Einstein and "spooky actions"

[Mr Spinkles]

He won it for his formulation. I think Poly's point is that the Born rule is one of the fundamental postulates of quantum mechanics. It's not a consequence which is derived, such as Fermi's golden rule.

Bingo! Copenhagen, like MWI, doesn't let you derive it at all - it just magically appears as an axiom for no theorectical reason.

Fair enough. For the record, I personally view MWI as an unhelpful and incomplete interpretation of QM at best. In that respect I probably agree with Legion. But I haven't read your posts carefully enough to weigh in on the central point of dispute between you two. I just thought I might help clarify a this (seemingly minor) point where it seemed you and Legion were talking past each other. (But perhaps I only made it worse?)

 

[idav]

Fair enough. For the record, I personally view MWI as an unhelpful and incomplete interpretation of QM at best. In that respect I probably agree with Legion. But I haven't read your posts carefully enough to weigh in on the central point of dispute between you two. I just thought I might help clarify a this (seemingly minor) point where it seemed you and Legion were talking past each other. (But perhaps I only made it worse?)

Whether going with MWI or with what Legion was saying about the photon knowing and doing some sort of time jump still leaves me with, "how in the hell did it do that?". Sure the copenhagen interpretation works but we still don't know how, which is why there are alternatives.

 

[Mr Spinkles]

[T]o say that it [the Born rule] isn't derived, while saying Fermi's golden rule is, seems to be rather odd.

I think when Poly and I talk about the Born rule we are talking about the interpretation, which is the key thing. As you said:

One cannot derive the Born rule itself, in that the "only one interpretation" part in the above scan is by no means proven.

Right. That is why it is generally taken to be one of the fundamental postulates of QM, right alongside the Schrodinger equation. From the fundamental postulates various consequences can be derived, such as Fermi's golden rule. IOW, the Born rule is an assumption that goes into deriving Fermi's golden rule, but not vice-versa.

 

[LegionOnomaMoi]

I'm not talking about how the relationship was produced originally, but how it relates to the wider model. Fermi's rule is not axiomatic, whereas Born's is.

If the fundamentals of QM weren't axiomatic, we wouldn't be discussing this. Physics would be using proofs for things like whether or not some statistical model was more robust or equivalence in computational complexity. Because the sciences derive "axioms" from observations, as reality is the set of axioms in any empirical approach. It's what empiricism means: observation and measurements that are used to construct models and theories.

Derivations in QM are uncomfortably close to mathematics. By 'uncomfortably", I mean that the scientific endeavor was and largely still is about constructing models from reality and only then making use of derivations (usually tangentially). When physics, the oldest science and the one that has as its name a word related to "physical" and derived from the Greek word for physical/natural, uses the equivalent of mathematical axioms, we run into issues. If I'm like basically all of those in social and behavioral sciences, and have nothing other than the generalized linear model and few add-ons that I know how to make my software use, I'm still using a model derived independently to apply to a specific study. The observations are not the model. In quantum physics, the distinction between the model and the system just isn't there in the way it is for every other scientific field.

Clearly, we've gained a lot from quantum physics. Which Is why I've lost count of the number of monographs, papers, and volumes that begin with saying how successful QM is and yet how there's so much missing and part of what's missing is knowing what the missing parts are.

Everett saw, correctly, an inconsistency in QM (I believe I quoted his dissertation for you). He wished to resolve that by getting rid of the "extraneous" stuff. He couldn't, and neither has anybody else. The born-rule works. We get results, and have been for a very long time (as far as science goes, anyway). The issue with MWIs is that they don't succeed In explaining why it is that the "standard model" works, and they require either the same things it does (mathematically), or ad hoc additions for which they lack a reason. Even worse, the theory requires that (as you say) everything is a wavefunction (or at least everything is "quantum" and there is nothing like a collapse), but cannot then demonstrate why we get results at all. We perform measurements on systems using experimental methods and what is basically statistical mechanics with a twist (and shaken, not stirred). But while we have both classical and statistical mechanics, we only have QM as a statistical mechanics which (unlike its classical counterpart) is related in ways we don't know to the systems we apply it to. The many-worlds interpretations, like me, don't like this. We have a scientific discipline devoted to understanding physical reality "works', and we finally dug down deep enough to discover probability functions.

That said, just because I don't like this irreducibly statistical view doesn't mean I'm prefer to replace it with another theory that can't reproduce QM without contradicting itself just because it means I don't have to deal with the measurement problem. This just leaves me with either stealing QM and re-arranging or re-naming certain things such that I basically have QM, the born rule, and an added model that I can't really explain because I have to use probabilities and a basis for measurement, and I can't show it is true either. Instead of trying to remove the interpretation part and just leave the math, I've added a theory with an at best tenuous relationship to experimental procedures and that I can't show evidence for. I don't see the point.

 

[PolyHedral]

Physics would be using proofs for things like whether or not some statistical model was more robust or equivalence in computational complexity.

...They don't? I would've thought the computational complexity and stability of, say, different weather models would be quite important. (That doesn't tell you anything about reality mind you...)

Derivations in QM are uncomfortably close to mathematics. By 'uncomfortably", I mean that the scientific endeavor was and largely still is about constructing models from reality and only then making use of derivations (usually tangentially).

That's the sign of a good model, though - when you deductively derive conclusions, and they turn out to be right.

When physics, the oldest science and the one that has as its name a word related to "physical" and derived from the Greek word for physical/natural, uses the equivalent of mathematical axioms, we run into issues. [...] In quantum physics, the distinction between the model and the system just isn't there in the way it is for every other scientific field.

I'm trying to point out that this was never true - the only reason it's become evident in QM when it isn't evident elsewhere is because clasiscal models are so intuitive and so easy to understand that we don't think of them as models, but instead of how reality "really is."

Every system is described in terms of some model. The Newtonian solar system? The concept of moving masses in 3D flat space requires a model defining what "mass," "3D space" and "time" all are. (A model formulated such that the shell theorem necessarily follows from it) The nucleus and gold foil Rutherford experimented with? That requires a model that includes an inverse-square electric field and a "proton" object, and implies a model of chemistry that says that gold atoms are quantintively different from other kinds of atoms. His results make no sense without thinking of nucleii, electric charge, and things called "protons". The combined gas law makes no sense without the model of a gas as a collection of non-interacting atoms, etc, etc.

There's nothing more "real" about imagining a proton as an elecrtically charged billard ball than there is imagining it as a complex-valued field that interferes with itself. They are both models, and they both describe things. Talking about a physical system without some model is equivalent to writing a book without using language - completely impossible and nonsensical.

Even naive physics is a model, because it specifies things that aren't immediately evident from raw sensory data humans can experience. It says the world is 3D, there is a defined "down", and solid objects do not intersect one another. As anyone who's ever wrote a physics engine can tell you, these things need specifying - they do not appear "automatically" in any way. It doesn't feel like a model, and it took a long time to work out it was a model, and not just an instrinsic property of the world, but it is in exactly the same way Newtonian gravity, atomic chemistry and quantum mechanics are all models.

Or would you like to tell me how it makes sense to talk about the orbits of the 8 planets in the Solar System without a model?

The issue with MWIs is that they don't succeed In explaining why it is that the "standard model" works, and they require either the same things it does (mathematically), or ad hoc additions for which they lack a reason.

Since MWI is an interpretation of a theory, rather than a theory unto itself, I would expect it to be mathematically identical to QM - by definition, anything that isn't is a seperate theory, not merely an interpretation of one.

As for the "why," which why do you mean? Why do the equations of QM produce accurate predictions? Or, why do we experience the naive-physics world when the objects described by quantum mechanics are so much weirder?

Even worse, the theory requires that (as you say) everything is a wavefunction (or at least everything is "quantum" and there is nothing like a collapse), but cannot then demonstrate why we get results at all.

Didn't I already explain how that works? Or did you disagree with the explanation's validity for some reason?

We perform measurements on systems using experimental methods and what is basically statistical mechanics with a twist (and shaken, not stirred). But while we have both classical and statistical mechanics, we only have QM as a statistical mechanics which (unlike its classical counterpart) is related in ways we don't know to the systems we apply it to.

Just like the spoon, there is no system apart from that in the model. There is nothing to apply the statistical mechanics to, because the only thing it makes sense to talk about are the entities described by those same mechnaics.

You can talk about observations, and whether models predict observations correctly, but once you start talking about the system geneating those observations, you have some (perhaps very general) model, and it makes no sense to ask how we are applying that same model to the system.

Instead of trying to remove the interpretation part and just leave the math, I've added a theory with an at best tenuous relationship to experimental procedures and that I can't show evidence for. I don't see the point.

I suspect you misunderstand MWI then.

 

[LegionOnomaMoi]

...They don't?

I meant to imply that this would be all that they would use proofs for, not that they don't. My bad.

That's the sign of a good model, though - when you deductively derive conclusions, and they turn out to be right.

1) MWI rids itself of the deductively derived conclusions
2) Knowing how your model relates to your system is even more important

classical models are so intuitive and so easy to understand that we don't think of them as models

I can link you to a particular event devoted to models, journals devoted to modeling in particular fields, conferences that are always on modeling, and which run the gamut from nanotechnology to musicology, all without looking at my bookshelves. It seems people are pretty aware that classical models are models.


To start, a simple definition (from the source I'll use as an example):
"A model is a simplified mathematical representation of a system" Boccara's Modeling complex systems (Graduate Texts in Physics).

I'll use their intro example: predator-prey models

Every variable corresponds with something that was either observed to construct the model and/or will be replaced by values for specific observations when used.

H, for example, is for herbivores, K for carrying capacity, -cP is the rate at which predators will die, etc. There are many such models, and the way one would determine which to use (of if a new model was needed) is by e.g., going out to some place, setting up observation equipment, and seeing how well the model describes what we actually see.

This doesn't change if we get really big (stellar astrophysics) or much smaller, like a neuron:

Why is it that I can take a book I have on modeling in neuroscience, climate science, systems science, or computer science and find so much of the same stuff, yet none of them treat models like QM? It's because to build a model of something, from cloud seeding to cloud computing, we observe the system and extract the relevant properties or features involved in the processes we're interested in. Then we can see not only if it works, but also alter, eliminate, or add features of the system to see if it improves the model. If I predict a neuron will fire at around x threshold, and it doesn't, I can check to see what I left out and/or if the various components of the model weren't adequate approximations of the system's components.
That's what correspondence means. I can't do that with QM. If I prepare a system and transcribe it into a model using e.g., Schroedinger's wave function, I cannot say even say why I describe the state as I do because pretty much all quantum "weirdness" is a logical consequence of using amplitudes to calculate probabilities. Also, whether or not it is an accurate interpretation, my model at least appears to describe one system as having more than one state at the same time. And because nobody knows how the "state" of the system corresponds to the model, it's impossible to say whether it does or doesn't.

The Newtonian solar system? The concept of moving masses in 3D flat space requires a model defining what "mass," "3D space" and "time" all are

"A model is a simplified mathematical representation of a system"

In QM, the model is the system. Which, if we are comparing to classical physics, either means it isn't a model, or that their isn't a system it models.

However, we aren't dealing with classical physics, but as you seem to think there's no difference, then perhaps you can explain why the definition given doesn't apply in QM.

When computational neuroscientists develop models, they observe neurons (mostly via imagining equipment rather what Hodgkin and Huxley had available). They then construct a model which they think captures the essential features of this type of neuron. Finally, they (and others) can repeatedly test this model to see if the parts and processes of the neurons it describes are good approximations of what actually happens.

Schroedinger's wavefunction, Heisenberg's matrix mechanics, and every other formulation of QM systems are constructed through outcomes alone. That's because in order to prepare a quantum system, we've disturbed it in non-trivial ways we can only approximate.

Talking about a physical system without some model

Models are essential. Recall my mention of Rosen's metabolic-repair model that (according to many, and in disagreement with many) purportedly shows biological systems aren't computable. You objected to the model because it made use of a function that wasn't explicitly related to the system. That's all of QM.

I would expect it to be mathematically identical to QM - by definition, anything that isn't is a seperate theory, not merely an interpretation of one.

One could argue that QM isn't a theory, but I don't think that's useful. It's much more important to note that it isn't an interpretation of QM, and cannot be. Because QM uses a particular set of postulates which include the very thing MWIs wish to be rid of: the variously described projection postulate. From Everett onwards, the goal involved solving the measurement problem. None have.

did you disagree with the explanation's validity for some reason?

Yes. We derived a way to get experimental outcomes (a model) that is very successful at telling us what we will find under the assumption that the same preparation will result in a system that is similar enough to give us similar outcomes. It works. Once you say that every measurement is simply a new branch, then the assumption that systems which are prepared in the same way will tend to give the same results has no basis. We know for a fact that for a given preparation our predictions can be way off. Otherwise we would have a statistical mechanics analogue to QM, rather than QM as a statistical mechanics without any classical mechanics analogue.

In particular, there is no classical analogue to the QM measurement process. Probability is at the heart of QM, as is probabilistic reasoning: given that I've prepared this system in this way, I should tend to get the same results as others do when they prepare it (akin to statistical mechanics). If I build a model of a neuron based on measurements of that neuron, the reason I expect my model to be reliable is because I'm assuming that repeatedly observing the neuron means repeatedly observing that neuron.

There is no reason for this assumption in any MWI. It's as if every time I measure the point at which a neuron fires, I'm measuring a different neuron without any reason for thinking it should behave the same as another (and they don't).

and it makes no sense to ask how we are applying that same model to the system.

Because all we have is a model that we call our system. Where else is this done?


We prepare a particular system in a particular way under the assumption that it will tend to result in outcomes similar to identical systems and preparations, and under that assumption we're right. However, MWI tells us that we aren't getting any outcomes particular to a particular system prepared in a particular way, because there are no outcomes. Just a resulting branch. There aren't even particular systems, as we can't prepare a quantum system if that's all that exists.

I suspect you misunderstand MWI then.

Possible. But you've said similar things about physicists:

...Those are completely different statements...they're not two ways of saying the same thing at all.

And physicists have said much the same as I have:
"no known version of the theory (unadorned by extra ad hoc postulates) can account for the appearance of probabilities and explain why the theory it was meant to replace, Copenhagen quantum theory, appears to be confirmed, or more generally why our evolutionary history appears to be Born-rule typical."
perhaps you could explain what I misunderstand, and one or more sources you use so that I can understand it as you do.

 

[idav]

MWI attempts to explain how all the weird stuff is happening at all. There are different ways of approaching MWI and Poly is saying it is some sort of quantum teleport by use of entanglement. I like that approach because it fits nicely with the spooky action at a spooky distance. When entanglement occurs it makes it seem as though the particle "knew" what state to be, even at a spooky distance, when in fact the particle was prepared that way as soon as it entangled with a pair.

 

[LegionOnomaMoi]

MWI attempts to explain how all the weird stuff is happening at all.

That's true of any interpretation. This particular one not only fails to give an explanation of that "weird stuff" that is consistent, it's "weirder" than the standard model. It assumes that the reality of classical physics doesn't exist.

There are different ways of approaching MWI and Poly is saying it is some sort of quantum teleport by use of entanglement.

MWI is a description for all of QM. It can't be "some sort of quantum" anything, as it is supposedly an explanation of all quantum observations/measurements.

I like that approach because it fits nicely with the spooky action at a spooky distance.

Instead, there is no action, no distance, and no classical reality.

 

[idav]

That's true of any interpretation. This particular one not only fails to give an explanation of that "weird stuff" that is consistent, it's "weirder" than the standard model. It assumes that the reality of classical physics doesn't exist.

On the contrary MWI is saying all realities exist at least in potential.

MWI is a description for all of QM. It can't be "some sort of quantum" anything, as it is supposedly an explanation of all quantum observations/measurements.

If it is in the qm world then it is using the mechanics like entanglement.

Instead, there is no action, no distance, and no classical reality.

Since when? Spooky action happens regardless of distance as shown in the delayed choice experiment. The spooky distance thing was predicted to happen because we already saw that type of behavior in the delayed choice experiment and distance wasn't the issue then either.

 

[LegionOnomaMoi]

On the contrary MWI is saying all realities exist at least in potential.

From the guy who actually performed on of the most cited "spooky action at a distance" experiments:
"Basically the solution proposed by the many-worlds view of quantum physics, also called the multiverse, is to deny that experiments have unique outcomes... According to this view, everything is quantum, once and for ever. Hence, the entire reasoning of Sect. 3.2 collapses: there are no inputs and no outputs! Actually, the motivation for many-worlds is not nonlocality, but the fact that today’s quantum theory offers no answer as to when a quantum measurement is finished. Hence, they conclude: quantum measurements are never finished, everything gets into an enormously complex state of superposition. Somehow, the only real thing is the Hilbert space and the linearity of Schrödinger’s equation."
from "Are There Quantum Effects Coming from Outside Space–Time? Nonlocality, Free Will and “No Many-Worlds" in Is Science Compatible with Free Will? Exploring Free Will and Consciousness in the Light of Quantum Physics and Neuroscience (Springer, 2013).

 

[Mr Spinkles]

Legion,

I actually tend to agree with your overall point about the unhelpfulness of the MWI.

But I don't understand what you are trying to say when you compare QM to population models. I think you may be missing Poly's point and reaching for disagreement unnecessarily.

PolyHedral is simply saying that a particle is "really" some strange object, and it is accurately described by the mathematics of QM. Naturally, a strange object requires strange mathematics to describe. That's what he means when he says QM is a "model", he's simply saying a particle is a particle and an equation is an equation. He's not saying QM is a simplified model the way population models, and most models outside fundamental physics, are simplifications. What you seem to be (correctly) pointing out is that in the world of models in science, QM is unique because it is a potentially complete model, which in principle specifies everything about the system it is describing. Other models in science, outside of fundamental physics (like population models) are typically incomplete simplifications from the get-go. So for example, when we make a population model we all know that there is more to the universe than just X = number of rabbits and Y = number of foxes. We are not even trying to capture all of reality in this model. But in fundamental physics, we are trying. In principle the Schrodinger equation (or some other equations) capture everything about reality. If we lived in a classical universe, then in principle classical equations would capture everything. That is what you seem to mean by "the system is the model", and I agree with that, but to be precise what you want to say is "the system is completely described by the model". This does not contradict Poly's point that the system is distinct from the model.

And I think Poly is right that the same things could be said of classical mechanics, which was in principle also a complete description of the entire universe. Now of course we know classical mechanics is actually a simplification of reality (like a population model). But the difference between a hypothetical classical world vs. the actual quantum world we seem to live in is, as Poly says, not a difference of elevating the status of models. In either case there would be "reality" and then there would be models which we believe to be complete descriptions of that reality. The key difference would be, as Poly says, that the classical case would be easier for primate brains to intuit.

 

[PolyHedral]

Sadly, the system will not let me frubal that post.

In either case there would be "reality" and then there would be models which we believe to be complete descriptions of that reality. The key difference would be, as Poly says, that the classical case would be easier for primate brains to intuit.

Also, it should be noted that reality (as distinct from any given model of reality) is a black box. It produces data, and the entire purpose of science as a discipline is to build some description that fills in the black box and predicts future data. That's why we can get away with a probabilistic model without saying its a simplification of some deterministic model - reality is just probabilistic, end of.

 

[LegionOnomaMoi]

But I don't understand what you are trying to say when you compare QM to population models..

The models are a demonstration I used to answer PolyHedral's query:

I still don't know what correspondence you're looking for. In the era of Rutherford messing around with gold foil, we said that the nucleus was a collection of charged spheres, even though we hadn't actually seen them and later on saw there were no spheres to be found.

The correspondence I refer to is this:
"In classical physics, the notion of the “state” of a physical system is quite intuitive...there exists a one-to-one correspondence between the physical properties of the object (and thus the entities of the physical world) and their formal and mathematical representation in the theory...With the advent of quantum theory in the early twentieth century, this straightforward bijectivism between the physical world and its mathematical representation in the theory came to a sudden end. Instead of describing the state of a physical system by means of intuitive symbols that corresponded directly to the “objectively existing” physical properties of our experience, in quantum mechanics we have at our disposal only an abstract quantum state that is defined as a vector (or, more generally, as a ray) in a similarly abstract Hilbert vector space."

The quote continues, but the rest is contained in PolyHedral's response:

The conceptual leap associated with this abstraction is hard to overestimate. In fact, the discussions regarding the 'interpretation of quantum mechanics' that have occupied countless physicists and philosophers since the early years of quantum theory are to a large part rooted precisely in the question of how to relate the abstract quantum state to the 'physical reality out there.' (pp. 14-15)
from Schlosshauer's Decoherence and the Quantum-to-Classical Transition (from Springer's monograph series The Frontiers Collection; 2007)

"Exactly as it appears." It's not like the universe can't be a line through Hilbert space.

I think you may be missing Poly's point and reaching for disagreement unnecessarily.

Wouldn't be the first time. But when I quoted Gisin, and Polyhedral stated explicitly that the position Gisin was describing is his own:

Is your quote supposed to imply something on its own? Because as far as I can tell, it's merely explaining the position I'm arguing

and this is the position:
"Basically the solution proposed by the many-worlds view of quantum physics, also called the multiverse, is to deny that experiments have unique outcomes (for a long list of various versions of the many-world view see Kent 2010). According to this view, everything is quantum, once and for ever. Hence, the entire reasoning of Sect. 3.2 collapses: there are no inputs and no outputs! Actually, the motivation for many-worlds is not nonlocality, but the fact that today’s quantum theory offers no answer as to when a quantum measurement is finished. Hence, they conclude: quantum measurements are never finished, everything gets into an enormously complex state of superposition. Somehow, the only real thing is the Hilbert space and the linearity of Schrödinger’s equation."

I don't see how what you say below can be true:

PolyHedral is simply saying that a particle is "really" some strange object,

Nor do other statements made in this thread (and others) seem to me to be saying anything simple about particles:

The mistake is to interpret the wavefunction as the probability of finding a particle there - there is no particle, only an entanglement.

When you have a wavefunction that describes the entire contents of the universe, there are no observables, because there are no observers which are not also part of the wavefunction. (Which means that the "observations" they make don't collapse the system at all but instead superpostion them.)

The observer does observe classical-esque results because the observer is superpositioned. Once that superposition happens, the branches evolve indepdently, and because they [usually] don't interact with one another, nobody ever observes themselves to be in a superposition, even though they are. (Although they can calculate backwards and conclude that they were.)

We still observe things in the everything's-a-WF model, it just shows up differently. The answer is now not "The wave collapses when you open the box and the cat is definitely alive or dead," it's, "The cat is both alive and dead, and your brain just got entangled with it."

I can see how some of these, particularly the first, do what you say. But taken together and understand as a sample with many other such statements, it's hard for me to see such views as "simply stating" anything about particles.

He's not saying QM is a simplified model the way population models, and most models outside fundamental physics, are simplifications

I know this. I brought up models that weren't like those in QM because of a lengthy tradition (which will one day be put to verse to in an epic far superior than Homer, and written in lambda calculus) between us on how much of a difference exists between QM and classical physics as well as other sciences:

That depends on what you mean. I think MacKinnon put it as concisely as possible in the preface to Interpreting Physics: Language and the Classical/Quantum Divide (Boston Studies in the Philosophy of Science, Vol. 289): "In contemporary particle physics conclusions from a theory are never tested against observations. They are tested against inferences based on observations and a network of presuppositions supporting the inferential process of experimental physics."

...What?
The sentence you quoted appears to either misunderstand how inferential science works, or is suggesting that physics is not science.

That was not his complete response, but I am still confused about the rest:

The more advanced theories cannot be directly tested against observation, because our squishy biology does not have the equipment to observe the results. We can't observe that quarks or muons or even individual atoms exist directly; they're inferred to exist from theory - or from "a network of presuppositions."

That seems pretty much what my quote said.

but to be precise what you want to say is "the system is completely described by the model".

I have said that, but again there is context here that makes communication difficult. I said "is the model" because the only thing one can point to when asked about the physical system is some formal description. There is of course something that is described by e.g., a wave function, but what we call the physical system is not that something. Not because it is simplified, but because only in quantum physics does experimental preparation ensure the the system is affected in non-trivial ways. So while the quote below is quite true:

This does not contradict Poly's point that the system is distinct from the model.

it has little to do with my point. I was emphasizing the qualitative differences between models in QM vs. models in general (or trying to do this).

EDIT: It should go without saying, but I would ask that Polyhedral correct any statements I've made which misconstrue his views, or ask me to correct or even remove any, I will of course do that (I have lots of posts where I have said things poorly, or have inaccurately expressed what I believe, or were just nonsense because I'd gone days without sleep and was using only alcohol as fuel). I am trying express my understanding of another's view, and there are any number of reasons and ways that I can fail do to this.

 

[Mr Spinkles]

I have said that, but again there is context here that makes communication difficult. I said "is the model" because the only thing one can point to when asked about the physical system is some formal description. There is of course something that is described by e.g., a wave function, but what we call the physical system is not that something. Not because it is simplified, but because only in quantum physics does experimental preparation ensure the the system is affected in non-trivial ways.

If I understand you correctly, you are saying that the fact that "experimental preparation ensure the system is affected in non-trivial ways" implies that "what we call the physical system is not [described by e.g., a wave function]". Is that correct or am I misinterpreting you? I don't understand why the former implies the latter. Obviously the former is true, and it is a significant conceptual difference between QM and classical physics.

I'm also confused as to what you mean by "the only thing one can point to when asked about the physical system is some formal description". It seems evident to me that if asked about the physical system, I can "point to" a formal or an informal description. Do you mean, a formal description is more important in QM than in classical mechanics, because we can often "intuit" the results of classical mechanics in an informal way, whereas in QM our intuition is more likely to lead us astray unless we stick to the mathematical formalism? That's certainly true ...

 

[LegionOnomaMoi]

If I understand you correctly, you are saying that the fact that "experimental preparation ensure the system is affected in non-trivial ways" implies that "what we call the physical system is not [described by e.g., a wave function]". Is that correct or am I misinterpreting you?

I'm not sure, because the answer is yes and no. I guess the best way to approach an answer is by looking at what is meant when, in experimental quantum physics, the state of a "system" is described (italics in original): "What is called a ‘‘state of a system’’, whether it is represented by a wavefunction, a ket or a density operator, refers to a statistical ensemble of systems, all prepared before the initial time under the same conditions as the system in hand."

If we are describing the state of a system, but it is not really a system, then we are describing something that relates somehow to what we want to measure, but lacks the one-to-one correspondence between that something and the way it is transcribed mathematically using theory and the experimental design.

Or, alternatively: "Rather than grappling with these [QM measurement interpretation] difficulties, in this book we take an operational approach. That is, we treat quantum mechanics as simply an algorithm for calculating what one expects to happen when one performs a measurement. We treat uncertainty about future measurement outcomes as a primitive in the theory, rather than ascribing it to lack of knowledge about existing hidden variables.
We will still talk of a quantum state as representing our knowledge about a system, even though strictly it is our knowledge about the outcomes of our future measurements on that system"
Wiseman, H. M., & Milburn, G. J. (2009). Quantum measurement and control. Quantum Measurement and Control (Cambridge University Press)

Once again, we "talk" about a quantum state that isn't, strictly speaking, related to any state of a physical system. And as I'm sure you are aware, there are hundreds upon hundreds of books and papers that either argue for a particular way of relating the formalism (wavefunction, ket, etc.) to the whatever-it-is we intend to measure, or that simply go through the issues and the various ways it is approached.

My point is not that the mathematical models do not describe the system at all. They must (otherwise, we'd have chucked QM out the door long ago). When we call the "state of the system" something that isn't even a system, then whatever formal (notational) schema we use to represent/model the system, we cannot say how the state of the system in the model corresponds in a one-to-one way with the system itself (or even what that really means).

We can say quit a bit, and a lot more thanks to much better technology, but we do not have the classical bijective relation between properties in the model and properties in the system that exists everywhere else in the sciences.

In fact, while models tend to be simplified version that retain only the essential features of interest, in QM the state of the system is complete. All the information of the QM system is contained (or encapsulated) in the state. Somehow, we have a perfect model (at least in the way it is described, as it is not just essential features but everything there is to know), yet we have so many different interpretations of how the model relates to measurement that these different theories can be grouped by type.

I don't understand why the former implies the latter. Obviously the former is true, and it is a significant conceptual difference between QM and classical physics.

That's more or less my point.

Do you mean, a formal description is more important in QM than in classical mechanics, because we can often "intuit" the results of classical mechanics in an informal way, whereas in QM our intuition is more likely to lead us astray unless we stick to the mathematical formalism?

Yes.

 

[idav]

Qm is different but isnt some crazy thing going against physics. For example thermal light can be calculated using the classical approach via schrodingers equations.

Theoretical and experimental justification for the Schrödinger equation - Wikipedia, the free encyclopedia

This implies something missing.

 

[LegionOnomaMoi]

Qm is different but isnt some crazy thing going against physics.

Crazy physics is rather relative. Aristotle's mechanics (theory of motion) held sway for almost 2,000 years. Why? Because it took more than philosophy to motivate formalizing a system of methods, tools, and connections between these (particularly mathematics as a tool and a "science"). That something more was a worldview in which there was an a priori reason to consider that the cosmos operated according to consistent laws. And while early Christianity completely rejected this, the later scholastics eventually turned "faith through reason" into proto-science. Hence the early modern statements like these:

1678 Court of Gentiles: Pt. IV iv. iii. 36 "Some of our Opponents resolve Gods certain prescience of sin into the infinitude of his science"
1700 N. Rowe "What makes Gods divine But Power and Science infinite"

Newton studied the bible more than he did mathematics and physics. Galileo's most controversial work in general wasn't what had him under house arrest, but challenging Aristotle's theory of motion. Galileo posited a clearly "crazy" theory of physics: the idea that, if one is moving in a ship (or car, or train, etc.) and throws a ball, the speed of the vehicle makes the ball travel faster. Why was it crazy? Because everybody knew Aristotle said that's not how motion works.

Fast-forward to the early 20th century, and Einstein dismisses QM as "spooky" clearly not physics. He was wrong. The history of science is littered with examples of the dismissed minority which turned out correct, as well as "absurd" ideas that were empirically verified.

Gravity isn't weird because we experience it every day. But psychic powers which enable invisible forces to move is clearly nonsense. I'm not arguing that telekinesis is real, of course, but rather to impress upon you a point: what seems completely rational to you would be dismissed as nonsense or magic to others of a different time.

QM is physics. Period. If it seems counter-intuitive, then we can look back in time and see a history of visceral reactions to such counter-intuitive claims that we now accept with ease.

For example thermal light can be calculated using the classical approach via schrodingers equations.

Theoretical and experimental justification for the Schrödinger equation - Wikipedia, the free encyclopedia

This implies something missing.[/quote]

 

[idav]

I know it's physics, that was my point. Measuring a photon may be counterintuitive but it isn't much different from the classical wave which can be measured classically. There is just something more to measuring a photon, I can accept that, I've known this since the 80's when they were saying light has a wave particle duality. The classical wave is just a wave which is the major difference. Light is both.

 

[idav]

The quantum particle of light is called a photon. Light has both a wave-like and a particle-like nature. In other words, light can appear to be made of photons (particles) in some experiments and light can act like waves in other experiments. The dynamics of classical electromagnetic waves are completely determined by Maxwell's equations, the classical description of electrodynamics. In the absence of sources, Maxwell's equations can be written as wave equations in the electric and magnetic field vectors. Maxwell's equations thus describe, among other things, the wave-like properties of light. When "classical" (coherent or thermal) light is incident on a photographic plate or CCD, the average number of "hits", "dots", or "clicks" per unit time that result is approximately proportional to the square of the electromagnetic fields of the light. By formal analogy, the wavefunction of a material particle can be used to find the probability density by taking its absolute-value squared. Unlike electromagnetic fields, quantum-mechanical wavefunctions are complex. (Often in the case of EM fields complex notation is used for convenience, but it is understood that in fact the fields are real. On the contrary, wavefunctions are genuinely complex.)

Theoretical and experimental justification for the Schrödinger equation - Wikipedia, the free encyclopedia

 

[LegionOnomaMoi]

I know it's physics, that was my point. Measuring a photon may be counterintuitive but it isn't much different from the classical wave which can be measured classically.

It is completely different. It's so different that an entirely new set of fields in physics had to be created and centuries of work became approximations. Pretending otherwise is just denial.