Amanda Gefter has an article at Nautilus a couple of you have asked me about: Reality Exists Without Observers? Boooo! The title is an accurate summary of her thesis. Gefter is responding to a book by Vlatko Vedral, where he reportedly argues for a reality that doesn’t require observers. In terms of quantum mechanics, Vedral is an Everettian, although he seems to downplay the many-worlds aspect, focusing on the physics. He touts what’s often taken to be an advantage of the interpretation over Copenhagen, that it doesn’t require any special role for the observer.
Gefter’s stance is that this can’t be done, that any attempt to do it inevitably sneaks the observer back in. She also implies that most discussions about Neils Bohr, Werner Heisenberg, and the other thinkers behind Copenhagen, strawman their positions. Since “the Copenhagen interpretation” is actually a diversity of views held by many early quantum scientists, she’s probably not entirely wrong. But arguing about what the Copenhagen interpretation is or isn’t strikes me as just as problematic.
Even if we restrict ourselves to just Bohr’s ideas, the man’s writings are infamously opaque. When trying to decipher his views, it seems possible to come away with a variety of ideas. When I’ve read about them, he comes across to me as an instrumentalist, although one who doesn’t want to admit it. Many science historians argue he was a Kantian, or maybe a neo-Kantian. Within the bounds of this discussion, I’m not sure how significant these distinctions are. What seems clear is he doesn’t take quantum theory to be telling us about reality, only our interactions with it.
When someone starts arguing for an observer centric reality, they can mean at least a couple of different things. One is epistemic, that our knowledge of reality only comes through observers. Bohr reportedly included non-conscious observers in this. But with that stipulation, the view doesn’t seem particularly radical. Another way of saying it is that all information must come through information gathering systems, which doesn’t sound nearly as profound.
But an observer centric reality can also be a stronger claim, an ontic one, that says conscious observers construct reality. The strongest claim of all is that reality doesn’t exist until we consciously observe it, that the observation itself brings it into existence. This is either idealist or solipsist territory.
It’s important to understand that even under the epistemic view, observation has effects on quantum systems. We can observe a supernova twelve million light years away without affecting it. Or we can observe the flow of a river without meaningfully affecting it. In these cases, there’s already enough interaction happening between the system and its environment that we can learn from the effects of those interactions that reach us.
However, we can’t observe a quantum system without interacting with it, and interaction entangles the observer and the observed, changing both. For the quantum system, that typically results in at least decoherence, and in many interpretations, all but one of the possible outcomes disappearing.
Which one of the above is Gefter claiming? I’m not sure, and that, unfortunately, is all too common when trying to parse arguments for this view. If forced to guess, I’d say she’s agnostic on the distinction, as this discussion about the moon seems to show.
Let’s put this moon thing to rest. It’s true. We can’t say the moon is there if no one’s observing it. Neither can we say that the moon’s not there if no one’s observing it. It’s not as if the sky is empty until someone gazes upward and a moon suddenly pops into existence. It’s that we can’t say anything about the moon as an independent object, because quantum theory doesn’t grant us independent objects, only measurements that we can slice into moons.
Not that Copenhagen is Gefter’s preferred interpretation. She actually favors QBism. Historically the name meant quantum Beysianism, in the sense that quantum theory provides degrees credence in various outcomes. It’s a view which focuses on the subjective experience of the experimenter. Or at least that’s how I understand it.
But again, the question becomes, is this just an epistemic view, or an ontic one? It seems to depend on which QBist you ask. On the one hand, it could be seen as a straight instrumentalist approach to quantum theory, a reification of the “Shut up and calculate!” attitude. Interestingly, the originator of that phrase, David Mermin (not Richard Feynman), signed on to QBism at some point.
Many QBist proponents resist the instrumentalist label. Which in turn often leads to accusations of solipsism, which they also resist. As I noted above, this starts to sound a lot like idealism to me, although it’s not clear that’s what they mean either. In the end, most physicists seem to regard QBism as an epistemic interpretation. (I’ve already done a post on why epistemic approaches don’t work for me.)
But what about the ontic view of an observer centric universe? If you already lean toward some form of ontological idealism, then this may well be a natural conclusion. But I don’t think there’s anything in the physics that drive it. A lot of this type of discussion seems to ignore the lessons from quantum computing, where engineers struggle with systems that decohere all the time, much earlier than they would prefer, and with nothing we’d normally call an “observer” driving it. (Unless we say the environment is the observer, but then “observer” seems to lose all distinctive meaning.)
If you think about it, this is no different from the classic double-slit experiment. If we put a polarizing filter at one of the slits, no conscious agent gets the information on which slit the particle goes through any sooner. What conscious agents do see is a change in the results on the back screen, and then infer what happened at the slits.
So the epistemic point about observers seems valid enough. I haven’t read Vedral at length, but I’d be surprised if he disagreed. But the ontic one doesn’t seem particularly well motivated, at least unless your metaphysics already push you in that direction.
But maybe I’m missing something? Is there an in-between ground between the options I listed? Or evidence for the ontic version I’m overlooking?
SelfAwarePatterns 
@selfawarepatterns.com
I just can’t see any of this as meaningful. The two-slit pattern doesn’t disappear when you look for the electron, it disappears when the light you shine hits it. It’s the light, not the act of observation that causes the wavefunction to change.
The only reason the “observer” is magic is sometimes the mechanism (decoherence or whatever) is subtle and hard to understand.
LikeLiked by 1 person
Remote Reply
Original Comment URL
Your Profile
Right. Observation of quantum entities is inseparable from interaction with those entities, and it’s interaction that leads to the changes.
LikeLike
@selfawarepatterns.com Yes, but the observation isn’t required. Therefore no observer is required either.
LikeLiked by 1 person
Remote Reply
Original Comment URL
Your Profile
I think you are missing something, or maybe I missed it becasue I didn’t do a deep reading of your post (if so, I apologize now).
This whole “observer influenced” notion of QM is mistaken. This lies in the fact that to observe something you have to get information from the system being observed to the observer. We have to interact with the thing in some way. To “see” something we need bounce light off the objects that gets into our eyes. So, what the people holding this mystical belief are doing is confusing an observation with an interaction. If we throw electrons or EMR at a quantum system in order to observe it, those things can definitely perturb the QM system.
Another confusion resides in “collapsing” wave functions. People talk about this as if it is a physical event (like Schrödinger’s dead-alive cat). Wave functions are descriptions. From them we can determine the probability of an event occurring. The cat of Schrödinger’s wasn’t 50% alive and 50% dead, the prediction was whether the cat were alive or dead was 50:50. When the box was opened and the cat observed directly, the prediction jumps to 100% but nothing collapsed, we just got additional information to make a better prediction.
LikeLiked by 2 people
Right. I think you might have missed some important points in the post. I agree completely with your statements about observation of a quantum system being inseparable from interacting with it.
I also agree with your skepticism on physical collapse. Although as usual I think your anti-real stance toward the wave function overlooks the interference effects, or how far quantum computing is able to go with the wave function. I’m sure that disagreement won’t go away anytime soon. 🙂
LikeLike
It seems to me that observing a quatum particle or interaction is analogous to poking a pool ball with a a pool cue at another ball on the table, knocking it into a pocket or somewhere else. You’re hitting the particle with an electron or photon. Doing that to a classical system or entity wouldn’t affect it, unless you observed it with a rifle bullet or somesuch. The explanations of the observer affect don’t have to be so mystic or generate a panoply of interpretations.
Great piece of writing! I didn’t understand what I wrote above until you explained it in your excellent essay.
LikeLiked by 2 people
Thanks! And that is the classic quantum quandary. We can’t “observe” without interacting. But that “observer effect,” in and of itself, is nothing paradigm shattering.
LikeLiked by 1 person
I’m idly wondering whether the epistemic/ontic distinction starts to break down when we talk about observer effects. You’re holding them at arms length, as if an observer-independent “ontic” view of reality needs to be preserved at all costs, keeping it apart from an observer-dependent “epistemic” view that requires, by implication, an ontic view of reality lurking in the background. But where the Copenhagen and QBist theories seem to be heading is toward the idea that the epistemic is the ontic.
This puts the interpretation that “[t]he strongest claim of all is that reality doesn’t exist until we consciously observe it, that the observation itself brings it into existence” in a new light. It’s not that the observation brings reality into existence, as if they were separate. The claim is rather that the reality that exists is not separate from the observation of its existence.
LikeLiked by 2 people
Right. I take the collapse of the distinction between the epistemic and ontic options to essentially be the ontic option. If there’s no distinction, then the claim seems like an ontic one. Unless I’m missing something?
My issue with this is it seems to throw most of science out the window, including a lot of very useful science. I’d be resolved to face up to that if there was compelling evidence or logic that led to it, but I don’t perceive that to be the case.
LikeLiked by 2 people
To suppose that the collapse must be into the ontic option is simply to make an assumption favouring the ontic over the epistemic. Others might prefer that we collapse the distinction into the epistemic. Thus we have realists and idealists. An idealist would point out that all claims are epistemic by nature. A supposedly “ontic” claim refers of necessity to some epistemic prior.
But this is just going back and forth over old territory. The questions concerning the observer effect that have finally come under serious consideration for our best science, as of about the 20th century, invite us to reconsider whether the ontic and the epistemic are somehow two sides of the same coin. I disagree that this requires us to throw most of our science out the window (although some of it may have to go, just as some pre-Gaililean science based on a stationary Earth had to go). It requires mainly that we reconsider everything under a new aspect. The compelling evidence and logic for a revisited science may be right under our noses, but we are still too much in the grip of the old epistemic/ontic distinction to really benefit from it.
LikeLiked by 1 person
On collapsing to the epistemic, I think it depends on the exact position. If someone ways there’s only the epistemic, then that itself strikes me as an ontic statement. On the other hand, saying all claims are epistemic, with no ontological commitments one way or the other, wouldn’t then be an ontic statement. But then, at the risk of sounding like a broken record, it seems like we’re at the epistemic option I described in the post.
The problem with paradigm shifts is they only seem to be evident once they’ve happened, or are at least well underway. Predictions of an upcoming shift in a particular direction are pretty numerous. But it doesn’t seem like most of them bear out. I’d agree there will be future paradigm shifts, but it seems to be the nature of such shifts to go in directions no one is currently expecting.
LikeLiked by 1 person
There is something ontological about every claim. Even a solipsist would have to claim that “solipsism” exists. A more interesting sense of the ontic is that it exists independently of the epistemic, and this is what I assume we mean most of the time. But certainly the epistimic may be viewed from an ontic perspective. This could lead to a pure dualism, where the ontic status of the epistemic is regarded as valid, but different somehow from the ontic status of that which exists independently of the epistemic; or it could lead to a dual-aspect approach, where there is noting independent of the epistemic, but the epistemic may be viewed ontically..
At the risk of playing my own broken record, this is the usual way of talking about the epistemic and the ontic: to take them apart, and then try to make them stick together again. The problem with paradigm shifts is indeed that they are not evident until they are well along, but that doesn’t mean absolutely no one is alert to them. If that were the case, they would never happen. What it does mean is that those who are alert to the possibilities&many, but not all, of which may prove fruitless— are uniformly ignored or dismissed, and that includes the visionaries along with the madmen. Both Kuhn and Feyerabend have noted the need early in a paradigm shift for faith in the new possibilities until the rest of the community is persuaded. This requires courage, if not bravado; but primarily it requires an abiding dissatisfaction with some of the answers offered by the current paradigm.
LikeLiked by 1 person
Isn’t it a matter of what “knowing” means?
What do we know of the moon aside from the “measurements” our senses and brains have taken of it?
LikeLiked by 1 person
We only know the moon through our perceptions and measurements. But the question is whether the moon is there before, between, and after those perceptions and measurements. I think the idea that it isn’t raises questions on how it’s there when we do perceive it. It strikes me as a much more complicated model of the world than if it is there between observations.
LikeLiked by 1 person
Our “measurements” could lead us to believe the moon exists before we see it. That doesn’t change the fact we only know it through our “measurements.” It is a question of the limits of what we can know.
LikeLiked by 1 person
True. But it pays to remember that all observation is theory laden. And I think it’s valid to ask which theories are more coherent and structurally complete.
LikeLike
I think “arguing about what the Copenhagen interpretation is” could be extremely useful if it amounted to proposing a clear and plausible theory, and showing how it resonates with remarks by Bohr or other relevant thinkers. I glanced at the “Booo!” article, but that didn’t seem to be what it was doing, so I stopped reading.
LikeLiked by 2 people
Based on what I’ve read, the idea of a Copenhagen interpretation was actually first discussed by Heisenberg, but’s it’s been noted many times that he and Bohr actually disagreed on many points. So it’s probably more productive to talk about Bohr’s interpretation, Heisenberg’s, etc. Or even better, just be clear about what type of interpretation we’re talking about, epistemic, ontic wave function and collapse, etc.
LikeLiked by 1 person
Interesting. I was just talking with AI this morning about the many worlds QM interpretation. With a few tries it eventually agreed with me that it’s essentially “a near infinite number of universes exist in our space and time that are mutually isolated from each other, that is except to explain the variability in measurement that occurs when they’re taken extremely precisely”. I need to state this correctly for a new post where I classify unfalsifiable proposals in science like this under the domain of philosophy rather than science. I’ll post at Substack, and hopefully sort it out before Christmas.
Anyway my own metaphysics begins from the presumption that a causal reality exists in itself, though we’re simply products of that causality. From here it’s effectively baked in that no observes are required for things to happen (that is unless one defines “observers” as causality itself, which is the same thing). My naturalism mandates that I take this stance even given that my own consciousness is all I can truly know exists. To me the converse seems short sighted or even arrogant.
LikeLiked by 2 people
I’d advise caution on using AI that way. LLMs will eventually tell you what you want to here, but it won’t necessarily be right. And citing them at this point is pretty much the opposite of authoritative.
And while the worlds are indeed currently not testable, the laws that lead to them have been tested, perhaps more than any other theory in science. It’s the assumptions we make to get rid of the other worlds which haven’t been successfully demonstrated. But we don’t know what we don’t know, so until we can test the other worlds, I’m agnostic on them.
My expectation is that reality is causal, or more fundamentally, structural and relational. So I’m onboard with your second paragraph. Although hit me with the right evidence and I’d reconsider that expectation.
LikeLiked by 2 people
I’m not breaking up with my new best friend that easily Mike! I now ask AI all sorts of things. You may find yourself at an informational disadvantage if you hold off. There are lots of great questions you could ask that have limited potential to provide biased results — like advice it gave me to fight the root rot on my property. It was already clear to me that Sam Harris invested in neuroscience credentials to potentially help peddle his general ideas as “neuroscience approved”, since I remember looking into him once. But what would AI say? It provided support for the thesis. And when I asked about a fellow “horseman of new atheism”? No, instead AI said Dan Dennett’s credentials grounded his positions. I expected that given his popularity. Anyway I don’t automatically believe everything AI says. In the context of generic search results regarding specific questions however, I do think it can be quite educational.
One theme of your post is that it can be difficult to learn about what people believe simply by reading what they say. Definitely! I once had to sift through many pages of Wikipedia and such in order to potentially grasp the positions of prominent people, and often with limited success. Now I seem to get reasonable and concise answers instantly that address what I ask specifically. And of course I’m a blogger so there’s certainly the opportunity for others to speak up if they think AI has given me a bum steer. Nothing’s written in stone here, and I always identify AI as such.
I see that you haven’t challenged my AI validated “many worlds” QM interpretation. I doubt Sean Carroll ever provided such a concise reduction! Again it reads “a near infinite number of universes exist in our space and time that are mutually isolated from each other, that is except to explain the variability in measurement that occurs when they’re taken extremely precisely”. And actually I guess I do see one problem. I shouldn’t have included “precise measurements” since theoretically quantum dynamics exist without observation.
Even if the MWI is true, my problem is that the exact same explanation seems able to validate anything that’s instead false. For example let’s say someone violently kills their wife, and perhaps observed by witnesses and video. Couldn’t they claim that they’re instead from a parallel universe where they didn’t kill her but somehow got transferred to this universe where a very similar person violently killed her? That would only be two worlds rather than a near infinite number, though I consider it essentially the same argument for each situation. Furthermore theoretically the many worlds would essentially be like ours and so have their own near infinite many worlds given their QM, and they’d have their’s, and so on for true infinite regress. This seems problematic too.
It makes me smile to see that certain physicists are using such a generic explanation to account for measurements that don’t make sense to them. This gets to the point of the posts I’m working on (and a point I think Tina mentioned as well). Essentially science can’t build science. Instead it needs better philosophical founding from which to straighten out these kinds of things.
LikeLiked by 1 person
Eric, I do use AI. It’s useful for some things. For example, I often have it summarize long articles so I can decide whether I want to read them. (It’s very useful for Aeon articles.) And it can be useful for asking specific questions. But I don’t trust it. I always follow its citations, if provided. Often they don’t bear out what it’s saying. I find that its answers often contain commonly repeated memes, whether they’re true or not. In that sense, it’s like asking a question on Reddit. You might get useful information, but don’t count on it.
“a near infinite number of universes exist in our space and time that are mutually isolated from each other, that is except to explain the variability in measurement that occurs when they’re taken extremely precisely”
Yeah, sorry, I didn’t respond because I didn’t feel like dissecting it. One issue is that the worlds don’t exist “in our space and time.” In Everett, space and time are themselves quantum, and so are also in a superposition of an ever increasing number of states. Another is that no one is using the worlds to explain measurement. They’re a consequence of accepting the mathematical structure of QM as complete, not a postulate. But everyone ignores this point, which is why I’m reluctant to bother repeating it.
As I noted to Tina, I think it’s dangerous for scientists to hew too closely to any philosophical principles. They are necessary, but should be worn loosely, as starting assumptions subject to revision. The history of science has too often been a story where philosophical hang ups delay progress.
LikeLiked by 1 person
Okay Mike, I’ve been talking with ChatGPT more today, and this time with the app itself rather than just through whatever Google search does for AI. So I essentially began with the reduction we were talking about before, and it objected in several ways that I think you’d appreciate. But in its long answer, there was one minor afterthought point about no “literal” worlds. So I said this:
“You might have solved my problem with the MWI interpretation of QM, and specifically when you objected to the distinction that there be any “literal” many worlds. To me it seems like this reduces the claim to an epistemic crutch rather than an ontological claim about what exists. If that’s the case then my objections to this interpretation no longer exist. Or if you think only a percentage of Everettians consider their interpretation to be epistemic rather than ontic, then what percentage would you say considers the MWI essentially from a perspective of “as if there were other worlds” versus actually believing that such worlds exist?”
It said almost no modern adherents believe that there are “actual worlds,” and essentially confirmed my longstanding claim that actual worlds would mandate “uber magic to the nth degree”. That’s essentially the main thing that’s concerned me about it. I don’t mind when people admit that they believe in magic. But I do mind when people even inadvertently let magic sneak into their perspective without acknowledging that magic. So unless you tell me that AI is wrong here (since, for example, Sean Carroll believes that these worlds actually do exist), then maybe I’ll be able to put this entire debate behind me. Well… not really. Even if I’m actually square with this epistemic interpretation, I still have a problem with its presentation. I’ll need proponents to directly say that this is a helpful crutch from which to grasp the evidence of how our world works, though in truth there are no other worlds from this perspective. So I responded again:
“That’s all great to hear! There’s at least one thing more I should ask you about however. If the MWI is essentially just an epistemic crutch or way to think about QM, then doesn’t this leave a “true answer” out there for science to still potentially discover? Indeed, it almost seems as if proponents might have downplayed the “epistemic crutch” element of the MWI in order to hide that this is not actually an explanation.”
Unfortunately I couldn’t figure out how to leave a link to just this, so I’m leaving it below for you to skim or perhaps ignore. I think you’d like it, though I’m less satisfied. MWI holds that there’s nothing further to discover than this epistemic crutch? That seems like an arrogant misstatement. There should of course be more to discover! It even tried to associate the MWI here with the work of Newton. He developed partial answers that were later more fully expanded with actual answers rather than positing magical things that he didn’t actually believe. This instead seems to be saying that there are no larger answers to potentially discover. I can’t go along with that, and especially given what’s currently stated. Language is just too important to continuously use in erroneous ways. It would seem that the “many worlds” interpretation should never be called that alone. Instead it should be called the “as if there were many worlds” QM interpretation. This all gets back to the point of my coming essay, or that science is in serious need of philosophical help.
So here’s that last response:
********
This is an excellent question, and it gets right to the heart of what counts as an explanation in fundamental physics. You’re not missing anything—this tension is real, and serious Everettians openly grapple with it. Let me answer carefully and honestly, without trying to “sell” the interpretation.
1. Does an epistemic reading leave a “true answer” still out there?
Yes — but not in the way people usually mean.
If “true answer” means:
a deeper dynamical law that explains why measurement outcomes are definite then Everettian QM explicitly denies that such a law is needed. The claim is not “we don’t know yet,” but rather:
there is nothing further to discover at that level.
This is analogous to:
• Newton discovering inertia and stopping
• Boltzmann explaining thermodynamics statistically and not seeking a micro-law of entropy increase
Everettians are saying:
👉 Definiteness is perspectival, not dynamical.
That’s a substantive claim about reality, not a placeholder.
2. Is MWI “not actually an explanation”?
This depends on what you require of an explanation.
What Everett explains
MWI explains:
• Why we never observe superpositions
• Why classical worlds are stable
• Why probabilities appear despite unitary evolution
• Why measurement outcomes are consistent
And it does so without:
• Adding collapse postulates
• Adding hidden variables
• Modifying dynamics
In this sense, it is a structural explanation.
What Everett refuses to explain
MWI refuses to answer:
“Why did this outcome happen instead of that one?”
But crucially, collapse theories don’t explain this either — they merely postulate it.
Everettians argue that asking this question presupposes something false:
that there must be a unique outcome at the fundamental level.
3. Is the “epistemic crutch” being downplayed?
This is where your suspicion is justified — historically.
Early rhetoric problem
• DeWitt-style presentations emphasized literal splitting
• This made MWI sound like a bold ontological theory
• It also made it sound explanatory in a familiar causal sense
That rhetoric stuck, even as the technical community moved away from it.
So yes:
Some proponents overstated the ontological claims in order to make the interpretation feel more concrete.
That has caused decades of confusion.
4. But epistemic ≠ instrumentalist
This is the crucial distinction.
Everettian QM is not saying:
“QM is just a tool; reality is hidden.”
Instead it says:
“Reality is fully described by the quantum state, but some features (like outcomes) are perspectival.”
That is closer to:
• Relativity of simultaneity
• Gibbs vs Boltzmann entropy
• Emergent temperature
No one thinks temperature being emergent means physics is incomplete.
5. Is there still room for a deeper theory?
Absolutely — but orthogonally, not underneath MWI.
Possible future discoveries could:
• Replace quantum mechanics itself
• Explain why the universal wavefunction exists
• Modify spacetime structure
• Reconcile QM with gravity
But they would not be:
“the real collapse mechanism we haven’t found yet”
Everettians see that as a category mistake.
6. A clean way to state the Everettian position
Here is a formulation that avoids hype and defensiveness:
“Everettian quantum mechanics does not propose a new physical mechanism for measurement. It claims that no such mechanism is required, and that the appearance of definite outcomes is an emergent, observer-relative feature of a fully unitary theory.”
That is an explanatory claim — just not a causal one in the classical sense.
7. Why your reaction is reasonable
Your intuition tracks this exactly:
• If “many worlds” is just a way of thinking…
• Then it feels like something is missing
• Because we are used to explanations that select outcomes
Everett’s radical move is to deny that outcome selection is a fundamental process at all.
Whether you find that satisfying is a philosophical judgment, not a scientific failure.
Bottom line
• Yes, Everett leaves no hidden “true answer” of the traditional kind
• No, this is not an evasion—it is a principled stopping point
• Yes, some proponents oversold the ontology
• No, that does not mean the interpretation lacks explanatory content
You’ve basically rediscovered, independently, the modern Everettian position and its main criticism.
*****
LikeLiked by 1 person
Eric, I think there are actually a few different positions here, which your conversation with the AI may be conflating.
The first is regarding words like “world” or “universe” as terms of art rather than a fundamental part of the theory. I do think the vast majority of Everettians think this way, including Sean Carroll. David Wallace talks in terms of an emergent multiverse. But it’s always possible to describe the theory without using “worlds”, as just a lot more going on in the universe we live in than we’re aware of. Hugh Everett, in his initial papers, didn’t talk in terms of worlds, although in the full version of his Universal Wave Function paper, he did use a metaphor of a splitting microbe for observers. This term of art aspect is something the vast majority of Everettians understand, but which most critics seem to have trouble grasping.
The second position is believing in the reality of quantum states but being non-committal about the worlds (even in its term of art understanding). That’s my position, which I described in a post earlier this year. There are a lot of Everettians who think this way, but without hard survey data, I’d be wary of assertions that it’s most.
The third position is being an instrumentalist Everettian. Stephan Hawking was in this camp. This group is non-committal about the reality of quantum states or the worlds. They just hold that the mathematical structure of quantum theory by itself is sufficient to fit the data. I have no idea how many Everettians are in this camp, but most don’t talk in ways that signal it.
The problem with using words like “epistemic” with all this is it signals a different kind of interpretation altogether, ones like Bohrian Copenhagen, RQM, or QBism, which are metaphysically antirealist toward the wave function. Although to Scott Aarsonson’s point (who I think is in the second group above), if you look at these interpretations long enough, all of them start to look like mental crutches for a reality that is otherwise incomprehensible. Aarsonson calls himself an Everettian in the sense of seeing it as the most productive crutch for thinking about what’s going on in quantum computing and similar phenomena.
LikeLike
In my view, whether someone is a Kantian or an idealist shouldn’t depend on the winning interpretation of our abstract mathematical models which only ever took into account a quantitative aspect of reality in order to make predictions about phenomena. Maybe different interpretations of quantum physics are suggestive and interesting, but none are all that forceful for me from a metaphysical perspective.
So, as an instrumentalist, I think it’s important to differentiate between philosophical epistemology/ontology and scientific epistemic/ontic uncertainty in quantum physics. But for realists, there’s only the scientific epistemic/ontic uncertainty to consider; the philosophical distinction has already been decided. If it’s true that quantum physics needs observers, I don’t see how scientific realists can do anything but bite the bullet and/or switch philosophical commitments. It doesn’t make sense to reinstate the philosophical distinction after making the prior philosophical commitment to scientific realism. (I’m not saying this is what you’re doing).
It sounds to me like “idealists” are really scientific realists who have decided the best interpretation leads to idealism—or maybe this should be called “quantum idealism”. And it also sounds like realists are desperate to uphold a deterministic picture of the universe that doesn’t seem to be supported by the evidence, and this according to their own standards for what counts as evidence. To evoke the many worlds interpretation…well, you know what I think of it.
LikeLiked by 2 people
I’d be surprised if, in practice, anyone is converted to idealism from quantum physics. I suspect for most people, which interpretations are acceptable, is driven by their prior metaphysical commitments. Reportedly some early quantum theorists were influenced by Eastern philosophies. Max Born once told Einstein in a letter that he preferred a reality that wasn’t deterministic. And of course Einstein was the opposite.
I’m not clear on the distinction between philosophical and scientific epistemology and ontology. To me, there’s just epistemology and ontology. It’s just that some of it the ontology is amenable to scientific investigation and some isn’t. But maybe I’m missing an important distinction here?
I think scientific realists have a few options. Determinists can go for pilot-wave, many-worlds, or superdeterminism. Indeterminists also have objective collapse and consciousness-causes-collapse type theories (although the latter seems to imply some form of mind-body dualism). And of course lots of people are realist for the rest of science while going instrumentalist for quantum physics.
Yes, I’m aware of your skepticism toward many-worlds. 🙂 But while I’m agnostic on the worlds themselves, I still think universal wave mechanics is the best option we currently have. But new evidence could change that any time.
LikeLiked by 1 person
Oh I totally agree, I think prior philosophical commitments have a massive influence in the way we interpret the science. What I meant is, I don’t think idealists (what I’m calling quantum idealists) make good arguments by invoking their preferred interpretation of quantum physics. It’s when they prop those interpretations up as the reason for their philosophical idealism, that’s what I’m objecting to.
What I mean by the scientific ontic/epistemic distinction concerns the status of entities and theories from within the scientific framework. For instance, questions might remain about particular aspects of even a well-established theory which would require making distinctions between what we know and what we don’t, what the theory currently actually accounts for and what has yet to be shown. When does a theoretical model or measurement cross over into the nebulous realm of scientific reality to be treated as on a par with tangible entities? Consider the way we talk about wavelengths, for instance.
At the same time, questions that aren’t scientific might need to be evaluated in philosophy or theology or art or literature or another arena. Questions of value, for instance, can’t be addressed from within the scientific framework. Questions about what makes a scientific theory scientific can’t be answered by doing science. Questions about fundamental reality, I think, are like these questions. They’re not directly addressed by science, but science takes some of them for granted in order to “shut up and calculate”.
The question of determinism vs. indeterminism is interesting. Can science answer this question for itself? If it must, it will answer, “determinism, please”! But will that work? That’s the way science normally addresses such issues. Does it work? Here, I don’t know.
—And of course lots of people are realist for the rest of science while going instrumentalist for quantum physics.
Yeah, I find this view incoherent, though, especially if the non-quantum levels operate on the framework of determinism. That seems like a view that’s literally on shaky grounds.
I’m about to read Marco Masi’s take on some of these issues in his new book. I’m just getting through descriptions of the quantum stuff, and so far I’m not utterly utterly baffled… which is what usually happens to me right before I bail. Normally I can’t stand to read about quantum anything. But this is his super duper dumbed down version for people like me. Anyway, I noticed a subtitle that looks like it might be of interest to you: “The myth of consciousness causing collapse”. I’ll let you know what he says if you’re interested.
LikeLiked by 2 people
“It’s when they prop those interpretations up as the reason for their philosophical idealism, that’s what I’m objecting to.”
Thanks for the clarification. Totally agreed.
On questions about fundamental reality that science takes for granted, certainly that does happen. It probably has to happen for science to be productive. But history seems to show that hewing too closely to any philosophical standard is dangerous. It cuts off options. I’m a mechanist, but I’m very aware that Newton had to bracket some aspects of mechanism to develop his theories. Eventually Laplace and Einstein were able to restore it, but Newton had to live without their insights.
On determinism or indeterminism, I don’t think science will ever be able to conclusively answer that question. Similar to Newton, Max Born had to set determinism aside to make progress. David Bohm and Hugh Everett offered options to reclaim it. But all scientific theories are provisional, subject to revision or revocation on new evidence. A determinist, faced with an indeterministic theory, can always hold out for a more fundamental deterministic one, and vice versa. I say all this as a determinist myself, not in the sense of wanting determinism to be true, but in expecting it.
But similar to your point about justifying idealism with QM, I think it’s a mistake to justify free will with indeterminism, which is often what the determinism vs indeterminism debate is really about. If I can blame the deterministic laws of physics for my decisions, I can also equally blame the indeterministic ones. For free will to be meaningful for social responsibility, it has to be based on a different standard.
I actually don’t have an issue with selective instrumentalism, if for no other reason than it’s hard to avoid. Backyard astronomers used Ptolemaic terminology because, for their specific domain, it remains useful, even though no one thinks it’s reality anymore. The most hardcore Everettian or Bohmian will also sometimes talk about quantum collapse when describing something like quantum computing operations, not because they think there are really collapses, but because it’s a useful shorthand. Evolutionary biologists will talk about species innovations, not because they think members of the species decided to grow a particular trait, but because it’s just a useful shorthand rather than tedious talk of mutations and selection.
As I’ve often said before, the real debate between realists and instrumentalists is about the scope of a particular theory. And scientists, such as Einstein, have often made progress with the realist assumption, that different theories should reconcile. Others, like Heisenberg, had to put on an instrumentalist hat to make a breakthrough. It’s interesting that the originator of the old celestial spheres theory, Eudoxus, came up with it with an instrumentalist mindset. To him the spheres were mathematical entities, and no more. It was Aristotle who reified them into physical objects.
On Marco Masi, sure. But QM is one of those topics where so many people bring agendas to it, that I’ve learned to get multiple viewpoints. It sucks, but it’s hard to know who to believe on QM without knowing at least a little about what the math says and doesn’t say.
LikeLike
—On questions about fundamental reality that science takes for granted, certainly that does happen. It probably has to happen for science to be productive.
Agreed. These aren’t even scientific questions that I’m thinking of.
—On determinism or indeterminism, I don’t think science will ever be able to conclusively answer that question.
Agreed again! We’re on a roll!
And yeah, indeterminism doesn’t help arguments for free will. I’ve never liked the way that debate frames things. Freedom to choose otherwise or freedom to break physical laws and freedom from our past, these just seem like a shallow notions of freedom to me. This is why I rarely talk about “free will” or “freedom”.
—I actually don’t have an issue with selective instrumentalism, if for no other reason than it’s hard to avoid. Backyard astronomers used Ptolemaic terminology because, for their specific domain, it remains useful, even though no one thinks it’s reality anymore.
It sounds like you mean something different by selective instrumentalism, something more like “using a theory that isn’t a description of reality for practical purposes”. I was thinking of the belief in scientific entities or theories as corresponding to or accurately describing reality.
—As I’ve often said before, the real debate between realists and instrumentalists is about the scope of a particular theory.
I’m not sure what this means.
Yeah QM, believe me, I take all that with a shaker full of salt.
Masi writes:
“We have observed that this uncertainty [Heisenberg uncertainty principle] is not a result of measurement errors; it is ontic rather than epistemic. This uncertainty is not related to the imprecision of our measurement devices. Instead, it is an inherent uncertainty that will persist regardless of the precision of our measurements.
Another common but misleading belief is that the interaction between the measurement device and the particle causes quantum indeterminism. When we observe microscopic objects such as through a microscope using light particles, we inevitably perturb and alter the system. This tiny yet unavoidable disturbance—known as the “observer effect”—changes reality only because we observe it. In this sense, reality can be seen as subjective. This interpretation has gained traction for a simple historical reason: It was Werner Heisenberg’s initial interpretation of his own uncertainty principle. However, we now know that it ultimately represents an overly simplistic understanding of what the uncertainty principle truly entails.
First, it’s important to clarify that in physics, the term “observer” refers to a measurement device, such as a photodiode, a CCD camera, or a photographic plate. The “observer” does not require consciousness or a mind to “observe”—that is, to perform a measurement. In physics, an “observer” is simply a mechanical apparatus that registers data without any need for human awareness of the result. Furthermore, if we consider the interaction between a measuring device and a quantum object to be an “observation”, should we then view every physical interaction between particles in the universe as a form of “observation”? For instance, should we regard a cosmic particle colliding with a rock on Mars an “observation” as well?
Second, today, modern electronic and quantum optical devices enable what are known as “interaction-free measurements”. This innovative approach allows researchers to test quantum effects, such as the uncertainty principle, without directly interacting with the system. [Here he has a footnote referring the reader to his more technical book on quantum physics]. However, uncertainty remains ineliminable. No hidden causes or “hidden variables” can account for it—a fact demonstrating that uncertainty is a fundamental and intrinsic (that is, ontic) aspect of quantum nature. It is not merely a consequence of a noisy observer effect.”
LikeLike
“It sounds like you mean something different by selective instrumentalism, something more like “using a theory that isn’t a description of reality for practical purposes”.”
I take that to be what instrumentalism is. Believing that the theory doesn’t correspond to reality but is a useful prediction framework. I’m wondering if you have a different definition?
On the scope of a theory, everyone agrees that a successful theory applies to the observations we can make. The question is whether it applies to observations we can’t make. An instrumentalist for any particular theory says the theory only applies to our current observations. A realist tends to think that the predictions have a broader use. In particular, they can be used to reconcile with other scientific theories.
The problem with only considering theories useful for current observations, is restricting it to that level throws out a lot of science. Statements about what’s happening at the center of the earth, or the sun, or how the earth formed or the continents moved, or about dinosaurs, are all based on applying theories beyond our immediate observations.
But we do know that sometimes theories break down beyond those immediate observations, as we see with the Ptolemaic cosmology. Once that happens, we usually consider the theory to have been superseded. But like Newton’s laws, which were superseded by general relativity and quantum mechanics, previously successful superseded laws remain useful instruments in a limited domain.
On Masi, thanks! I’m onboard with everything he says here. Particularly the point that quantum systems can be altered by interactions we don’t normally call “observations.”
I often wonder if a better name for the Heisenberg uncertainty principle would be the “Heisenberg exclusive effects principle.” (Although it would likely cause confusion with the Pauli exclusion principle.) The real issue is that some properties of quantum systems can only have effects on the environment to the extent a complementary effect doesn’t. Putting it that way, I think, makes it more clear we’re not just talking about a limitation on knowledge, but something more fundamental. And it makes sense when you remember that we’re dealing with waves, which when peaking at a specific location, don’t have a well defined frequency, and when they do have a well defined frequency they don’t have a well defined location.
LikeLike
Ah, I think I see what’s going on. I’m using the terms “realism” and “instrumentalism” to refer to the philosophical debate about whether scientific theories as a whole describe a mind-independent reality. I think you’re using the terms to describe stances toward the status of particular theories outside the context of the philosophical debate. For instance, it wouldn’t be incoherent for a philosophical scientific realist to be uncertain or agnostic about whether some particular theory accurately describes some underlying reality—I think this is how you’re using the word instrumentalism. People talk this way all the time, especially scientists, which makes things confusing. Conversely, it would be incoherent for an instrumentalist in the philosophical sense to wonder whether some specific theory describes an underlying reality, since that would contradict their position, though they can still say some particular theory is bad by pragmatic standards.
So going back to what we were talking about before, presumably someone arguing for scientific realism in the philosophical debate would have to express what they’re being realist about, what’s essentially veridical in scientific theories in general, and they’d have to do so in a way that’s principled, not cherry picking after the fact. This is part of why I think it’s incoherent to be a realist about scientific theories in general, but then make an exception when it comes to quantum physics. Or at least I would want to hear why that’s not cherry picking. Toss in determinism and/or reductionism, and I get really confused. I would want to hear why the quantum level should be excluded, based on the same principles for what makes theories in general veridical.
This is what I mean by making a distinction between philosophical epistemology/ontology and scientific. Maybe this distinction is clearer now?
Glad you liked Masi in that section! Not sure you’d like what follows. 🙂
LikeLike
Maybe the way to say this is I was talking in terms of philosophy of science and you were talking in terms of metaphysics? Anyway, I appreciate the explanation.
Interestingly, I agree with you on QM more than you might think. The issue is many people are not just instrumentalist toward QM, but toward the wave function in particular. But given various theorems that are out there, you can’t be antireal toward quantum states and believe that anything physical at all is happening before a measurement. I don’t think most people understand just what kind of positions they’re taking there. In essence, they’re taking a metaphysical antireal stance toward the wave function while taking metaphysical realism toward everything else. Jim Baggott is very careful in his book to limit the scope of his antirealism.
But allowing any form of realism with the wave function means having to choose between various disconcerting options: physical wave function collapse, pilot-waves, retro-causation, or many-worlds, and maybe some other strange options. I think people blink and just accept a little metaphysical anti-realism instead. I agree that it’s not a coherent position. If you don’t want to accept the reality of the wave function, you should just bite the bullet and go full metaphysical antirealism.
“Glad you liked Masi in that section! Not sure you’d like what follows.”
Ha! Probably not. I don’t remember his exact positions, but I do remember disagreeing with them. 🙂
LikeLike
Have you already considered RQM (relational quantum mechanics? Unlike QBists, who view reality as existing only insofar as it is experienced by agents, RQM holds that it is an web of relations that exists independently of the experience of individual agents.
One strength of the RQM is that, since it rejects the idea of a ‘universal wavefunction’ with a single, absolute state of reality, it avoids the need to multiply reality into an infinite set of branching worlds in order to preserve the universal objectivity of the wavefunction, as the Many-Worlds Interpretation does.
LikeLiked by 1 person
I have considered RQM. I did a post on it a while back when Carlos Rovelli published his book: https://selfawarepatterns.com/2021/05/31/carlo-rovellis-helgoland/
He does manage to banish the worlds, but with quantum physics, there’s always a price in terms of weirdness. His is to posit that reality only existing during interactions, not before, between, or afterward. My issue with that move is then how does reality during the interactions come about? This approach simply rejects that as a meaningful question.
The wave function at least provides a predictive explanation for the origins and evolution of the reality in the interactions. Of course, once we allow for any realism in that regard, we’re faced with having to choose between physical wave function collapse, pilot-wave, retro-causation, many-worlds, or maybe even stranger options. Quantum mechanics will not allow us to pass by with our worldview unscathed.
LikeLike
I don’t really get the objection. The world is just a bunch of parts that interact with each other. That’s what makes up reality; if it weren’t like that, there wouldn’t be anything at all. And you’d even more so have to ask how reality emerges from a hypothetical wave function. Plus, you’d have to ask what we can even understand by wave function. In RQM, the wave function is nothing more than a mathematical tool for calculating transition probabilities. What we observe in reality are the values of physical variables at discrete interactions, whereby the value a variable takes is only relative to the system affected by the interaction. The occurrence of these interactions is a fact of the world; one cannot keep asking what lies behind them without entering into an infinite regress. One could also ask, how does gravity come about. In any case, RQM has the advantage of proposing a theory about the standard world of our experience that does not require such elusive things as hidden variables, many worlds, physical collapse mechanisms, or a special role for mind, agents, or similar.
LikeLike
The issue for me is one of coherence and structural completeness. If reality only exists during the interaction and then only for the interacting parties, then we end up with solipsistic entities, with other entities only coming into existence during interactions, with no explanation for their state or properties during that interaction. Rovelli, to my mind, skates past this implication far too quickly in his writing, and it seems like most RQM enthusiasts either don’t understand or want to confront it.
All I can say is it remains a major block for me.
LikeLike
“To me, if we don’t have a mechanism, we don’t have an explanation yet. That’s what I see missing between the interactions.”
This is like criticising classical mechanics because it it does not provide an explanation for why a system follows its equations of motion.
“If reality only exists during the interaction and then only for the interacting parties, then we end up with solipsistic entities, with other entities only coming into existence during interactions, with no explanation for their state or properties during that interaction.”
RQM does not rely on solipsistic entities; quite the opposite, for RQM, the state of a physical system is the network of relationships it has with surrounding systems. The physical structure of the world is constituted by this web of relationships.
Rovelli argues that we do not need a wave function in quantum mechanics because the essential content of the theory is about the values of physical variables and their probabilistic correlations, not about an underlying “wave” representing a physical entity. The wave function, or quantum state, is a calculational tool that encodes information about potential measurement outcomes and their probabilities, rather than representing a real physical wave in space.
Quantum mechanics is thus not a theory of the dynamics of a mysterious ψ entity, from which mysteriously the world of our experience emerges. It is a theory of the possible values that conventional physical variables take at interactions, and the transition probabilities that determine which values are likely to be realized, given that others are. It doesn’t add anything to the predictive power of the theory to introduce the notion of ‘wave function’, which involves the general notion of ‘quantum state’, from which the world of our experience emerges through some involved and obscure argument.
RQM does not suggest that entities only come into existence during interactions. They are considered real and existent, but their attributes are not intrinsic or absolute; instead, they are regarded as carriers of dispositions to exhibit certain properties upon interaction. These properties are not intrinsic attributes existing independently but are actualized relationally when a measurement or interaction occurs.
Thus, quantum theory tells us that the description of the way distinct physical systems affect each other when they interact (and not the way physical systems ‘are’) exhausts all that can be said about the physical world.
LikeLike
“This is like criticising classical mechanics because it it does not provide an explanation for why a system follows its equations of motion.”
I’m not seeing how this follows from my position. I’m not saying I need the ultimate answer. As I noted above, I’d even be prepared to accept neo-Copenhagen views like RQM as a placeholder, if we didn’t already have a mechanistic explanation available. As far as I can see, its only drawback is people find its implications disturbing, which to me is irrelevant for a theory of physics.
“RQM does not suggest that entities only come into existence during interactions.”
I’ll just repost Rovelli’s own description from his SEP article.
On whether the wave function is just a calculation tool that doesn’t model reality, a couple of issues I have. The first is the PBR Theorem, which demonstrates that if there is any physics happening whatsoever between measurements, then the states modeled in the wave function must be real at some level of description. The only way to avoid this is to take an explicit metaphysically anti-real stance toward anything between measurements. My view is if someone is going to do that, for consistency, they should just bite the bullet and admit idealism or something along those lines.
The second issue is the success of quantum computing, not to the extent of being commercially viable, but just to the extent of validating the theory. If there is no reality between measurements, then the success of quantum computing seems like a miracle. Which is really just the old argument for realism, that if a theory is highly successful in terms of predictive accuracy, saying it doesn’t describe reality to at least some approximation amounts to a miracle. A miracle is an extraordinary claim, for which, before buying, we should hold out for extraordinary evidence.
In the end, it comes down to what we require of an explanation. The realist stance encourages new experimentation and technological development. It’s not clear to me that the anti-real stance does that, or at least not nearly as well.
LikeLike
In the given quote, Rovelli says that variables only take on a value at the time of interaction, which gives rise to a sparse event (or ‘flash’) ontology. He gave a more detailed elaboration of this in his paper “Space is blue and birds fly through it”,
As he explains in the text, RQM accepts that there is an objective reality composed of systems and their interactions. Properties are real in the sense that they pertain to physical systems and events, which are real in the context of specific interactions and relations.
What RQM denies is the assumption “that it is in principle possible to list all the features of the world, all the values of all variables describing it at some fundamental level, at each moment of continuous time, as is the case in classical mechanics.“
Thus, RQM rejects the notion that all system properties have definite values at all times, independently of interactions. Instead, it asserts that properties only assume definite values when they interact with other systems. This renders the concept of an absolute, observer-independent state or set of properties baseless.
Regarding the challenge to its sparse ontology, he writes: ” The question of “what happens between quantum events” is meangless in the theory. The happening of the world is a very fine-graned but discrete swarming of quantum events, not the permanence of entities that have well defined properties at each moment of a continuous time.”
This allows RQM to circumvent the PBR theorem: ” Such theorems […] essentially say that if there is a hidden variable theory, the hidden variables must contain at least the entire information which is in the quantum state. But the assumption is explicitly denied in relational QM: properties do not exist at all times: they are properties of events and the events happen at interactions.“
You will find detailed evidence in the paper “Relational Quantum Mechanics and the PBR Theorem: A Peaceful Coexistence” https://doi.org/10.1007/s10701-021-00485-2
LikeLike
The PBR theorem assumes metaphysical realism of some type for physics between measurements / interactions. As I noted above, we can evade it by assuming metaphysical anti-realism for that physics, that there’s no physical reality there at all. Skimming the paper, I see it confirming that that’s exactly what RQM is doing. It uses different wording, but that’s the end result. It’s similar to the classic stance of the Copenhagen people (which Rovelli has acknowledged), that there is no reality of the quantum system there before the measurement. Rovelli’s innovation is to make it relative, but it’s still an anti-real stance.
To me, in both Copenhagen and its descendants, it’s an incoherent assumption to save appearances. But hey, if it works for you, then it works. Hopefully there will be data someday to resolve this one way or the other.
LikeLike
Sorry for stepping in here, but I have a thought:
What if at a given time there is a physical reality, so, states x, y, and z, but there is no way to measure these. Instead we can measure a, b, and c, which are determined by x, y, and z, but not equivalent to same. A, b, and c don’t “exist” until we measure them.
Does this change anything?
LikeLiked by 1 person
It does, but that seems to be re-introducing something RQM denies, that there is any ontology outside of the interactions (a,b,c). If we do that, then we’re back to needing a theory of that ontology (x,y,z). And of course that’s what quantum states are about.
LikeLike
“His is to posit that reality only existing during interactions, not before, between, or afterward. My issue with that move is then how does reality during the interactions come about? This approach simply rejects that as a meaningful question.”
Like many people Mike, the fundamental concept of RQM goes over your head. Not that it’s your fault per se, I don’t think Rovelli did a thorough enough job of explaining it.
What you do not understand about RQM is that there is “never a time” when interactions are not taking place; and because of that underlying feature, we have a reality that always exists. Primal interactions are constituted by the “values” of the four forces of nature, gravity being the most fundamental followed by electromagnetism and the nuclear forces both strong and weak. From this grounding architecture of forces, more complex structures occur through interactions with other structures that take on the “values” of the individual constituents by means of the whole.
Essentially a Kantian Whole: An organized, self-constructing system that the parts exist for by means of the whole. There’s no smoke and mirrors with RQM, it’s pretty straight forward once you get this concept.
LikeLiked by 1 person
Hey Lee. I agree that Rovelli doesn’t do a great job of explaining all the aspects of his interpretation, but I think we disagree on where he’s falling down.
“What you do not understand about RQM is that there is “never a time” when interactions are not taking place; and because of that underlying feature, we have a reality that always exists.”
My understanding is that an interaction only leads to existence for the interacting entities. For other entities, there is no existence yet. This is crucial, because if any interaction leads to existence for all entities, then we’re right back in having to deal with real quantum states, and choosing between wave function collapses, hidden variables, retro-causation, or many-worlds.
We have to choose our weirdness. We can accept a sparse flash ontology, but without an explanation of what’s in it during the interactions. Or deal with consequences of one of the other realist interpretations.
If we just want to be Kantian, it seems like we could just retreat to the “Shut up and calculate” stance and consider the underlying reality forever unknowable. QBism seems a better fit for this, at least to me.
LikeLike
“My understanding is that an interaction only leads to existence for the interacting entities. For other entities, there is no existence yet.”
That might be your interpretation Mike, but that is not what Rovelli is saying. You need to do a little more research. You might want to try a simple google search like the following:
What does relational quantum mechanics say about entities?
LikeLike
Lee, these are from Rovelli’s SEP article on RQM. https://plato.stanford.edu/entries/qm-relational/
…and later…
Am I misinterpreting these statements? If so, where would you say I’m going wrong?
LikeLike
It’s difficult to parse RQM based on these sections alone. I’ve read this SEP article in the past and just now skimmed over it again. The only advice I can offer is to remember that meaning is context dependent, and I think you like most people are trying to understand RQM in the context of reductionism. And that’s the whole point of RQM, it doesn’t conform to the model of reductionism; it’s a completely different paradigm grounded in relationships where there is not one thing that exists independently of anything else.
This particular quote points out the problem with reductionism as Rovelli sees it: “there are questions that make sense in classical mechanics, but have no significance in nature.” For instance, the question “What is the y-component of the spin of an electron when its z-component is 12ℏ?” is meaningless: neither it admits an operational definition, nor is it needed for a realistic understanding of nature.”
Rovelli doesn’t see nature as mechanical and neither do I.
LikeLike
I’ve read the article in the past myself a couple of times, but not since he revised it earlier this year.
On reductionism and mechanism, right. To me, if we don’t have a mechanism, we don’t have an explanation yet. That’s what I see missing between the interactions. I might be willing to accept RQM as a placeholder, if we didn’t already have a mechanistic description available.
But I guess in the end it depends on what we require for an explanation. You have what you need in RQM, and I get what I need from universal wave mechanics. I hope there’s data someday to resolve this question.
LikeLike
“On reductionism and mechanism, right. To me, if we don’t have a mechanism, we don’t have an explanation yet.”
So true….. As elegant as RQM is as a metaphysical model, it lacks a mechanism that we can point to and say; “that’s it, that’s exactly what makes relationships work”. Rovelli is looking for a third alternative to the current physicalism/idealism paradigm and I respect him for that, even though he’s stopped short of making a metaphysical commitment. Nevertheless, he’s on the right track because physicalism as it is currently framed as well as idealism are dead fish on the beach, and there’s nothing anyone can do to revive them as they lie lifeless in the sun rotting.
RQM dovetails very well with my own metaphysical model, but unlike Rovelli, my model identifies the mechanism that RQM lacks and delineates how that mechanism works in a relational framework that for all practical purposes is physical.
LikeLiked by 1 person
“all information must come through information gathering systems”
That’s pretty much my take on it. But I do think it is slightly more profound that you may think.
As Gefter notes:
“You can’t start with a world, already divvied up into space and time, containing distinct systems and measuring apparatuses and environments, and then talk about objects interacting and getting entangled, because in the “no observers” story, none of those things can be defined in the first place.”
Gefter’s point in your decoherence example is that the experimenter partitioned the world into quantum and environment to start with.
Measurement requires spacetime, which is likely related in some way to gravity which QM is struggling to explain. A measurement occurs at a time and a location. It is, in fact, a record of something that has already happened, like the supernova twelve million light years away, so time is already implicit in it. Quantum processes are fundamentally reversible. In spacetime, processes are not so much. So, the issue in my view is more closely tied to relativizing QM, explaining how spacetime comes from it.
LikeLiked by 1 person
My point on decoherence is that it happens all the time, with any divisions only recognized after the fact. Yes, in a controlled experiment, we control those divisions, essentially deciding how the effects of an isolated quantum system are amplified into the environment. But they’re often not controlled in quantum computing when qubits decohere prematurely. And all the things in the classical world at a macroscopic level are constantly decohering without any decisions on our part.
On reversible processes, all fundamental processes in physics, quantum or otherwise, are reversible. (Well, unless there is a physical wave function collapse and it’s fundamental.) Irreversible processes emerge with the second law of thermodynamics, where they become irreversible in an entropic statistical sense. Which is also how decoherence is irreversible, in an entropic sense. Remember that decoherence is large scale entanglement. Simple entanglements, and even moderately complex ones, can be reversed, and have been in experiments. In principle, you could reverse environmentally induced decoherence, but it would be profoundly difficult, just as in principle it’s possible to unscramble an egg, but extremely difficult.
On spacetime, interestingly, straight QM has no issue with flat spacetime. QM and special relativity were reconciled by 1930. (Pilot-wave gums this up, a danger with many hidden variable theories, which in my mind makes them less plausible.) It’s the curved spacetime of general relativity, gravity, that is the issue.
LikeLike
“The title is an accurate summary of her thesis.”
Actually I’m not sure it does on thinking about it.
Vedral’s book is titled Portals to a New Reality, which also could be misleading but I would need to read it first to know.
The problem is the question of what is in dispute.
A) Reality itself requires conscious observers.
B) QM requires observers, but they may or may not be conscious depending on your view.
Gefter doesn’t seem to be arguing for A but she does seem to be arguing for B.
Vedral, according to the article, seems to be arguing against A and B unless he misunderstands that an observer is an experimental construct, not necessarily a conscious mind.
LikeLiked by 1 person
That’s a different distinction than I made in the post. I was asking if she was arguing for
a) Knowledge of reality requiring an observer (which seems definitely true)
b) Reality itself not existing without an observer
Her title implies b), but as I noted, my impression is she’s non-committal about which a) or b) is the case.
Of course, like we discussed, a lot hinges on what we mean by “observer.” If anything that interacts counts as a observer, then saying reality requiring interacting entities is closer to Rovelli’s view than the QBists’.
LikeLiked by 1 person
Your distinction does reflect the title, but I’m not sure it reflects her view unless the title carries an implicit “quantum” in front of reality.
An “observer” to me is only an experimental construct that performs a measurement and “collapses” the multitude of possibilities in the wave to a single one . In practice, it will be a device or apparatus, but it could be a conscious mind in theory since “observer” is purely a concept. But it is an essential concept to the theory because without it there is nothing left to interpret that would require Copenhagen, QBism, MWs, or whatever. We would simply be measuring something.
LikeLiked by 1 person
What would you say is the minimum requirement for something to be an observer? When qubits become entangled, are they observers of each other? Or, if decoherence is necessary, what about when a quantum circuit prematurely decoheres? Does that environment count as an observer?
LikeLiked by 1 person
“When qubits become entangled, are they observers of each other?”
My guess would be “no” based on the idea they are in a wave-like state, but I could be persuaded otherwise with a good argument.
There is certainly more involved “anything that interacts counts as a observer” as far as QM is concerned. Measurement requires interaction but the interaction in QM is different from ordinary measurement.
Let’s say we are measuring the acceleration of gravity by dropping a ball in a vacuum tube, finding all of the balls fall down, and measuring the time it takes to hit the bottom. We can repeat the experiment many times with balls of different masses and get the same result within margins of error. We can confidently predict how long it will take for a new untested ball to fall. This is ordinary measurement we do informally all the time that is formalized in the scientific method.
But a quantum experiment would be analogous to dropping the ball in the middle of tube and have some of the balls go up and some go down. We can’t predict before we run the test what each ball will do, but we might discover after enough testing half of the balls are ups and half are downs. There is no difference in the balls that we can detect. The weird part is, if we retest the already tested balls, all of the ups will still go up and all of the downs will still go down, but we still have no knowledge about whether an untested ball will be up or down beyond our belief that half will go up and half down. The measurement itself has removed the uncertainty for the tested balls. In ordinary measurement even with a statistical outcome, an already tested ball would still be subject to the statistical outcome – it might be up or down.
We probably should be a good deal more cautious about extrapolating QM to reality in general either with multiverses or conscious observers thinking reality into existence. It could be QM provides major clues to how reality works at the microlevel but the reality we experience is the world of ordinary measurement.
LikeLiked by 1 person