Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Re QM: Why Not Knowledge

  1. Jan 21, 2006 #1


    User Avatar
    Science Advisor

    One of my goals in participating in this forum, is to get my physics chops back, after a long period in the hinterlands of commerce. I've learned a great deal from many bright and interesting people here.

    My introduction to QM happened almost 50 years ago, at which point I became fascinated with the issues of interpretation. Nobody then worried much about interpretation, because the heyday of particle and solid state was in full swing by the late 50s -- Schwinger's Radiative Corrections Paper came out in 1949. There was more opportunity for theoreticians than there were theoreticans -- many of us found working with experimentalists fascinating and rewarding -- dealing with Nature face-to-face if you will. QM interpretation was not on the A-list; too many other things to do.

    After reading many posts about interpretation, some clear and straight forward, some logically and factually challenged, some that are over the top, I feel comfortable in laying down my views on interpretation, many of which I owe to Sir Rudolf Peierls' writings.
    To be direct: QM probabilities represent states of knowledge, just as classical probabilities do. When I do a measurement, indeed the wave function collapses to provide the knowledge of the particular outcome. The collapse is a physical phenomena, as the brain goes from knowledge of probabilities to knowledge of the outcome.

    This approach has many advantages, and requires no mysticism nor methodological hand-waving. And it recognizes that, at the probability level, classical and quantum physics are very similar. That is, probability is probability, whether applied to electrons or football games. Why? The formal measure-theoretic approach is applicable to both flavors of physics. The probabilities at issue are often governed by similar differential equations for the systems probability density. In physics we use the evolution of the density matrix, in classical cases we use some form of a stochastic differential equation. (See An Introdution to Stochastic Processes by M.S. Bartlett. This is a tough book.But, if you can get through it, you'll have a better understanding of why probability is probability -- QM and CM just use different dynamics and assumptions. )

    It's relevant to note that classical and quantum meet in Rutherford's experiment. The alpha-nucleus cross section can be computed classically or quantumly (sorry 'bout that) with identical results.

    This means, of course, that there are many wave functions, one for each observer if you will. Naturally, when doing a scattering experiiment, all observers will agree on the outcome probabilities. It's crucial at this point to reinforce the almost obvious: the outcome probabilities of the various observers must be the same: they are all observing the same event. And, Einstein is our authority. He says that events are invariant. (I'm assuming all the observers are in the same inertial frame. But unitary transformations are around to help any referehnce frame generalization required.

    There's another way to view this interpretation, based in the Shannon-Weiner Theory of Communication(or Information) -- discussed in Bartlett's book --. Roughly speaking, a measurement generates a signal, which ultimately one or more of the experimenters receive. If you push through, you can identify the transmitter and the receiver, the channel(s), the aplphabet, and the error patterns. One of the coolest results is that it is always possible to discern a message from noise -- but it may take a long time. The basics are simple: Shannon, sort of in so many words, uses the law of Large Numbers to average out the errors, which may require lots of repititions.

    (I'm fairly convinced that all of physics, at the minimum. can be fully described as a communication system -- nature speaks to us through experiments and observations. What I'm less than clear about, is the value of such an enterprise. if it has not been done, it might well be a great thesis topic for a budding theorist.)

    A look at EPR, and then we're done. EPR is probably unparalleled in physics for the amount of nonsense generated, the amount of profoundity generated, and the amount of passion generated. Let's look at EPR from the knowledge perspective. Let's do the simplest experiment-- electrons or photons, each with opposite helicities. and each travelling down a tube toward a detector, which will measure helicities. We'll make a configuration that precludes quick communication between Mr. A, and Ms. B -- there is no chance they can compare notes afer one measurement.

    Mr. A finds helicity UP. He knows from appropriate theory, that Ms. B, measuring after Mr. A will measure spin down. he is projecting by means of established theory, that the second measure ment MUST be spin down. (We do this sort of thing all the time.) He knows that a single helicity measurement is sufficient to determine the spin state of the two particle system.

    Ms. B knows zip about Mr. A. As far as she is concerned, there's an equal chance for UP or DOWN. However, she does know that if she gets UP, then he got DOWN, and vice versa. Again, this is aplication of fundamental theory.

    Everything is local, and there is no need to invoke faster-than-light communication. Also note, that this experiment can be done with numerous classical systems. The difference beteween QM and CM show up when, for example, we measure spin in a direction other than that of the particle's momentum, the outcomes of which are controlled by Clebsch-Gordan coefficients. (The closest classical situation I can imagine is a pair of spinning tops with nutation and precession somehow paired.)

    If there is any magic here, it has to do with physics itself. Physics tells Mr. A about things outside his immediate environment, by inference rather than by direct observation. It is our cumulative knowledge that allows Mr. A to transcend space and time with his knowledge.

    Reilly Atkinson

    What about simultaneous measurements with Mr. A and Ms. B? Anybody know anything about this issue?
  2. jcsd
  3. Jan 22, 2006 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    This is the positivist viewpoint. It takes that there is only a "state of knowledge". But the problem I have with it, is:
    knowledge about WHAT ? If there is no state of nature, what does it mean for me to have "the knowledge that the electron is in position x" ? If there is no "position of the electron" in the state of nature ? There *is* then no nature (or at least not the possibility of a mathematical description of the state of it) ?
    And from the moment you DO introduce a (classically-like) state of nature of which we learn stuff, you run into troubles, no ?
  4. Jan 23, 2006 #3


    User Avatar
    Science Advisor

    I view myself, as do, I think, many scientists as a pragmatist. At the minimum, this pushes us toward less than perfection in the big picture, and modest comfort in the day-to-day picture. Said directly: I assume there's a world out there, it provides the signals we perceive. As a physicist, I'm cautious in making inferences, so I say, "It is as if Nature is out there." Looks like a "State of Nature", sounds like a "state of nature", walks .....

    So, outside of my scientific perspective, I say, "It's a duck" -- like my grandkids are real, my piano is real, ....." God bless Occam.

    As a matter of fact, this Peirels - Knowledge approach to QM really parallels our day-to-day encounters with reality -- you "see", you know, you don't see, you don't know, of course, unless .....'" What is more practically real than our knowledge?

    I think, I might have been a positivist once -- but then I enountered Theloneous Sphere Monk, ee cummings, and Paul Klee, and decided, as. pointed out by Archibald MacLeish, "A poem should not mean but be."

    That's a rather oblique answer to your question about knowledge. But it is indeed an answer. You truly know what you know.

    Reilly Atkinson
  5. Jan 23, 2006 #4
    Dear sir,

    I am Greek and there is an ancient greek scientific rule that says: "I know only one thing, that I know nothing".

    This phrase belongs to Socrates. While the contemporary wisdom names the ancient Greek scientists as philosophers, actually they were scientists. So was Socrates too.

    Maybe, this could be the answer to your question: Science always proves the ignorance of the scientists. This is the essence of science. QM does this in a more profound way, neverthelles it is "science".
  6. Jan 23, 2006 #5


    User Avatar
    Science Advisor

    leandros-p -- What you say makes sense to me, with, however, one exception. My family and I have greatly benefited from modern medical practice, which is permeated by what might be called, at least by some, scientific knowledge. (In fact, as I think about it, without something to know, there can be no ignorance. )

    And, I think the essence of science is courage: to seek order in the face of ignorance and stuff going every which way. Perhaps another take, courage to follow one's curiousity. And, to do all of this with a respectable level of humility.

    Reilly Atkinson
  7. Jan 23, 2006 #6
    So have I, and that includes from you, and the other mentors, and some unofficial mentors as well. My comments below are just my current thoughts on this stuff ... and are, as always, amenable to modification.
    Any corrections or alternative perspectives are appreciated.
    This is sort of the way I've learned to think about it. Whatever else the wave equations and wave functions might be conjectured to represent wrt an underlying quantum world, it can at least be said that they represent our quantitative apprehension of events that occur at the level of our sensory perception.

    The probabilities, and our knowledge of quantum systems, refer to the formalism(s) and the frequencies of occurance of certain specific qualitative (macroscopic) phenomena at the detector end of a setup --- things that anyone with normal sensory capabilities can observe directly.

    The detector outputs might be assumed to be amplifications of phenomena occuring at the level of an underlying quantum world not amenable to our sensory perception. This assumption of an underlying realm seems to be a reasonable one given what has actually been observed by physicists. Nevertheless, it is a metaphysical assumption --- and because the definition of this assumption entails that the underlying quantum world isn't amenable to our sensory perception, then the precise relationship between an underlying quantum reality (whatever it might be and however it might be behaving) and the detector outputs that the formalism of the theory quantitatively describe is, and must remain, a matter of speculation.

    Thus arises the quantum measurement problem.

    It's difficult to sort out exactly what quantum theory, being an assemblage of mathematical abstractions (some of which correspond to our ordinary experience and some of which don't), corresponds to in the real world (which, presumably, includes an underlying quantum world as well as the worlds of microscopic, macroscopic and telescopic phenomena) --- except that it seems clear enough that it contains symbolic elements that correspond to the data, the instrumental outputs, that you or I or anyone can perceive.

    We ascribe meaning to statements about the world of our sensory experience by pointing to the objects and events that those statements correspond to. But we don't, by its definition, perceive an underlying quantum world. So, there's no way to ascertain how closely, or to verify even if this or that qualitative or quantitative symbolic construction corresponds to an underlying reality that is precipitating the causal chain of amplifications that eventually results in the data. We can only ascertain how closely the quantitative predictions correspond to the instrumental output.
    This seems like it might be an unnecessary complication. The basis of empirical science is that we are (us humans) essentially the same (sort of) detector. So, for a given instrumental setup, you don't need to have different wave functions for professors Smith and Jones' brainstates. The instrumental setup is the observer of (or perhaps more precisely, the producer of --- or perhaps more precisely, as far as we can know, it is) the quantum system, and what is observed by us is the instrumental setup's output (via data tapes, screens, clicks, cloud chambers or whatever).

    The order of detection of an AB pair doesn't matter in the optical EPR-Bell situations you're considering, afaik. The pairings are made via coincidence intervals as well as time to amplitude conversion. There is one joint polarizer setting associated with a pair. In practice, again afaik, simultaneous detection doesn't actually occur.
    Last edited: Jan 23, 2006
  8. Jan 23, 2006 #7
    The positivist viewpoint is the basis of empirical science. It's all about meaning. The basis of meaning is our sensory perception. Any statement that purports to describe something in the physical world elicits two questions: 1) what do you mean?, and 2) how do you know? If the answer to the first question has no physical referent other than a set of symbols and the manipulation of these symbols, then the statement is said to be metaphysical. The world-branching of the MWI is, by its definition, a metaphysical construct.

    The photons and electrons of quantum theory are, by contrast, labels that correspond to specific experimental preparations and results, as well as symbols and symbolic operations.
    "The electron is in position x" corresponds to a specific instrumental result. QM assigns probabilties to the various possible instrumental results. The "state of nature" pertaining to some preparation prior to an experimental run is a statement regarding the possible and probable instrumental results of the run. The "state of nature" pertaining to the preparation following the run is the raw data (which might of course require some post-processing to make theoretical sense of).
    No. The "(classically-like) state of nature of which we learn stuff" is the stuff that we perceive --- from the results of quantum experiments to observations of the world at large. Using this as the evaluational basis for statements about the world provides an unambiguous criterion for deciding which statements about the world are more or less accurate --- the positivist approach is the one that characterizes science precisely because it avoids the troubles that accompany metaphysical statements.

    The troubles that accompany the MWI interpretation are unavoidable given the two assumptions on which it seems to be based: 1) there is an underlying quantum world wrt which we're necessarily perceptually ignorant, and 2) the formalism of qm is in 1-1 correspondence with the makeup and behavior of this underlying quantum world.

    Even though assumption 1) seems reasonable and arguable enough, it isn't necessary.
    Wrt assumption 2), you could just as well say that the formalism of qm isn't in 1-1 correspondence with an underlying quantum world. Either way, it's just metaphysical speculation.

    On the other hand, the minimalist probabilistic interpretation makes no such assumptions and is therefore consistent with the tenets of empirical science. It concerns what we know to be true (ie., even if assumptions 1) and 2) are true, which we don't know, it is still also true that qm is a mathematical algorithm for predicting the probable results of experiments.

    Anyway, lets assume that assumption 1) is true. Then any troubles that might accompany qm (or any theory of quantum phenomena) have to do primarily with our perceptual ignorance of an underlying quantum world. That is, it's primarily a hardware (biological as well as instrumental) problem rather than one of merely interpreting the formalism (since our perceptual ignorance of an underlying quantum world entails that we have no basis on which to evaluate how closely the qm formalism corresponds to it).

    I've been reading over your recent replies to me regarding MWI, as well as some other stuff. At this time the main thing that I'm getting from them is that MWI advocates want quantum theory to be more than it actually (at least the way I see it so far) is.
    Last edited: Jan 23, 2006
  9. Jan 24, 2006 #8


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Warning: highly philosophical content. If this continues, I think I'll move the thread to the philosophy forum.

    Of course it is metaphysical! But so it the hypothesis of the existance of an objective world. Even the phrase "the voltmeter reads 7V" is metaphysical because it includes in it, an implicit assumption of an objective existance of a voltmeter and of its reading. The only phrase that is not metaphysical is "I perceive a voltmeter reading 7V". But we are so very used to making this identification that we don't even bother.

    That's my point: this is just as well metaphysical. The only thing you really know is that you perceive a specific instrumental result. You have extrapolated here the "I perceive an instrumental result" into "there IS a specific instrumental result" ; but we're so used to do so this metaphysical interpretation, that you don't even notice.

    Entirely, but as such, it is a viewpoint which is "minimalistic" (and compatible with solipsism in fact! You equate the "state of the world" to the "state of your subjective perception" - the difference with solipsism is that you do not postulate the non-existance of a real world, you simply don't want to consider it). So a positivist interpretation denies the possibility of a description of an objective world. Although this is possible, I find it too early to deny the very well working hypothesis of a real, objective world.

    YES! That's the very meaning, to me, of a physical theory: you ASSUME for the sake of interpretation, that the theory is describing correctly an objectively existing world, and then you point out what are the relationships between the formal elements of the theory, and the (metaphysically postulated) external objective world.

    There are 2 possible denials of an interpretation:
    1) there IS no objectively existing world
    2) the theory at hand is NOT describing it

    Now, if we take 1) (which is the solipsist viewpoint), I think we can stop right away. If we take 2) then, by definition, we are not going to interpret a (wrong) theory, are we ? So in our quest to interpret a theory we must assume that there *IS* an objective world, and we must also assume that the theory describing it *IS* going to be describing it correctly. So of course, for the sake of interpreting quantum theory as a description of an objective world, we have to assume 1) and 2). Any denial of this starting hypothesis puts into jeopardy the very exercise of an interpretation.

    The only way out is solipsism! However, the hypothesis of an existing external world IS possible, and it has always been a very fruitful hypothesis, so there's no reason to drop it now.

    This is what I don't understand, btw. People arguing against an MWI interpretation of the quantum formalism often do this because they "find this assumption of parallel worlds in which copies of themselves wander around too weird to be true" ; but then they have no difficulties of kicking out ENTIRELY the assumption that there *IS* an objective world!

    Well, as I tried to point out, this statement is simply that QM could be the WRONG theory, which has a WRONG description of the objective world. I ALWAYS POINTED OUT THAT THAT WAS A POSSIBILITY, but then you should come up with a better theory! But this is NOT the starting point of an interpretation of a theory.
    Take Ptolomy's epicycles. It is empirically CORRECT. It is, however, a WRONG description of the solar system (with its spheres and so on), and Newton's theory does better. Once we have Newton's theory, we can UNDERSTAND how it came about that Ptolomy's epicycles gave good empirical results, and we also can re-interpret them as terms in a perturbation expansion in Newtonian dynamics. However, as long as you didn't have Newton's theory, there was no point in viewing Ptolomy's theory as a perturbation expansion of an unknown dynamics. That was not a helpful way to work with Ptolomy's theory, was it. If you wanted to interpret Ptolomy's theory, you had to stick to the spheres. It would have been difficult to reason physically in Ptolomy's theory without the spheres thing. So the *right* thing to do was to accept the hypothesis of reality of the spheres. This can, one day, happen to quantum theory. Maybe we'll find an underlying theory which will explain the formal success of quantum theory. Then we will not have to worry anymore about interpretational problems of the quantum formalism, because they will now be interpreted in function of the new theory. At that point, MWI will be of no use anymore. I never claimed anything else. But, there's no other such theory around ! (at least, that doesn't bump into other problems)

    The problem I have with seeing QM as "just a mathematical algorithm" is that you do not know anymore what the different parts in the calculational procedure are supposed to mean. It is not unthinkable that 1) and 2) are false, and as you point out, the assumption of their correctness is essentially a metaphysical hypothesis. But WHY SHOULD WE NOT CONTINUE TO ACCEPT THOSE METAPHYSICAL HYPOTHESES, IF THIS IS SOMEHOW POSSIBLE ?
    I mean, we never did that with classical physics, did we ? We always assumed that the state description of classical physics corresponded to an objective world outside, and it helped us in using classical physics, and we saw it as having a strong explanatory power. The moon really was there, and was not just "a construct of our knowledge", right ? It was a strong intuition builder to take the material points in classical physics are "really out there". Why deny this faculty to QM ?

    But I fail to see WHAT troubles. If I offer you a view in which the theory to be interpreted can:
    1) accept the existance of an objective world (a metaphysical hypothesis)
    2) assume that the formalism of the theory we're trying to interprete is a correct description of an objective reality (the correctness of the theoretical framework)
    3) be in accordance of how we should deduce from it, our subjective experiences

    isn't that better than an "interpretation" that is based upon:

    1) the assumption that there is no objective world (but only our subjective experiences, organized around some principles of knowledge)
    2) that the formalism of the theory is in fact fundamentally misguided but results nevertheless in correct empirical predictions ?

    Isn't 1) far more outlandish than the claims MWI makes and isn't 2) a priori denying the theory you're trying to interpret ? Of course all these considerations are metaphysical, but isn't a priori taking 1) and 2) as working hypotheses, a far more reasonable POV, which will pay off in a much better intuitive understanding of the theory at hand ?

    The point is: why can't we assume that quantum theory is more than you think it is, as a working hypothesis, to construct a view on it ? You can only *know* that quantum theory is less than it claims if you have a better theory, no ? Must we necessarily start by assuming it is WRONG, even before thinking about it ? Is that the best way to interpret it ?
  10. Jan 24, 2006 #9
    This is just my opinion so feel free to disagree.

    Is anyone surprised that when you start looking at things that are smaller than we can percieve, that we no longer have a means to absolutely prove anything, therefore QM is always going to be a hypothesis driven science that leads to theory rather than what we see we interpret as with most other sciences, until we can absolutely say that whatever maths and evidence we come up with is inference rather than stated fact.

    With the advent of QM theory there was a move away from experimentation into speculation that has continued unchecked to this day, when you can't see directly what it is your looking at then what do you expect? Yes QM has a more than a few bewildering concepts to get your head round, but for now I see it as the best fit theory, and untill someone comes up with some proofs for the other theories floating around I'll be looking at that not dreaming of membranes or strings. Fascinating though they are.
  11. Jan 24, 2006 #10
    Dear Reilly,

    We all have the same experience of medical treatment, but let me say that this is not the effect of "scientific knowledge". Science works with "information".

    Both Physics and Medical sciences are conducting their function by processing "information". It is a common misunderstanding that a pill, or a medical operation, produces a result in a deterministic way, but every scientific medical effort to cure is just an attempt to provide the best treatment. "Best medical treatment” is that which has the best probability to provide cure, therefore the "medical practice" is an attempt for treatment.

    So, like any other science does, medical science just makes "attempts", just offers a "prospect", just measures the "odds". Medical practice just handles information. There is no "knowledge" in science.

    It is human custom to reckon the information with the best uncertainty factor as knowledge. But this is a mistake. Knowledge should have no element of uncertainty.

    If you study a medical publication, it becomes evident that medical practice is about statistics and probabilities. Certainty and absolute knowledge is not there at all. Any well defined medical therapy, which has the best probability to cure, is always proven to fail to provide the remedy in certain instances. These “certain instances”, in which cure fails, are undefined until the moment they are being observed and it is possible to define them only by a statistical definition. This makes Physicians to “look” like non-scientists. But they are Scientists; they do not know the cure, they calculate the cure.

    Quantum Physics constantly reminds to us this axiom: Science is the mathematical/logical process of information.

    The scientific “peculiarity” of Quantum Physics was expressed by the ancient Greek scientist Heraclitus: “You cannot step twice into the same river, for other waters and yet others go ever flowing on”.

    And a final comment for vanesch: let me say that this is not "highly philosophical content". This is the most physical scientific content. It is the empathy within ourselves towards a specific classification that makes the content to "look" metaphysical or physical. And it seems to me that the forum is using a specific classification, although this is what Quantum Physics is trying to prevent us from doing. It is all about scientific methodology and philosophy has nothing to do with that.

    Last edited: Jan 24, 2006
  12. Jan 24, 2006 #11


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Just for the sake of it: string theory and variants ARE quantum theories which do not touch upon the general axiomatic structure of quantum theory.
  13. Jan 24, 2006 #12


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Just as a general remark: when I say that something is "philosophical" in content, I do not mean this in any kind of negative way. I know that in the culture of some physicists, "philosophy" is another word for mystical, meaningless blathering ("it is nothing but philosophy" Feynman style of talking), but I am not one of them. I simply mean that the discussion goes towards the more conceptual or primordial viewpoints, and goes away from the discussion of actual scientific practice, which is the discussions of the mathematical technicalities of the formalism and/or the operational technicalities of experimental work.

    I really think that I'm going to move this thread to the metaphysics forum...
  14. Jan 24, 2006 #13


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    I read this before I went away, and I have been turning over in my mind. The knowledge interpretation doesn't say anything silly like there is no state of nature. It just denies that quantum mechanics has anything to say on the subject. And why should QM do that? If you consider its history, beginning with Planck and going through Bohr and Heisenberg and Schroedinger and so on, none of them were concerned to erect an ontology, they were trying to explain how subatomic particle could behave in the apparently paradoxical way they did; how electrons seemd to orbit but without radiating. How X-rays could behave now like waves and then like little bullets.

    And the theory these founders erected DOES that! It accounts for our observation of these subatomic events (and of course nowadays of macroscopic events as well). Why does it have to carry an ontological burden too?

    Knowledge of what? Of the outcome of the experiment. As Bohr maintained, there is always an experiment in the discussion because QM is not a theory about the world but a theory about how we see the world.
    Last edited: Jan 24, 2006
  15. Jan 25, 2006 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The problem with that POV, is that it doesn't allow you to posit it as a universal theory of nature. In other words, it is not fundamental physics.
    I repeat the 3 possibilities we have:
    1) There is no objective reality
    2) There is an objective reality, but it is somehow not logical (in that there is no mathematical structure that can model it entirely, that you will always run into contradictions)
    3) There is an objective reality, and it is logical (in that there exists a mathematical object that models it perfectly).

    I would say that physics takes as its basic hypothesis, (3). (1) is solipsism, and (2) is illogical and as such, not open to rational inquiry on a deep level (this is often the hypothesis of religious world views, where a kind of deity decides, in an unfathomable way).

    So once we accept (3), all physical theories are supposed to be approximations of that ultimate mathematical model (the correct laws of physics). As such, they need elements in them that are supposed to at least try to come close to the mathematical model of nature. Now, in certain cases, we have found good models, just to find totally different, and better models afterwards. We can understand the good models as aspects of the better models. But when we "reach the bottom" of our theoretical constructs, in the sense that we arrive at our currently most fundamental description, we have no choice but to seek in them the things that are supposed to correspond to the mathematical object that is supposed to represent nature. For the moment, QM is our "bottom" theory. We have no underlying theory. If we want to interpret it (in the sense of turning the mathematical machinery into a physical theory) we have no choice but to take certain theoretical constructs in the formalism of QM as corresponding to that sought-for mathematical object that describes nature. Maybe it does this in a clumsy way, but it is always better than to simply give up on it, because if we do that, we LOOSE THE LINK BETWEEN experiment and theory.

    Bohr can only pretend to escape this necessary interpretation by assuming
    that QM is not a universal theory, nor is classical physics, but that there is some kind of barrier between both. He takes it that classical physics can pretend to a description of "macroscopic nature" (and he NEEDS this description to associate to a measurement apparatus, a certain operational meaning) WITHOUT having an ontological description of any microscopic nature. For instance, a light flash coming off a screen is probably seen as having a classically ontologically correct description (using classical EM). An electron, however, is to have such an ontological description (it hit the screen) when we want to make sense of the light flash in the first place, but is NOT to have such an ontological description without the screen. This is inconsistent. The electron WAS there and DID produce a flash, or it doesn't make sense to talk about the electron as being somewhere. But you can't have it both ways. You cannot have it that "an electron was ontologically responsible for the emission of a flash", and "we're not supposed to talk about the position of an electron, or even of an electron in the first place". Bohr seems to take the first view when it suits him (in order to be able to give a meaning to the quantum formalism, and to define the relationship between his observables and an apparatus), and take the second view when confronted with the consequences of his first view.

    1) "is there a flash ?" (yes, because macroscopic and thus classical, which has an ontology). :cool:

    2) "is the flash caused by an electron ?" (yes, because otherwise I loose my link between my quantum description (position operator of electron) and my experimental reality) :shy:

    3) "was the electron at the screen then ?" (eh... yes, we saw the flash) :uhh:

    4) "was an electron emitted from the electron gun ?" (yes, otherwise it wouldn't have arrived at the screen, would it ?) :blushing:

    5) "was there an electron between the electron gun and the screen ?" (sure, but we cannot talk about its properties, it is now described by a wavefunction, which is simply supposed to be a tool to link our knowledge from the emission of the electron to the flash seen on the screen) :approve:

    5) "is an electron an excitation of the Dirac field ?" (the Dirac field is a construction in our theory structurising our knowledge about the different ways of preparing and observing experiments, and we should not attach any reality to it, nor to its excitations) :approve:

    6) "so finally, there was no electron, because it is only a concept in our theoretical construction of our knowledge organization, namely "excitation of the Dirac field" ?" (indeed, the electron, as a excitation of the dirac field, is not supposed to correspond to a certain ontology, but just an explanatory concept in the organization of our knowledge, as proposed by quantum theory) :rolleyes:

    7) "But if the electron doesn't correspond to anything, why do you say that it was causing the flash ?" (eh... an electron is nothing else but a "flash causing entity" in our theory) :grumpy:

    8) "Are atoms made up of flash causing entities then ?" :bugeye:

    In Bohr's view, we then have: a classical current heats somehow a classical wire in a classical electron gun, "magic happens" and then a classical flash is emitted by a classical screen. The "magic happens" is the quantum description of the wavefunction of the electron (which is not supposed to correspond to anything real) evolving through the setup, which is not, in any way, supposed to correspond to any ontological description. But it needs suddenly to come into ontological existence when it activates the measurement apparatus, because otherwise, there is no link between the outcome of the measurement apparatus and the quantum mechanical description. This is where Bohr is cheating. He DOES give an ontological existance to his quantum chimera, but only when it suits him (as the proper activation of the measurement apparatus), and then denies it when it doesn't suit him. That's what I call "inconsistent".

    Also, I find this view highly incompatible in the framework of (3), because NO consistent mathematical description of nature is proposed. There is a dancing left and right between classical and not classical, real and "just formal"...
    Nevertheless it is not consistent with (1) either, because SOME ontology is given to macroscopic descriptions in classical theory. It's closer to (2): an illogical universe. An inconsistent world view is probably a suitable description of such a universe :yuck:
  16. Jan 25, 2006 #15


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    Sure. And so what? Maybe some other physics will do the job or maybe not.

    Do you really characterize your work with MWI as "fundamental physics"? How about careful's backward causation or the transactional interpretation? They all sound like sci fi bafflegab to me. "Built up logically" yes, but based on wild and crazy (i.e. implausible to me) premises.

    (added) One of the constant features of the history of physics is the temptation to reify. From Descartes' vortices to Kelvin's and Maxwell's mechanical ethers, this is something a prudent physicst (like Newton!) will do well to avoid. Let your theory do what it does, explain what it explains, and don't try to push it into being a fundamental account of reality.

    (second add) For the rest of your argument, my sense is somewhere between your options 2 and 3. I would plump for 3 except for that intrusive word "perfectly". But assuming that 3 is true and there is a perfect mathematical model of reality, why would you assume that quantum mechanics is it? Notice that QM is subsumed in field theory, that field theory has only shaky localization and problems representing interaction, and that "thingness" seems not to be preserved very well by the proposed physics of the standard model. You can't REALLY fillet non relativistic QM out of all this melange and treat it as a standalone philosophical construction can you?
    Last edited: Jan 25, 2006
  17. Jan 25, 2006 #16
    You are right, if you are accepting that science is about the "description of nature".

    But science is about the "information about nature".

    It is the difference between the phrases "description of an object" and "information about an object" that makes the difference in forming a conclusion about the "knowledge" of the object. Science is a common language that is used to share information. It is impossible to share knowledge about something, because knowledge is non-communicable.

    The failure in descrying the difference between “description” and “information” produces the confusion. Let me use an analogy: if I will describe to you the whether of tomorrow I am providing knowledge, I am sharing my knowledge with you. If I will give you a forecast for the whether of tomorrow I am sharing with you scientific information. This example seems obvious, because we are talking about the future and it is impossible to know the future, while it is only possible to calculate the future based on information with calculated uncertainty.

    This uncertainty is not based on our weakness to gather the proper scientific information. Whether forecasting depends on many physical elements, each one being distinct information, interacting with each other and the problem of describing the future whether is a problem of making the proper measurements and the proper calculations of an object that does not have one “picture”. We can not calculate the whether by seeing a single “picture” of information.

    Bohr said about quantum theory: “Evidence obtained under different experimental conditions cannot be comprehended within a single picture, but must be regarded as complementary in the sense that only the totality of the phenomena exhausts the possible information about the objects”. ( http://www-groups.dcs.st-and.ac.uk/~history/Mathematicians/Bohr_Niels.html )

    It is important that Bohr was talking about “the possible INFORMATION about the objects”. He was NOT talking about the “the possible description of the objects”. In Science, there can only one “description” of an object; while there can be a number of possible information about it. The “totality” off the possible information constitutes the “picture” of the physical object as a multi-folded description.

    Regarding the description of a physical object as the “evidence obtained under different experimental conditions”, according to Bohr, it “cannot be comprehended within a single picture, but must be regarded as complementary in the sense that only the totality of the phenomena exhausts the possible information about the objects”.

    So, Bohr said that the totality of information about “the flash”, “the electron on a screen”, “the electron between the screen and the gun”, “the emission of electrons”, “the excitation of the Dirac field”, “the flash causing entity”, “the electron as atomic particle”, which you enumerate in your post, in the extent that they “exhaust the possible information about the electron” must be regarded as “complementary pictures” in order to comprehend the “evidence obtained under different experimental conditions”.

    Quantum theory makes clear that a physical object, like an electron, is not possible to be comprehended/described by a single picture/information. The picture of physical objects is not univocal. Physical objects are “realities” that produce different information, which are being expressed by Science discretely because information/evidence is obtained under discrete different experimental conditions.

    The “inconsistent world view” is valid only if you assume a single pictured view.

    But it seems that the “World view” should not be comprehended as a single picture, but each “world view” must be regarded as complementary picture of a multi-pictured view.

  18. Jan 25, 2006 #17


    User Avatar
    Science Advisor

    Well, when this thread got moved, I got lost.

    Pragmatic and practical (ex)physicist that I am, I have litle clue about most of the discussion that has emerged from my initial post -- way above my pay grade. And, for me the knowledge interpretation is practical, simple, consistent with actual practice in the physics community -- and it is totally consistent with the way most people operate on a day-to-day basis. (I'll leave that idea as it is. I tend to avoid statements like, "It is clear,...." "It is obvious,..... But in this case, to me it is obvious and clear. To me wave functions that physically collapse, MWI-type interpretations, make no sense, and seem overly complex, non-Occam if you will.

    I will say that much of the impetus for this statement comes from my experience as a business man and consultant working with probability and statistics, simulation models, market and survey research, and from my experience as a research economist "bringing physics methods" to housing and urban economics. Probability is probability is probabiliy; it's all the same.

    What sciences are not hypothesis driven?

    In fact with the advent of QM theory, experimentalists went nuts, behaved as if they were in a candy store. QM is based on experiment. My terribly biased view is that physicists go into massive speculation when they don't have enough to do, when they can't interpret or design experiments.

    leandros_p -- What's the difference between information and knowledge? Funny, as a parent, a teacher, a manager, I always thought I was communicating my knowledge -- still do, for that matter.

    To me, knowledge is what you know. Information is that which informs, unless you want to reevisit Shannon and Wiener. Info and knowledge can be the same, or different.

    As far as I'm concerned, for example, my Doc's knowledge about the efficacy of various medications is exactly that. And, of course, he is highly aware of the problems and uncertainties. But, I pay him because he knows the ins-and-outs of treatments and medications.

    For example, we know that left untreated, an infection can and will kill you. We know that a bi-polar individual who does not take Lithium (or equivelent) wiil start climbing the walls in a manic episosde. If you drink no liquids for long enough you will die. So, in my humble, simple-minded way, I conclude we know a lot about medicine, including some certainties.

    Back to physics. But thanks for the discussion, even if much of it is way over my head.
    Reilly Atkinson
  19. Jan 25, 2006 #18
    Dear Sir,

    Let me present a schema of information and knowledge; doing that I will try to use your "language".

    The process of "knowing" is the following:

    Source -- Transmitter -------- Receiver -- Destination

    Between source and destination travels a Message – this message is called knowledge. Between the Transmitter and Receiver there is “noise”, always.

    The reception of message and the possession of the message thereafter are called knowledge status, at the destination.

    This is a general logical schema. The absolute “knowledge”, in the way that you understand the term, is when in the above procedure the destination accepts the message in the original, unchanged and complete form, which had at the source.

    Unfortunately, the intermediate stages of transmitter and of receiver, in being imperfect, combined with the noise that exists through the propagation media introduce alteration to the message, either by addition, or by subtraction. Therefore, the original message arrives at the destination modified, that is, the message at the destination is true but has a degree of uncertainty, while the original message at the source had no degree of uncertainty. The message at the destination is not the same as the message at the source. There is an uncertainty factor introduced in the message at the destination, during the process, which was not present at the source. This uncertain knowledge is called “information”.

    You say that you are communicating your knowledge, which in a similar way was communicated to you in the first place. Because you do that according to the above logical schema, your "audience" does not get knowledge; it gets information, that is, a message with a degree of uncertainty. This is not a metaphysical interpretation of knowledge or of information. This is a scientific way to describe physical communication and physical understanding.

    But then, don't we ever possess knowledge at all? Well, not in an absolute way. We possess knowledge in a relative way: First, we calculate the degree of uncertainty for the information we posses and we classify them accordingly. Then we accept the convention that knowledge is the least uncertain information.

    This process of transmitting a message between a source and a destination is the way we process the reality of our everyday life, which is full of continuous messages which have their source in the outside world and their destination is “within” us.

    So, scientists used the same methodology to get to know the messages that were transmitted by the physical objects and phenomena, which were being received by the proper receivers, either in the lab by instruments, or mentally by our logic. In the process, the message becomes information, in the context that it is modified compared to the original message, acquiring a degree of uncertainty.

    While it is perfectly normal to suggest that we “know” and that we transmit/communicate our “knowledge” to others, the words “know” and “knowledge” have only conventional meaning, in defining the least uncertain possessed information by us.

    There was a time, when Science also used such conventions. Today, Science of physics can not accept these conventions. It has to use literal terms in defining the “message” that physical objects transmit to our “receivers” as information, because the physical sources that are being examined, such as the atomic and subatomic particles and phenomena, send inconsistent messages.

    The objects of classic physics are sending messages to our scientific receivers that during the above process are altered by the imperfection of the stages of the system and by the noise. Scientists have learned how to overcome these problems, which are called measurement problems. BUT, the objects of quantum physics are sending messages to our scientific receivers that are inconsistent, not just altered or modified. So scientists felt obligated to treat these messages in their “true” form, that is, as information. This is the “truth” of physical reality that we refused to face in classic physics, in the past: We can not know physical reality; we can only be informed about physical reality.

    This is hard to accept, because for some strange reason it makes us feel like second class beings. Our certainty of “being” is staggered.

  20. Jan 25, 2006 #19


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    And given the fact that ALL objects are "objects of quantum physics" in a certain way, this simply means that our universe is inconsistent...
    I think that that step is slightly too fast, and I have difficulties seeing why people want to accept so very easily that the universe is INCONSISTENT (can you figure that ?), simply because the obvious view is "too wild and crazy" :bugeye:. So quantum theory, in this view, is a mathematical algorithm that allows us to find the statistics of inconsistency ?

    I mean, quantum theory forces us in a certain way, to led go some of our preconceived ideas. That always happens with a new physical paradigm. Relativity had us let go the idea of absolute time, for instance. (To me, quantum theory just learns us something else: the relativity of perception.)

    But I think there is a hierarchy amongst our preconceived ideas:

    1) there exists an objective world
    2) that objective world is logically build up
    3) our subjective world is defined by the objective world
    4) physical theories try to model the objective world
    5) relativity is a principle (time is relative)
    6) time is absolute
    7) newtonian physics works
    8) Ptolomean system works
    9) the earth is flat

    We've been changing a few times our view of the world through physical theories, and had to let go some of our previously cherished ideas. But I think that there is a difference between letting go "the earth is flat" and "there is no objective world", or "the objective world is inconsistent".

    I think that letting go things like "the objective world is logically build up" is a far greater step, than to say "the relationship between my subjective experience and the objective world became relative, and is not 1-1". This goes in the same direction as the "relativity of simultaneity". And it goes in the same way as a long tradition: the earth is not the center of the universe, the sun isn't, our galaxy isn't, even your perceived universe is not THE universe, but just one amongst many.
    As such, we can keep all the rest of our hypotheses. Is that not much simpler, more consistent and elegant way of viewing things than "the universe is inconsistent", or "quantum objects are not described by quantum theory" or so ?

    As to the remark that QM is maybe NOT containing the final principles of nature, I say: SURE! The day that we know more, we'll change again our world view. Maybe we'll have to take a step back. Maybe, after all, time IS absolute and relativity WAS wrong in its conceptual framework too. (Hey, maybe the earth IS flat too! :tongue2:) But what I don't understand is that people go all out of their way in order NOT to have to face the simple interpretation of the QM formalism: that (as all physical theories have always done) it *describes* nature, up to a point. That its mathematical structure *comes close* to what really happens out there. Ok, it takes on a very peculiar description of nature we are not used to (namely a difference between an objective reality and what we perceive of it, which is only one small part). But is that not conceptually simpler to swallow, than to go and claim that nature is *inconsistent* or that we don't have the ability to *describe* nature ?
  21. Jan 26, 2006 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think I understand what you say. You say that people using quantum theory, only use it in a limited way, and then apply the Born rule, get out results (which can be the starting point for a classical calculation) and that's it. As such, QM was a tool to delve into some microscopic aspects, why not.

    What I want to say is that there IS a possibility to view the entire quantum mechanical theory as a genuine, real description of the world and that it is *conceptually helpful* to do so when you are working with it, especially in tricky cases such as EPR. I do not find it conceptually helpful to say that the mathematical concepts I'm manipulating are NOT related to anything "real" out there: it kills all intuition one can potentially have about it. The electron as a flash-causing conceptual entity in an electrongun-screen setup without any reality attached to it, not very helpful, is it. Why not little spaghetti monsters throwing light balls at the screen ?

    Same with EM: you could somehow claim that the EM field is "not really out there", and is just a calculational tool to find the correct cause and effect relationships between moving charged particles. But that is not the right mindset when you are doing classical EM calculations. When you DO accept that there's a arrow attached to each position (for instance), you get a much better view on what is happening.

    It is in the same spirit that I'm pushing the MWI view. I'm not pushing it as "being ultimately true" ; rather I say that it is a view (IMO, the ONLY view) that goes naturally with the formalism of quantum theory, and which allows us to stick to some more fundamental hypotheses, such as the existance of a logically consistent objective reality, and the idea that we are somehow describing that objective reality. If the price I have to pay is that I have to accept the working hypothesis that I subjectively experience only PART of that objective reality, I do not find that so disturbing (many people do, apparently). If this is what *really* happens, I don't know, and up to a point, I don't care. I just want to have a coherent conceptual description of nature, that's all, and do so with what we have at hand. It does not need to be "ultimately right", as long as it is consistent, logically, and with observation.

    I find all other alternatives too pricy: "the universe is inconsistent", go figure. "we are unable to describe nature, just to know relationships between an experimental setup and measurement apparatus", what a defeat !

    The only one that is potentially acceptable, is: "QM is simply a wrong theory, but empirically accurate in those domains where it has been tested". But then one should put one's money where one's mouth is, and consider all efforts of a unitary quantum gravity and related to it, all quantum cosmology and so on, as a total waste of time. One should then look for the better theory right away! I'm however, rather pessimistic that a more modern theory will give us a SIMPLER conceptual picture. Usually, things get conceptually farther and farther from our intuitive ideas (the very argument to refute MWI). So I'm not sure that, if ever we find out that QM is wrong, the replacement will be easier to interpret, and closer to our intuition.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook