.

If you believe in free will, you hang your belief on an indeterministic collapse of the wave function, because there is no room for free will in any other physical law.

[Disinfo Agent]

I don't see what human choices, which are very likely decided at the molecular level, have to do with Quantum Mechanics.

[Peter Wilson]

Molecules are quantum-mechanical systems.

[Disinfo Agent]

Well, maybe I just don't know enough chemistry, but I thought molecules could be studied with classical models alone. Everything is ultimately quantum mechanical, of course, but you don't need QM to launch the space shuttle, for example. For all practical purposes, there is no quantum uncertainty in the launch. Interpretations of QM are totally irrelevant to it.

[gzhpcu]

Well, maybe I just don't know enough chemistry, but I thought molecules could be studied with classical models alone. Everything is ultimately quantum mechanical, of course, but you don't need QM to launch the space shuttle, for example. For all practical purposes, there is no quantum uncertainty in the launch. Interpretations of QM are totally irrelevant to it.
This is addressed by the correspondence principle formulated by Niels Bohr in 1923, which states that the behavior of quantum mechanical systems reduce to classical physics in the limit of large quantum numbers.

[Ken G]

If you believe in free will, you hang your belief on an indeterministic collapse of the wave function, because there is no room for free will in any other physical law.

Actually, neither the deterministic nor the indeterministic view allow for free will. The former for obvious reasons, but the latter involves randomness in an inescapable way, so that's hardly free will either. How is it free will if the choice is random? The problem is with an incorrect definition of free will. Free will means that the decision process is entirely personal, with no outside controls forcing the issue one way or another. That's all it is, freedom to make a personal choice. There is little or no connection between this definition and quantum mechanics, and until we have a theory of cognition, the connection will be completely unknown.

[Nereid]

If you're into this kind of thing, I recommend the works of Daniel Dennett (http://ase.tufts.edu/cogstud/~ddennett.htm) - in one of his books (I forget which) there is a very clear demolition of many kinds of "QM permits free will" arguments, with some surprising twists (wrt 'free will').

[worzel]

If you're into this kind of thing, I recommend the works of Daniel Dennett (http://ase.tufts.edu/cogstud/~ddennett.htm) - in one of his books (I forget which) there is a very clear demolition of many kinds of "QM permits free will" arguments, with some surprising twists (wrt 'free will').
Would that be "Freedom Evolves"? I was going to mention it but couldn't make a decent summary of any of his points despite finding him very convincing myself. For me, his demonstration of freewill (operationally at least, which is all we can really fathom) in a determistic world via ever increasingly complex chess programs was very illuminating.

[Ken G]

For me, his demonstration of freewill (operationally at least, which is all we can really fathom) in a determistic world via ever increasingly complex chess programs was very illuminating.

Yes, I agree that an "operational definition" of free will is the key issue for science. Philosophical definitions are much harder to confront with any kind of scientific theory, so I think that's usually the disconnect when people use physics to try and understand free will. Apples and oranges.

[Nereid]

Would that be "Freedom Evolves"? I was going to mention it but couldn't make a decent summary of any of his points despite finding him very convincing myself. For me, his demonstration of freewill (operationally at least, which is all we can really fathom) in a determistic world via ever increasingly complex chess programs was very illuminating.Yes, that's the one (though there are also some good parts in Darwin's Dangerous Idea).

One take-away for me was that the space left for 'free will', should you choose to try to tackle it with a mixture of logic and the results of modern science, is very small indeed. Or, saying it slightly differently, 'free will doesn't exist, in its naive formulation; but why should you care anyway? - its non-existence doesn't take anything away from your perception of choice, nor any morality you choose to build around/from that perception.'

[worzel]

I also agree that quantum mechanics' truely random events are no more a get out for freewill than than the randomness of the starting variables in the nth decimal place of a deterministic world. They're both unknown and effectively random to us.

To argue that someone is responsible for a decision that was derived from a random quantum wave fuction's collapse but not for a decision that was determined by the creator choosing 1.0(may zeros)01 instead of 1.0(many zeros)02 for some initial variable is just silly. It's not like they had any choice in either of those cases.

[Ken G]

Along those same lines, I would argue that the concept of free will is itself the problem, not anything that we have learned from science. Even an intelligent caveman who knows no physics at all can tell that the concept of free will, in its naive description, makes no sense. Because if one defines a free choice as one that has no influences on it at all, then what is left but pure randomness, and how is that free? What most people mean by free will is the ability to follow their own preferences, rather than someone else's, but then the choice is affected by the preferences so is not free. Thus it must be the preferences that are chosen freely-- but this only pushes the question back one level: what influences the preferences that we choose to value? Even before there was physics, there was either influence or randomness, and neither are "free". So we don't need to find a way to make physics compatible with free will, we need to find a definition of free will that makes some kind of basic sense. I would say that it has to do with being personal-- free will should be called personal will instead.

[Peter Wilson]

Actually, neither the deterministic nor the indeterministic view allow for free will. The former for obvious reasons, but the latter involves randomness in an inescapable way, so that's hardly free will either. How is it free will if the choice is random? The problem is with an incorrect definition of free will. Free will means that the decision process is entirely personal, with no outside controls forcing the issue one way or another. That's all it is, freedom to make a personal choice. There is little or no connection between this definition and quantum mechanics, and until we have a theory of cognition, the connection will be completely unknown.(Emphasis added)

Right. I have no theory of cognition. All I'm saying is, the indeterminism of QM is the only place to look.

If you say, Peter, you say you're an engineer. Design me a computer, I will say, Ok. What kind of logic gates do you want me to use?

You see, a computer requires logic gates. Computers can (and have) been built using mechanical logic-gates. But you cannot build a computer from wires, resistors, capacitors and coils, because these alone cannot perform a logic function. At the heart of all of today's most sophisticated, complex computers lie the same thing: a bunch of transistors acting as logical AND gates, wired together into a fantastically complex, yet functional, logical system. But at the heart of all this complexity lies the basic "conditional-if:" the transistor produces a valid output if--and only if--both inputs are valid.
All logic systems, from the basic two-input transistor to the most complex computer on earth, require a conditional-if mechannism in order to operate.

As an engineer, I know a logic system cannot be built without a conditional-if component at its heart. Neurons, by the way, are conditional-if components. They produce an output if--and only if--certain inputs are present.

In the same way, I reason that a large, complex system, say an ant, cannot make an unconditional choice (exercise free will), unless it is built from components capable of making unconditional choices. We know ants, at some level, operate molecular machinery, and molecules are QM entities, and we know QM systems are capable of unconditional choices.

[worzel]

Logic gates make computers predictable whether the underlying physics is inherently random or deterministic chaos (and analogue computers have been built too). Maybe ants and us are deterministic but chaotically so, so as to give the impression of freewill. Or maybe ants and us are truely governed by truely random events which also gives the illusion of freewill. Both alternates would appear random not only to the observer but to the one making the decisions, so I don't see why only one is the place to look for freewill, whatever that is.

[Ken G]

The problem is not that free will is an illusion, it is that it is undefined. What is an "unconditional choice" anyway? It, like free will, is an oxymoron, because if something is unconditional, then it is random, and if it is random, it is not a choice. Choice is the expression of conditions, and free choice simply describes where the conditions come from-- places that are defined to be "free". So it's all in the definition of free, and in practice, that definition is personal.

[worzel]

Yes, as usual you make the point better than me Ken G.

I would add that if intelligent agents came about through natural selection then what use to evolution would random processes be anyway? Surely a useful brain (for suvival) would actually harness physics in a way analogous to digital computers. That is, whatever the underlying randonmess and/or chaoticness of the physics, the brain, like the computer, would only be useful if it could overcome this to produce somewhat reliable results.

[Nereid]

Yes, as usual you make the point better than me Ken G.

I would add that if intelligent agents came about through natural selection then what use to evolution would random processes be anyway? Surely a useful brain (for suvival) would actually harness physics in a way analogous to digital computers. That is, whatever the underlying randonmess and/or chaoticness of the physics, the brain, like the computer, would only be useful if it could overcome this to produce somewhat reliable results.There are many fascinating tangents to this, not involving the brain at all (well, only peripherally).

Vision is one - the more we learn about vision, in critters large and small, ancient and modern, with big brains and hardly any brains at all, etc, etc, etc, the more intricate it becomes. (Somewhat OT, for this point, is that 'eyes' have been produced by convergent evolution a great many times). One aspect - vision systems are far from 'perfect' (however you define the concept), and can be 'fooled' ... and how they are so fooled, and how 'fooling' differs between species, classes, families, etc, ah such amazing diversity!

[worzel]

The question of primary colours (what they are, and why) came up on another thread. Just as well our sight isn't so good, really, or we'd need far more technologically advanced monitors and printers to enjoy "true" colour :)

[worzel]

I've found programming over the years particularly inspiring in wondering how sentience might arise. I don't think the programs of today are any where near sentient even in kind, let alone complexity. But with programming you've got this universe that the program "sees" via the computer's sensory input and the programming telling it how to react. I doubt our brains operate in anything like a procedural manner (or particularly functional or declarative come to that) but some rather trivial programs can give rise to some suprisingly sentient appearances.

Silly example (10Mb) (http://mboyd.demon.co.uk/Rovorbot_agk.avi)

[Ken G]

Yes, computer programming is certainly the initial effort at understanding intelligence. This raises two new questions: Which is the more difficult to achieve: the software or the hardware required for sentience? And: which will we achieve first, artificial intelligence, or a mastery of our own?

[Peter Wilson]

The problem is not that free will is an illusion, it is that it is undefined. What is an "unconditional choice" anyway?
Okay, jumping ahead a little, call it a "rational choice." Jumping way ahead, call it a sentinent choice.

It is impossible to build a rational machine, capable or making sentinent choices, if its entire mechanism is deterministic.

[Peter Wilson]

That is, whatever the underlying randonmess and/or chaoticness of the physics, the brain, like the computer, would only be useful if it could overcome this to produce somewhat reliable results. Right, but something else is required.

Consider this "ant-test," which I used to use as a basic ant IQ test: Place a stick on the ground infront of a walking ant, any ant. When the ant clambers over it, lift the stick. The ant will stay on the stick, and walk to one end. Seeing it goes nowhere, it will turn around, and begin hiking towards the other end of the stick. While it is doing so, switch which end of the stick is held. The ant will arrive at the 2nd end of stick, and quickly determine it goes nowhere, also. At this point, the ant will often turn around, and march back to the 1st end. Arriving there a 2nd time, it will usually jump to the ground.

Everything the ant does, up to jumping, can be considered "automatic," instinctual, programmed by its DNA, "robotic," whatever. Somewhere in the ant's machninery, there is a conditional-if mechanism: if you come to the end of stick, go back other way. The ant-anatomy also has another conditional-if built in: If at an unknown height above unknown terrain, do not jump. It is easy, in Darwinian evolutionary terms, to see how these two conditional-if mechanisms could evolve.

Nonetheless, the "ant-test" provides the ant with a problem which it could never have faced in its evolutionary history. Free-floating sticks that do not go anywhere do not exist in nature, so it has no "programmed response" for this test. Yet, ants do not get "stuck," going from one end of the stick to the other in an endless loop. They synthesize a new choice, jumping to the ground, which is the "rational choice," given the situation, and usually on 2nd or 3rd try.

The ant's choice is not random. Otherwise, conducting this test on a large number of ants would yield a random number of "tries" before the ant "gets it," and jumps to unknown fate. But ants generally "get it" rather quickly, and opt to make "a new choice." So the ants response is not "deterministic," nor is it random.

Maybe it is a stretch to call it free-will, but there is something going on in the ant that is more than mechanism. And again, I have no idea, absolutely no idea, how this is possible, or what is going on, but QM is the only place I can imagine to look. If it is "determinism all the way down," I cannot imagine how an ant could ever make anything but an automatic, pre-programmed choice. Ergo, I conclude QM is essential to "life-as-we-know-it," thus a proper "interpretation" of QM is essential to understanding life.

[worzel]

I'm glad you agree that real randomness wouldn't be much use, although it could be useful in some circumstances. But I just don't see any justification for stating that sentience can't be deterministic.

I don't think your ant test proves anything. I can well imagine evolution producing something far more subtle than simple if-then rules for each situation. [ slight tangent ] Especially when you look at how little information is needed to program incredibly complex fractals and the fact that naure is largely fractal in geometry - I can imagine some ecomony of encoding of decision making processes such that you get back far more than you put in. A simple boredom rule would suffice in your example: "if x has been tried twice already to no avail then try something else." [ assuming a probably grossly simplistic ant-brain wiring ]

BTW: I think ants can safely jump from any height.

[Robert Andersson]

Peter, even an ant brain is adaptive, so it is entirely possible that the number of tries depends on that particular ant's previous experience. Also, it is complex, so factors such as the stick's texture, length, angle, wind, what it ate for breakfast could play a role. It could well be "deterministic" (not saying it is).

I don't think conditional ifs (programming kind) is a good description of the mechanism of a brain.

[worzel]

If it is deterministic it could theorectically all be done with "if then's", couldn't it? It may not be the best way to do it, though.

[Ken G]

It is impossible to build a rational machine, capable or making sentinent choices, if its entire mechanism is deterministic.
This is an assertion, not an argument. You have to specify what it is about "sentient choice" that contradicts "deterministic". This will require a definition of the former, that does not explicitly rule out the latter, or else it is a tautology. This is the real problem-- how can one meaningfully define a "sentient choice"? I maintain that everything we know about sentience comes from the personal experience of it, not from analytical thought about it. We are at the level of someone trying to understand thunder when all they've experienced is bowling. (And personally, I don't think it's at all clear that sentience has anything to do with nondeterminism or quantum mechanics. A dead person is just as "quantum mechanical" a system as an alive person.)

[HenrikOlsen]

Okay, jumping ahead a little, call it a "rational choice." Jumping way ahead, call it a sentinent choice.

It is impossible to build a rational machine, capable or making sentinent choices, if its entire mechanism is deterministic.
Part of the problem here is that you have a builtin assumption that the human brain is rational, it may just as well be rationalizing ie. some process makes the decision, then the process of conciousness invents a reason why that choice was made.
You wouldn't be able to differentiate between the experience of those two situations.

I'd rather say that it's entierly possible to make a completely deterministic rational thinking machine, being rational isn't the problem.

Whether the deterministic process used to make the decisions is simple enough to predict it with a simpler process, may actually be the key, and as for sentience, you'll have to define it first.

[Ken G]

You may find words similar to that effect elsewhere in this thread, in fact.

[Peter Wilson]

I have opined as to why the indeterminism of QM is essential to certain philosophical belief systems. In particular, most religions have pegged humans as "free" to choose between good & evil, and I have argued this cannot be possible, without indeterminism. Objections have been raised to my arguments.

Let us assume all the objections are valid, and indeterminism of QM has nothing to do with mind/free-will/sentience. If so, is there any "interpretation" of QM that does matter, philosophically?

[Ken G]

Let us assume all the objections are valid, and indeterminism of QM has nothing to do with mind/free-will/sentience. If so, is there any "interpretation" of QM that does matter, philosophically?

Not that I know of, but then, I'm always quite suspicious of efforts to extrapolate physics into philosophy. It's almost always done in a way that presumes questionable elements of what physics is meant to do.