Because when people say realism they tend to mean a completely unique God's eye view of reality, which is a much stronger realism. Once you start to be able to view things in different perspectives then people start to use that as arguments against realism.

But then, what's your account of the 'unreasonable effectiveness of mathematics in the natural sciences' (Eugene Wigner). If they were purely tautological, how could they be exploited to discover things that otherwise would never have been known? The example I often give is Dirac's discovery of anti-particles, which was predicted solely on the basis of mathematics, with no empirical evidence forthcoming till much later. How could tautological statements yield genuinely new observations? Not to forget the many predictions arising from Einstein's theories that took decades to empirically validate ('Einstein Proved Right Again'). — Wayfarer

Maths is like writing. Its a language that describes structure. The unreasonable effectiveness of writing is not magic, its that you can invent words to represent anything in the world, anything in imagination you want. Its the same with math. Math is a gigantic field with many many different topics where you can describe many different facets of structure and use math to invent structures that nicely fit things you observe. Its not miraculous at all.

I genuinely have never understood why people find it miraculous that people can invent a model that makes predictions, some of which havn't been observed yet, and they turn out to be the case. I don't understand why people find that miraculous or interesting. I don't need a special explanation of why that sometimes happens. All that maths does is describe structure in terms iof quantities. You observe stuff in the world with a structure and you fit math to it. Its very simple. What about maths that works well is it is flexible and diverse so you can invent math that describes a huge number of things completely disparate.

. Both structures can be accounted for wholly and solely in terms of physical and chemical principles. — Wayfarer

So is a brain.

But even very rudimentary organisms already instantiate order on a different level to that of the physical. — Wayfarer

Complexity doesn't make something not physical.

living tissue is 'nothing but' physical matter, but that is highly contested and besides not in itself an empirical argument. — Wayfarer

Outrageous statement.

I see no reason to believe that it can be described in terms of, or limited to, physical principles, nor to describe the brain as a physical object. — Wayfarer

Yes, we can describe it in terms of things like statistical inference and machine learning, neither of which assume anything other than the idea that learning is embodied by physicla stuff: i.e. cells, biochemistry, fundamental physics, all of which there is some substantial understanding.

It is an embodied organ, embedded in a body, culture and environment — Wayfarer

Yup, no conflict here.

For the nature of mathematics, there is no reason to believe that this is grounded in or determined by any physical laws or relationships. — Wayfarer

All I am assuming is that a physical structure called a brain or perhaps, another kind of machine, can learn to do math purely in virtue of its physical structure and the kind of learning or inference it can perform as describable via statistical inference and machine learning.

This is a good point that made me think and want to look up about how smell works more, but couldn't you also say we can distinguish our emotional reactions about something from our ability to identify it through the senses? Albeit, maybe they interfere.

I think color has good points too. Nervous system structure will affect how we detect color, albeit one might still say all the distinctions we make map to physical events in the world. Maybe this needs to be unique to be realist? Rather than a brain that may be structured in a way to do the job more efficiently, resulting in possibly more convoluted or context-dependent ways of mapping to the world. There is also the sense that if one identifies blue colors more similarly than yellow or red or whatever then this seems also some added kind of detection of similarity purely a byproduct of the cone system - but then again, greens will also be more similar than reds because they are mapping to physical structures that are more similar. I guess would need to explore what is going on more though maybe. I could make an argument that maybe the ways different cones affect perception just can be seen similarly to how we might bound the same events in the world in different ways. But then maybe to some this is quintessential antirealism.
Maybe one could say perceptual differences reflect the fact we see different parts of the same reality, obtaining different partial information about actual physical events, but animals with different cones or more resolution of vision are just detecting more stuff or different stuff than we are. I guess this again is a very weak realism still.

Ultimately there always does seem to be some kind of arbitrariness somewhere. For me its a question of whether that still preserves reliable information, which albeit is always depends on whether you happen to live in an environment where your senses are reliable. In other contexts they may not be (i.e. light [in the correct ambient environment reflecting off objects] is not the only thing that can stimulate a retina).

I have said in other threads I think indirect vs direct realism is actually arguably kind of ambiguous. And I feel comfortable with some kind of minimal realism I think because maybe I think a fundamental metaphysical characterization of intrinsic reality is unintelligible. The best imo we can do is some weak floppy notion of structure, perhaps informational structure. While intrinsicness isn't accessible at all, and perhaps fundamentally meaningless, I think structure is accessible by us, even if in a convoluted or perhaps idealized or compressed way. The issue is that no creatures have access to all the kind of structure in reality that one might be able to plausible detect.

But maybe reality "as it is" is nothing more than the structure of reality "as it is" which we can access to some extent because we can all navigate the world correctly - but maybe my ability to find my way home was in fact some kind of accidental heuristic - we just don't have access to all of the structure. My thoughts are that phenomenal experience is just informational structure (or isomorphic to it if you want to be more precise) which itself maps to the world structure at least partially. When we see the world "not as it is" we have different mappings that miss stuff out.

These are not “things” in the physical world, but they constrain what can be true of that world - hence their designation 'laws'/ The very framework of physics, for example, depends on mathematical structures that don't exist materially. — Wayfarer

But mathematical structures are effectively tautologies so I don't see any reason for them to be meaningfully instantiated in some realm of their own or something like that where they magically affect the rest of reality.

The only fallback against that is to try and show that ideas are somehow identical with neural structures - as indeed D M Armstrong and other materialists insist. — Wayfarer

But there is overwhelming evidence that physical structures like brains are sufficient for all our reasoning, including mathematical. Why do you need to invoke anything else?

Well, probably unlike most, I put some stock in spiritual insight. The archetypal sage - whom is most likely not an actual person - has the ability to see 'how things truly are', which exceeds the scope of mere objectivity. Again from the entry on Pierre Hadot: — Wayfarer

What's your point?

which says that, were reason to be understandable purely in naturalistic terms, as an adaptation to the environment, then how could we have confidence in reason? — Wayfarer

I've always found this point quite strange because from what I see, people reason "badly" and get things wrong literally all the time, including scientists and academics. I feel like, even though we are very smart, human progress in terms of knowledge is not some direct consequence solely of our ability to reason but this long process of trial and error where people throw stuff at the wall and get it wrong all the time, and some stuff just sticks for one reason or another and we remember it over the generations. Its like a another form of Darwinian selectionism.

Apologies, very long reply. Again, I don't expect any replies to these kinds of thoughts because when I come to these replies I am just ending up writing down going through my rambling thought process about how to produce a coherent view for these things, which goes way beyond being a self-contained reply. I am not restricting these thoughts as I would if expecting and requiring a reply to them. I am just going through my thought process.

I believe that realism is more like an epistemic position rather than an ontic one. — boundless

But realism is more a claim that we can have knowledge of that 'mind-independent reality' and it's where things get murkier. — boundless

Yes, very true. Its totally reasonable to have the thought process that: there is a reality out there independent of us even when we are not looking; when we do look, what we see is dependent entirely on out biology had how that biology relates to the world in a specific and non-unique perspectival way (based on the physical interactions mediating the relation between things in the outside world and our brains). It is then fair to say how we see reality and information we gain that can be put to use is dependent on a perspective of a mind.

For me, its acknowledging this fact but also arguing that this exact same situation can be also equally viewed as a brain receiving genuine information about the outside world which depends entirely on whats going on out in the world. I think there is wiggle room in deciding what constitutes mind-independent knowledge, or at the very least knowledge that is in some sense real.

Again, the motivation is that it seems paradoxical to say that all knowledge is false or not real, yet our mastery of the world is very good. And I think taking anti-realism to its logical conclusion, it makes sense to me to say that all knowledge, whether scientific or just your everyday knowledge, is not real in that anti-realist sense. That logical conclusion is why I tend toward a total deflationary attitude toward all knowledge and epistemic activities. However, apart from the fact that deflating knowledge and epistemic behavior actually requires a tentative story of what is objectively happening with regard to minds and brains and consciousness, I think just for the sake of a coherent story, there should be a kind of compromise position accounting for the fact that while knowledge and epistemic activities don't support the most extreme, almost ridiculously naive realist position, the aforementioned paradox makes the notion of anti-realism a bit misleading. The deflation of the anti-realism vs. realism dichotomy itself is part of a solution; we might even say that one can only view the construct of "real" perspectivally in a way that requires adding your own assumptions about what constitutes "real" that are not straightforwardly unambiguous. At the same time, the paradox can only be fully resolved imo by a story about some "real" engagement with the world. Often our engagement is completely erroneous; at the same time, such error is not dichotomous but a continuous, fuzzy gradation.

If I am not mistaken, ontic structural realism is the position that, while we can't know the intrinsic properties of mind-independent reality, we can, at least in principle, know some structural aspects of it. — boundless

That I believe is actually epistemic structural realism. Ontic structural realism is that there is nothing more than structural properties to reality. I think an ontic structural realist might say there is some objective, uniquely describable set of structures. My view would be considerably weaker than that.

I think part of my view is changing the standard of what constitutes "real" metaphysics and "real" knowledge.

Some say that what is "real" metaphysically is a world of objectively, uniquely defined stuff that kur words and sentences map to. For me, its sufficient just that our words consistently map to stuff in a self-consistent way. I don't need some notion of objective boundaries in the world beyond my senses, just that if I see stuff or say stuff, it always maps to the same part of the world with the same relations to other parts of the world.

Because what is knowledge but the ability to predict what happens next? In that sense, knowledge is entirely about structure. The idea of intelligible intrinsic properties separate from structure then doesn't make sense - there is simply nothing to know about that kind of thing and it doesn't even logically make sense unless it led to some perceptible structural distinguishability. Indistinguishable intrinsic properties are meaningless.

I would say it is this kind of argument which could be used to attack the idea that we cannot knoe about the intrinsic nature of the world mind-independently - it doean't really make any sense the idea that there is anything to know or anything intelligible about that. Everything that is meaningful and makes a difference in the world is about structure that makes a difference - a boulder is meaningful because it has structural relationships to everything else in the world and makes a difference to how things around it behave, whilst other things evince its existence by enacting change upon the boulder; when you push it it moves. If intrinsic properties cannot make a difference structurally then not only are they meaningless but they give no reason for us to even speculate on them. A Bayesian might say we should update out beliefs only as much as required by the evidence. If no difference is made, there is nothing to update.

There is then the possibility of knowledge being mediated by different structures that produce exactly the same predictions counterfactually. Like how Newtonian physics can be formulated dynamically or in terms of least action or complex Hilbert spaces. But then if there is no way of distinguishing what different models say about the world then how do they make a difference to our knowledge? How do they make a difference to our mastery? They don't. It then doesn't make much of a difference that we describe the same thing in different ways; we are still making a consistent mapping to the same world. Our conscious experiences are just that - informational structure about what is going on in the world, albeit form a specific persepctive limited by specific physical interactions with a small contained part of the world - nonetheless, when not going haywire they map to the world in a way that in principle would be vindicated by habitual engagement with the world. Sure we can be wrong or incorrect about how we see our mapping to the world but this is not so interesting if it is possible in principle for us to be errorless (albeit one could also be a radical skeptic about errors).

Ultimately, though we are often wrong and models often do make considerably different predictions. Even in something like quantum interpretations. Different interpretations make the same empirical predictions but clearly they do not make the same predictions, fullstop. Many worlds predicts a completely different universe to Bohmian mechanics or relational mechanica structurally; its just that physics so far has hidden the means from us to actually distinguish those different ontologies.

We are wrong all the time. We make idealizations that are often wrong in some parts, albeit vindicate the important predictions we are intetested in in other aspects.

I think many realists are not interested in the possibility that our theories are mistaken. For many realists, I think it is sufficient that it is in principle possible to have a model or maybe sets of many models that predict everything one could predict about the structure of the universe correctly. They would then say that many theories are wrong now, but the fact that they predict things correctly means some of the structure is correct and that those predictions will get better over time.

At the same time, I think most anti-realists would say this is nothing more than empirical adequacy, or empirical structure, especially if one ditched the idea that "real" requires unique, objective deacriptions.
I guess here it makes salient that my views about the issues of realism in regard to indeterminacy are no different to any anti-realist, and I embrace that anti-realism because I believe all knowledge and epistemic activities can be deflated, albeit deflated under a kind of scientific or scientifically-amenable description of how exactly we perform those activities.

But my concern I guess is that the upholding of predictions via empirical adequacy requires a form of real engagement with the world, albeit one that can be mistaken.
So in that sense many theories are just plain wrong on some level or some aspects; if they make acceptable predictions in some places, that needs elucidating about how it captures empirical structure or if it does so only by luck or too thinly to be interesting.
I guess it could be vacuous when one considers that structure must be scaffolded on other structures and that they could be plausibly scaffolded on many different incompatible ones.
I suspect many attempts at explanations are like this. Flat out wrong.
It is then valid to be truly skeptical of scientific theories that could be flat out wrong. But I think something like classical mechanics actually makes too little metaphysical assumptions to be vacuous in this way. In some ways, Newtonian mechanics is actually just a thin description of behavior we see in empirical structure. Its not like saying that the earth is flat and then finding out it is rouns - which invokes considerable extra metaphysical, structure depth beyond what we see in empirical behavior. Quantum thwory may be the same as Newtonian mechanica in that regard.... but quantum interpretations isn't as it goes metaphysically deeper.

I guess under my view is the idea that maybe there can be nothing more to say about reality than what could be perceived or distinguished in empirical structures counterfactually; albeit ones that can be mistaken and do scaffold on each other in some sense.

Again, I think because of the complicatedness it becomes difficult to unambiguously define real and not real in regard to descriptions and theories that themselves can be deflated in terns of physical activities. Nonetheless there is maybe a fuzzy gradation between consistent mappings, engagements with the world and ones that are erroneous, or at least our predictions of our own knowledge are erroneous

But then, if we accept that 'mind-independent reality' is intelligible, we might ask ourselves how is that possible. — boundless

And what is even more interesting is that if we do accept that we can know (part of) the mind-independent reality it is because it shares something with our own mental categories. So, it would imply that, say, mathematical platonists are in some sense right to say that mathematical truths are mind-independent, eternal and so on. — boundless

From my perspective, all it requires is a mapping so it is sufficient for a physical reality were things behave in consistent ways that structure of some parts of reality can be mapped to the behaviors of other parts (e.g. like say a mirror reflecting the image of the structure of a room). So I disagree about platonism.

The empirical knowledge that science gives us is undeniable. But, in a sense, we can't 'prove' in any way that this means that we do know the structure of 'reality as it is'. — boundless

On the other hand, basically everything seems to tell us that we can know something about a mind-independent reality. On the other hand, however, there is no logical compelling argument that we can. — boundless

Yes, I think skepticism is always real and healthy. Knowledge, or rather, beliefs can be and often are outright wrong. They cannot be compelled to be correct. I think maybe this though goes into a question of agnosticism about theories rather than anti-realism. If it were anti-realism then it would be: even if our predictions were correct would it be not real? Now, I have said that I believe all epistemic activities can be deflated as complicated, even instrumentalist, constructions, maybe they all are a bit erroneous too. But again, I have shifted my standard for realism - I think if one views realism in terms of unique mappings to reality then ofcourse there are always many possible descriptions of the world and none of them would be real. That is valid. But I think it is also valid to say that if our epistemic constructs are all deflations and all we have to go on is whether our predictions and mastery are correct, then there is in some sense a realism to it because it reflects some real engagement which - via the purview of the free energy principle - would reflect some genuine statistical coupling between us and reality; at least that is the story, and all we have are stories. But then again, those models can be wrong, they can only be valid in a small part of reality and turn out incorrect when we come across a novel context.

I think my main contention is the status of the connection between pluralism and realism or anti-realism; we may choose to construct different tools for describing reality, nonetheless they are describing or pointing at the same thing given that they are used appropriately. I think this is inherently ambiguous. Maybe this inclines to borderline paradoxical aspects about the relation of theories to reality. Maybe we can deflate all realism and truth but nonetheless still use those words meaningfully in a deflated context. But maybe we should all be scientific agnosticists though justified in choosing theories we believe are either the best of a bad bunch now or that we believe most likely to not become erroneous in the future.

Aha, I actually removed the reference to your post on purpose because what originally was going to be just a reply to your post ended up as just complicated train of thought that I couldn't be bothered to edit into something more comprehensible. Sometimes I like just posting those trains of thoughts about some complicated topic as they come to me but I don't expect them to be easily comprehensible for others so I decided to not ask for a reply, as it were, because I would need to excessively edit and re-think the post in order to do so. But I think one of the last quotes of that post gives a summary of my perspective:

I think my perspective is similarish too Otavio Bueno's structural empiricism masquerading as a very weak ontic structural realism... so weak that they are interchangeable. This comes from my anti-realist inclinations to deflate things but the desire to acknowledge a mind-independent reality in a way that is not totally divorced from what we do and think — Apustimelogist

Link to structural empiricism
e.g.
https://scholar.google.co.uk/scholar?cluster=14237184630099891718&hl=en&as_sdt=0,5&as_vis=1

At the very least, when I create stories about the world - that may end up being erroneous - these stories are about real things, or supposed to be. That may be wrong.

I do uphold all of the anti-realist arguments against realism though regarding indeterminacy of our models - leading to the conclusion that models are nothing above how they are manifested or used within our own cognitive behaviors (e.g. a physics textbook is just squiggles on paper; it means nothing unless someone is engaging with it and comprehending it and then able to perform acts or calculations utilizing it - physical and cognitive events).

But then there is an interesting loophole in that this applies equally well to the use of words like "truth" or "real". They are indeterminate. They are manifest in use when I say things like "truth is what is the case; what is true is the case; that thing is true if and only if it is the case" and acts of identification of things "as the case", whether there is any deeper, profound, intrinsic, unique to that sense or reference - or not at all. And when the anti-realist says things are not real, that is equally indeterminate in the same sense which they use skeptical arguments to attack realism.

So then the question is: if truth and reality is in use, Why do I need to change my use of those words?

They work very well and coherently most of the time. I am sure almost every anti-realist maintains their use of those terms at least in some contexts within everyday life. I may want to change the use in some contexts, but not in any radical way - e.g. "the quantum wavefunction isn't real"; "I just discovered yesterday that ghosts are not real"; "Harry Potter isn't real".

Are scientific theories true?

Well, unlike some of the easier cases above I just gave, I think this is much more ambiguous and the answer appears different from different angles. Is Newtonian mechanics true as a unique metaphysical theory? No. Does Newtonian mechanics capture actual structure of the world in terms of empirical structure that satisfy predictions? I would say arguably yes. Is Newtonian physics as appears in those empirically verified events an idealization of fine-grained eventa? Arguably yes. Is that coarse-grained structure real though? Maybe, if you think of realism in terms of overlapping structure rather than unique objects. Are there many different formulations of classical physics with the same predictions? Yes. Are the differences non-trivial? Maybe, because some involve dynamical causality, some are about least action with fixed final boundary conditions, some describe classical mechanics as waves, some describe classical mechanics with complex numbers using a Hilbert space analogous to quantum theory.

We can say there are different ways of enacting the same empirical structures that we can distinguish non-uniquely through our sensory apparati but nonetheless map onto the world beyond those boundaries.

At least, that needs to be the story for my models about the universe to seem coherent.

I think there is a kind of Wittgensteinian aspect here in relation to his famous quote about "throwing away the ladder" or something like that. Once I have deflated all meaning to use, I can throw away the ladder and just use those words how I normally do, because my meaning of the words didn't depend on some kind of intrinsic magical ontology in the first place. There is nothing to be changed, just the acknowledgement of how meaning is nothing above use - and we cannot step outside of that as it were regardless of how much we try.

But in that use there is still some engagement with a real world, even if in a minimal sense. Something like active inference in the sense of Friston's free energy principle. To model the world entails a statistical coupling between the internal states of a system and some external states across its Markov Blanket from which there is a conditional independence. And from this perspective or story, any useful model implies some real meaningful interaction with the external world, even if in a limited, perspectival sense - even if I want to deflate representations, I could consider them in some sense real if there is a meaningful sense of a consistent statistical mapping (maybe only approximately, and idealized) in some context where information is being communicated between the external and internal states via the senses. This may not be unique, but then that would imply a very very thin kind of realism which is about empirical structure which anti-realists may not generally disagree with. Maybe they won't say what they see are real intrinsic objects of the world in a direct aquaintance; but maybe they would agree it is real information about the world that is consistent, albeit the plurality of ways we can extract consistent information is potentially huge.

Yes, I don't think a fundamental metaphysics of the universe is intelligible; but it doesn't mean the structures we perceive are not genuine information, even if there is always something a bit ineffable about that and we can artifically draw boundaries in various ways.

And there the tangent went, again!
But I think maybe that was describing how my position is kind of a structural empiricism masquerading as a weak ontic structural realism.... or rather maybe it is in fact a kind of very weak epistemic structural realism that I am espousing where I am embracing the kinds of trivialities you get from Newman's objection, whilst thinking about truth and realism and structure itself in this thin deflationary way related to use within a context, and perhaps ambiguous in a multi-scale reality within which our ability to engage with information is also intractable, convoluted (in the sense of a literal convolution), context-dependent and fuzzy (ambiguous even to ourselves often, think Quine jungle gavigai and Kripkenstein quus, but also very effective). Things can be the case in that their empirical consequences follow, counterfactually speaking. Things can also be the case even if "things" do not reflect rigid cookie-cutter boundaries. Even something like the self is like that - completely illusionary if we want to go down the road of deconstruction and deflation. Yet my holistic experiences must be an actual occuring structure of the universe (without implying anything deeper about what experience means - its just a thin word, a label).

I think I sit on the fence between preferring an epistemic or ontic structural view given that I would be more inclined to say that not just a notion of intrinsic fundamental reality is epistemically unintelligible but also that in some sense there is nothing more to know about reality than structure - an intrinsic aspect completely distinct and separate from structure seems kind of redundant, even nonsensical to me. The closest we could get to that is conscious experience. But the thing is that from my perspective, I am inclined to say that the only intelligible description of conscious experience is as informational structure. Like, without defining information in any kind of profound metaphysical or ontological way, it seems completely coherent to me to say that my visual experiences can be equated to some informational structure originating from my retina due to light hitting it. At the same time, the fact that all world structure we can engage with is contingent on a perspective is more in line with the epistemic view, even though I would say that this perspectival structure may have a consistent mapping to whatever is outside our sensory boundaries and so is veridical in some sense, albeit a weaker sense of veridical than someone who believes that perceptions have to capture some unique way the world purportedly is. Similar to as Banno sometimes says, I like conceptualizing our engagement with a real world in terms of kinds of "views from anywhere" as opposed to a kind of "view from nowhere" (god).

I think though that just because we do not have a unique, objective "god's eye view" though, doesn't mean that the information we obtained in any given perspective cannot reflect genuine information that consistently maps to the world with consistent relationships to other parts of the world. That is how perception works. And in some sense I think descriptions are second order - putting boundaries around "things" and giving them names is second order. If anything, pointing at and giving "things" names is more of a metacognitive description of my own perceptual abilities - I am describing or pointing out my own ability to make distinctions.

What is first order is perception itself, which reflects real information about the world directly in your perceptual experiences, without having to put a boundary around or label something. I don't need words to catch a ball, I don't necessarily need the concept of a ball to catch a ball; nonetheless, I am engaging with real information about and mapping to the world. It just happens to be information through a limited purview.

I would argue that a big reason why there are different purviews - that is not strictly about mind-dependence - is simply because we have physical structures that relate to other structures in the world in specific ways (e.g. the structure of our eyes and how they interact with light which interacts with other objects). Obviously though, even with physical similar capabilities, different animals may be better at abstracting patterns from sensory information. Is this a kind of mind-dependent confabulation or a detection of actual statistical structures in the world which are captured in our perceptual tools? Or some animals just cannot abstract certain kinds of relationships between percepts from the world.

Now, I would say that obviously, a brain dping statistics depends on the brain's capabilities, so it is mind-dependent in that sense. But, if there are some kind of criteria where one can evaluate through perception a correctness in detecting or mis-detecting higher-order patterns, and some kind of consistency in which one can react and behave and which they relate to other parts of the world, then surely these actually reflect some kind of meaningful structure out in the world that makes a difference, even if this structure is inherently fuzzy and perhaps convoluted.

What I would say is that we can make valid distinctions about the world which are meaningful in terms of consistent relational structures; but at the same time uphold a pluralism. There are many different ways we can partition the world statistically - infinitely many, perhaps. But surely, if one can engage with and distinguish these structures in a way that is consistent, then these cannot be arbitrary. The only reason I can make arbitrary boundaries around objects in my perception is because I can actually distinguish structural information about the world in my perception in order to draw those boundaries. And I don't think we even draw strong boundaries anyway really - the idea that we kind of have this rigid repertoire of concepts which we apply to the world is an idealization imo. Anyone tomorrow could invent an entirely new conceot or objects which is physically meaningful and catches on for one reason or another.

Yes, one could say conceptual pluralism is just anti-realism; at the same time, our engagement with the world is arguably real - or at least, if I were to tell a story about how we do that, it would have to be in some sense "real" - and putting into use these concepts may be enacted in some real engagement with the world, or the structure that comes at us directly in perception that maps to the world.

I guess question is about how apt a description being "mind-dependent" can be if it is clearly non-arbitrary. You could call the drawing of a boundary itself arbitrary, but if I can repeatedly identify this bounded "thing" in perception that comes up again and again due to information from the world at my sensory membranes, then is this really non-arbitrary?

If uniqueness of description is a criteria for realism then sure these things are very problematic. But why does uniqueness have to be a criteria for realism? You can say describe theories in multiple different ways or formulations which are deemed as equivalent in a way that people would just say its multiple descriptions of the same thing. Thats not to say that you couldn't argue there are non-trivial differences that distinguish them, but maybe there is no definite boundary of when different formulations are part of the same theory or different theories. Someone could arbitrarily decide formulations are different based on what others deem a small trivial difference. Like in the description of bounded objects then, perhaps the boundaries between saying "different" vs. "the same" are complicated graded structures of difference and similarity over different scales. Once there is this arbitrariness in saying some things are "different" or "the same", then its not clear there is a definitive way of sying whether a plurality of descriptions should be deemed indicativeness of real or non-realness. The whole issue might be deflated. "real" is a second-order description we apply to structures that we come across in percepts, make inferences about.

Yes, you could use indeterministic pluralities to say things are not real. Perfectly valid if you have your own criteria for saying something is real or not - i.e.it must be unique.

What is interesting to me though is why these arguments have little or no effect on realists often. It seems from reading that realists actually have moved their goalposts, based on anti-realist arguments like underdeterminacy and the potential prospect of things like scientific theories being wrong. But then you have to ask what it is that realists cling to - and I think it is this very fact that, regardless of the kinds of pluralities and indeterminacies and fuzziness, there is still this kind of non-arbitrary nature of our perceptions that map to some structure in the world and we engage with. You could say our descriptions are not real but it seems superficial when we can still engage with the world perfectly well. Obviously some descriptions are obviously wrong - which then occurs when our engagement with the world results in errors. But I would say errors is not so interesting to arguments on realism. Everyone can plausibly be wrong; anto-realists can probabilistocally be vindicated in the beliefs that there is more to learn, that theories are idealizations ans incomplete, that some past theories are outright wrong. Maybe some current theories too if we find new predictions. But then again, unless they make additional metaphysical claims someone can still claim a theory is still valid in some purview even if it does not capture the world uniquely - Classical physics is still widely used because it captures and describes relational structures we can engage with in the world without needing to make excess metaphysical assumptions. The question is whether it is impossible in principle to have a meaningful, consistent engagement with the external world, which we can do from various purviews without comong up with errors. If we have a plurality of ways of describing the world that don't come up with wrong predictions, does that mean none of these things capture real structure in the world?

But at the end of the day, from my perspective, yes this perceptual thing may be a bit too minimalist for realism from the perspective of anti-realists. But I think the issue of pluralities could come under attack for being weak by realists and at the same time if you are something like Van Fraasens form of anti-realism then this issue comes up for empirical perception not just theories - e.g. theory ladenness. But then, how can we engage with the world so well when even perceptual categories are idealizations and theory-dependent and even wrong.

I think the whole issue should be deflated maybe since the coarse distinction of real vs non-real doesn't adequately reflect the subtlety and nuance when it comes tothe balance between arbitrary boundaries and yet our very real, skillful engagement with the world regardless of such boundaries. Descriptions are red-herrings if we inflate them because really all there is is perceptual experiences in flux - we engage with structural information from the world in our senses constantly and we instrumentally, enactively use that information to guide our actions. Even a description may just be indicative of a metacognitive ability to make higher-order inferences about our own epistemic actions - the behaviors of thought and perception. We are extremely complicated machines that far outstrip literal words and descriptions in our epistemic activities and perceptual abilities. In some sense then all descriptions are "not real" because they can be deflated in this sense. Yet there is a mins-independent world and we engage with it to prosuce the useful behaviors that constitute knowledge, even when our abilities actually outstrip our descriptions. My ability for instance to distinguish different faces far outstrips any use of words I could make up to classify them - apart from the use of proper names which don't even have a description attached (i.e. they are effectively just pointing).

Maybe though the realist is just moving the goalposts for "intrinsicness" but then I guess the realist would also say that standards of intrinsicness that are too high are vacuous. We could plausibly say nothing is real because no descriptions, not even perceptions satisfy some ultimate criteria of intrinsicness (e.g. vision would have to be independent from any perspective, any brain, any intermediate physical process)

But then what is the outcome of saying nothing is real? The paradox that a world where nothing is real often seems perfectly coherent.

There is even a pragmatic limit to anti-realism in some sense.

But even then that depends on intuition since some people just don't have that intuition and they think some notion of anti-realism intuitive makes sense. The issue is we can gerrymander these borders of realism and anti-realism very easily without them having strong, consistent empirical consequences everyone can agree on like in everyday life or sciences or archaeology. And obviously this is graded. Maybe discussions about realism and anti-realism are just naturally inclined to kinds of contextual paradoxes almost - at least, if nothing is real, then "anti-realism" is also a false label. The problem is that usually when we sraw boundaries around objects, they aren't usually mutually inconsistent; but we have decided such for realism and anti-realism even though what is more apt may be some kind of gradation. Is "real" a label for a certain kind of abstract consistency in our experiences, in contrast to misapprehension? Its about intrinsicness? But maybe only kind of intrinsic things about fundamental reality that are intelligible are structures in some weak sense - and there is nothing more from the perspective of intelligibility or meaningfulness. But that is a significantly weaker sense of intrinsicness given the structure is weak and has a perspectival aspect. But neither would I say it is viewing the same thing from different ways (in a subjective sense). Rather we can view overlapping structures within reality on our sensory boundaries. Maybe those two statements can be seen as equivalent though.

Maybe what I have done is shifted something that an anti-realist would not view as intrinsic (i.e. perceptual, empirical observation - perceptual, empirical structure) and upgraded it to something that mediates intrinsic information.

I think my perspective is similarish too Otavio Bueno's structural empiricism masquerading as a very weak ontic structural realism... so weak that they are interchangeable. This comes from my anti-realist inclinations to deflate thingsbut the desire to acknowledge mind-independent reality in a way that is not divorced from what we do and think - realism should be in the story, at least for a person that wants to assert things about the world, or have theoretical preferences, even if they may be incorrect.

Actually the conspirational nature is not to be invoked here. One might still assume that our cognitive functions are useful, i.e. have a pragmatic goal. Practical usefulness does not lead to accuracy.
The problem I see here is that one can't claim knowledge about the 'mind-independent world' if one doesn't make some assumptions that can't be proven empirically.

I don't think that anyone believes that newtonian mechanics gives us a literal picture of the world nowadays. Still, it is still immensely useful and in a sense a source of valid knowledge, if knowledge is interpreted in a pragmatic way. — boundless

I think the disagreement is that what you are attacking is some kind of unique objective description of the universe (e.g. Newtonian mechanics, falsely speaking). However, from the beginning of the conversation, I have just been talking about information about the world we gain from perception or observation. And we may put boundaries around objects in perception in different ways if we really want to; but, nonetheless, what appears on our retinas and other sensory boundaries are patterns that map to events or structures out in the world, mostly in a consistent manner. And this kind of consistent mapping (at least in some restricted relevant context) I think is actually the minimal requirement for pragmatism and use.

I think it depends on the person. These views are perfectly mutually consistent although different people may be more or less inclined to enactivism; but the manner in which fep is structured really has some inherent enactivism. The fact that fep can apply to anything also is perfectly consistent with embodied and extended cognition, especially when you consider that the enactive nature of fep as instantiated by Markov blankets with active states can be nested - i.e. Markov Blankets within Markov blankets - cells, neural populations, brains, social systems, eco-niches, etc, etc.

And lack of a rational answer to that question makes me ask a different question instead. — noAxioms

But again, why assume that such a vague, abstract, distal question can be given a coherent answer? Doesn't make sense to me, and you're never going to be able to replace it with a surrogate question which is simultaneously near equivalent but less vague and detached from the current capabilities of our knowledge.

Take the debate between free active inference and embodies enactivist approaches in neuroscience. — Joshs

What is "free active inference"? Neither would I say these views are having a debate; they are perfectly mutually consistent. One thing I like about active inference is that emphasis on formality helps clarify, imo, some of these philosophical positions on biology that are often stated by other authors in ways which can comes across as vague or imprecise or over-inflated.

Overall, especially closer to the beginning, this was fairly good albeit I would say it saidsome of its claims very overconfidently imo.

But there is some nuance regarding descriptions like the following:

Alternatively, stochastic mechanics could abandon realism entirely for observables other than position, treating them as inherently probabilistic or emergent from the stochastic dynamics. This would align with quantum mechanics' rejection of pre-existing values but might undermine the "realist" motivation of stochastic mechanics as a classical-like interpretation. — sime

In essence, the KS theorem forces stochastic mechanics to compromise on realism to remain viable, aligning its reciprocal processes with quantum mechanics' contextual or non-realist nature while retaining a locally causal framework. — sime

Realism of Velocities:
If velocities (e.g., the current velocity ( v )) are considered real properties, they are assumed to have definite values at each point along the particle’s trajectory, consistent with the realist assumption that the particle has a well-defined position and motion. — sime

For example, attempting to define definite values for spin or momentum observables alongside position in a way that reproduces quantum predictions would lead to contextuality, contradicting non-contextual realism. — sime

The realism of particle configurations in stochastic mechanics has nothing to do with the realism related to the KS theorem. From the stochastic mechanical perspective, KS theorem and similar are about statistics. Particles can then always be in definite positions but their statistics respect the KS non-realism.

I actually used to refer to realism completely in this statistical sense until someone suggested to me it is more intuitive to talk about realism in QM interpretation in realistic fundamental ontology. So I started changing my use of the word more in line with this intuitive way of thinking about what realism is. Stochastic mechanics has realistic ontology of particles but is statistically non-realistic.

The A.I. is right though in line with what I said in my previous post that in stochastic mechanics, momentum and spin are defined statistically so they are not actual properties of individual particles but statistics that could only be related to lots of particles under repetition. You actually can have definite spin or momentum at points in space or for measurement results, just it doesn't apply to any specific particle and there may be a statistical spread over all results. The inclination to call these exact same statistics in other interpretations as "indefinite" in some scenarios (e.g. measurement results indicate equal superposition of spin up and down) comes from the assumption that all this stuff is talking about the property of a single particle. If you do that you cannot help but say that momentum and spin or momentum in these situations are indefinite. But under a statistical or stochastic interpretation, no claims are being made about a single particle so the idea of "indefiniteness" doesn't necessarily hold up except for in the kind of trivial notion that statistics have a spread, which is incontroversial and mundane. Conversely, particle ontology may always take definite positions, but their statistics can have a spread which is what in conventional interpretations would seemingly look like "indefinite" position. But again, this is not about a specific particle but statistics. When you take this into account, the A.I.'s claim that stochastic mechanical definite position is linked to KS realism is false; particle positions in stochastic mechanics can be indefinite in the statistical KS sense (i.e a statistical spread or uncertainty related to measurement interactions) while being definite ontologically for each particle.

Then there is also the following:

In stochastic mechanics, the current velocity v=ℏmIm(∇ψψ)v = \frac{\hbar}{m} \text{Im} \left( \frac{\nabla \psi}{\psi} \right)v = \frac{\hbar}{m} \text{Im} \left( \frac{\nabla \psi}{\psi} \right) depends on the wave function, which encodes global information about the quantum system. This raises the question of whether such a velocity, if real, implies non-locality. — sime

If ( v ) is a real property, this dependence suggests non-locality, as the velocity of one particle is instantaneously influenced by the state or position of another, without a local physical mechanism. This is analogous to the non-locality in Bohmian mechanics, where the velocity of a particle is guided by the non-local quantum potential or wave function — sime

But in entangled or multi-particle systems, the wave function’s global nature introduces non-local dependencies, even though the stochastic evolution of each particle’s position is locally governed. — sime

Yes, wavefunction and velocity fields are global and carry global information but they are not physical things. They are epistemic descriptions of statistics.

It is right Markovian stochastic mechanics is non-local as described in terms of instantaneous influences. But the non-Markovian reciprocal process version by Levy & Krener does not have this property at all, and explains it away as an artifact of an Markovian idealization. It then reproduces the correct behaviors without explicit Bohmian non-locality also responsible for configuration space descriptions where distant particles depend on each other.

You need to do better than that. — sime

I linked you the EPR paper and the whole centra idea in stochastic mechanics is that you can derive the Schrodinger equation which means that all predictions can be reproduced. The diffusion process in Levy & Krener has conservation laws identical to the Schrodinger equation.

they are shifting to reciprocal processes to provide a non-classical account of non-locality , as per the explanation provided by Chat-GPT , in the hope of explicating the presupposition of non-locality in stochastic models that is hard-encoded in the latter's reliance upon configuration space. — sime

I don't really understand where you are Chatgp is getting this which just suggests to me that this based on the aforementioned unreliability of these A.I. explanations.

What is interesting about Levy & Krener's theory is that it is based on realistic point particles but does not have the non-locality that Bohmian mechanics has, whilst fulfilling all predictions.

It is still Bell non-local which is different. It is widely believed that a kind of Bohmian non-locality is required to explain Bell non-locality in these kinds of models, but Levy & Krener direcrly refutes this thought while also explaining ehy Bohmian non-locality emerges in Markovian stochastic models.

Do you agree or not agree, that any theory dependent entirely on local causality cannot be a full explanation of QM? Secondly, how do you propose physically interpreting the use of time-symmetric reciprocal processes for guiding a collection of particles in a way that that is compatible with local realism? — sime

A full explanation of QM need to violate Bell inequalities whoch falsify non-contextual hidden variable models. But stochastic mechanics is not a non-contextual hidden variable model. The subtlety is that the fact that the model is contextual doesn't necessarily have to mean non-local causation, but such an explanation is vwry intuitive in a scenario like spin measurements.

I think that what sets apart stochastic mechanics from other interpretations of QM is that in all other interpretations, spin is a property of individual particles. In stochastic mechanics, spin is a property of particle statistics that only describe the behavior of particles counterfactuslly under infinite repetition of an experiment.

You can then imagine an infinite ensemble of particle trajectories between some initial preparation and final spin measurement. The final spin measurement just divides the infinite ensembles into pairs of sub-ensembles with different statistics. It can be shown that the spin statistics of these sub-ensembles would have to remain constant between initial preparation and final spin measurement which means that if you introduce another spin measurement on a different ensemble of particles, you could correlate their respective final measurements by allowing them to share an initial preparation which fixed a correlation between them.

For what its worth, I'm finding vanilla ChatGPT especially helpful with regards to navigating in a sourced way the nuances of the stochastic mechanics interpretation. As an outsider to the physics research community who nevertheless has a vested interest in understanding the mathematics and logic of a wide range of theories for purposes in relation to computing and category theory, I'm generally finding LLMs particularly useful for getting to grips quickly with unfamiliar theoretical ideas and for understanding the tone and the context of research papers, without which it can be difficult to understand what authors are selling versus what they are claiming - a very common problem indeed. — sime

Completely disagree. I mean, what your A.I. gave me about stochastic mechanics is false in a way that can be easily checked. And I have my own experiences of sometimes using A.I. to help find some kind of an answer and finding that they were misleading or incorrect when checking sources myself.

A.I. will give you explanations that are easy to understand; it doesn't mean they are always accurate, just like in this case!

For instance, I notice that certain physicists who are prominent members of the PhysicsForums.com were almost automatically dismissive of stochastic mechanics for the same obvious reasons that i opined earlier in this thread, but they also suspected that the authors selling stochastic mechanics were dishonest, doing pointless metaphysics, or failing to own up to the problem of entanglement. — sime

These physicists may not have reliable opinions either if they are not at all familiar with and actually understand stochastic mechanics and perhaps are even invested in their own preferred interpretation. Obviously I cannot actually evaluate their opinions if what they say isn't given.

whilst also stressing the fact that stochastic mechanics cannot be an explanation for non-locality for obvious Cohen-Specker reasons — sime

False.

the model assumes non-locality in the form of the configuration space upon which the model places a quantum diffusion - namely the space describing the joint positions of all of the particles that cannot be decoupled into independent diffusions satisfying local causality if non-local entanglement is to be describable by the model. — sime

The Levy & Krener paper doesn't have this and explains why it occurs in a model which uses an artificial Markovian assumption.

There is no mention let alone explanation of entanglement anywhere in that paper — sime

I just gave you a quote earlier that says that it is in accord with Bell inequalities but they actually do have a paper where they produce the Bell violations with stochastic mechanics:

https://scholar.google.co.uk/scholar?cluster=15973777865898642687&hl=en&as_sdt=0,5&as_ylo=2024&as_vis=1

Stochastic mechanics can reproduce all behavior of quantum mechanics.

Now, the above paper has a kind of non-local behavior, but the authors interpret it epistemically. And this is completely justified imo because the Levy-Krener paper shows that this non-locality is a byproduct of using an artificial, idealized Markovian assumption. Their model alsocan reprosuce all quantum behavior without any non-locality at all in the formalism by not using the idealized Markov assumption.

So your following statement is false:

for the non-locality of their background hypothesis. As it stands, it is a metaphysical interpretation of the Schrodinger equation that reproduces a fragment of the least problematic parts of Quantum Mechanics with deafening silence on the most critical aspects of QM that the interpretation either fails to address, or helps itself to by appealing to unstated non-local premises. — sime

Because stochastic mechanics reproduces all quantum behavior under assumptions about point particles that always are in definite places at any time and move along continuous paths. The fact you can do this suggests it is metaphysically possible; and stochastic mechanics gives a very simple physical explanation: energy conservation. All quantum behavior manifests from this. You don't need a non-local background to do this. No where is it implied that explicit non-locality is needed, only this energy conservation property; it follows that non-local entanglement behavior would follow regardless of exactly how you enforce this energy conservation, aslong as it is achieved. Non-local correlations are a consequence of initial local interactions where the resulting behavior does not dissipate but maintains its initial correlations.

is a non-earth shattering exercise in using stochastic differential equations to simulate whatever one wishes. — sime

You can't just do this though. These stochastic processes work under the same physical constraints people would normally assume are impossible in regards to producing quantum behavior. Its not an arbitrary formal reconstruction, it has strong physical interpretation implied directly by the formalism.

They were actually my own opinion in my own words, prompted by my understanding that the authors of the other paper you mention were reconstructing quantum diffusion out of time symmetric diffusions that is reminiscent of the symmetric casuality inherent in the transactional interpretation of QM. Personally I think that more modest paper is much more informative. — sime

Well its completely polar opposite to what the transactional interpretation.

What do you mean by "modest"?

A problem here, I believe, is that you are assuming that there must be some kind of correspondence of our mental constructs of the world and the world in itself. The structure of the model must somehow reflect the structure of the world. But how can we verify this assumption? — boundless

Well, my reply would be that if this were not the case, then it would suggest a picture of the world and metaphysics which is much more inflated than I currently believe, where there is some kind of conspiratorial aspect of nature that deceives our senses. Even though this could be the case, I don't see any positive evidence to believe this over a simpler story of how the world works and how we relate to it like the one that has been built up through physics, biochemistry, neuroscience, etc.

I cannot access the Levy & Krener (1996) paper — sime

Btw, you can see a pdf download link by just searching something like "Levy & Krener (1996)" in google. The pdf link comes from the author's website associated with their university.

However, their inability to fully capture quantum non-locality (e.g., entanglement) reinforces the uniqueness of quantum mechanics, prompting deeper inquiry into what makes quantum systems distinct. — sime

This is absolutely false and A.I. do not reliably give you information. Often when I research something, I will read the A.I. summary thing on google just as to give me some indication if what the answer might be, but I absolutely cannot trust this. I look at the sources it says it gets the information from every time and I find that it's relatively common that the A.I. will mix things up or confabulate ideas from various sources that actually don't make sense together, ending up in wrong or misleading answers.

I think the A.I. can be used as a research aide, but it cannot be trusted to reliably give you answers to things.

For instance, I can tell that the following phrase came from an A.I. :

that implement non-local aspects of the time-symmetric transactional interpretation of QM. — sime

Because the transactional interpretation has nothing to do with stochastic mechanics and is almost a polar opposite interpretation.

I have a suspicion that the authors you mention aren't intending to address foundational questions of QM ,and are instead focusing on the technicalities of constructing laws and diffusion models that cohere with the Schrodinger equation, with potential relevance to the subject of modelling quantum decoherence, by which classical diffusion can emerge in the limit of quantum diffusion, but without relevance as to the question of the nature and ontology of quantum states and quantum measurements. — sime

I am pretty sure this is not the case. For instance:

https://scholar.google.co.uk/scholar?cluster=856861870672922375&hl=en&as_sdt=0,5&as_vis=1

"Nelson’s stochastic mechanics formulation of quantum mechanics, which started it development with his article On the derivation of the Schrödinger equation from Newtonian mechanics in 1966 [6], is completely in accord with the requirements of Bells inequalities. By now, it has been developed to a mathematical rigor that completely parallels the formulation of classical analytical mechanics. It thus provides sufficient mathematical structure to suggest a clear physical picture of quantum phenomena."

cannot be explicated in terms of the local interactions of a regular stochastic process such as Brownian motion. — sime

Stochastic mechanics shows mathematically that entanglement follows from a non-dissipative / conservative diffusion.

Ignoring the issue is an option, sure. There are solutions (at least two), and some problems still have no solution, room for further study.
The goal isn't to 'know' how the universe works, but rather to find some valid ways that it might work. — noAxioms

I wouldn't say its necessarily ignoring an issue but expressing skepticism that anyone can sensibly tackle this topic without more context and knowledge about the universe. All the options I have heard are extremely speculative shots in the dark, sometimes bordering on nonsense (e.g. God did it). And again, I think this issue is not really a technical problem but one of subjective incredulity unless there is an actual physical contradiction here whoch is not just about rarity. Obviously we will just have to agree to disagree; from my subjectove perspective the issue is borderline close to "why is there anything at all?", just leading to my skepticism of a sensible answer.

I see what you mean, but IMO isn't enough to reject what I am saying. — boundless

I don't think so, because I didn't make any assumptions about labelling other than inadvertantly in order to convey the point about perspective. Regardless of whether or how you label things, there is an image there on your retina or other sensory epithelia that is in some sense can be said to be carrying information about the world. I don't think you need to stipulate boundaries for this to make sense, and in any case, I don't think people really view the world like this anyway. Sure people talk about objects like televisions and cars, but when push comes to shove I think the way people engage with reality is far more fluid and flexible than the idea that we uphold some fixed ontology with lists of well-defined objects. Now you can say this is kind of an anti-realism about objects, which couls be true to some extent, but its also kind of vacuous in a way because ultimately we are talking about different ways to effectively point at arguably veridical information about the world.

That is only true if the so-called background hypothesis, which is typically assumed to consist of a random field locally perturbing the motions of the particles, is assumed to have supplementary non-local Bohmian character as necessary to explain the statistics of quantum entanglement ... in which case your preferred interpretation becomes a variant of the Bohmian interpretation. — sime

If someone wants to call it a variation of Bohmian mechanics, I don't really see an inherent issue but you don't explicitly need Bohmian kind of non-locality for the theory to produce Bell violations. There is at least one version by Levy & Krener (1996) which is does not have Bohmian non-locality, produces all the correct predictons, and even explains that non-locality only comes when an artificially idealized assumption is used for constructing the theory.

quote="sime;983839"]The stochastic interpretation provides a conception of wave-particle duality with an uncertainty principle, but without additional Bohmian mechanicks it cannot explain Bells Theorem, for there is no getting around the fact that classical particle-field interactions that communicate slower than the speed of light cannot explain the 'action at a distance' of Bells inequalities; either the medium denoted by the background hypothesis is local, in which case we do not have quantum semantics, else the medium has non-local effects, in which case we have Bohmian mechanics.[/quote]

Well Levy & Krener's stochastic mechanics van in principle do so without explicit non-local communication.

I think the stochastic interpretation is pedagogically useful for providing a common-sense physical explanation for potentially classical aspects of complementarity that are often mistaken for inherently quantum phenomena, analogous to how Spekken's toy model of quantum mechanics is useful for providing common-sense epistemic intuition for understanding complementarity without assuming a physical account of the Schrodinger equation. But in neither case is there either a physical or epistemic explanation for entanglement. — sime

False. It can produce all quantum behavior and explains it in terms of a non-dissipative diffusion. The issue is that its depiction of QM is radically different from what people are used to so ita difficult for them to imagine. For instance, on stochastic mechanics, spin is a statistical property and doesn't belong to individual particles. An interesting possible consequence is that we no longer have to think of measurement in entanglement experiments as actively changing properties of individual particles.

But notice that embodied unstated realist assumptions about 'what the world is like'. And as Sabine Hossenfelder points out in Lost in Math, there's this tendency in today's physics to rationalise posits on the basis that they supposedly make intuitive sense and then to devise the mathematics to make them stand up. So given your realist predilections, then this approach seems natural to you. — Wayfarer

Very true, we all have different dispositions for intuition; albeit, I would say that the only reason these intuitions are open to us in physics is because of results like quantum theory. Consensus would have opted for a realistic interpretation had it been conceivable in the first place, so I would say a realistic interpretation should be preferable if available. There would be no QBism if a realistic interpretation has heen initially available to us.

And this is being borne out by experimental validation of 'Wigner's Friend'-type scenarios. — Wayfarer

Yes, but a stochastic interpretation has its own viee of Wigner's friend which isn't subjectively perspectival.

But it still relies on a hypothetical substrate — diffusing particles and a non-dissipative background — that isn't observable and must be posited as a metaphysical assumption (presumably subject to further investigation. — Wayfarer

Sure, but I would say it is arguably still better than many other interpretations given it provides an explanation for quantum behavior, it completely deflates the measurement problem and classical limit, it returns metaphysics to what is intuitive and commonsensical. I would say from a standpoint of rationality this is a preferable theory because arguably we shouldn't update our beliefs about the universe (or anything) any more than required given the evidence. And despite the background assumption I think this theory is clearly the least ontologically radical of any alternatives in the sense of diverging as little as possible from pre-quantum beliefs about how the world is, what everyday experience suggests the world is like, what other sciences suggest the world is like.

It seems almost completely conspiratorial to me that the world would have some bizarre, inscrutable ontology when mathematically there is a theory, maybe even multiple theories, that can produce quantum behavior through the kind of common-sensical, realistic, classical-like manner thag the world otherwise presents itself to us as.

And yes, obviously the background is a big assumption; but we already know - or at least quantum field theory tells us - about a kind of background because the vacuum is not empty: vacuum energy and fluctuations. We just don't know if that kind of background could be part of or indicative of the same kind of background as you would want in stochastic mechanics. But at the very least this isn't such a big leap as something like Many Worlds where you are assuming something else radical (and a bit ridiculous) exists that we can never observe and has no basis anywhere else in science. We know a kind of background exists (or you are at least justified in believing it based on quantum field theory):

https://en.wikipedia.org/wiki/Vacuum_energy

We just don't know if it could be the stochastic mechanical one.

It is an interpretation in the sense of Bohmian mechanics, in fact their underlying mathematical structure is very similar.

Entanglement and Bell inequalities are direct consequences of the non-dissipative diffusion like all other quantum behavior. The background hypothesis gives a conceivable way in which the non-dissipative diffusion could be physically realized, albeit without specific details. This is an interpretation explicitly intended to give a physical account of quantum theory.

Right. The theory accounts for the observed statistical patterns of quantum mechanics (similar to the Born rule), but it does so by modelling outcomes, not necessarily by explaining the underlying quantum structure. So it’s phenomenological in the scientific sense of being descriptive, not necessarily explanatory. — Wayfarer

As I said in my first post, it derives quantum theory from non-quantum assumptions about classical particles undergoing diffusion. It proves that in principle quantum mechanics can be instantiated in normal classical particles given the correct specifications which basically just amounts to conserving energy on average.

There are indeed others, but are there others that fall under methodological naturalism?

The problem is considered real in the scientific community, despite your expressed apathy on the subject. — noAxioms

Again, my point is that this issue is so abstract and we know comparatively little about thr universe works that I don't trust anyone's reliability in offering an explanation which is even close to correct.

At the same time, the problem is not an actual technical problem or one of errors in predictions. The problem is subjective personal incredulity which I don't really share because I have no inherent problem with the issue of very unlikely events occurring, especially in scenarios where we have no context to reliably and assess the issue, like with the question of "why is there anything at all?"

The selection of paths followed is clearly not random. Not asking, being pretty sure there is no answer (yet). For whatever the particle is or is not, the account for the diffraction pattern is MIA - and so far a great mystery. — tim wood

There is actually a mathematically rigorous theory called stochastic mechanics which shows that you can produce all quantum behavior from classical particles that undergo a non-dissipative diffusion.

https://en.wikipedia.org/wiki/Stochastic_quantum_mechanics

What is a non-dissipative diffusion? Well a dissipative diffusion is something like a pollen particle being pushed about seemingly randomly by H2O molecules in a glass of water. As the pollen bashes into H2O molecules, they impart a drag or frictional force on the pollen meaning it leaks or dissipates energy into the environment.

In a non-dissipative diffusion, energy does not dissipate in these interactions; any energy lost would be returned on average to the pollen by the H2O molecules. The system is frictionless in some sense.

What is really unique about stochastic mechanics is its the only interpretation / formulation I know of that actually derives quantum theory de novo from other assumptions. It then suggests that it is sufficient for quantum behavior to be executed by regular classical point particles so long as you enforce the absence of dissipation in its behavior so that energy is conserved on average.

But if you want a kind of common-sense realistic ontology, it probably means you want a model similar to the pollen floating in the glass of water. There are experiments which actually produce quantum-like behavior in this kind of way:

https://en.wikipedia.org/wiki/Hydrodynamic_quantum_analogs

What isn't mentioned in the article but you can find in the papers is that the reason the quantum-like behavior occurs is that vibrating the bath reduces viscous dissipation in the bath, 'viscous-ity' being related to friction due to interacting molecules in a fluid. Effectively, vibrating the bath replaces energy that is lost due to friction, and this makes the bouncing droplet (video at top of the page) exhibit behavior that looks quantum-like.

If hypothetically reality were to be like this then it implies particles are floating around in stuff. Very big assumption, but not so big considering the fact we know that space isn't really empty but filled with a vacuum energy and fluctuations. This provides a very plausible home for this mechanism and the fact particles are floating in a kind of fluid of stuff would give a mechanism for what you are talking about in the post, about how they seem to be guided along paths in a double slit experiment that appear as an interference pattern. Closing a slit then affects whats going on in the fluid which would then pass along to the particles move through it. You no longer have to think about a particle "interfering with itself".

So I think there is no reason to give up on a classical, realistic universe just yet. It is absolutely mathematically possible.



You see in the image trajectories from this approach produced by a mathematical simulation (far right image). Bohmian mechanics also uses classical particles but it effectively just takes the quantum wavefunction and puts deterministic trajectories on top - it doesn't explain anything about why quantum behavior occurs. In contrast, stochastic mechanics starts with a classical description of particles being pushed about like the pollen in a glass of water, and shows that under specific conditions related to energy conservation, as I previously described, all quantum behavior occurs for regular classical particles.

Yes, 'qualia' might well be about mind-dependent objects but they are certainly not mind-independent objects. — boundless

So, maybe, we are encountering an antinomy here: on the one hand, positing a mind-independent world seems necessary to make sense of our experineces. On the other hand, however, there is no epistemic guarantee that our cognitive faculties can step outside from our perspective and give us a non-mediated knowledge of the mind-independent world. So, it seems that we are stuck in an antinomy here.

So, I guess that the question is: can we really assume that we can make a description of a mind-independent world when we are 'inside' our own perspective and it is not obvious we can really step outside of it? — boundless

For me, i think one might be able to say that even though we view the universe from different perspectives, they arguably all procure information about the world that is still mind-independent. If I view a tree from one angle then another, then through a microscope or through infrared goggles, through the echolocation of a bat, through the chemoreception of an insect on the bark; all of these perspectives produce information that maps onto the world consistently due to the way the external world is. It just happens there is a plurality of ways one can engage with the world and extract consistent information about it.

There is an intelligible solution. Read the OP. — noAxioms

And there may be others but imo we don't know enough about the universe to give any a substantial construction. There are some cases with theories or ideas where I think there is good enough reason to put it out there and advocate it with reasoned arguments or evidence. For me, I don't think this is one of them. For this particular case I much prefer being conservative; albeit, it is also conforming to my intuition that I don't think unikely events inherently need an explanation. In some contexts maybe they do, but there is no context here for me to make that judgement imo.

Roll a 10000 dice. Any outcome that comes up is just as extremely unlikely as the next. So no, that's not the problem. The problem is that it came up 6's on all dice, first try. That is a problem. Not being bothered by it is the choice made by most, but that doesn't make it a problem not in need of solving if one wants a valid answer to 'why is reality like this?'. — noAxioms

But this is part of my point. Like you've started using analogies like this when it isn't really clear if this is even a fitting analogy because we just don't know enough.

And even then, I still don't think the analogy is necessarily saying much. The fact that 6's came on all dice came up first try doesn't necessarily warrant an explanation because its perfectly possible.

You can keep asking 'why, why, why... ' but these aren't interesting questions unless there is a kind of reasonable potentiality of an intelligible solution. You may ask why anything exists at all... clearly an example of a question where at least with what we know now does not have a reasonable, even conceovable solution.

I think some people are sometimes too eager to make everything fit into a neat box right this moment. To me, trying to answer these kinds of ultimate questions is just very premature.

Well, I believe that some properties we assign to 'external objects' are not mind-independent even in this sense. I am thinking about colours, sounds, smells etc in the way we percieve them. — boundless

I think they do capture mind-independent information though. When you see red, it is generally related to actual structure in the world that is being communicated to. Same with sound or smell, albeit there is probably a lot of nuance. And if toy think about it, all I see is color, or "shades" so in some ways I think color an't be any more remarkable subjectivity-wise than anything else we see. Its more difficult to articulate a deacription about color though, which I think may be part of why it often gets special attention philosophically as a kind of paradigmatic example of qualia.

I'm just noting that human biases tend to slap on the 'real' label to that which is perceived, and resists slapping that label on other things, making it dependent on that perception. — noAxioms

But surely this is nothing to do with the reality outside our heads which is mind-independent, and only about our beliefs about the world and how we should attach labels to it which trivially come under the umbrella of 'mind'. I think there must be a mind-independent reality but obviously any attempt to describe it necessarily is an act inside your head. At the same time, just because you cannot 'access' reality without a purview from inside your head doesn't mean you cannot navigate it accurately in principle; and I think the very consistency in how different people see the world implies mind-independence. I think there is a sense in which we are directly aquainted with information about the world in our perceptions even if it is inherently perspectival, just in the sense that the information you get from the world depends on your position within it and relation to it - but this is non-arbitrary because how you relate to the world is mediated by how the mind-independent world behaves (e.g. you see things because of the laws of physics and the physical structure of your nervous system and body). I do agree we can very easily disagree on what things are "real" - and there may not be a substantial definition - but I believe these discussions are usually so abstract they do not have much interesting implication.

I do subscribe to a perspective where you can basically deflate everything in regards to our minds and what minds are doing - beliefs, concepts, etc - which could be interpreted in terms of a kind og anti-realism. But what is left after such deflation? I don't think such deflation even makes sense without a mind-independent reality that scaffolds what is left post-deflation, and everyday experience and scientific observation tells me that it is there beyond experience. Again, we cannot describe mind-independent reality in a perspective-free way (non-threatening to the notion of mind-independent reality though) which leaves us in a kind of strange loop - there are inherent logical difficulties in self-description, self-reflexivity to the extent that I simply don't think minds ... or brains ... or whatever ... can do it, and we are simply stuck with accepting a limit to what we can talk about regarding fundamental reality, the nature of one's own consciousness, our descriptions and explanations (which can be deflated to acts). But this limit is about us and doesn't extend to reality itself as say someone who supports a relational quantum mechanics might think. Maybe what I am saying is still quite close to what some people are thinking about when they dismiss a mind-independent reality though; but I think describing it using that specific phrase would be misleading, personally.

Part of what has been learned is the incredible unlikelihood of our universe's fundamental constants being what they are. — noAxioms

I have personally never understood the fascination with this topic. I has never bothered me that extremely unlikely things can happen. I have never felt the need to explain it. I am not entirely sure reality warrants an explanations, a priori, and even thinking about a posteriori, I don't think we know enough about reality to be confident that we can narrow down a reasonably accurate explanation and not miss out on some plausible explanation that relies on information we simply don't have yet. However I look at it, it doesn't seem like an interesting topic to me - one that can wait and shouldn't be used to inform beliefs about metaphysics of the universe.

But the exact 'current' state of the moon is not in any way fact.

Bohmian mechanics takes that principle as a premise. Almost no other interpretation does. — noAxioms

For me, I advocate this kind of counterfactual definiteness so this segment I don't think is powerful to me.

It's hard to think it matters that most interpretations don't advocate counterfactual definiteness is you son't subscribe to them, ha.

:100: :up:

What bothers me, though, is that there is no reason to believe that consciousness cannot reoccur again. It already happened once – I’m conscious now. Why wouldn’t this phenomenon occur again? — Zebeden

But would the fact that it can happen again be any different from the fact that it has happened for both you as well as me (it has happened "again" spatially rather than temporally, as it were), only I cannot experience what you experience because wwr are two different individuals?

Also, what are the merits of eternal reincarnation. You and I probably have it pretty good right now - not just pretty good considering this isn't neolithic era or 14th century France or something like that. Then you look at how lukcy we are to be humans over other organisms.

Surely, it is only overwhelmingly likely to go downhill after this life? Or if not downhill, to another form of life one wouldn't necessarily want to live. I am not sure about life as an aphid. I guess they probably don't live very long anyway.

I think neither determinism nor randomness have anyrhing to do with free will; they are both equally conducive or non-conducive to it. The concept of free will in its prima facie conception makes absolutely no sense when you try to place it in the framework of any kind of scientific framework. The only kind of "free will" that makes sense is something that is kind of trivial like the psychological experience of agency or choice-making under regular psychological conditions.

The distribution of an unknown random number generator could equal anything. If an analyst knows that he doesn't know the rng, then why should he represent his credence with a uniform distribution? And why should the ignorance of the analyst be of interest when the important thing is determining the function of the unknown distribution? — sime

I feel like this kind of issue can still be talked about in the same kind of framework; for instance, Bayesian model selection where you are using Bayesian inference to select priors and models you want to use; and things like hyperpriors and hyperparameters.

Ever heard of imprecise probability? — sime

I don't think it rings a bell

It makes no sense to represent ignorance. To me that's a contradiction in terms. — sime

But surely, ignorance is directly related to probabilities. If an event has a probability of 1, you can predict it perfectly; if all the probabilities are equal, then its like maximal unpredictability.

for what does it mean to say that " Hypothesis A is inductively twice as probable as Hypothesis B when conditioning on an observation"? — sime

The probability that some hypothesis was the cause of your observation; and even if your prior is wrong, probability theory is the only logical way of changing probabilities when you see the evidence if you know the likelihood afaik.

The best way of expressing ignorance with regards to the likelihood of a possible outcome is simply to refrain from assigning a probability — sime

Then how are you supposed to update your ignorance when you encounter new evidence?