The typical and the exceptional in physics

[ddd123]

Maybe a less controversial term would be Intersubjective Reality Principle: since we share the same basic sensory-cognitive apparatus, we can agree on / synchronize our subjective information through language.

Digression: it's often said that intelligent aliens would understand the same physics / maths we do, like in the movie Contact when they find ET radio signals with prime numbers. But an alien with a completely different sensory-cognitive apparatus may not translate reality in that way (e.g. using numbers), but in a way so radically different as to be mute for us (we may even not be able to recognize such a species as intelligent). That's because we are "trapped" in that set of basic thinking categories (try to do away with quantities...).

 

[A. Neumaier]

This sort of debate has occurred often in history of science. GR says that absolute position doesn't exist. Zeno argued that velocity (momentum, motion) wasn't real. Ontological status of energy was questioned by many. Read "On Action At A Distance" by Maxwell http://www.informationphilosopher.com/solutions/scientists/maxwell/action_at_a_distance.html where he argues that the electric field is objectively real, against the majority opinion of the day (led by Ampere). It sounds a lot like this thread, in places.

Nice paper!

 

[secur]

...Intersubjective Reality Principle [IRP]...

Good point. SRP threatens to get into hard-core philo: solipsism, "Cogito Ergo Sum", Leibniz's Monads, that sort of thing. As soon as we engage in discussion, much less science, we move to IRP. Discussion makes no sense without inter-subjective reality principle.

However SRP is still relevant to science here and there. We can use both acronyms. BTW these days everybody seems to want their own acronym, principle, or theory - but that's not my motivation. We just need a brief way to refer to the principle(s).

SRP, as opposed to IRP, is relevant to the practice of science: how do we know we can trust other's results? Brings up topics like peer review process (PRP :-), replication of experiments, and fraud/fudging. But far more relevant is the following.

In typical Bell gedanken, the "weird" part is that Alice and Bob's results seem to depend on each other at the moment of observation, even though they're spacelike separated. But there's no problem locally: each gets random sequence of 1's and 0's as expected. We don't know they're correlated until the results are brought together. Now, according to SRP, no possible observer can know both sets of results until they're brought together. The illegal observer who can somehow do that has been derisively referred to as the "God's-Eye view". Thus SRP provides a way to dismiss the whole puzzle. It never happens that a single observer, at one point of spacetime - the only allowed type of observer -, knows something that requires FTL influence or similar.

I don't agree that solves the conundrum, although it does ameliorate it. But that's not the point anyway. The point is, this is one example showing SRP's relevance to the quantum ontic/epi question.

IRP is like co-moving observers in GR. SRP says you've got to be careful with that concept. In truth there's only one observer at one spacetime point. Co-moving observers are very useful, but when you get right down to it, remember they are an abstraction. Their results must be brought together in one place, for one observer (scientist) to make sense of. That explains some GR oddities, which I won't get into.

I'll show SRP/IRP's relevance to science, QM, and ontic/epi wavefunction question in many other ways, if I get around to it. The only problem is organizing the plethora of examples! Anyone is welcome to come up with a few. The best attempt wins a surprise prize.

...an alien with a completely different sensory-cognitive apparatus may not translate reality in that way (e.g. using numbers), but in a way so radically different as to be mute for us...

That's certainly conceivable. But, as you know, the hypothetical alien's alien mode of thought is NOT conceivable to us.

Nice paper!

It really sounds a lot like today's controversies in places. Here's a quote from Maxwell, where he paraphrases his opponents:

"If we are ever to discover the laws of nature, we must do so by obtaining the most accurate acquaintance with the facts of nature, and not by dressing up in philosophical language the loose [my bold] opinions of men who had no knowledge of the facts which throw most light on these laws. And as for those who introduce aetherial, or other media, to account for these actions, without any direct evidence of the existence of such media, or any clear understanding of how the media do their work, and who fill all space three and four times over with aethers of different sorts, why the less these men talk about their philosophical scruples about admitting action at a distance the better."

It's amusing that the word "loose" is also used by Murray Gell-Mann, with the same pejorative intent, in characterizing his opponents: "People say loosely, crudely, wrongly ...". There's nothing new under the sun; furthermore, plus ça change, plus c'est la même chose.

 

[ddd123]

I'll show SRP/IRP's relevance to science, QM, and ontic/epi wavefunction question in many other ways, if I get around to it. The only problem is organizing the plethora of examples! Anyone is welcome to come up with a few. The best attempt wins a surprise prize.

I would really like to hear them. I can think of examples in physics where SRP/IRP can be brought in for comment, but none in which they have a clear impact.

 

[stevendaryl]

If you're going to suggest that the phase space point (p,q) of classical physics is an example of an epistemic state, then we'll just have to agree to disagree :-)

I'm not sure if this is missing the point (no pun intended), but while I agree that a point in phase space is ontic in classical physics, in practice, we don't have a point, but a region of phase space or a probability distribution on phase space. We don't know precisely where the actual system is, we only know it probabilistically. The distribution is epistemic, rather than ontic (it reflects our subjective knowledge).

The corresponding thing in quantum mechanics to a "point in phase space" is a pure state, and the corresponding thing to a distribution in phase space is a density matrix. The weird thing about quantum density matrices is how they mix up ontic and epistemic. There is no unique way to determine which parts of the density matrix are due to our lack of information, and which parts are due to objective facts about the system. What makes quantum measurements strange, and what it makes it impossible to prove that they violate causality, is that in a measurement, the same uncertainty seems to shift from ontic to epistemic. After the measurement is done, you can pretend, after the fact, that it was epistemic all along.

Concretely, if you have a pair of particles in an entangled state (measured by Alice and Bob at a spacelike separation), then the density matrix that you obtain from one of the particles by tracing out the degrees of freedom of the other particle is a density matrix. This density matrix is not epistemic, but ontic---you started with a pure state, where presumably you know everything there is to know about the composite two-particle system. After Alice measures her particle's spin and "collapses the wave function", you can think of Bob's particle as being in a definite spin state. But Bob doesn't know WHICH definite spin state, so he would describe his particle using a density matrix---the same density matrix that he originally constructed from a pure two-particle state by tracing out Alice's particle's degrees of freedom. So the wave function collapse has no effect on the density matrix used by Bob, it just changes it from being ontic to being epistemic.

 

[Simon Phoenix]

I'm not sure if this is missing the point (no pun intended), but while I agree that a point in phase space is ontic in classical physics, in practice, we don't have a point, but a region of phase space or a probability distribution on phase space. We don't know precisely where the actual system is, we only know it probabilistically. The distribution is epistemic, rather than ontic (it reflects our subjective knowledge).

Absolutely

Totally agree - I only mentioned the phase space point because it seems to be the 'archetypal' example of what is meant by an 'ontic' state according to the foundations guys.

The weird thing about quantum density matrices is how they mix up ontic and epistemic. There is no unique way to determine which parts of the density matrix are due to our lack of information, and which parts are due to objective facts about the system

Absolutely again

Let's suppose that Alice prepares a spin-1/2 particle in a pure state according to
(1) an eigenstate of spin z
(2) an eigenstate of spin x or spin z or spin y
(3) an eigenstate of spin in any direction

She gives the particle to Bob and only tells him that she has prepared it according to (1), (2) or (3). In each case Bob's density operator (the one that he assigns) is the same (just half the identity operator) but we would, I think, be justified in claiming the 3 situations could be significantly physically different from one another - at least according to Alice who knows what physical state has been prepared. Interpreting the density matrix - even as a 'proper' mixture has this ambiguity as you say.

What is different in each of these 3 cases is the measurement strategy Bob must perform (on an ensemble of identically prepared particles) to determine which eigenstate with say 90% probability in the least number of measurements.

 

[ddd123]

I only mentioned the phase space point because it seems to be the 'archetypal' example of what is meant by an 'ontic' state according to the foundations guys.

To think of it, I'm starting to have second thoughts about this. Are we sure that the foundations guys actually agree on this? Can we have a clear reference?

I can think of an example in general relativity, which is classical mechanics. By taking the point in phase space in general relativity as ontic, you can derive that the total energy of the Universe is zero, i.e. the Universe is empty. See here: https://arxiv.org/pdf/gr-qc/0605063.pdf .

"The equivalence principle, says that at any location, spacetime is (locally) flat, and a geodesic coordinate system may be constructed, where the Christoffel symbols are null. The pseudotensors are, then, at each point, null. But now remember that our old Cosmology requires a co-moving observer at each point. It is this co-motion that is associated with the geodesic system, and, as RW's metric is homogeneous and isotropic, for the co-moving observer, the zero-total energy density result, is repeated from point to point, all over spacetime."

But the "flaw" seems to be something similar to the problem of the test charge for fields: a point particle's rest mass which the comoving observer is attached to is not considered in the argument. Here the SRP comes into play because the subject's role (the "test charge") is ignored in depicting the object (the "field") and you come to an "absurd" conclusion (although the author of the article takes it seriously, as well as many others).

So @secur's point about comoving observers has appeared here, it seems.

 

[stevendaryl]

"The equivalence principle, says that at any location, spacetime is (locally) flat, and a geodesic coordinate system may be constructed, where the Christoffel symbols are null. The pseudotensors are, then, at each point, null. But now remember that our old Cosmology requires a co-moving observer at each point. It is this co-motion that is associated with the geodesic system, and, as RW's metric is homogeneous and isotropic, for the co-moving observer, the zero-total energy density result, is repeated from point to point, all over spacetime."

Okay, but isn't this an argument that the pseudo-tensors are not actually physically meaningful (since they can be made to be zero at each point)? Of course, you can use the pseudo-tensors to compute things that are physically meaningful, like the curvature. So maybe that example is closer to the vector potential in electromagnetism. By a suitable choice of gauge, you always make it vanish at any given point. So it's not physically meaningful. But it can be used to derive the electric and magnetic fields, which are physically meaningful.

There definitely is some kind of connection (no pun intended) between the wave function and gauge transformations, in the sense that the wave function transforms under a gauge transformation in a way related to the way the vector potential does. So maybe the wave function and the vector potential are approximately equally physical (or equally unphysical).

 

[ddd123]

Okay, but isn't this an argument that the pseudo-tensors are not actually physically meaningful (since they can be made to be zero at each point)?

That's not what I understood to be the article's thesis: "the Universe is singularity-free, and was created ab-nihilo" immediately precedes that quoted part. It's also what some popularizers are saying at the moment, their arguments reconduct to this one and the quantum fields' ZPE.

But of course I could be misunderstanding something.

 

[secur]

In typical Bell gedanken, the "weird" part is that Alice and Bob's results seem to depend on each other at the moment of observation, even though they're spacelike separated. But there's no problem locally ... It never happens that a single observer, at one point of spacetime - the only allowed type of observer -, knows something that requires FTL influence or similar.

I'll show SRP/IRP's relevance to science, QM, and ontic/epi wavefunction question in many other ways, if I get around to it. The only problem is organizing the plethora of examples! Anyone is welcome to come up with a few.

I would really like to hear them. I can think of examples in physics where SRP/IRP can be brought in for comment, but none in which they have a clear impact.

Is "relevance" the same as "clear impact"? Many troublesome points, already known, can be seen as an application of S/IRP. Thus it can be used to organize these types of objections, perhaps even reveal new troublesome points that people have missed. But S/IRP can't come up with new physics. Would you call that having a "clear impact"? Like the example given above re. "God's Eye view" of Bell experiment. It's been mentioned before, to dismiss nonlocality claims, and flows naturally from analysis of the experiment using S/IRP.

The key reason I haven't already given more examples: it first requires a closer look at what S/IRP really means - boring, but necessary. Need a firmer foundation before proceeding. I'll give it a shot but it's going to be painful!

But the "flaw" seems to be something similar to the problem of the test charge for fields: a point particle's rest mass which the comoving observer is attached to is not considered in the argument. Here the SRP comes into play because the subject's role (the "test charge") is ignored in depicting the object (the "field") and you come to an "absurd" conclusion ... So secur's point about comoving observers has appeared here, it seems.

S/IRP insists that there MUST BE a real physical observer involved somewhere along the line. Could be a human, computer, whatever; but S/IRP won't allow you to get too pie-in-the-sky. Note, in this example, the GR analysis makes statements about all co-moving observers on a Cauchy surface. But we must hypothesize a lapse of 7 billion years (actually a lot more than that, but you get the idea), allowing all the data to come together in one place for analysis by one observer, before drawing conclusions. Otherwise you're making the God's-Eye view mistake.

Note I'm not objecting to gedankens. We can imagine spaceships at almost speed-of-light, observers falling into black holes, etc. We all know such things will never happen in the history of the human race, but at least they're physically allowed. Only truly impossible things are no good: like God's-Eye view, re-running an experimental observation in the past with a different detector setting, and so on.

Many think this is mere nit-picking (at best). Who knows, maybe they're right, but that's S/IRP analysis. By tethering physicists to what's actually possible it makes physics a lot less fun, but meaningful. Whether that's an acceptable trade-off can be disputed.

 

[Eye_in_the_Sky]

My issue with going into detail about perceptions of reality and so forth is that we end up talking about mental constructs of mental constructs (our mentally constructed models of a perceived reality which is itself a mental construct) and so we end up with this tortuous nesting of mental constructs.

So I like to keep things simpler and just assume that there is an external reality, that our mental construct of that reality, derived from sensory inputs, is pretty faithful. So the question then becomes one of whether the models we adopt to explain things have a mapping to some external reality or whether those models merely describe what we can know about that reality.

Here is an alternative 'mind set':

The matter at hand is one of "ontological status" in connection with the CONSTRUCT of 'spacetime'. Specifically, the question is:

Regarding the CONSTRUCT of 'spacetime', what is it that 'IS'?

Or, to use Bell's words: What are the 'beables'?

To my reckoning, I would say:

beable: a 'causative' agent that acts in spacetime

So, for example, a 'classical system' that resides in spacetime and that can have interactions with other such 'systems' in spacetime will then bear the DESIGNATION 'beable'. Such a 'system', we will say, has "ontological status" with respect to the CONSTRUCT of 'spacetime'.

But, of course, this TYPE of 'system' is a special case of 'beable' – for it is a 'local beable'.

To my reckoning, I would say:

local beable: a 'beable' that resides in spacetime

The above definitions are my own WORD DEFINITIONS.

From what I have so far read of Bell's words on this notion, I have not yet seen a clear ARTICULATION.

 

[zonde]

I take it we all agree on this fundamental fact: "Reality" is (when you get right down to it) subjective. This principle needs a name, how about "Subjective Reality Principle" (SRP).

I sort of understand what you mean but I can't agree to your choice of word "subjective" and "reality".
Objective/subjective does not mean that we have/don't have some absolute knowledge. Subjective means that particular thing can freely vary from observer to observer while objective means that one observer can fix the state of thing so that for other observers respective thing is fixed as well.
And "reality" is not any of our models but rather imaginary concept on which all our successful models of reality converge. Reality is like a value of a function which is not defined at particular point but function converges when we take limit to that point.

In typical Bell gedanken, the "weird" part is that Alice and Bob's results seem to depend on each other at the moment of observation, even though they're spacelike separated. But there's no problem locally: each gets random sequence of 1's and 0's as expected. We don't know they're correlated until the results are brought together. Now, according to SRP, no possible observer can know both sets of results until they're brought together. The illegal observer who can somehow do that has been derisively referred to as the "God's-Eye view". Thus SRP provides a way to dismiss the whole puzzle. It never happens that a single observer, at one point of spacetime - the only allowed type of observer -, knows something that requires FTL influence or similar.

There is hierarchy of models. There are more basic ones that we utilize for building higher level models. The most basic models are not product of science but instead we have them at birth or learn very early at childhood. These very basic models define "society view" on reality and while it is not exactly "God's-Eye view" it comes very close to it. And this "society view" on reality is fundamental for science (but not for philosophy).
Bringing results together can be done in practically countless different ways and result of correlation is supposed to be independent form chosen method. This is possible because our chosen method does not alter results. This comes from very basic model that takes measurement records as factual (objective) and it's the same model on which we base "society view" on reality. If we object to this very basic model then we have to replace it with alternative model in a way that gives us alternative version of "society view" on reality. I don't know about any such proposal and I am skeptical that it's possible.

 

[RockyMarciano]

There is hierarchy of models. There are more basic ones that we utilize for building higher level models. The most basic models are not product of science but instead we have them at birth or learn very early at childhood. These very basic models define "society view" on reality and while it is not exactly "God's-Eye view" it comes very close to it. And this "society view" on reality is fundamental for science (but not for philosophy).
Bringing results together can be done in practically countless different ways and result of correlation is supposed to be independent form chosen method. This is possible because our chosen method does not alter results. This comes from very basic model that takes measurement records as factual (objective) and it's the same model on which we base "society view" on reality. If we object to this very basic model then we have to replace it with alternative model in a way that gives us alternative version of "society view" on reality. I don't know about any such proposal and I am skeptical that it's possible.

But you are saying that this skepticism has a "psychosocial" basis, i.e. it is not scientific or philosophical, right?

 

[zonde]

But you are saying that this skepticism has a "psychosocial" basis, i.e. it is not scientific or philosophical, right?

My skepticism has philosophical basis. I would regard any idea about hypothetical alternative version of "society view" on reality as philosophical.

 

[RockyMarciano]

My skepticism has philosophical basis. I would regard any idea about hypothetical alternative version of "society view" on reality as philosophical.

Fine, but it is rooted on the trust in a "society view", which suggests something related to social groups ways of thinking.
I mention it because I would find such philosophical stance hard to defend as a premise from which to decide about issues related to physics.

 

[zonde]

Fine, but it is rooted on the trust in a "society view", which suggests something related to social groups ways of thinking.
I mention it because I would find such philosophical stance hard to defend as a premise from which to decide about issues related to physics.

Easy. Scientific knowledge is possession of society. No society, no scientific knowledge.

 

[vanhees71]

I think now this thread really goes way off the track. Of course, the findings of the natural sciences about nature are not just a "society view". This nonsense is sometimes claimed by sociologists, but it's plain wrong. It's the opposite of what's natural sciences are about, namely about reproducible objective observations of nature.

 

[stevendaryl]

Easy. Scientific knowledge is possession of society. No society, no scientific knowledge.

Not that I disagree with anything being said, but there are those who dismiss all talk about the foundations of quantum mechanics as "just philosophy". I don't think that's true, because I think that there is a lot of physics involved in understanding what goes on in quantum experiments. However, when you talk about scientific knowledge being a social construct, I would have to agree with the complaints--at this point, you're really talking about philosophy, not physics.

As I said, it's not that I disagree, but it increases the danger that the thread will be closed due to being off-topic for a physics forum.

 

[vanhees71]

It's not even philosophy, it's plain wrong, and I must say at this point, the thread in fact should be closed!

 

[ddd123]

It's not even philosophy, it's plain wrong, and I must say at this point, the thread in fact should be closed!

Only because of what zonde said? I agree that it's off topic. I was waiting for secur's examples.

 

[zonde]

I think now this thread really goes way off the track. Of course, the findings of the natural sciences about nature are not just a "society view". This nonsense is sometimes claimed by sociologists, but it's plain wrong. It's the opposite of what's natural sciences are about, namely about reproducible objective observations of nature.

Hmm, you are probably missing the context. I proposed "society view" as an way to define objective observations about reality as opposed to definition of "objective" as some sort of absolute knowledge about reality that an individual could posses and that is basically void.

Btw you yourself used the phrase "objective observations of nature". So, in what sense have you used it?

 

[vanhees71]

It's perfectly the opposite of what you claim: It's independent of "society views". Anybody can reproduce the observation and will find the same result, independent of his or her cultural background.

 

[zonde]

It's perfectly the opposite of what you claim: It's independent of "society views". Anybody can reproduce the observation and will find the same result, independent of his or her cultural background.

Hmm, then my only conclusion is that my choice of the term "society view" was very unfortunate as it triggers very strong associations with something rather irrelevant to my point. My idea was that objective is something about what different individuals would agree. And this is basically what you say (you just leave out communication part).

 

[A. Neumaier]

I think now this thread really goes way off the track.

It is completely off topic since post #59 - and should have from there been a different thread with a title such as
''Ontology and epistemology of quantum mechanics''.

 

[secur]

I sort of understand what you mean but I can't agree to your choice of word "subjective" and "reality". Objective/subjective does not mean that we have/don't have some absolute knowledge.

Subjective means what's known or experienced by an individual, a "subject". We have "absolute knowledge" of this, and nothing else.

And "reality" is not any of our models but rather imaginary concept on which all our successful models of reality converge. Reality is like a value of a function which is not defined at particular point but function converges when we take limit to that point.

We can't know "reality", ontology, for certain. In a sense, it's a myth invented by our subjective minds. Having said that, of course I believe in real reality as much as anyone - can't help it.

Bringing results together can be done in practically countless different ways and result of correlation is supposed to be independent form chosen method. This is possible because our chosen method does not alter results. This comes from very basic model that takes measurement records as factual (objective) and it's the same model on which we base "society view" on reality.

Yes "society view" is standard. We realize our subjective realities coincide, so we all believe in objective reality.

If we object to this very basic model then we have to replace it with alternative model in a way that gives us alternative version of "society view" on reality. I don't know about any such proposal and I am skeptical that it's possible.

You might ask: if reality is ratified (and reified) by all, why dispute it? Ok, initially it's an invention of our subjective minds. But since there's so much evidence for it, why not just accept it?

Well, as mentioned above, I do accept it FAPP, for non-relativistic classical physics. But QM and other modern science force us to question "reality" closely. Objective reality is constructed from sense inputs, consensus opinion, and a few more ingredients, notably memory. Every step of the construction makes hidden assumptions. Some of them, it turns out, aren't entirely right.

 

[ddd123]

Subjective means what's known or experienced by an individual, a "subject". We have "absolute knowledge" of this, and nothing else.
[...]
We realize our subjective realities coincide, so we all believe in objective reality.

There's a reason we call it objective and not absolute. Well, of course there's that the latter is stronger, but to clarify in what sense could help:

http://www.etymonline.com/index.php?allowed_in_frame=0&search=object
'from Medieval Latin objectum "thing put before" (the mind or sight)'
http://www.etymonline.com/index.php?allowed_in_frame=0&search=subject
'from Medieval Latin subiectare "place beneath"'

Well, how about something like this. The subject has its mode of perception and cognition, a grid within which he frames the world: this grid he puts underneath an object, what stands in front (maybe, initially, he imagines himself in place of the grid or as the grid, the "eye view": if it's all-encompassing then it's "God's eye view" that can witness a Bell type experiment unfolding, for example). For example, it may well be the Cartesian grid with its pointlike or extended objects within it in Newtonian mechanics.

So objective and subjective aren't really excluding each other. Once an inter-subject is agreed upon, objectivity follows - that's objective! But we choose the grid. That can be a model, criteria...

 

[secur]

We have a Shared Model of Objective Reality (SMOR, let's say), which is constructed from our individual Subjective Realities, via a number of steps. It's not entirely right. In classical physics, for the most part, it was fine. But modern science often contradicts its assumptions. When that occurs we call it "weird" and tend to have trouble dealing with it. Here are seven examples.

1. SMOR assumes 3-d Euclidean space, with independent 1-d time (as Kant, correctly, said). But "real" reality is Minkowskian, even Reimannian. That's part of the "weirdness" of relativity.

SMOR assumes that when we observe the property of an object, it really has that property. Even that can be doubted, but let's suppose it's true.

2. However SMOR also assumes that the object had that property just before we observed it. We don't know that, and it could be wrong. That's part of QM "weirdness". Or I should say it's one way to explain part of QM weirdness, namely the collapse interpretation.

SMOR assumes the past is fixed and unchangeable. OTOH the future is uncertain and doesn't yet exist.

3. But actually the future could "already" (whatever that means in this context) exist, and be unchangeable, just like the past. That's the Block Universe.

4. Or, the past could be changeable for all we know. That's retrocausality, another so-called QM interpretation.

SMOR assumes there's exactly one spatial reality existing along one unique time axis, that we all share.

5. But there might be none. That leads to the present discussion of epistemology vs. ontology.

6. Or, there might be different versions of the one "real" reality for each subjective observer. That's part of relativity "weirdness".

7. Or there might be lots of other "parallel" realities, with their own separate observers. Admittedly there seems to be just one that we experience, but we wouldn't know about the others. That's the Multiple-Worlds interpretation.

And so on. This random listing is haphazard and tedious. The better way is to simply go through the steps by which SMOR was constructed. The "hidden" assumptions are easily uncovered, and listed systematically. Every one can be questioned. It's easy to find alternatives that don't contradict any human experience or experiments. Many of them point directly at a modern theory, or at least some ontological interpretation thereof. Indeed, they cover all so-called "weird" aspects.

Many key theoretical breakthroughs in modern physics have happened simply because someone stumbled on the fact that one of SMOR's hidden assumptions was wrong. There are other hidden assumptions of SMOR, easily revealed by a little systematic analysis, which no scientist has yet doubted. They may be key to future breakthroughs.

Bottom line, the first thing to do is to systematically analyze how we came up with SMOR. @ddd123, that's what you're beginning to do in your post above.

 

[zonde]

You might ask: if reality is ratified (and reified) by all, why dispute it? Ok, initially it's an invention of our subjective minds. But since there's so much evidence for it, why not just accept it?

You have misunderstood me. I say that you can question reality as you like but please consider all the consequences. And I insist that by questioning reality you are questioning scientific approach. So if you accept scientific approach then you implicitly accept reality as well.

But QM and other modern science force us to question "reality" closely.

No. Look, there are two options we can consider:
1) there is more elegant model but it is self contradictory when we get down to it (we assume reality at the start but the model at some point contradicts this assumption).
2) there is less elegant model but it is consistent (we assume reality at the start and we do not run in any contradiction with reality).

You are making implicit assumption that more elegant model must be more valid than less elegant model. But this assumption is wrong as far as scientific approach is considered. Say if more elegant model is falsified by experiment but less elegant model is confirmed by experiment you would not suggest that there is something wrong with scientific method, right?

Objective reality is constructed from sense inputs, consensus opinion, and a few more ingredients, notably memory. Every step of the construction makes hidden assumptions. Some of them, it turns out, aren't entirely right.

Do you suggest that some necessary assumption is provably false? I think you somewhere have taken too strong assumption where we actually need weaker assumption.

 

[secur]

... by questioning reality you are questioning scientific approach. So if you accept scientific approach then you implicitly accept reality as well.

Well, yes and no. Scientific approach, or method, makes most sense assuming a "real" objective reality. Also, it's part of reality, itself. So yes, the two go together. OTOH one can question reality, at the fundamental level, and still engage in normal activities like going to the store for milk, posting on PF, and following the scientific method.

You are making implicit assumption that more elegant model must be more valid than less elegant model. But this assumption is wrong as far as scientific approach is considered. Say if more elegant model is falsified by experiment but less elegant model is confirmed by experiment you would not suggest that there is something wrong with scientific method, right?

I guess the "more elegant" model would be SMOR - is that right? Regardless, I'm definitely not suggesting there's anything wrong with scientific method.

Let's back up a bit. There are two models involved here: SMOR, and the scientific or physical model. The latter is the one you consider "really real", and, if I understand correctly, less elegant. Physical model is right according to scientific approach, of course; that's how it's generated. SMOR is right according to a priori categorical intuition, as defined by Kant.

BTW I'm not citing Kant as an "authority" - not asking you to believe it just because Kant said it. I mention him because, for those familiar with him, he presents this "a priori" concept pretty well, saving me the trouble of explaining it. If anyone is not familiar with him I'll be happy to explain the concept, just ask.

Both models are valid, or "right", in their own sphere. From one point of view, one's better; from another pov, the other is. By comparing them, finding their differences, we obtain insights that are key to understanding modern physics. But we can't learn any new physics! It's "philosophy of physics", or metaphysics, ontology. It make sense of, and sorts out, QM interpretations. It also points the way to potential fruitful future investigation in physics.

Do you suggest that some necessary assumption is provably false? I think you somewhere have taken too strong assumption where we actually need weaker assumption.

Give me an example of a necessary assumption. Or, an unnecessary one. I don't see the distinction. They're all necessary, since some sort of assumption is needed to cover that area. They're all unnecessary, since there are always possible alternatives. An example would clarify the distinction you're making.

Anyway, any "hidden" assumption of SMOR might be provably false - meaning, it might disagree with the scientific physical model. Some have already been proven false, in that sense. Others may never be proven false. But there will never be a time when we can know for sure that a given assumption is definitely compatible with the physics of the "real" world. It must always remain open to contradiction by future experiments. That's the nature of assumptions.

I hope that addresses your questions but am aware it might not, please let me know.

 

[ddd123]

Give me an example of a necessary assumption. Or, an unnecessary one. I don't see the distinction.

This reminds me of Peres' book again, at page 168:

Salviati. [...] The crucial point in Bell’s argument is that although the individual results are unpredictable, their correlations, which are average values, can be computed by quantum theory, or can simply be measured experimentally, irrespective of any theory. The amazing fact is that it is possible to prepare physical systems in such a way that the inequality (6.30) is violated, and therefore the identity (6.29) cannot be valid.

Simplicio. An identity which is not valid?

Salviati. This is of course impossible, therefore there must be a flaw in this argument. Either it is wrong that the observers have a free choice among the alternative experiments (namely, for each pair of particles, only one of the four experimental setups is compatible with the laws of physics—the others are not, for reasons unknown to us), or it is wrong that each photon can be observed without disturbing the other photon. Take your choice.

Simplicio. Both alternatives are distasteful. I prefer classical physics.

 

[zonde]

I guess the "more elegant" model would be SMOR - is that right? Regardless, I'm definitely not suggesting there's anything wrong with scientific method.

Let's back up a bit. There are two models involved here: SMOR, and the scientific or physical model. The latter is the one you consider "really real", and, if I understand correctly, less elegant. Physical model is right according to scientific approach, of course; that's how it's generated. SMOR is right according to a priori categorical intuition, as defined by Kant.

Ok, my guess was wrong and so you misunderstood my replay too. Here is what you said and let me give you different (and hopefully less cryptic) answer:

But QM and other modern science force us to question "reality" closely.

This is not so and here is why. There is pilot wave theory that gives all predictions of standard QM and is consistent with "reality" and "other modern science" (meaning relativity).
So as long as there is at least one valid option that allows us to keep "reality" we can't say that we are forced to question "reality".

Considering your point 1. in list of failures of SMOR it seems you have impression that special relativity rules out preferred reference frame. This is not so. SR is fine even if we declare some inertial reference frame "preferred". And the same applies to FTL of some hypothetical physical phenomena that is currently out of our experimental reach. Such hypothetical FTL phenomena could simply establish domain of applicability for relativity while still be consistent with relativity in the domain where it is tested and verified.

Give me an example of a necessary assumption. Or, an unnecessary one. I don't see the distinction.

Necessary assumption:
- there is universal time;
- past is fixed and unchangeable.
Unnecessary assumptions:
- object has a property even before we observe it;
- future is uncertain and doesn't yet exist.

 

[secur]

There is pilot wave theory that gives all predictions of standard QM and is consistent with "reality" and "other modern science" (meaning relativity). So as long as there is at least one valid option that allows us to keep "reality" we can't say that we are forced to question "reality".

Sorry, that's wrong. Pilot wave requires FTL influence or communication. This is really important so if you don't believe me ask Demystifier :-) The pilot wave must influence the particle's position (beable) faster than light.

Bohmian mechanics is not quite entirely accepted by mainstream, there are unresolved issues, IMHO. However it's probably viable - i.e., can reproduce all QM experiments - so let's assume so.

In that case, if it didn't require FTL influence, it would have to be considered "right" since all other interpretations violate some normal principle of physics. And QM mystery would be solved.

De-Broglie Bohm theory proves that the statement "there is no nonlocal influence" is unjustified. Since this viable interpretation does depend on that, it's impossible to rule it out. And if anyone rejects Bohmian, collapse interpretation also shows this. Gell-Mann claims there's actually something wrong with both those interpretations. I disagree, collapse at least is sound. But he's got the right idea: to justify his statement against nonlocal influence, he must attack these. As long as any viable interpretation exists with nonlocal influence, it can't be ruled out.

The same BTW is true of other things such as Multiple Worlds. If we suppose MWI is truly a valid interpretation (which isn't clear) it proves that multiple worlds might exist. One value of any interpretation lies in this ability to demonstrate the possibility of whatever ingredients go into that interpretation. Thus, we can't say "there is no collapse" until proving the collapse interpretation isn't viable. Which means, can't match some experimental result.

Considering your point 1. in list of failures of SMOR it seems you have impression that special relativity rules out preferred reference frame.

No, not at all. I'm referring to Minkowski space vs. Euclidean. The most important aspect is dependence of time on speed. That's not in SMOR but is in physical world as revealed by modern science.

This is not so. SR is fine even if we declare some inertial reference frame "preferred". And the same applies to FTL of some hypothetical physical phenomena that is currently out of our experimental reach. Such hypothetical FTL phenomena could simply establish domain of applicability for relativity while still be consistent with relativity in the domain where it is tested and verified.

I agree. My previous posts never contradict these facts - on purpose, at least.

Necessary assumption:
- there is universal time;
- past is fixed and unchangeable.
Unnecessary assumptions:
- object has a property even before we observe it;
- future is uncertain and doesn't yet exist.

Thanks, you've given the requested examples and I appreciate that. Unfortunately now you'll have to explain them! I see no fundamental differences here. Any might be right or wrong: viable alternatives exist. In that sense, none are necessary. And each fills in an obvious "hole" in SMOR. That is to say, answers a question which is bound to arise. In that sense, we could say they're all necessary.

Sorry to be obtuse, but why is any of these more, or less, necessary than another?

 

[zonde]

There is pilot wave theory that gives all predictions of standard QM and is consistent with "reality" and "other modern science" (meaning relativity). So as long as there is at least one valid option that allows us to keep "reality" we can't say that we are forced to question "reality".

Sorry, that's wrong. Pilot wave requires FTL influence or communication.

Yes, of course Pilot wave requires FTL! Any realistic model of entanglement requires FTL influence or communication! That's the point. There is no way around Bell inequality violations that is local and consistent with SMOR.

Pilot wave theory is the less elegant model that I meant in my earlier post. More elegant models are some of those that try to keep locality at the cost of being inconsistent with SMOR.

Thanks, you've given the requested examples and I appreciate that. Unfortunately now you'll have to explain them! I see no fundamental differences here. Any might be right or wrong: viable alternatives exist. In that sense, none are necessary. And each fills in an obvious "hole" in SMOR. That is to say, answers a question which is bound to arise. In that sense, we could say they're all necessary.

Sorry to be obtuse, but why is any of these more, or less, necessary than another?

Yes, any assumption can be right or wrong. That's what is meant by "assumption".
But you say that some assumptions have alternatives, right? Are these alternatives consistent with SMOR?
And do you mean that assumptions that I listed as unnecessary are actually required for SMOR? Can you provide your arguments? I will try to provide arguments why they are not necessary.
As I see assumption "object has a property even before we observe it" can be relaxed to assumption "object has a position even before we observe it". This will give us enough certainty to speak meaningfully about SMOR.
Assumption "future is uncertain and doesn't yet exist" is related to assumption that we have "free will". But this requires some philosophical arguments why "free will" is such a game changer as far as SMOR is concerned. As I see it's more important to have some sort of independence of our consciousness and I think this can be achieved even without "free will".

 

[secur]

@zonde, I understand your view better now. We have somewhat different ideas of SMOR. That name "Shared Model of Objective Reality" is perhaps misleading. To you it's sort of "common sense physics". You want to fine-tune it, as little as possible, to match experiments. Whereas I mean a very basic model which reflects the way our brains are wired. It can't be changed (without major brain surgery). It determines and defines how we perceive reality: space, time, objects, motion. It must be dealt with as is. The problem, it doesn't match advanced experiments, but physics must. Physics doesn't have to be common-sensical, elegant, Occam's razor compliant, etc: those are luxuries, extras.

So there's a fundamental tension between SMOR and physics. I'm proposing the way to deal with that situation. We can't change, or ignore, either SMOR or physics (i.e., experimental results). Their inter-relation must be exposed and regularized. I'll try to show what I mean in another post, and think I'll change that misleading name.

Any realistic model of entanglement requires FTL influence or communication! ... There is no way around Bell inequality violations that is local and consistent with SMOR.

True.

Pilot wave theory is the less elegant model that I meant in my earlier post. More elegant models are some of those that try to keep locality at the cost of being inconsistent with SMOR.

Ok - now we're on the same page. You're even a step ahead, because I haven't yet explained why FTL is consistent with SMOR.

But you say that some assumptions have alternatives, right? Are these alternatives consistent with SMOR?

No, they're not, from my point of view. Note, I have firm ideas about SMOR but haven't yet defined it much. I've just mentioned Kant and left it at that. No doubt my view is close to yours and other's, but there's bound to be disagreement. However ignoring that, when we change one of its assumptions to match advanced physics, it's no longer SMOR.

As I see assumption "object has a property even before we observe it" can be relaxed to assumption "object has a position even before we observe it". This will give us enough certainty to speak meaningfully about SMOR.

Yes, that's the Pilot Wave alternative. Or we can suppose object has no properties at all before observation. I would say any relaxation of the original assumption violates SMOR.

Assumption "future is uncertain and doesn't yet exist" is related to assumption that we have "free will".

SMOR does, indeed, assume free will. But physics works without it; we can use Block Universe and still get the right answers.

I'm beginning to see where our confusion lies. I could say this: All the assumptions are necessary for SMOR; none are necessary for physics.

BTW this illustrates an important general lesson. These concepts are original philosophy (not science) so we don't yet have a common language for them. Therefore statements will inevitably be misunderstood. It's necessary to keep talking, even when - especially when - the other guy seems to make no sense. After a while the terminological mismatch is bound to come out. But if you just give up, (as so many do), it never will.

We need a firmer foundation. I should define SMOR more. But my main point is more general: how to analyze physics considering SMOR. I'll try to give a clearer, high-level, statement of my overall approach. If you lose interest, and no one else cares either, that's alright: you win a few, you lose a few.

 

[secur]

A Priori Reality (APR)

Shared Model of Objective Reality (SMOR) apparently doesn't convey the right meaning. Let's call it A Priori Reality (APR) instead. It's the intuitive fundamental version of reality people all share, unless they're damaged. Cats and monkeys also have it, for the most part. All animal brains (more or less) work that way. Any information we sense, know, or remember, has to exist within the APR framework or model. See Kant's categories, etc, for more on the subject. Someday I need to specify it more exactly if anyone's interested. APR governs all human experience.

Physical Reality (PR)

PR is the model of reality derived from science, physics. It's always being updated as new physics happens. It's much more detailed than APR. It matches all experimental results.

APR doesn't match PR (the so-called "real world") in a number of ways, like relativity. Conversely, the physicist's model fails to match some human experience. Both models are valid in their own domain.

Relation between APR and PR

At first glance you may think APR has nothing to do with science. We can simply experiment and build our PR model. Who cares about APR which, after all, is something even cats understand. But science definitely can't ignore APR, nor does it. APR dominates (one way or another) half of science.

Most advances in physics involve rejecting one or another assumption hidden in APR. Simple physics accords with a priori intuition but before too long, with more refined experimental technique, it doesn't. It's very useful to categorize theories according to which APR assumption they invalidate, and what they replace it with. One example, from dozens: relativity proves APR's Euclidean space is wrong; actually it's Minkowskian or Riemannian. "Wrong" in this context means: doesn't match the "real" universe we live in.

But APR plays another, even more important, role in science and physics. The key fact you may not realize is: we can think only in terms of APR. Nothing else can or ever will make sense to us - until the brain itself gets modified. That means all physics must ultimately be expressed in terms of APR, else it's gibberish.

Of course good physics already follows this rule. One example, GR violates APR in a few ways, like curved space. But GR calculations are more-or-less always done on a tangent plane, with a time axis - which does match APR. We tame the curved Riemannian space, which can't make sense to us, by envisioning an APR-compliant space tangent at each space and time point. Etc. A lot of physics reasoning, and intuition, involves expressing a non-APR aspect of the PR model in terms of APR.

Experiments are the best example. All facts of physics no matter how non-intuitive must ultimately be demonstrated and proved by experiments (or, observations). As we all know, we understand experiments from the point of view of our macro, classical world. We read a pointer on a dial, or something equivalent. Thus all facts are grounded in the APR. APR doesn't know about relativity, atoms, EM fields, wavefunctions, or even a rotating Earth orbiting the sun. APR thinks the Earth is flat, and holds still. The proof of such effects must be, and is, understood in simple APR terms that even a monkey can appreciate. The dial points here, not there. That's the only way we know about quantum spin. The Foucault pendulum plane is seen to rotate during 24 hours. That's the closest we get to sensing Earth's spin.

Theoretical APR Analysis

Here's how to analyze a physics theory to ensure it's "APR-legal". First find where it violates APR. (Note, even in advanced topics a lot of it is intuitive as it stands.) Wherever it's non-APR, you have to translate the non-APR concept to APR. For example, for GR, Einstein (actually, Riemann, or originally Gauss) translated the non-intuitive concept of curved space into APR via tangent planes linked via Christoffel connections.

All classical well-accepted theories of physics already are APR-legal but it's very instructive to analyze them and see that's true. Quantum Theory, however, isn't, at the moment, leading to an interpretation quagmire. Then there are radical theories that aren't APR-legal at all. They really need work.