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

Making sense of observer participation in physics

  1. Jun 18, 2010 #1


    User Avatar
    Gold Member

    It’s now taken for granted among physicists that whatever the fundamental structure of physics may turn out to be, it's nothing at all like what we experience. Both Relativity and Quantum Mechanics are thought to describe a universe that's so radically different from the world we deal with in everyday life that it may never be possible for us to understand it in any intuitive way.

    But there are good reasons for thinking the opposite may be true – that is, the reason it’s so hard to grasp what QM and Relativity are telling us is that the world they describe is the world we experience – the world from the standpoint of “the observer”. The problem is that we still haven’t learned to conceptualize the world the way we actually see it.

    When we look at the world around us, our brains automatically interpret it in terms of real objects that last through time and move in space – basically the same way Classical physics pictures the universe, the same way philosophers have imagined it since ancient times. But at a deeper level the world we actually experience is not a world of objects laid out in space and time, but a world of physical interaction, centered on our own point of view in this ongoing present moment.

    Note that I am not talking about our subjective experience of the world, but about the nature of the physical world itself, from an observer's point of view. Nor do I want to deny the reality of objects moving in objective space and time, with properties independent of the observer. We know perfectly well that this Classical description of the world is extremely accurate – within limits. But we also know that it breaks down at the fundamental level addressed in Relativity and QM – the reality of objects “in themselves” is not absolute.

    Now obviously the point of science is to get beyond the subjective viewpoints of individuals and develop an accurate description of the world we all live in together. We’ve taken for granted that this means an “objective” description in which the observer disappears from the theory altogether.

    An alternative would be to describe the world as a system of interaction and communication between individual viewpoints. Not only the viewpoints of human beings, of course! Again, what we’re talking about isn’t something going on in our heads, but the actual physical environment “out there” in which we and everything else participate, interactively.

    And again – it’s certainly no coincidence that the information we get from this environment can be so well interpreted in Classical terms – that it makes so much sense to see it in terms of bodies with objective properties, moving in space according to very precise mathematical laws. But we should know better now than to take this picture for granted. We should be trying to understand how and why a system of interacting viewpoints might evolve to support something that looks so much like an “objective reality” of made entities with observer-independent properties.

    The main difficulty with this is that the philosophical tradition gives us almost no help in conceptualizing a world that’s neither “objective” in the Classical sense nor in any way “subjective”. Heidegger’s early (unfinished) work Being and Time was an explicit attempt to develop an ontology of relationships – but he made no progress after that, and none of his so-called followers even seemed to grasp the idea. In physics, Carlo Rovelli’s “Relational QM” seems very close to this approach, but it operates at a purely technical level, and his work on Quantum Gravity doesn’t address the role of the observer at all. So the work of conceptualizing the physical world that we (or anything else) actually experiences is still barely begun.
  2. jcsd
  3. Jun 18, 2010 #2

    The other point of science will be to break away with the inherent human naivety. The description laid out by classical physics was a dream that evaporated completely in 1964.
    Last edited: Jun 18, 2010
  4. Jun 18, 2010 #3


    User Avatar
    Gold Member

    The metaphysics of CS Peirce is your best bet here. Although he is famously difficult to understand.

    His synechism - http://en.wikipedia.org/wiki/Synechism - was an attempt to generalise semiosis, a logic of meaning-developing interactions, so that accounts of the material and the mental realms could be put on the same footing.

    See for example this essay...

    The key idea is that reality develops hierarchically. It begins as a state of vague formless potential, a sea of directionless fluctuations (firstness). Then by chance, there are fruitful interactions between some of these fluctuations and things can begin to develop (secondness). Then there is a generalisation over all these concrete interactions so that a whole persisting system emerges with its habits or laws (thirdness).
  5. Jun 18, 2010 #4
    Relativity and qm aren't really fundamental. The former might be characterized as a theory of measurement, and the latter as a theory of measurements.

    If we assume that our sensory experience, our shared reality, has emerged from a deeper reality, then should we also assume that this deeper reality operates according to different fundamental dynamics? That wouldn't make much sense. Why not assume that there's a fundamental dynamic at work, and that it pervades all scales of being and becoming? This is actually what qm, via wave mechanics, seems to be indicating.

    I think that at the fundamental level there is no point of view. Just a fundamental wave dynamic in a fundamental medium of unknown structure -- from which has emerged a hierarchy of 'particulate' media, with variable manifestations of the fundamental dynamic, interspersing and interfacing to create the complex persistent standing wave structures that constitute our 'reality' of ponderable objects.

    If you substitute 'waveforms' for "viewpoints", and add observer-dependent properties, then I agree.

    This approach is similar to the way I like to imagine the unfolding or evolution of ... everything. Just a bit too anthropomorphic (or, anthropocentric?) or vague in it's terminology (fruitful?, habits?). Maybe not. (Certainly no more vague than what I proposed.) I don't know. Anyway, one idea is that you define some properties of a fundamental medium, define a fundamental wave dynamic, write a computer simulation, and let er rip. Of course, a really (really) big computer screen will be required to truly 'understand' how the various emergent 'regimes' (and things like, us) ... emerge.
  6. Jun 19, 2010 #5


    User Avatar
    Gold Member

    Ever since Descartes posed this dichotomy between objective, material reality and the subjective “mental realm” of signs and representations, there have been efforts to resolve it back into a single coherent view of reality... Spinoza and Hegel being the key players. And it seems apt to compare Peirce with the NeoPlatonists, who brought ancient philosophy to its final phase in a transcendent vision of reality as a dynamic / evolutionary whole.

    But this kind of vision is almost opposite to that of “the observer” – seeing the world always from inside, from a particular point of view, in this moment, now.

    It’s remarkably easy for us – whether as philosophers or physicists or normal people – to imagine the world “from outside”, as a vast object extending through all of space, that’s been around a very long time. Then we think about what it’s made of and what kinds of principles operate in it. We’ve gotten very sophisticated at this, over the centuries.

    But it’s remarkably difficult for us to look at the world we actually see and feel around us, in real time, and imagine its structure. This isn’t a matter of reconciling the “mental” and “material” aspects of reality – the world of present-time physical interactions doesn’t correspond to either of those categories.

    Descartes made a very important and radical step, by conceptualizing for the first time the subjective inner world of the observer. That posed the dual problem philosophy has wrestled with ever since – (1) how can we know anything about objective reality when all our actual evidence is subjective experience? (2) How should we understand a world that’s made of these two utterly different components, subjective consciousness and material objects?

    Heidegger’s point was that what’s fundamental in our actual experience – and also, he believed, ontologically fundamental – was neither the objective reality of “things-in-themselves” that no one can ever experience, nor the isolated subjective “consciousness-of-itself” discovered by Descartes. We exist first of all not “in our heads” but “out there” in a world of relationships between us and other people and things, a world we all participate in by communicating from our respective points of view.

    Understanding the structure of this “between-world” is what’s so difficult for us, I believe. We automatically picture it “from outside” as something like a vast web of interconnections extending over all space and time – a kind of “relational” vision of reality first invented by Leibniz. This kind of thinking is almost inescapable for physicists. Heidegger’s followers mostly took the opposite tack, taking his analysis of “being-there” as a description of our subjective human experience, with no relevance for physics.

    I find it strange and wonderful that what’s so familiar to all of us – existing as points of view in communication with other points of view, in real time – also seems to be the hardest thing for us to conceptualize.
  7. Jun 19, 2010 #6


    User Avatar
    Gold Member

    Relativity and QM may only be steps toward a fundamental theory... but just because they are both theories of measurement, they open up a view into physical foundations that has yet to be understood.

    Your point of view makes sense, and is certainly the traditional one – i.e. that the ultimate description of the world should be independent of any point of view on it. Ultimately it is what it is, and the question is just how to describe its “fundamental dynamic”.

    With Relativity, a way to eliminate the observer’s viewpoint was discovered early on, in Minkowski’s reformulation of the theory in 4-dimensional spacetime. The observer’s location and state of motion still had to be taken into account in applying the theory in any specific case, but spacetime itself became the object of the theory, with properties independent of any observer’s viewpoint.

    With QM, it has also been possible to eliminate the observer, but only by making very bizarre assumptions, that have no other justification but to preserve the traditional notion of a world that’s real in and of itself, without regard to any point of view. I’m thinking here of Bohmian mechanics and also the many-worlds hypothesis. This kind of theory seems to help many people get more comfortable with quantum physics, without actually explaining anything.

    My own sense is that the way toward a deeper understanding is in taking “point of view” seriously as a basic feature of the physical world.
  8. Jun 19, 2010 #7


    User Avatar
    Gold Member

    I’ve mentioned Derrida following on from Heidegger before here, with "differance" lying in the fuzzy boundaries between dichtomies, a changing "aporia" of potential, which might be helpful?
  9. Jun 19, 2010 #8

    There isn't ever going to be a fundamental theory in an observer-dependent reality but only fragmentary models(partial glimpses) into the workings of reality. And it seems this is exactly the case we are forced to deal with. But there never was hope for a true TOE anyway, so this hardly changes anything. I'd say that in a participatory/relational/contextual reality, the prospects for reaching truths is very close to zero.
  10. Jun 19, 2010 #9


    User Avatar
    Gold Member

    Granted, we begin entangled with the world. And it seems like brute experience, and thus the thing that perhaps has to be explained in its own terms. But nothing can actually be explained - seen from a better vantage point - unless we step away from this naive beginning point in some reasoned fashion.

    And I would actually class Peirce as an internalist perspective as it is a view that remains within the system (whereas the classical mechanical view is an externalist discourse, putting the laws of nature outside the system they govern, rather than emerging naturally within the system as part of the general process of development and self-organisation).

    Another view on this is modelling relations. And here I think the interesting thought is that the "retreat into our heads" is in fact what is required to disentangle ourselves from the world that we want to make "objective". The modelling relation dichotomises formal models and informal measurements. And the better the model, the fewer the measurements it needs to make of the world. The informality (ie: subjectivity) is reduced as much as possible so as to maximise the formal representation (ie: the object knowledge).

    Of course the model is all in the head. And it should not be confused with the reality it describes (the map is not the terrain). But it is also as disentangled as possible from the reality which it describes. It sees the world in the most reduced and simple terms.
  11. Jun 20, 2010 #10


    User Avatar
    Gold Member

    I entirely agree with you here. The “stepping away” from entanglement is not only basic to science but to ordinary human experience as well. When we look at the world around us, what our minds “see” are objects located out there in space – not the physical interaction via photons that our eyes actually detect.

    This automatic translation from the “between-world” to the “object-world” is what makes Classical physics seem so in tune with everyday experience, whereas Relativity and QM seem paradoxical.

    I think though that "stepping out of the system" to a theoretical perspective is only one step, and that we also need to find a way of stepping back inside -- conceptualizing the point of view on the world we all actually begin from -- the structure of interaction we actually experience.

    I think I see your point. Some models – including essentially all the mathematical models used in physics – emphasize purely formal structure. To the sense that a model includes some kind of internal creative dynamic, it does seem closer to the “existential” perspective we have on the world “from inside”. (And since I was making historical notes above, I’ll add that the inventor of this style of philosophy was Jacob Boehme, an early 17th-century German shoemaker whose work Hegel much admired.)

    But I don’t think this perspective works, to make sense of “the role of the observer” in Relativity and QM. Specifically, because the philosopher “steps away” and remains outside the system, envisioning it as a complex dynamic system sitting there on his desk, so to speak, doing its emerging.

    There’s absolutely nothing wrong with modeling a system this way – it’s what science and philosophy have always done, with resounding success. But with the problem of Relativity and QM and how to combine them into a coherent vision of the world, we’re in a domain where the traditional approach has completely failed – not at the level of technical modeling but when it comes to producing insight into what the physical world is doing and why.

    Of course GeorgCantor may be right that “There isn't ever going to be a fundamental theory in an observer-dependent reality...” But since we’ve barely begun the task of modeling the physical world “from inside” – that is, the world of communicative interaction we actually experience, rather than the object-world we imagine we experience – that judgment seems very premature.

    The touchstone of such a model would be that it starts from the viewpoint of “the observer” – from a certain place, in this ongoing present moment now, in this particular interactive context. I don’t get that in Peirce or in the dynamic systems-modelers that you’ve referred to elsewhere, e.g. Salthe and Rosen. That’s not to say their work has no interest! – but it doesn’t seem to aim at conceptualizing the world made of interaction between observers, where everything participates both as observer and as something observed.
  12. Jun 20, 2010 #11


    User Avatar
    Gold Member

    Let me try putting this in the form of a philosophical argument.

    One of the most basic things we know about the physical world is that it works extremely well to model it as an objective reality – i.e. as a system made of things (like particles and fields and space and time), that have definite characteristics “in themselves”, independent of the context in which they’re observed in any particular case.

    But Relativity sets specific limits on this type of observer-independent description, and QM very strongly suggests that it doesn’t work at all, at the fundamental level.

    Now an equally basic thing we know about the physical world – assuming we know anything at all – is that it’s observable... or at least, some aspects of it are.

    So the argument goes like this --

    1. Though we normally interpret our experience as giving us information about objective reality, the physical world we actually experience consists of physical interactions that give us information about the world.

    2. So even if the world is objectively real, it is also something else, namely a network of interactions that communicate information between physical systems.

    3. Since we – or any other physical system – get information about other systems only through this communications network, that network must itself contain all the empirical information that any system can obtain. In particular, it must contain all the background-information necessary to define the information it communicates.

    4. So whether or not there is an objective reality underlying this system, it must be structured so that it provides an adequate reference-context for all its information – including the information that constitutes the reference-context for other information.

    5. Ergo, since the world is observable, it must have a certain kind of self-defining structure that is in principle independent of the existence of any underlying reality of things-in-themselves with definite properties.

    Now given QM, it seems to me the most likely conclusion is that there is no underlying reality “in itself” and that the inter-referential structure of the communications web is the structure of the physical world. Then as suggested above, “We should be trying to understand how and why a system of interacting viewpoints might evolve to support something that looks so much like an ‘objective reality’ made of entities with observer-independent properties.”
  13. Jun 21, 2010 #12

    Consider the fact that we are assuming an observer-dependent reality. In 1, 20 or 50 years i will be gone and gone with me will be the observer-dependent reality. It takes a leap of faith to say that:

    1. The observer-dependent reality will continue to exist(without the observer, from my POV)

    2. The description of the "physical world "from inside" "will be discovered within my lifetime.
  14. Jun 21, 2010 #13


    User Avatar
    Gold Member

    So how do you picture the nature of these interactions exactly?

    To me, following thermodynamic principles, they would seem to acts of equilibrating. When two particles interact, it is all about a lowering of the universe's entropy. So it is not so much that information is communicated, as information is being degraded. A tensegrity or constraints satisfaction approach.

    So particle B does not learn something about particle A via the exchange of a photon - as implied by "observation". Instead, the system involving A and B equilibrates an energy gradient, making the universe energetically flatter (more homogenous) and thus less observable. A locus of energy against a generally cold background is highly visible. And the acts of "observation" - the interactions among particles that are a collapse of a wavefunction - are in fact acts of thermalisation or decoherence.

    But how are you viewing interactions here? Also in terms of the mutual information?
  15. Jun 22, 2010 #14


    User Avatar
    Gold Member

    From the standpoint of an “observer” – in the broad sense, including any system that can react in some way to its environment – the “here and now” is the factual situation set up by its past interactions, in which various things can happen. This is what’s described by the wave-function in QM – the structure of possible outcomes within a given context.

    Interaction with another system then “selects” a certain subset of possibilities as “what actually happens” – which then modifies the observer’s situation, making a new context for what can happen next. The so-called “collapse” of the wave-function, creating a new wave-function based on new information. The new information selected in each interaction not only changes the local situation but also gets communicated on to other systems through further interaction.

    I think that’s the basic picture of what happens in physics, as seen “from inside”. I don’t see it mainly as the degradation of pre-existing information, along an entropy gradient. Rather I think of it as an evolutionary process, where what evolves is the ability of systems to create specific information interactively, in the context of given information inherited from the past, and then also to pass new information on as the basis for a new context.

    So to respond to your question... I don't have a specific notion of what interactions are, but rather a rough idea of what it is they need to do -- that somehow they have to contribute to a context in which other interactions can make a difference, contributing to another context, and so on... in order to make it possible for any sort of information to be "observed" or "communicated" or physically defined, through interaction. I'm trying to imagine an process through which random interactions can evolve into "measurements".

    In the highly-evolved universe we see now, measurement happens in a lot of different ways. Every way of observing any particular kind of physical information – position, momentum, mass, charge, spin, etc. – requires other observations of different kinds of information. The basic pattern is still the same – each interaction makes a difference to the local system, that modifies its receptiveness to future interaction, and also gets communicated on to other systems.

    This works so extremely well, now, that except at the quantum level, physical information seems to be completely well-defined “in itself”. Physical objects seem to exist with exactly definite properties, independent of any observer. And they seem to interact according to precisely defined mathematical laws that are exactly the same from any observer’s point of view. So when people first began thinking about the nature of the physical world, they took the notion of observer-independent reality as a basic assumption.

    QM argues against this, based on empirical investigation. One can also argue against it philosophically, by pointing out that if the world were at bottom an intrinsically well-defined mathematical structure in which observation played no role, there’s no reason to suppose that it would happen to support anything like the communication of information between local subsystems. Apart from QM itself, with its bizarre “collapsing” wave-functions, I don’t know of anything in mathematics that resembles the interactive selection process outlined above.
  16. Jun 23, 2010 #15


    User Avatar
    Gold Member

    Hmm, it seems to me that this is exactly what you rightly criticise - an external view of the idea of an interaction. QM is a model that puts the observer outside the continuously evolving system, the wavefunction, and then says suddenly everything is collapsed by an extrinsic act of observation. To be an internalist, we would have to see observation, whatever that ends up meaning, as an intrinsic aspect of the collapse. Which to me is what the decoherence interpretation, based on thermodynamic reality, is stretching towards.

    I prefer to see an interaction as an act of equilibration rather than observation because this says both sides of a wavefunction collapse, or however we are modelling the interaction, are changed. In some sense each end of the interaction is observing the other. There is no intrinsic direction dividing the observer and the observed. It is all relative!

    The idea of an observer says that there is a stable entity which undergoes a particular kind of change - from a state of uncertainty to a state of certainty. The entropy of the observer is reduced, and so it seems fair to speak of some information having been transferred from the observed system to the observing entity. That is the gist of the idea, agreed?

    From the second law point of view, that seems a dangerous idea as it appears to say that the more interactions we have, the more negentropy we create. When somehow it has to come out the other way round.

    So instead, I think it makes more sense to see an interaction as a process of mutual change - one in which information is not transferred but wasted. Minimised because where you once had two particles say of widely different energy levels, after the exchange of a photon of whatever, the difference is much less.

    Of course, what is then different about this mutual observation approach is that it has to take place across time - nonlocally - rather than happening locally in time. Which would be the transactional approach to QM. Past and future are not an intrinsic part of what is happening as it "all happens at once". Instead, a time-like path is what gets created - read off from the outside perspective as a photon having to follow a gradient from the hotter to the colder particle.
  17. Jun 23, 2010 #16


    User Avatar
    Gold Member

    Apeiron – I appreciate all your comments on this. And it’s interesting that there’s a question as to what counts as a description of the world “from inside”. To me this means, from the standpoint of someone or something, always at a certain place in the world, in real time. Our normal interpretation of the world translates from the “first person” view to the objective view – metaphorically, from radial coordinates to Cartesian coordinates. This works very well in classical physics, where everything can be assumed to be determinate “in itself”. But I think both Relativity and QM are showing us that this assumption no longer works at basic level. Instead of a reality that can be described “from outside”, from no particular point of view in space and time, at bottom the world is being woven, moment to moment, through the connections between particular viewpoints.

    You’re getting at a central issue here. We do want to get beyond the “first person” view to an understanding of the physical world we all participate in. And if at the base-level we can’t abstract from the observer’s viewpoint altogether, then we need to consider what it means for systems to observe each other, mutually.

    Now per Relativity, there certainly is an “intrinsic direction” to communication between the observer and observed. To the extent the two are spatially separated, it takes time for information to go from one to the other. But it does seem that at the quantum level all interaction is “transactional”, involving both parties. I think this is the meaning of the “Born rule” – the probability of an event is the square of the “probability amplitude”. So when two systems interact, the probability of a particular outcome is the same as if each system independently happened to choose that particular outcome. Even if communication does go from one to the other, there’s a mutual aspect to it – there’s only communication between us if I hear what you say, and we both understand the message.

    Actually – I’m thinking of Rovelli’s “Relational QM” here – there’s only a communication between us if I hear what you say, and then what I say about it to someone else agrees with what you say about it to someone else, and if what they both say about it to someone else agrees, etc. That is, an isolated interaction between two systems doesn’t collapse their wave-functions, so far as any other observer is concerned – it only “entangles” the two systems. Possibility becomes actuality only from a particular observer’s viewpoint. The weaving of “objective reality” at the level of classical physics is an ongoing process, and "facts are facts" only from some point of view within this process.

    I agree that decoherence – and particularly Zurek’s “quantum Darwinism” – are relevant here. These theories explain something about the emergence of a classical world on the basis of quantum statistics, but they don’t address the nature of measurement-interaction itself.

    This doesn’t make sense to me, because I’m thinking that interaction actually creates information through a kind of accidental mutual agreement. It’s not as though the energy levels of two particles are intrinsically well-defined (on some external scale) both before the interaction and after. I think there's a confusion here between the quantum mutuality involved in measurement and the classical mutuality of momentum-exchange. And it's not as though the "superposition" described by the wave-function is a real structure containing a lot of actual information that gets "thrown away" when it "collapses".

    The thermodynamic viewpoint operates at a classical level – it assumes information is well-defined in itself. Then to the extent a system has structure, there is a statistical “tendency” for the order to degrade, while there are situations in which structure can be built up and entropy gets reduced, locally. But I don’t think quantum measurement-interaction involves either increase or decrease of order / entropy.

    To me, this is the essence of the view “from outside” space and time. The picture I’m working toward is one in which everything has its own point of view in its own present moment – this ongoing “now” in which we see new facts coming into being all the time. The so-called “collapse” being the only thing we (or anything else) experiences. All the information anyone can have about past history is woven into this web of present-time interaction, which sets up new situations all the time making new things possible.

    Of course I’m not thinking of a universal simultaneous “now” that pertains to the entire universe at once – we know from Relativity that’s not how this works. I have my present moment as I write this, and you have yours as you read it -- obviously not the same moment, but physically connected. When events are separate in space, they’re also separate in time. But there’s a complicated web of ongoing physical communication through which we’re all participating, in the “real time” of our our respective viewpoints, in determining new facts and setting up new possibilities for each other.
  18. Jun 24, 2010 #17


    User Avatar
    Gold Member

    The way I am looking at it, the interaction does collapse the wave function. There is environmental decoherence. But the crucial extra bit is that the collapse is the least mean path across all the intervening spacetime. So it is a collapse that includes all that happened "along the journey", which is the retrocausal or time symmetric aspect.

    So say some photon from a distant star has crossed a million lightyears to arrive at your eyeball as you look up at the night sky. I would see this as a connection which coalesces in nonlocal fashion to produce a local, classical, outcome. A thermal gradient exists, which makes the emmission/absorption exchange very likely, even though it is also very unlikely in terms of the space and time across which the wavefunction must arise. And when the collapse does happen, in probablistic fashion, it has to take account of everything that may have been happening inbetween, along the way.

    So if some experimenter has set-up a twin slit array, or there is some star doing gravitational lensing, all this other classical level stuff becomes part of the probabilities. Which is the apparently retrocaual aspect, but really is more a matter of mutual satisfaction of constraints that spans spacetime so as to in fact create a concrete path taken by an event through spacetime. A possibility of an inteaction collapses to the actuality of an interaction. And it has an intrinsic temporal direction as the thermal arrow points from the hotter to the cooler side of the interaction.

    Now most interactions will be far shorter range than this of course. Most thermalising possibilities become actualities within split light seconds. Events like a photon crossing millions of lightyears would be vanishingly rare. And so - the key point - the bulk of the universe would be knit of very local interactions. The universe would be mostly decohered and classical at any one time. There would be a solid block of thermal history that makes it sensible to talk of a classical arrow of time. Everywhere you looked, there would be a broadly classical universe sliding down the entropic gradient of cooling and expansion. The quantum weirdness, the still unresolved wavefunctions weaving classicality across unusually steep thermal gradients, and so across unusually large expanses of spacetime, would be a remnant feature of what we can actually see.

    Where does this leave observers? It says observers are classical points of view which generally see classical worlds. But with luck, or special experimental set-ups, glimpses of the underlying quantum non-locality can be caught. Although in being caught, they are being rendered classical.

    In the view I am describing, the collapse is created by the wider observing environment in that between any two ends of an event, there exists some already concrete block of decohered reality that must be taken into about as the context. So it is an intrinsic act of observation in this sense. The world exists inside the interaction and must be taken into account in determining the probabilities involved - in shaping the wavefunction.

    Every interaction or act of collapse would move the general classical story forward in time. In terms of the information, this would be that thick block of accumulated actual events - choices made. And then the longer range events, like the ones that span unusual distances or bridge steeper thermal gradients, would be taking into account a denser intervening mass of this history. It would be equilibrating a potential for a fragment of new information as the least mean path through an already existing thicket of information.

    You do have that problem that information is regarded as inherently "crisp" - actual variety rather than potential variety. So a further part of my view requires a pre-informational state - a quantum vagueness. Information is what gets created locally as a concrete collapse to form a definite event. Though conservation principles still operate as in some sense this information was all spread around vaguely (as a wavefunction, a collection of probabilities). Then collapse puts it all definitely in the one place (or path). And importantly, not at any of the other places/paths that might have been possible before the event coalesced.

    The star had a potential to emit a photon and the photon could have ended up being absorbed anywhere. The wavefunction spread out across a wide area. But in the end, that smear of "informationalness" snapped into being as the event registered by an eye - and as the absence of any other kind of event anywhere else.

    I agree about the weight of accumulated history defining what is past and creating the constraints on the future - setting the context for collapses that follow on afterwards.

    But I am extending this view so that the "now" is more a statistical idea, a developing centre of gravity. As I say, events have no individual now until they have happened. And that "happening" is something that can extend far to either side of the bulk of what has already happened. So there is no real present moment. But for the bulk of events, being very short ranged, they will all be sharing much the same "now". Only occasional events will have "nows" with scales of millions of lightyears in their formation rather than split lightseconds.

    So most events will see the same world - not much will have changed in the time it takes them to coalesce. But for other rare events, huge intervening change would have to be taken into account as context for their quantum statistics. The universe would have expanded for a start, and so the photons would have to red-shifted. All sorts of gravity fields and other distortions may have arisen in the spacetime that is being bridged. The wavefunctions would look pretty complex.

    So not all events experience the same history. Just the statistical majority.

    Yes. But I am putting that felt weight of accumulated history inside the interactions - the collapsing wavefunctions - rather than outside.

    The observer effect, the extra information that can make a wavefunction more complex, is found within as the well-formed classical terrain a path must be found across. So in the case of the travelling photon, what an experimenter did or did not do with his twin slit apparatus, is a fact to be found inside the collapse as part of the context taken into account. The experimenter claims to be an observer standing outside the event - seeing it as a photon moving through his time. But the photon sees it the other way. The experimenter was one of the bumps in the road that had to be negotiated in finding some thermal gradient minimising connecting two spatiotemporally distinct locales.
  19. Jun 26, 2010 #18


    User Avatar
    Gold Member

    It’s a grand vision, for sure, that you're sketching here – but I don’t think it really addresses the basic issue, i.e. why does something have to be observed in order to be “real”? Why does the determinacy of a fact depend on that information being physically accessible?

    First, environmental decoherence has to do with how the wave function of a system evolves – it has nothing to do with the “collapse”. Decoherence explains why, when you measure a system, the result of the measurement will be one of a number of “classical” possibilities, but it doesn’t try to explain which of those possibilities you get. Essentially, it explains why a measurement never shows you a system being half one thing and half another – even though its wave-function consists mainly of such half-and-half “states”. It doesn't explain why QM describes things as "superpositions" in the first place, or how observation works to get a specific outcome out of that.

    Second, even though it’s tempting to think of the “collapse” as a real, objective event – and even if you’re willing to accept that such an event can have a “retrocausal” aspect that can extend over millions of light-years, and take into account all kinds of things “happening along the way” – that still doesn’t fit the observational evidence of QM.

    There are situations in which information can be obtained about a system (“collapsing” the wave-function) without interacting with it. And you can interact with a system in a way that results in a measurement, but then “erase” the information obtained, or make it physically inaccessible – and find that the superposition was not in fact collapsed by the interaction.

    So QM really will not let us off the hook, by making “observation” into an objective, physical event. (This is why many reputable physicists have gone so far as to involve the observer’s consciousness in the process of “collapse”. But that “solution” explains nothing – it only refers us back to the one thing in the universe that is less well understood than quantum physics!)

    I think the real solution is that at a basic level, the physical world is a communications system that defines all its own information through different kinds of interaction among different kinds of systems. In Rovelli’s words, “Physics concerns the information systems have about other systems.” Such a system can be described objectively, "from outside" -- but only through notions like "superposition" and "collapse" that literally defy logic.

    I won’t say that empirical evidence proves your idea is wrong. It’s always possible to adapt, no matter how “weird” the evidence – look at Bohmian mechanics! But I think that you're really bypassing the issue of observer participation in order to treat this situation as analogous to classical thermodynamics. Essentially you’re assuming that quantum “weirdness” is something that becomes “vanishingly rare” in a complex interacting system, that can still be described from an objective standpoint “outside the world.” I’m thinking of it as a pretty clear indication that we need a radical shift in perspective.
  20. Jun 26, 2010 #19

    What is left after GR and QM for 'structure' of the universe but information(as in relations between concepts) and/or mind?

    Yes, the foundations of the house are gone, but the house is still persistently there and it's time to investigate what the house really is. If the shift is not radical, there is no hope for it(was that Bohr who said something to this effect?).

    It all comes down to what we can say about reality without making unwarranted assumptions. The most we can safely say is the combination "observer-observations". All the rest is assumptions about whether the observations cause the existence of the observer or vice-versa. Does it really make sense to say that observations cause the existence of observations?
    Last edited: Jun 26, 2010
  21. Jun 26, 2010 #20


    User Avatar
    Gold Member

    Two things here.

    The first is that "observation" is being generalised to the notion of top-down contextual constraint. So you are probably thinking of observers as particular points of view (preserving local causality), whereas I am taking a fundamentally "non-local" approach where the constraints that shape up what is observed, what is real, are the global "point of view".

    So this is a way of transforming vague information to crisp information (what you probably mean by "physically accessible" - information that once crystalised, becomes part of that global constraining context, the collection of events that have definitely, classically, happened, and so now acting as a strong historical constraint on all future possibilities).

    Second, this is a limits based approach. So we don't have to argue for complete, 100%, transitions - as for example, a "classical collapse" as such. Classicality (completely crisp physical existence) is an ideal, a boundary state, which reality can approach, but not arrive at. This is key to removing the ontological mystery.

    This not a no-collapse view, because there is an asymptotically as-close-as-you-like approach to a state of crisp collapse. But equally, there is no actual collapse in that troubling usual sense. There is just a constrained as possible state which occurs when as much as possible context is acting to constrain matters.

    You will appreciate how this is an internalist perspective. The externalist perspective wants to stand to the other side of the collapse, viewing things from a classical realm. The internalist perspective is the view from inside the constraints, where things may become asymptotically classical, but the "observers" still remain within the system.

    Yes, decoherence sees the information leaking away into the environment. So there is no collapse as such. But the superposition becomes so generalised that it is a global context, asymptotically close to what we would mean by a classical realm in practice.

    And yes, decoherence needs more work as an approach. I like the way it is going (better than any other brand of QM) but I can see the ways it still tries to assimilate itself to traditional reductionist thinking. It has has not let go of certain ontological presumptions.

    As I said, the view I was attempting to describe sees reality as a mass of interactions which becomes 99.9999% classical where the interactions are densest (of small spatiotemporal scale/low energy) and then becoming increasingly fuzzy, vague, sparse, ragged, around the edges - the largest scale interactions. So reality becomes stiff and thickened where interactions are dense, thin and uncertain where they are sparse. And from a temporal point of view, this can be seen as a moving mass, whose centre of gravity defines the "now" of things. So a temporal story emerges.

    Prigogine, for example, took a view somewhat like this. Kauffman talks of the adjacent possible. Gell-Mann takes the consistent histories approach. All these systems thinkers would tend to view the puzzles of QM in terms of global constraints and fuzzy edged attractors existing within a phase space. The problems of QM have not been solved by a systems approach, but we can see some of the general thinking that would follow.

    All the experimental manipulations seem consistent with the view I am taking I feel. The retrocausal approach allows anything that happens "along the way" to be included as part of the global constraints that then determine the local probabilities.

    Again, locality emerges because top-down constraints shape up a particular wavefunction of possibility. There is some final uncertainty about how things will fall - how the wavefunction will collapse. A weakly constrained interaction would collapse in what looks like a highly unpredictable fashion. But reality is formed of tightly woven histories and so most interactions are so highly constrained that outcomes are almost classically predictable.

    QM experiments are based on relaxing the normal prevailing state of constraint so as to loosen the stays and maximise the surprise.

    The fiction is that there was crisp local information ever there. What exists before constraint produces a crisp outcome is only a smeared potential for information production. So talking about partial measurements or quantum erasers is a way of preserving a non-contextual, externalist, view of matters.

    Again, observation is global constraint in my book. It has nothing to do with particular observers, not even complex ones like humans.

    Well, the environment - the general prevailing state of the universe - is an informational structure. It has followed some particular historical path (and not taken many others that were once available to it). And it is the information embedded concretely in that environment that then shapes all future events. It does not determine them (that would be classical). But it does constrain them. There is a weight of prevailing history that makes the next moment very certain in terms of how far QM can introduce any actual novelty. But far off in the distance, the future can remain more uncertain, fuzzy, vague, open to spontaneous surprises.

    The near term is strongly constrained and so is almost deterministic. The distant future is weakly constrained and so far more indeterminate. At least so far as predicting the course of particular events, as we are normally talking about with QM experiments.

    I see it the other way round. I start from taking nonlocality as basic, then seek to recover locality only in the limit. There is no whiff of hidden variables in this story.

    So QM weirdness is the ground of being (or quantum vagueness - vagueness being a legitimate technical term). And the universe is being treated as a dissipative system - a system that dissipates the energy of vagueness to construct a crisp, classical-looking, structure built of information (a history of paths taken).

    This is the essence of the dissipative systems view. Negentropy (information) is extracted from a flow of entropy to create a persistent structure, a persistent global context.

    So it is a thermodynamic view, a systems view. But not exactly classical. Prigogine may have got the Nobel, but the idea of dissipative systems is still not widely taught or understood.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook