↪boundless Linde says it changes our perception of what appears to be ‘the past’. — Wayfarer

I think that it is a good way to put it :up: That's would be an 'epistemic' claim, consistent with epistemic interpretations. So, I don't think that is a controversial statement.

What's odd is that this is a thread about justice and fairness, yet it contains page after page of speculative quantum physics. — Banno

Lol, yeah, I see your point. Is it possible to 'split' the thread?

Indeed, physics has its merits. I don't think anybody denies that. I was just pointing out what some of its problems are, and how these problems relate to mathematical logic. — Tarskian

Ok, I see. Thanks for the clarification.

↪Wayfarer Again, there is the bit where you give and the bit where you take back. You want consciousness to be the special thing that collapse wave functions, but you don't want it to be different to the other stuff of the universe. — Banno

Actually QBism is a form of 'consciousness causes collapse' interpretation as I understand it. But given that in such an interpretation the wave-function is merely a tool that allows one to calculate probabilities to have some kind of experiences, I do not see how it can be problematic or how can it give a 'special' role to consciousness. It certainly does not mean that when a conscious agent makes an observation something in the 'fabric of the universe', so to speak, changes. The only thing that happens is an update of knowledge (in QBism this is expressed as an upddate of a probability, which in turn is interpreted as a quantification of a degree of belief).

I do not think that QBism or similar views are in contrast with your own view.

On the other hand, Rovelli's RQM holds that all physical objects can be considered as 'observers'. To be fair, I prefer QBism or similar views because, after all, what a 'physical object is' is extremely vague. Is a table a physical object? Are its legs physical objects? Does each atom of a table have its 'perspective'? Do all its atoms and the whole table have their own perspective? How they are related?
My problem with such a view is its vagueness.

↪boundless That’s pretty right - I do hold to a form of epistemic idealism. But I also claim that what we can claim is real is inextricably connected to what we can know, which I think is a consequence of my training in Buddhist philosophy. — Wayfarer

Ok :up: Note, however, that if one thinks that the delayed choice experiment shows that a measurement can change the past (rather than our knowledge of it), then one enters in some serious difficulties. That's why I don't like the expression 'partecipatory realism' even if respected physicists use it.

I do not doubt their credentials. I am merely saying that an ontological reading of certain ideas can be misleading. Of course it also depends on how 'ontology' is defined.

Yes, for instance QBism does require that 'observers' can 'experience', i.e. are conscious. But this does not imply that consciousness 'affects' the 'external reality' when a measurement, observation or anything like that is made.

If 'consciousness causes collapse' is taken as meaning that 'an experiment updates the knowledge/degree of beliefs' of a conscious agent, I do not see anything 'weird'. On the other hand, if it is taken as implying that an observation of a conscious agent changes the 'fabric of reality' or that 'consciousness causes a drastic change in external reality' and so on, then yes it becomes a problem IMO.

Thanks for the link, btw, I'll read it. Anyway, I do think that your own idealism is actually a form of epistemic idealism, rather than ontological (based on what I read in the past). Ontological idealism treats consciousness as the sole substance.

So like gauge invariance vs Poincaré invariance? Constrain spacetime to a manifold of points and it still has degrees of freedom in that the points may spin rather than sit still. They may be vector and chiral rather than scalar. Quantum spin arises as an intrinsic property and the rest of particle physics follows. — apokrisis

Well, maybe 'symmetries' are the only intrinsic properties that can be discovered by physics.

But even quantum spin after all is a quantity that describes how an object responds to an interaction. So I am not sure that it can be said that it is an intrinsic property of a particle (I admit that this does not 'conclusively' shows that it isn't lol...).

(IIRC there was a paper by philosopher Michel Bitbol that discussed this kind of things. If I find it I'll link or quote the relevant parts)

Yeah, that bit. The principles of physics are to be formulated so that the frame of reference being used does not change those principles. Any frame will do. This was intended to head off the common notion that science seeks a "view from nowhere" - perhaps the view you described and disagreed with as "independent from any reference frame". Rather, science seeks a view from anywhere. A pont worth making in a philosophy forum. — Banno

Completely in agreement with this :100:

That seems to be making 'an ontological claim'. Or wait - is it an 'epistemological claim?' — Wayfarer

It depends on the interpretation of this type of things. In epistemic interpretations like, say, QBism or some form of Copenaghen-ish interpretations, such an experiment results in an update of the knowledge of the agent/observer, not in the 'coming to be' of some kind of reality.
At best, it is an 'ontological claim' in the sense that 'the world' that the agent 'observed' is not 'the world in itself', but the 'world observed'.

These kind of theories do not make ontological commitments, i.e. do not attempt in describing 'how the world is in itself' (from a 'view form nowhere', if you like).

Unfortunately, concepts like 'participatory realism' IMO kind of muddy the waters. They seem to imply that 'the observer changes reality'. This in QBism, as I understand it, actually means 'the observer gets a new experience'. Other similar interpretations say that 'the observer knowledge its updated' and so on.

I guess that 'participatory realism' was developed in other to defend QBism and similar views from the charge of solipsism. But I do not really see the point of that. It only complicates the issue.

So, physicists want a "grand unified" pattern instead. Physicists seem to view this effort as essential. — Tarskian

My point was simply that I think 'Physics' as a discipline has still a 'raison d'etre' if 'a theory of everything' is impossible to achieve. I myself worked briefly in condensed matter physics, a branch of physics that is quite independent to the search of a 'theory of everything'. To some physicists might be essential, maybe even famous ones, but this does not mean that physics becomes vain if a 'theory of everything' is impossible. There still much to be discovered about physical phenomena that can be 'modeled' with current theories.

Hi Banno, apokrisis,

The Principle of Relativity asks us to set out the laws of physics in such a way that they apply to all frames of reference. — Banno

Correct. But this only says that physical laws are the same in all reference frames. Invariant properties in all reference frames are not necessarily property of a 'physical world independent of all reference frames'

Hence this suggests to me that any true description of the physical world can be made from any perspective/frame of reference. — Banno

I disagree, if the 'phyisical world' here is meant the 'physical world as it is independent from any reference frame'. On the other hand, yes, I agree if this is taken to mean that any reference frame can be used to discover/find some truth that is valid for all other reference frames.

But we can have a theory of reference frames can’t we? We continue on as we see with holography, de Sitter metrics, or twistor space. We can have general arguments that pick out 3-space as special as the only dimensionality that has the same number of rotational degrees of freedom as translational ones. — apokrisis

Yeah, I guess that we can but I am not sure how this is an objection to what I said earlier, if you meant that way. I am perfectly fine, for instance, with what you say about 3-space. But this can be IMO understood as a pointer to a property in common to all reference frames. Regardin holography, de Sitter ant twistors, well as I said, they all seem all quite far as the 'physical world as it appears to us', so to speak. As I said:

'How the world is' independently from any perspective seems to get weirder and weirder as we get to more 'advanced' theories. — boundless

I see it as a sort of 'evidence' for this kind of tendency.

There may always be questions but they also can be new ones. — apokrisis

Agreed.

Anyway, to put in another way what I am saying, I think that a distinction can be made between 'intrinsic' and 'relational' properties. Relational properties are not properties of 'a thing in itself' so to speak, but is a property that arises in a relation. As an example, the apparent height of a tower seen from a distance is a relational property, not an intrinsic one (it cannot be assigned to the tower, without taking into consideration something else). If we do not know intrinsic properties of objects we do not know them 'in themselves', but only in relation.

According to, say, Galileo all 'primary' qualities were intrinsic to physical objects. But his own 'principle of relativity' actually showed that velocity is a relational property. In special relativity, for instance, even distances and temporal durations are not intrinsic properties*. In QM this 'relationality' is even more explicit.

*Yeah, the speed of light is the same in all inertial frames in special relativity. But it does not follow that it is an intrinsic property: after all, velocity is a concept that makes sense only when a particular reference frame is considered. I think that the same can be said even for rest mass. Rest mass is a quantity that tells us how an object 'responds' to some interactions. BTW, after all the operational definition of all physical quantities is in fact relational. So, maybe it might not be surprising if it turns out that they are not intrinsic.

It is actually the ultimate goal of science: — Tarskian

I disagree. Science can exist even if such a theory is impossible. It isn't essential to science IMO, so it cannot be its 'ultimate' goal.

One of those possibly pseudo-questions which may be sophistry; but, in your opinion do you think physics describes logic? — Shawn

No. In fact, one might say that the opposite is true. As Hume said, there is no 'proof' of, say, physical causality, we cannot be certain of it. On the other hand, it seems that any physical theory must be logically consistent. Logic is, I think, transcendental (i.e. a necessary precondition to any explanation) as the early Wittgenstein said.

Also, there is no conclusive evidence that physical laws are not contingent.

Doesn’t the same problem crop up in a relativistic context such as the simultaneity issue? No absolute reference frame and yet that can still be approached in the limit. — apokrisis

I agree with that. In fact, I believe that relativity has similar interpretative difficulties. On one hand, a 'literal' interpretation of relativity leads one to an 'eternalist block view', i.e. change and the 'now' are merely illusions. On the other hand, as Einstein himself said the 'now' is a great problem for relativity. After all, the experience of the 'now' is undeniable and so 'immediate' experience seems to contradict a literal interpretation of relativity. So what?

In relativity the 'branching' of space-time into space and time is associated to the choice of a particular reference frame, i.e. perspective. This is similar to what happens in QM with 'observers' (whatever one takes them to be): only at measurement/observation/interaction etc physical quantities assume a definite value. But hey, even in classical mechanics velocities etc have undefined value until a reference frame is chosen.

Like QM and like what I said about newtonian mechanics, this suggests to me that any description of the world must be made from a particular perspective/frame of reference. 'How the world is' independently from any perspective seems to get weirder and weirder as we get to more 'advanced' theories.

In a sense, I get that it can be seen as a disappointing view but, on the other hand, it is at the same time in a sense 'liberating' and 'awe insipring' (reality seems much more mysterious than it appears to be...).

I think that my view is close to the view presented by Bernard d'Espagnat in his 'On Physics and Philosophy' or of the late David Bohm (who wasn't an 'instrumentalist'). For the latter see e.g. this interview*: https://www.youtube.com/watch?v=Mst3fOl5vH0 (yeah, I know that the late Bohm is controversial but still...also his concept of 'active information' might be congenial to biophysics)

Yes. It is perfectly acceptable to me to go full Copenhagen and say all we can know is the numbers we read off dials. If a proper ontic interpretation isn’t available, quantum physics still works as instrumentalism. Copenhagen remains the sensible backstop epistemic position. — apokrisis

Same as all theories. It does not matter that the literal interpretation of newtonian physics is right for its applications.

IMO Physics seems to assume that there are regularities in phenomena (but any ontological commitment of the theory is unneeded)

Yeah. Heard quite a bit from him on Physics Forum some years back. But I can’t remember whether I was agreeing or disagreeing with him at the time. I will have to check that reference. :up: — apokrisis

Lol, I 'knew' him there too years back. Seems to be a very thoughtful researcher. Unfortunately I found many of his works very complicated and well above my level. I do not know if his interpretation is actually accepted as a 'canonical' one as say de Broglie-Bohm, MWI, Copenaghen etc

*I do not think that he is completely correct in that interview, but I agree with his main message.

Classicality comes to be in the limit. So reality never arrives at that ideal conception we have of it, but through decoherence, it approaches a classical state for all practical purposes. We can apply that brand of physics and logic to it. — apokrisis

Ok, I see. But IMO, while decoherence - for all practical purposes - explains why we see definite outcomes in experiments, it doesn't explain why experiments have a unique outcome. In other words, as far as I can tell, this is why decoherence is not taken as a complete solution of the measurement paradox. Decoherence gives the definiteness of the observed outcome but is not enough to explain the uniqueness of the outcome.

But the cat is a hot body in a warm place. It went into the box decohered and not coherent. It wasn't converted to a Bose condensate. It remained always in a "thermalised to classicality state". — apokrisis

OK, you are right.

MWI is the kind of nonsense to be avoided. Spontaneous collapse fails if you demand that reality actually be classical rather than just decohered towards its concrete limit. Zeilinger's information principle captures some aspects nicely. — apokrisis

Ok, sorry for the misunderstanding then. I fully agree with you about MWI. I referred to spontaneuous collapse theories because they provide a consistent 'ontologically interpretable' (or 'realist') explanation for the uniqueness of the outcomes.

But, anyway, I concede that, in a sense, one can say that decoherence is 'enough': after all, it is enough to explain, for all practical purposes, why we get definite outcomes. As far as observations, applications, and practical concerns are concerned, yeah, you don't need other assumptions. Still, from a theoretical point of view, I think that uniqueness of the outcomes is a crucial assumption in physics and I do not see how decoherence can explain that.

To be honest, I set the interpretation aside these last few years to let the dust settle. Youngsters like Emily Adlam are coming along and making more sense. — apokrisis

Ok, I see! I don't know Adlam, thanks for the reference.

In a sense I think I agree with you. IMO, I see QM as a practical recipe, useful for predictions and applications. I favor epistemic interpretations like QBism. I think that it is impossible to make a literal interpretation of the 'orthodox' quantum formalism that makes 'fully' sense, so to speak.

But as I say, biophysics puts it all in a new light. Something has been missing. It seems obvious to me that this is it. — apokrisis

Ok, I see, thank you.

Personally, I think that QM strongly suggest that we cannot describe physical reality as it is independently from a particular perspective and that physical theories are, in general, useful tools and 'fictions' (but as I said in my previous post, I think that this insight is actually present even in newtonian physics, albeit less explicitly). But this certainly does not mean that, in the future, it won't be replaced by something better.

P.S. I have a hunch that you might find interesting the Thermal interpretation by Arnold Neumaier. It is an 'ontologically interpretable' interpretation which apparently solves the measurement problem and other issues of other 'realist' interpretations (btw, it is also holistic: it sees entangled particles as a single extended objects, not reducible to its parts that have nonlocal properties...so some 'quantum weirdness' remain). Unfortunately, his papers are too difficult for me

Regarding the 'parallel' idealism-realism debate, I think that physics - and science in general - is, well, silent on that issue as well.

For instance, the concept of 'reference frames' is central to the theory. In fact, many physical quantities have undefined values before a reference frame is specified, i.e. this means that they are not intrinsic properties of physical objects but only relational ones. Reference frames are associated to physical objects but are actually abstract concepts of the theory - they more or less correspond to 'perspectives' where a given object is at rest. And all measurements happen in a given frame of references.


Anyway, even in classical, newtonian, mechanics the 'physical world as it is, independent from any perspective' is quite different from 'the observed reality', which is always associated to a given frame of reference.

Related to the above, I would ask: what constitues a 'perspective'/'reference frames' in physics?
And how is the physical world independent of all perspectives? Can we describe it?

Nope. Enzymes are large mechanical structures. Decohered and classical for all intents and purposes. But they can dip their toe into the quantum realm, exploiting tunneling to jump chemical thresholds. — apokrisis

Hi, I read your linked post and I enjoyed it. But still I don't understand how 'classicality' 'comes to be' in your view.

Let's consider a less crude version of the Schrodinger's cat experiment, where the cat is either awake or asleep. Decoherence IMO can only remove interference, not superposition, hence the cat is still, if we take the quantum formalism literally, awake and asleep at the same time. However, decoherence gives the appearance of classicality because it says that it is observed in a definite state.

Let's say that I observe the 'asleep' cat. In your view, what happened to the 'awake' cat?

In MWI, the superposition is always preserverd and each definite state is 'actualized' in a branch of the wavefunction (but observers have access only to one). The 'branching' can be seen as due to decoherence.
In epistemic Copenaghen-ish views, QM is not to be interpreted literally and the collapse is just an update of knowledge (these views do not make any ontological commitments to what 'happens' before the observation: the state of the 'unobserved' cat is beyond the range of descriptions).
In (non-local) hidden variables interpretations, the cat is always in an unique state and it is determined by these hidden variables
In spontanueos collapse theories (these are actually not interpretations as they make different predictions from QM), at a certain scale wave-functions collapse and give an unique definite outcome
In RQM, anything can be an observer and any interaction is a measurement. But, again, like in epistemic views the 'unobserved' cat is in an indefinite state.
And so on.

Spontaneous collapse theories - (Edit: or maybe some version of MWI) - IMO seem to me the most compatible to your views.

I make this same point all the time. :up: — apokrisis

:up: Glad to hear that! I think that, what is common to all physical theories is that they are predictive tools with an extraordinary range of practical applications. Physics per se does not give us any ontology IMO. I also think that many debates about topics like free will, reductionism* etc are due to an unneeded ontological interpretation of physical theories. In fact, I think that the most consistent form of 'realism' with all physical theories is actually what Bernard d'Espagnat called 'open realism' in his "On Physics and Philosophy" (p.28, emphasis in the original):

Realism (Open). This is - indeed quite 'open' - view that there is something the existence of which does not hinge on thought
Is this 'something' the set of all objects, of all the atoms, of events, God, the Platonic Ideas, still something else? Open realism is mute on this.
...
It just says 'something,' in the widest possible sense of the world.

Of course this does not mean that is the only viable form of realism.

*I do believe, in fact, that even newtonian mechanics is not really 'reductionistic'. Conservation laws seems to me properties of a whole isolated system, not reducible to the properties of its parts. Of course, if one does not make any ontological commitments, newtonian mechanics is neither reductionistic nor holistic.

I posted this general story here - https://thephilosophyforum.com/discussion/comment/679203 — apokrisis

Thanks for the link. I'll ask more questions after reading it :)

What you often hear from idealists (Kastrup and Hoffman are good examples) is that materialism and a physical world is debunked and quantum physics tells us reality comes into being by the act of observation. Therefore idealism is a more reasonable and parsimonious explanation for our experience. I've often thought that the arguments in favour for idealism are actually more arguments against old school materialism than any great championing of an 'it's all consciousness' style metaphysics. — Tom Storm

Note, however that Hoffman does not really say that. He says, more or less that QM suggests that 'physical reality as it appears to us' comes into being by the act of observation. And he says that this is consistent with his 'interface theory of perception', i.e. the view that perception gives us a simplified interface of the 'external reality' which is useful for our survival.

Unfortunately, this is then mixed up with his 'conscious realism', i.e. his Berkeleyan-like view that conscious agents are the only reality. To his credit, he never say that the 'interface theory of perception'necessarily implies 'conscious realism'.

Note that his 'interface theory of perception' is a kind of epistemological idealism (which is actually compatible with the existence of a mind-independent reality). His 'conscious realism' is of course a form of ontological idealism. Unfortunately, sometimes his two theses are mixed up.

Hi apokrisis, all,

I have outlined many times how biosemiosis now adds the epistemic cut to the business of quantum interpretation. As a mechanism, a modelling relation, even our enzymes and respiratory chains are actually doing that - preparing states of coherence with the intention of collapsing them and so ratcheting the entropy flows of a Cosmos guided by the telos of the newly NASA-rediscovered concept of dissipative structure. — apokrisis

I think that the problem of this view is that it does not explain how those 'complex' objects/processes like 'enzymes' or 'respiratory chains' arose in the first place. Was everything in superposition back then? What happens when these kind of object 'cause' a collapse?

I think that all positions of what consitutes an 'observer' are susceptible to this kind of objection and it is a serious objection if QM is interpreted ontologically. I think that it is best to interpret it epistemically: wave-function, Born Rule etc are all useful concepts, computing techniques etc that enable us to make correct predictions. In this view the 'collapse' is just an update of knowledge and QM is silent about what 'happened' before the appearances of 'observers' (and it is also silent on what constitutes an 'observer'). QM is IMO best seen as a theory that does not make any ontological commitments: it does not give a 'picture' of 'how reality is'* and it does not say to us what an 'observer' is but is still very useful to make predictions, applications and so on. (My favorite interpretation is QBism, althought I think that, unfortunately sometimes its proponents seem to present it as an ontological interpretation of QM, perhaps unwittingly... especially when they talk about 'participatory realism')

As an aside, I do not think that 'decoherence' alone is enough to solve the measurement problem. It does explain (at least 'for all practical purposes') an appearance of a classical world, but it does not IMO explain why we observe a single outcome in quantum experiments (MWI supporters like decoherence because it explains why 'branches' of the wave-function separate. But MWI makes the additional assumption that all outcomes are actually observed even if they are inaccessible to us)

*I actually think that we tend to do that also in classical physics. For instance, an ontological interpretation of the concept of 'force' in newtonian mechanics is clearly inappropriate but it is an useful conceptual fiction that helps us to make predictions, build things etc (like, say, the concept of 'sunrise' or 'sunset'). The same IMO holds for QM.

https://erlichdallc.wordpress.com/2019/04/11/my-current-understanding-in-a-nutshell-of-everetts-relative-state-formulation-of-qm/ — i aM

Nice!

Isn't a core idea of SR "relativity of simultaneity", i.e. simultaneity of events is entirely dependent on the reference frame of the observer? — i aM

I believe that it depends on how you define SR. In the usual definition, both preferred foliation of spacetime and retro-causality are incompatible with SR.

Yet, I think that they are both compatible with Lorentz symmetry.

By my understanding, in PWT the pilot wave controls the velocity of the particle. And that velocity depends not only on the position of the particle, but also the positions of all the particles it is entangled with; and that information is all available instantaneously to the pilot wave which controls the velocity of the particle. I don't see how that can be reconciled with SR. — i aM

Well, there have been attempts to reconcile SR and PWT. To my knowledge, they involve the use of preferred foliations of space-time or retro-causality. Of course, this is different from saying that PWT can be reconciled with SR in its standard formulation, but the point is that apparently it can be reconciled with Lorentz symmetry.

I was however referring to this way of reconciling SR and PWT via this defense of presentism proposed by NoAxioms:

I have done an advocatus diaboli thread defending the compatibility of relativity and presentism, so I maintain that they're not incompatible. SR says that the preferred frame cannot be determined given the special case after which it is named. But inability to detect such a frame does not mean that there isn't a special one. Presentism doesn't even require it to be a inertial frame, and no presentist that knows their physics seems to assert that it corresponds to such a frame. The foliation is always bent, which has the interesting paradoxical implication that no two stationary observers are simultaneous in each other's inertial frames. I find that hilarious, but not paradoxical. — noAxioms

Yes. With locality, which is essentially saying no FTL. — noAxioms

Ok! Fine, then we agree :smile:

Pilot wave is a form of Bohmian mechanics: Pro counterfactual definiteness (objective state) and denial of locality. So I wonder how they interpret spooky action at a distance using pilot waves. I don't know the official line on that. They certainly cannot reproduce spooky action using a classic pilot wave setup like they use for double slit. — noAxioms

Considering that the 'wave-function' is a 3N-dimensional (N being the number of particles) object and the theory predicts that the 'influence' is instantaneous, I do not think that there can be an explanatory dynamical model like - say - the one that uses mediators.

(As an aside, I prefer using the term 'de Broglie-Bohm theory' or 'pilot-wave theory' for various reasons. Among these, 'Bohmian mechanics' is actually the preferred term of a specific sub-school of the dBB-supporters and also downplays the role of de Broglie. It gained success and maybe it is the most used term in literature... :smile: YMMV )

I simply meant that without the selection postulate, it seems that RQM implies the splitting.

Anyway, I agree with you. RQM seems simply silent on this point. — boundless

Maybe it's embarrassed. :yikes: — Wayfarer

Well, possibly! :razz: — boundless

This is why I resist describing RQM under presentist terms. If time is external to the structure that is the universe, then such selection is an objective act relative to this realm under which time exists, and it isn't really RQM anymore if such an objective action takes place.

With time being part of the structure, no event/state (something to which a relation can be made) 'flows' to a different event, necessitating such a selection. Thus there is no selection postulate.
This isn't an embarrassment, just an implication of a relative interpretation. — noAxioms

Ok, I can see the problem! Even the presentism you referred before as compatible with SR seems to imply a unique, well-defined state of the universe (the 'unicity' referred in the SEP article on Consistent Histories) :smile:

(as I said it was not exactly what I had in mind. But I think I'll leave it at least for now...)

The Andromeda Paradox is about the ambiguity of what time it is elsewhere, not about the state being definite. The former is a frame dependent thing and the latter is a statement of superposition of something unmeasured. I think you meant the former but your wording suggested the latter. — noAxioms

Yeah, sorry!

Totally agree. Two observers at the same place but different frames might disagree about what is going on at Andromeda, but they'll agree entirely about what has been measured. The light cone from that location is a frame independent thing. — NoAxioms

Yes! In Relativity the ordering of events in every light cone is an invariant (unless one accepts tachyons or any FTL influence). — boundless

I didn't say that. I said the set of events in a given light cone is frame independent. The ordering of those events is still quite frame dependent. — noAxioms

I am not sure I am following you. In fact, I just am saying that the cause precedes the effect in all reference frames without FTL. Isn't it right? :smile:

Interesting corollary for a presentist, who by definition cannot observe any existing thing. In 8 minutes, the thing I observe will not be the present state of the sun. It will be an observation of something nonexistent.
— noAxioms

Yep! Presentism is somewhat problematic in Relativity. I would say that 'global presentism' is simply incompatible with relativity of simultaneity. Maybe a form of 'local presentism' can be saved but it is surely counter-intuitive (I personally lean towards some form of presentism and I admit that I am troubled by this). — boundless

I have done an advocatus diaboli thread defending the compatibility of relativity and presentism, so I maintain that they're not incompatible. SR says that the preferred frame cannot be determined given the special case after which it is named. But inability to detect such a frame does not mean that there isn't a special one. Presentism doesn't even require it to be a inertial frame, and no presentist that knows their physics seems to assert that it corresponds to such a frame. The foliation is always bent, which has the interesting paradoxical implication that no two stationary observers are simultaneous in each other's inertial frames. I find that hilarious, but not paradoxical. — noAxioms

Ok, I see. Interesting, thanks! I wonder if this can be used to reconcile SR with pilot-wave theory... :smile:

Note, however, that the 'presentism' that I had in mind was somehow different. I am not sure of how to explain it - so, I'll leave it for now.

As he puts it, decoherence gives us quasi-classical worlds (branches) but not actual classical worlds. Which means that decoherence can be treated as irreversible and the worlds as classical for all practical purposes. Nonetheless interference between branches continues to happen in accordance with quantum mechanics. — Andrew M

Ok, I see!

I think splitting might be implied only because Copenhagen and Consistent Histories don't specify any physical mechanism, whereas MWI does. But since some other unknown physical mechanism can't be ruled out at this point, then being silent seems a reasonable option (and treating interference as unactualized potential). — Andrew M

Well, yeah, this would explain the silence :smile:

Another curiosity: what do you think about the problem of interfering branches in MWI (and maybe in RQM if no selection mechanism is accepted)? As 'I aM' (see here) noted it is true that due to the decoherence the interference term becomes very small. Yet, rigorously, it is not exactly 'zero'. Given the fact that decoherence occurs a lot of times, it seems possible that - sooner or later - interference will be observed. In other words, it seems that decoherence gives (multiple but) definite outcomes only 'for all practical purposes' (I remember to have read that decoherence is said to solve the measurement problem 'only for all practical purposes' but I am not sure that this the reason why it is said so...).

Nit-pick: you think that maths, a human invention, is more fundamental than the stuff of which the universe is built?
— Pattern-chaser
I consider it something discovered, not invented. If invented, pi would not be the same value in another world.
— noAxioms

:up: — Wayfarer

+ 1 :wink:

(More precisely, I believe that there is something in maths that is discovered. It cannot be totally invented)

I simply meant that without the selection postulate, it seems that RQM implies the splitting.

Anyway, I agree with you. RQM seems simply silent on this point.
— boundless

Maybe it's embarrassed. :yikes: — Wayfarer

Well, possibly! :razz:

I see what you mean, but 'pre-measurement Alice' can predict that 'she' will be 'remembered' by both 'post-measurement Alici'. — boundless

Therefore the future does exist to Alice, just not a specific state. The cat exists to Bob, even when in superposition of dead and alive. — noAxioms

Ok! Agreed! :smile:

Ok, I agree. But my point was another. If you say that 'your' present exist (the 't=0' 3D hypersurface), then the Andromeda Paradox is unavoidable. — boundless

This hypersurface exists. So does this different hypersurface. That's just two different things, not a paradox. — noAxioms

Well, I think I see where you are getting at but I am not sure you can really avoid the paradox if you say that all events in the hyper surface are in a definite state. I am not saying you are wrong, I just do not know.

The answer to this objection is to not regard what is outside the light cone in the same way of what is inside from an ontological point of view. — boundless

Totally agree. Two observers at the same place but different frames might disagree about what is going on at Andromeda, but they'll agree entirely about what has been measured. The light cone from that location is a frame independent thing. — noAxioms

Yes! In Relativity the ordering of events in every light cone is an invariant (unless one accepts tachyons or any FTL influence).

On the other hand, it seems intuitive to accept the 'existence' of the present (e.g. I will observe the present state of the Sun at t=8 minutes). — boundless

Interesting corollary for a presentist, who by definition cannot observe any existing thing. In 8 minutes, the thing I observe will not be the present state of the sun. It will be an observation of something nonexistent. — noAxioms

Yep! Presentism is somewhat problematic in Relativity. I would say that 'global presentism' is simply incompatible with relativity of simultaneity. Maybe a form of 'local presentism' can be saved but it is surely counter-intuitive (I personally lean towards some form of presentism and I admit that I am troubled by this).

There is nothing physical that connects my current state to that past state as opposed to any other random arrangement of matter. Identity is abstract, not real. There are plenty of philosophical arguments that demonstrate this. — noAxioms

I sort of agree with this (but the reasons are not exactly the same...as I said I have a different view about mind) - it seems that there is some kind continuity without, however, a persisting identity (but we are digressing maybe...). This is not IMO however a complete denial of the existence of 'individuality' (and 'identity' in some sense). — boundless

Agree that what I said depends on my personal choice for philosophy of mind. Some interpretations do give identities to things. Mine just happens not to. — noAxioms

I see!

Not sure how you combine your mind interpretation with your QM one. Does the pre-Alice ontologically become one of the post-measurement Alici to the exclusion of the others because the mind-identity can only follow one of them? That's a very different QM interpretation. — noAxioms

Well, note that I do not currently accept RQM as 'my interpretation'. But it is one of my favorites.

Anyway, I do not believe that the mind is something immutable. So, for me, it is more like a 'stream of consciousnesses'. When I said that, in some sense, 'individuality' is preserved I meant that these 'streams' or 'continuums' are distinguishable. Yet, I do not believe that there is 'something' that 'persists' in the process (a 'substance') - in other words, I do not believe in a substantial identity.

Personally, I do not like the idea of the 'branching'/'splitting' - that's why I am insisting with the 'selection' postulate. On the other hand, I think that this kind of position about the mind is logically consistent with the 'branching' idea. Furthermore, I do not believe that the 'splitting' is a necessary feature of RQM. Before this discussion, I believed that there was a selection postulate in RQM. I now think that the theory is simply silent on it.

I agree. But that interpretation of RQM would only be a semantic difference from MWI, not a substantial one. My understanding is that RQM is a more abstract interpretation that captures what Rovelli considers to be the key elements of QM and nothing more. For example in his RQM paper he says, "From the point of view discussed here, Bohr’s interpretation, consistent histories interpretations, as well as the many worlds interpretation, are all correct." That is, they all share those key elements (albeit they commit to further things as well that differentiates them from each other, such as many worlds versus a single world). — Andrew M

:up: I completely agree!

But unless one adds a selection postulate, I believe that before the measurement 'Alice'/'Wigner's friend' can safely say that all 'Alice-s'/'Wigner's friends' will remember 'her'/'him'. What do you think? — boundless

I'm not sure I see the issue you're raising here. But I would agree that post-measurement, Wigner's friend (or friends on a MWI-style reading) would have a memory of themselves prior to measurement.

Regarding a selection postulate for RQM, I think it's just unknown and RQM doesn't commit to anything specific. — Andrew M

Well, I worded it badly. I simply meant that without the selection postulate, it seems that RQM implies the splitting.

Anyway, I agree with you. RQM seems simply silent on this point.

idk, by my understanding decoherence would render interaction between specific separate branches highly improbable. But because there are SO MANY separate branches, it would happen regularly. Sorta like what happens with quantum tunneling. — i aM

:up:

Well, that's another good argument IMO against the view that decoherence is enough to solve the measurement problem (even in MWI).

Yeah, sorry. I was a bit flippant.

The point is that in your example the interaction would give you some information of the other world(s). In MWI, you would observe superposition due to the interference of the branches (which would be very weird).

The fact we do not observe superposition is explained in MWI using decoherence, which suppresses the interference (well, technically, it renders it negligible...).

I'm not sure how weird it is. As an analogy in the classical world suppose that I forget to blow out a candle before I go to sleep and the house burns down. I'll experience regret because there is some world out there where I did not forget to blow out the candle and the house did not burn down. — i aM

...

If no alternative world in which the house did not burn down existed, it would never occur to me to be more careful in the future. — i aM

But you can explain regret in that way only if you accept the concept of parallel universes, i.e. if you accept the idea that whatever is possible, happens which is precisely what says MWI.

Honestly, I do not find any compelling reason to accept the idea of parallel universes/branches etc.

My reading of RQM (and Rovelli) is that RQM doesn't accept the existence of more than one Alice (or, at least, need not). Per RQM, all that is known to Wigner is that Wigner's friend has made a measurement and that the value is (physically) indefinite for Wigner until it is localized in his reference frame. — Andrew M

Well, more or less I always understood RQM in that way! :smile: ... After my dialogue with noAxioms, I am not sure about it. In fact, the 'relativization' of existence makes perfect sense in RQM. For each 'Alice' (each 'Wigner's friend') the other(s) cannot be said to 'exist'. But unless one adds a selection postulate, I believe that before the measurement 'Alice'/'Wigner's friend' can safely say that all 'Alice-s'/'Wigner's friends' will remember 'her'/'him'. What do you think?

I should reword. Yes, the odds are almost a certainty from the beginning that the unicorn will occur in some world, but I meant given a single measurement giving one random collapse. You only get one try. From the beginning of the universe, there's not even a planet on which a single measurement might hope to collapse a unicorn. I would presume an existing Earth with life already on it would raise the odds of a unicorn considerably from the odds from a blank slate. — noAxioms

Ok, I agree!

To summarize, in RQM, according to the pre-measurement 'Alice' both 'Alice-s' (or 'Alici' :wink: ) will exist. — boundless

Such statements are why I balk at A-series wordings like that. Under RQM, both post-measurement Alici (the plural is so stupid I am compelled to use it) consider the pre-measurement Alice to be part of their history. To pre-measurement Alice, the other two do not exist. The future is unmeasurable and thus doesn't exist to that instance of Alice. So there's no 'will-exist' except to indicate that certain future events (post-measurement Alici) consider certain past events to exist and others (like the one where Alice didn't measure it at all) to not exist. — noAxioms

I see what you mean, but 'pre-measurement Alice' can predict that 'she' will be 'remembered' by both 'post-measurement Alici'. This is not too very different from what MWI says. According to this view, MWI and RQM would be similar (not the same, but similar...).

I agree about the lack of contradiction. I know what you're saying and agree with it, but I don't like the A-series wording of it. 'Will exist' makes it sound like existence is something objective that occurs, and not the relation to something. The future Alici cannot exist ever to the pre-measurement one because there is no 'ever' to that version. She's an event, and events don't move into the future. — noAxioms

OK! That's fine. Note however that if one accepts presentism the past and the future do not 'exist'. Only the present does. It is true that the past can be said to 'exist' in the sense that there are data, in the present, about it. At the same time, it is also true that it is possible to make predictions about the future.

So, it seems to me that the relation between the future Alici and the past Alice is more or less the same of the reverse. So, before the experiment, Alice can predict the 'appearance' of the future Alici. So, for her, it seems legitimate to say that a 'split' happens.

SR is also quite consistent for the same reason: different orderings of events are not contradictory if they're from different perspectives. — noAxioms

Ok, I agree. But my point was another. If you say that 'your' present exist (the 't=0' 3D hypersurface), then the Andromeda Paradox is unavoidable. The answer to this objection is to not regard what is outside the light cone in the same way of what is inside from an ontological point of view. On the other hand, it seems intuitive to accept the 'existence' of the present (e.g. I will observe the present state of the Sun at t=8 minutes). If you follow your intuition, you end up with the Andromeda Paradox. I am not absolutely certain that the intuition is wrong, though (if not, dBB supporters would be very happy).

Well, 'I', from an RQM standpoint, am an event, despite my whole me being an abstract worldline. So in that event sense, I don't exist to myself, I only have memory of some past consistent state. From a pure event perspective, any two events (the table lamp and I at two specific moments) cannot exist in relation to each other. Neither exists to the other if the two events are space-like separated, and only one might exist to the other if not. It isn't paradoxical since no such mutual existence relation is ever posited. — noAxioms

Well, this seems also the implication of presentism plus SR/GR.

All different events, so not comparing the same thing. There is no 'the lamp' any more than there is a 'me' making that decision. We're both a series of events, any of which can relate to other events. The fact that a certain event in the past is considered 'also me, yesterday' is an abstract designation I make. There is nothing physical that connects my current state to that past state as opposed to any other random arrangement of matter. Identity is abstract, not real. There are plenty of philosophical arguments that demonstrate this. — noAxioms

I sort of agree with this (but the reasons are not exactly the same...as I said I have a different view about mind) - it seems that there is some kind continuity without, however, a persisting identity (but we are digressing maybe...). This is not IMO however a complete denial of the existence of 'individuality' (and 'identity' in some sense).

(But I am not sure about this :smile: we are probably digressing here)

I think I see what you are getting at*. But I do not believe that this really solves the problem that I have in mind. Unless you specify a duration for the events. — boundless

I'm sorry, but what was the problem? I thought the lack of duration was exactly what solved the problem. — noAxioms

Yeah, it seems so.

Not even familiar with the term Process Philosophy, but perhaps I am discussing it anyway. I'm a poet and don't even know it. — noAxioms

Well, I do not know very much about it. It is a kind of presentist ontology. 'Objects' are not regarded as substantial entities but rather as patterns in succession of events. On the other hand, existence is not defined in a relational way.

I think that works as well, yes. I seem to have a pretty weak grasp on the panpsychism idea. It doesn't seem to have a consistent interpretation from one person to the next. — noAxioms

Yeah, there are a lot of different versions of it. The parallelist variety, for instance, holds that everything has both a physical and a 'mental' aspect. The more complex an entity is, the more complex are both its physical and mental sides. In other words, mind and matter are like two sides of the same coin. But this is off-topic :sad:

Hi,

How does MWI handle probabilities in its branching of worlds? For instance if there are two possibilities (+ or -) and each has a probability of 50%, it makes sense to say that two separate branches result. — i aM

In MWI, you need an additional axiom AFAIK to include the Born Rule. I know that there have been some attempts to derive the Born Rule but I do not know if any of these attempts are regarded to be satisfactory. Note that some deny that such a derivation is necessary (I disagree, though - FWIW).

I'm not sure why, in MWI, the separate branches are said to not be able to interact with one another. Especially if, as you say, it is all one thing in Hilbert space. — i aM

In MWI, the branches can interact. But the likelihood of this interaction is negligible. (By the way, this is another really weird feature of MWI...)

Thanks for the very informative answer, again. I hope I'll can answer tomorrow (if not, I will on Saturday).

I am still confused about regarding 'observers' (as defined in RQM) as 'events' without duration but considering systems as abstractions denoting a stream of events is very interesting (note that I do not regard my confusion as an objection :smile: ...).

(sorry again for the edit... I misread a part of your answer)

For all those Alices (Alici? :confused:) to exist, you need to change the definition of 'exist' from the RQM one to the MWI one. The change of definition is what distinguishes the two, not that there's all these Alices. — noAxioms

Agreed! Until yesterday I did not fully understand RQM, I believe. My confusion was about the treatment of the ontological status of Alice: for me the 'Alice' that does the observation was the same as the 'Alice' observed by 'Bob'. Which is not true in RQM.

Or maybe it becomes a binary event if the measurement tells only if the decay happened during a certain time interval (I am not sure if you were saying this in the second paragraph). — boundless

Yes, I was forcing a (still imbalanced) binary event from a non-binary situation. — noAxioms

Ok!

I partly disagree. In Tegmark's view, there is no metaphysical split at the level of the universal wave-function. So, in that case there are indeed different 'Alice-s' there. It is not a 'real split' because what is truly real is the universal wave-function. — boundless

Sounds pretty similar to me. There is one universal wave function, some solutions including an Alice in one state or another. — noAxioms

I am not sure that I understand you here but I think we agree :wink:

In MWI, the 'Alice that observed decay' would know that, indeed, 'Alice that observed no-decay' exists. In RQM, however, the answer is, you say, negative. Why? The meaning of 'existence' is different in RQM: it is relational. — boundless

Yes, the other Alices don't exist, per definition. We can still, just like the MWI person does, say that the decay measurement doesn't exist relative to the no-decay Alice. She was a real (and even more probable) outcome of the quite real pre-measurement Alice. The unicorn is more difficult due to the vast improbability of one from a 100 million year ago wave function of Earth. Of course humans are near equally as (if not more) unlikely per that same wave function. — noAxioms

Ok!

Pop quiz: At what distance in the past does the wave function of Earth have the highest probability of there ever being a unicorn today? — noAxioms

I would say at the beginning of the history of the Universe (unless one believes to ancient mythologies that actual unicorns wandered on the Earth).

So, we cannot treat the 'pre-measurement Alice' as the 'Alice that observed decay'. So, a negative answer is perfectly fine. — boundless

Note that above I did treat the pre-measurement Alice as being real to either post-measurement Alice. Both have memory of that state and thus have taken a measurement. — noAxioms

Ok.

To summarize, in RQM, according to the pre-measurement 'Alice' both 'Alice-s' (or 'Alici' :wink: ) will exist. But both post-measurement 'Alice-s' regard the other one as 'non-existent' and the 'pre-measurement' as having existed in the past. There is no contradiction here because the states are perspective dependent.

I know that it is problematic within a relational framework, but it appears that there is no reason to believe that the other event did not occur. — boundless

It is not valid to state "the other event did occur" in a relational framework. The statement is an objective one, and has no meaning in a relational framework. It seems to constitute a counterfactual statement just like "The decay occurred". It didn't. It occurred relative to me, but it didn't just 'occur'. Thus just say that the other event occurred to the measurer of the other outcome. I tried to do what with the no-decay Alice above. I carefully avoided a wording like 'she exists' or 'the decay was not measured'. I tried to be careful to always include the relations. — noAxioms

Yeah, sorry! You are correct :smile:

It seems a bit 'solipsistic' for 'Alice that observed decay' to declare that 'her' counterpart that observed no-decay is simply real, in this case. — boundless

Non-mind idealism of a sort, but not solipsism. If existence hinges on an interactive relation with a subject, then that subject defines its own existence, which is interaction-idealism. But there is symmetry. Everything does it, so it isn't solipsism. The table lamp does it, so it isn't mind-idealism. — noAxioms

Ok, I see what you mean. But since yesterday, I am doubting that RQM is really consistent. But maybe the situation is the same as in the case of SR if one does not accept the 'block universe' (well, to be honest, I am not completely sure that even SR without the 'block universe' is really consistent...).

[But if one accepts some sort of panpsychism it is a mind-idealism :lol: Well, I believe that RQM and Process Philosophy can fit nicely together.]

A table lamp might measure a different version of me and thus I cease to exist in relation to it, but I always exist relative to myself, so I'm here. Under solipsism, I would not exist because only the table lamp (or whatever the one privileged thing is) counts. — noAxioms

Yeah in RQM, 'you' according to yourself and 'you' according to the table lamp are different. The table lamp according to itself is different from the table lamp according to you.

Furthermore, as I said before, I find RQM somewhat vague in the definition of 'perspectives'. According to RQM, every physical system is an 'observer'. Fine, but if we consider, for instance a pen, it can be argued that its parts can be considered a 'physical system'. — boundless

If you get right down to the details, an observer is an event, and a system is not. No pen is in a consistent state with itself since at any given moment, parts of it are separated by about a nanosecond of speed-of-light space, and thus the ball of the pen is in superposition relative to the clicker at the other end. So in that sense, I do not exist as a system with a state. Right now I am completely undefined since no point of me has had time to measure any other part. In hindsight (say a microsecond later), that state is mostly defined and immutable. I mean, suppose that my retina has just measured the decay result from the first photon from the device giving answer to my query. OK, several other parts of the front of me also measured that, but not yet the innards (brain in particular), which are still in superposition of decay happened or not. Relative to different parts of me, the measurement was taken or not. It isn't entirely correct to say that the measurement has been taken relative to me since 'me' is not an event. — noAxioms

I think I see what you are getting at*. But I do not believe that this really solves the problem that I have in mind. Unless you specify a duration for the events.

So, it seems that there is actually a very, very huge number of 'physical systems' (and, consequently, 'perspectives'). I believe that this is a legitimate criticism to RQM (as legitimate as the criticism to MWI to have too many 'branches'). — boundless

As legitimate, yes, but I find neither argument to have any teeth. Yes, it is an obscenely large number. Physics is full of those. One measurement such as a decay has seemingly infinite possibilities (not a discreet list), so even a trivial system already has infinite worlds. If one is to worry about how such a list can be instantiated (where do we find room to put them all?), then you're applying classic wording to a Hilbert-space problem. It is a good argument against the universe as computer simulation hypothesis since the implementation really would need to find room to put it all. — noAxioms

This is not true, however, for CI. In CI, only a specific class of entities can be considered an observer (which kind of 'entity' is subject to interpretation). At the same time, however, Relativity seems to imply something analogous.

I noticed that. It seems not to matter. In relation to me (or to anything else), what has happened is what has happened (fixed, in the past), and what will happen is meaningless since none of it can happen to me. Multiple future possibilities will be able to claim me as prior state, but that fact doesn't change if the list of those future possibilities is determined or not. Hence agnostic: it works either way.
— noAxioms

Ok, it seems so from a RQM point. — boundless

Still, the wave function is a pure function just like it is in MWI. Without interference from outside tweaking the terms, how can it not be deterministic? Sure, RQM doesn't care either way, but the same reasoning that MWI uses also works with RQM.
I'm now arguing against what I said above. I also do not know the meaning of the 'agnostic' designation on the wiki list for RQM. — noAxioms

From the perspective of the 'pre-measurement observer' if no 'selection' is made, then I'd agree it is deterministic. But if you consider the perspective of each 'post-measurement observer', the situation changes. For the 'Alice that observed decay' attributing the status of either 'existence' or 'non-existence' to the 'Alice that observed no decay' is meaningless (and vice versa). So, in this sense maybe we should understand the term 'agnostic'.

(sorry for the late edit!)

To summarize, I believe that RQM has two serious problems.

1) I believe - as I said previously - that there are indeed too much 'perspectives'. If every physical system defines a 'perspective'/'reference frame' (i.e. is an 'observer' according to Rovelli), then there is an incredibly huge number of perspectives.

2) If after a 'measurement' the measuring physical system becomes something else, we are, indeed, implying that after every physical interaction (for Rovelli, measurements are physical interactions) 'creates' new perspectives.

IMO, these two are very problematic features of RQM. YMMV!

Well, I was presuming their possibility. It's just a horse-like thing with an evolutionary feature currently found on a narwhal and arguably a rhino (both mammals). It can happen, no? The being a sucker for human female virgins is implausible if there are no humans around, but I think its still a unicorn without that feature (or the blowing of rainbows out it's arse). Pick a different example if you find them impossible. — noAxioms

Well, yeah in that case it might be a possibility. Who knows :smile:

BTW, regardless unicorns, I believe that the ontological status of possible yet unactualized 'things' it is a very interesting topic. Suppose that X is possible but it is never actualized in the past, the present or the future. Is that possible? Well, yeah, I believe. As I said before I am a free will believer. So, there are choices that I could have made in the past but I chose not to make them. Indeed, they were 'possible'.

Maybe those choices are simply 'unreal'. Yet, they are not 'unreal' in the same sense that dragons are (assuming that they cannot exist). Or in the same sense that the academic career of a dragon is. What makes something possible? What it means for X to be possible?
Using a MWI-like reasoning, one might say that whatever is possible is, in fact, actualized 'somewhere'. But for people like me who do not accept that kind of reasoning, it is an interesting conundrum.

Radioactive decay is a wonderful example of a lot more than two possibilities since it could decay any time, thus infinite possibilities. Alice detecting it after one second is different than the Alice that measures it after two, but there is one Alice that doesn't measure it at all yet. This isn't to say that the one Alice is less probable. That depends on the half-life of the sample.
For the sake of your example, the device measures the decay destroys the 'when' part of it, so all Alice gets is a yes/no from the device when she makes her single query as to if it's happen already or not. Cat dead or alive so to speak, which as I recall was done in this manner. — noAxioms

Ok, well as you say they are all indeed different cases. But suppose that as per above, not everything that is possible actualizes. Hence also in this case, only one 'event' happens. Of course, I am assuming that not everything happens. But note that if you, instead, accept the 'existence' of all those Alice-s, how RQM is really different from MWI (except for the universal wave-function)? I believe that Tegmark pointed this out to Rovelli.

Or maybe it becomes a binary event if the measurement tells only if the decay happened during a certain time interval (I am not sure if you were saying this in the second paragraph).

I knew where you were going with this. It seems solved by MWI by exactly what you quoted from Tegmark: There isn't actually any metaphysical split. There is but the one wave function with different solutions, and thus one Alice in two unequally weighted states. There are not two separate worlds, one metaphysically weighted more than the other. But the weight never changes from the '1' that it always was.
This argument is a great one against the whole metaphysical split interpretation of MWI. — noAxioms

I partly disagree. In Tegmark's view, there is no metaphysical split at the level of the universal wave-function. So, in that case there are indeed different 'Alice-s' there. It is not a 'real split' because what is truly real is the universal wave-function.

I thought we were discussing MWI there. Looking back at the exchange, maybe you mean RQM here. — noAxioms

Yeah, I was referring to RQM. Pardon the lack of clarity.

Under RQM, the Alice that measured no-decay does not exist in relation to the one that measured decay, so there is no weight problem. There is no selection since there is but the one world in relation to any particular state. The Alice that had measured the decay is not the same event as the pre-measurement Alice, so no selecting took place for either of them. That's at least how I've been wording it. — noAxioms

Ok, I think I see what you mean. Nice work!

In MWI, the 'Alice that observed decay' would know that, indeed, 'Alice that observed no-decay' exists. In RQM, however, the answer is, you say, negative. Why? The meaning of 'existence' is different in RQM: it is relational. So, we cannot treat the 'pre-measurement Alice' as the 'Alice that observed decay'. So, a negative answer is perfectly fine.
I know that it is problematic within a relational framework, but it appears that there is no reason to believe that the other event did not occur. It seems a bit 'solipsistic' for 'Alice that observed decay' to declare that 'her' counterpart that observed no-decay is simply real, in this case.

Furthermore, as I said before, I find RQM somewhat vague in the definition of 'perspectives'. According to RQM, every physical system is an 'observer'. Fine, but if we consider, for instance a pen, it can be argued that its parts can be considered a 'physical system'.
So, it seems that there is actually a very, very huge number of 'physical systems' (and, consequently, 'perspectives'). I believe that this is a legitimate criticism to RQM (as legitimate as the criticism to MWI to have too many 'branches').

And if we unite these two aspects we end up with an even more huge number of perspectives!

I noticed that. It seems not to matter. In relation to me (or to anything else), what has happened is what has happened (fixed, in the past), and what will happen is meaningless since none of it can happen to me. Multiple future possibilities will be able to claim me as prior state, but that fact doesn't change if the list of those future possibilities is determined or not. Hence agnostic: it works either way. — noAxioms

Ok, it seems so from a RQM point.

We are starting to agree more and more, making these posts a bit shorter. — noAxioms

Well, I believe that we are certainly understanding each other more and more. Not sure about the agreement :wink: Anyway, thank you very much for the very clarifying explanations. Now I feel that I understand RQM much better!

For those, like me, that are not averse to a Kantian-like sub-interpretation of CI, I suggest also this article by Cuffaro: http://philsci-archive.pitt.edu/14357/1/kant_bohr_hermann.pdf.
This paper also discusses the 'neo-Kantian' views of Grete Hermann (link to Wikipedia) who also much time earlier than Bell discovered that not all hidden variables theories are in conflict with the predictions of QM.

Another Kantian-like perspective as I said before is advocated by for instance Bitbol. For convenience I give again the link to Bitbol's paper: http://www.bourbaphy.fr/bitbol.pdf.

↪boundless Thanks for linking that. The comments from Demystifier and other members whose opinion I respect, align with my impression that this paper doesn't reveal anything new of significance. It effectively just affirms that if you insist on locality then you have to give up on CFD, which is in many ways the same thing as 'objective reality'. Bell told us that in the early sixties. — andrewk

Yes, I agree. This experiment does not 'reveal' anything new. This does not mean that it can be interpreted as an evidence that CFD is problematic, but we already knew that.

There is also a parallel discussion in Physics Forums about this experiment: https://www.physicsforums.com/threads/a-realization-of-a-basic-wigners-friend-type-experiment.968181/

I would not call what a table lamp does 'epistemic', so again, I do not personal hold view described there.
— noAxioms

Ok, I agree. Bad choice of terms on my part. For the sake of generality, let's just say 'not ontic' instead of 'epistemic' and drop using the word 'knowledge' in favor of 'addition of new 'information'' (for the lack of a better term). — boundless

The table lamp does acquire information (physics definition), so I can go with that. — noAxioms

Ok let's keep 'information', then! :smile:

Again, you are right! I should have said, instead: 'non-representational' means that the wave-function simply does not have any ontological meaning. — boundless

It may or it may not. Wasn't this the thing we said we're not sure about? It seems quite interpretation dependent, even to the point of interpreting the meaning of 'ontological'. — noAxioms

Yep!

Tegmark's mathematical universe says that the wave function is what the universe is. It doesn't just describe it, but it actually is it. That's one kind of ontological statement: mathematics is fundamental, not just descriptive — noAxioms

Correct! There is a spectrum of views here.
Tegmark's position is that the Hilbert space is the only true reality, whereas our 3 dimensional space is merely an appearance. That's why in his view it is perfectly safe to say that there is no 'real' splitting.
On the other hand, one can even IMO in MWI take a less ontological view about the 'wave-function' (and the Hilbert space). I believe that a purely descriptive position is consistent with MWI.

Another kind would be two different interpretations of this mathematical universe where the mathematical structure has the property of existing (MWI variant) as opposed to RQM, where 'exists' is a relation, not a property. That's a different sort of ontological statement. RQM doesn't say that the universe doesn't exist (isn't nihilistic). It just gives no meaning to the phrase. — noAxioms

Good point! Existence is relational. The 'universe as a whole' is not in relation with anything, so 'existence' here does not apply. This does not imply that the universe 'does not exist', as you say :smile:

In other words, if we take this reasoning seriously we can speak of 'existence' in the presence of relations. If there are no relations, we cannot speak in terms of 'existence' (maybe of a different kind of 'reality', if 'reality' is taken as a more general term than 'existence'...).

And no, neither MWI nor RQM needs to accept this mathematical foundation. I just used the mathematical universe example to illustrate two different kinds of ontology. — noAxioms

Agreed!

I am not sure if what you are describing here can be applied to the 'version' of RQM that I have in mind where after the observation, for the observer, the other branches do not exist. — boundless

Sounds like what I have in mind as well. For me, unicorns don't exist. For the unicorn, I don't exist. You seem to indicate that what I've described is something else. — noAxioms

Thanks for the clarification but I still do not understand how you say that we can avoid some sort of 'selection' here.

To make it simpler, consider a radioactive decay experiment. There are two possibilities: Alice either observes the occurrence of the decay or not. Let's call these two possibilities, 'decay' and 'no decay' respectively. Let's also say that the probability of 'decay' is much less than the one of 'no decay'. Alice performs the measurement. And, say, she observes 'decay'.
If we do not accept the selection, we should accept that there is 'another Alice' that observes, instead, 'no decay'. And - besides the existence of 'another Alice' that observers 'no decay' - we have the different weights problem that occurs in MWI. In fact, this scenario is not very different from MWI. The only difference is that here we do not have a 'universal wave-function'.

On the other hand, if a selection is accepted, there is only one outcome.

BTW, the table on the Wiki article on the interpretations of QM, says that RQM is 'agnostic' about determinism. So, maybe RQM is simply silent about the selection.

[Also, IMHO unicorns here are not a good example. For me, they are simply impossible (but as you say, better not to be too dogmatic about this :wink: ).]

BTW, I am sympathetic to some kind of realism for mathematics. I do not believe that mathematics is entirely a product of our minds but, at the same time, the usual version of mathematical Platonism does not convince me. So, maybe mathematical relations are 'real' (but not 'physical'). Interesting view — boundless

Or maybe they're physical but not real. — noAxioms

I believe that 'real' is more general than 'physical'. In fact, I just cannot understand how mathematical relations can be 'physical'. :wink:

In my opinion Wheeler's view is a bit ambiguous. At times he suggests some form of 'panpsychism'. In other places, he seems to suggest that an 'observer' can be a sufficiently complex physical object. By 'sufficient complex physical object', I mean that such an object must be able to store and process information. And maybe, he considers that these objects are somehow sentient.
But IMO, he does not give a 'special role' to human consciousness (or animal consciousness...).

Personally, I prefer either Bitbol's approach, where you can define perspectives to sentient beings, or Rovelli's approach where you can define a perspective to everything (and might relate Rovelli's view to a form of panpsychism). — boundless

I'll now explain why I believe that replacing 'conscious observers' with 'sufficiently complex physical object/system' does not solve anything IMO.

In CI, 'classicality', i.e. having definite values of physical quantities, arises due to collapse of the wave-function. But collapse itself needs in CI a classical physical system, i.e., in this view, a 'sufficiently complex physical system'.

This is a deep issue in CI. Classicality is both a pre-condition to explain 'measurements' and a consequence of measurement. So, the problem of conscious observers is now replaced by another problem: CI cannot explain the arising of classicality - after all, to work CI requires that something must be treated as classical in the first place. Hence, the first occurrence of a 'classical system' is left completely unexplained in CI. Hence as it is said in in an already quoted article by Bitbol (http://www.bourbaphy.fr/bitbol.pdf), according to CI:

As a well-known article about the measurement problem of quantum mechanics puts it: the quantum theory can describe anything, but not everything [Peres1982][Fuchs2000].

This also shows IMO that the problem of the validity of 'ancestral statements' is not solved by replacing 'conscious observers' with 'sufficiently complex physical systems'. CI needs some kind of classical object as a starting point.

If one wants to avoid completely the problems due to giving a special role to conscious observers of some physical systems, one IMO should simply choose another interpretation. After all, there are a lot of interpretations of QM. RQM for instance does not have this problem because it treats all physical systems as 'observers', so there is no problem to explain how 'classicality' arises.

Apologies for slow reply time. — noAxioms

No worries!

A tool (a map of Paris say) may be just a tool and not be the thing it describes, but it very much still describes Paris. Thus I object to the statement that the tool doesn't describe anything. — noAxioms

I see what you mean and IMO this is a good argument against a 'non-representional' reading of the wave-function: it is difficult to accept that the wave-function can have absolutely no ontological meaning and, at the same time, be so useful.

I actually am not sure. On one hand, I am inclined to say that when the 'observation' occurs the other 'branches' do not exist (for the 'observer'). On the other hand, I recognize that the 'non-representational' reading is problematic. Maybe thinking in terms of potentiality/actuality helps.

This is very compatible with the view that collapse is due to an increase of knowledge (i.e. an 'epistemic', not 'ontic' view). — boundless

I would not call what a table lamp does 'epistemic', so again, I do not personal hold view described there. — noAxioms

Ok, I agree. Bad choice of terms on my part. For the sake of generality, let's just say 'not ontic' instead of 'epistemic' and drop using the word 'knowledge' in favor of 'addition of new 'information'' (for the lack of a better term).

In other words, 'real' and 'representational' should be taken as synonyms (or very close to that) - the point is that there is a biunivocal correspondence between mathematical formalism and reality. — boundless

Disturbingly close, yes, to the point where no arbitrarily close inspection will yield a difference. This is not true of the paper map of Paris. — noAxioms

Again, you are right! I should have said, instead: 'non-representational' means that the wave-function simply does not have any ontological meaning.

I don't think there is just the two choices. It is certainly not representational in my opinion, but the wave function has meaning only in relations, not objectively, so it isn't ontological as in 'is real' but more like 'does relate to'. 5 really is less than 7 and that is not just a representational property of the tool that is the integers, and yet the integers don't need to have Platonic realism (ontological meaning) for that relation to be true. — noAxioms

Ok, I see. I am open to the view that there is a 'middle way' between the two positions. At the same time, however, I am not sure if what you are describing here can be applied to the 'version' of RQM that I have in mind where after the observation, for the observer, the other branches do not exist.

BTW, I am sympathetic to some kind of realism for mathematics. I do not believe that mathematics is entirely a product of our minds but, at the same time, the usual version of mathematical Platonism does not convince me. So, maybe mathematical relations are 'real' (but not 'physical'). Interesting view :smile: