Brent, You keep insisting that probability is justified simply because "that’s how physics works," but that doesn’t address the foundational issue. You rely on probability as an empirical tool, yet in a single-history framework, it describes events that were never real in any way. If probability is meant to quantify potentialities, but only one sequence of events ever occurs, then the unrealized possibilities were never actually possible—they were just numbers with no ontological status.
You say that if an event never happens, then we should adjust our prior probabilities. Fine. But in a single-history universe, every event only happens once, so there’s no way to distinguish between a genuine probability assignment and an incorrect one. If an asteroid has an "80% chance" of hitting Earth and doesn’t, was the probability wrong? Or did we just get lucky? In a framework where only one history exists, you can always retroactively claim the probabilities were correct, no matter what happens, because there’s no underlying structure to validate or invalidate them. That makes probability little more than a storytelling device. You mock MWI’s "imaginary worlds," yet in your framework, probability relies on imaginary possibilities that never had any reality. The difference is that in MWI, probability describes real distributions of outcomes across branches, while in a single-history world, probability is just a tool for reasoning about things that were never going to happen. You claim that probabilities in MWI aren’t "actualized," yet you rely on probabilities that, in a single-history world, have even less connection to reality—they refer to events that never happened and never will. As for the Born rule, you claim that all derivations assume it outright, but that’s simply false. The Deutsch-Wallace approach derives it from decision theory, Zurek’s envariance provides symmetry-based reasoning, and other work has shown that the Born rule emerges naturally from the structure of the wavefunction. Meanwhile, your framework just assumes it as a brute fact, with no deeper explanation beyond "that’s how physics works." If you’re fine with that level of arbitrariness, that’s on you, but don’t pretend it’s a superior foundation for probability. Ultimately, your position amounts to saying that probabilities "just work" and that we shouldn’t ask why. That’s fine if you’re satisfied with a purely instrumentalist view, but don’t pretend that’s a deeper understanding. MWI at least attempts to provide a foundation for probability, while the single-history view just asserts that numbers assigned before an event have meaning, even when they describe things that never happened. If that’s not an empty formalism, I don’t know what is. Quentin Le ven. 7 févr. 2025, 04:00, Brent Meeker <[email protected]> a écrit : > > > > On 2/5/2025 11:44 PM, Quentin Anciaux wrote: > > Bruce, > > You keep insisting that randomness "just is" and that no deeper > explanation is possible, but that’s precisely the problem with the > single-history view: it reduces probability to a descriptive afterthought > with no fundamental meaning. You argue that in a single-history universe, > we must simply accept that an event had an 80% chance of happening even if > it never does. > > You're making up strawmen. If it never does in many trials, then it's > simple Bayesian calculation to tell you it's much more *probable* your > 80% value contains an error than not. You've obviously never dealt with > the actual application of probabilities. > > But in what sense was that probability real if only one history ever > unfolds and the event never occurs? > > You say that probabilities are "real enough" because they are based on > prior experience. But prior experience in a single-history universe is just > another way of saying, "this is what happened before." That’s not an > explanation; it’s circular reasoning. > > No, it's the most basic form of prediction from statistics. > > You’re using past outcomes to justify probability assignments, but if > probability is supposed to describe potential events, then what does it > mean when a "possible" event never happens, despite being assigned a > nonzero probability? > > You keep harping on "it never happens". Bruce and I have both addressed > that. It you toss a coin twice and "it never happens" to come up tails. > If you toss it a thousand times and it never comes up tails, it's a > two-headed coin. > > > In a multiverse framework, probability describes the relative frequency of > events across actualized branches. > > You keep using the word "actualized" which according to my dictionary mean > "made actual". But that's false. The branches are just what, in another > context, you characterize as mere abstract mathematices. > > It is not just an abstract expectation—it is grounded in the structure of > the wavefunction itself. > > Which itself is an abstraction, an abstraction that does not even include > probabilities. > > You dismiss the idea that measure in the wavefunction corresponds to > probability, but this is not an assumption—it follows naturally from the > mathematics of quantum mechanics. The Born rule is not an extra assumption > in MWI; it emerges from decision theory (Deutsch-Wallace), from symmetry > arguments (Zurek’s envariance), or from self-locating uncertainty. You keep > demanding "proof," > > It's not just me. People have been searching for a way derive the Born > rule for decades and in every case so far the derivation have been found to > assume things equivalent to assuming the Born rule. And that goes for > Deutsch and Wallace. Zurek doesn't even pretend to derive it, he just > asserts that it must work such that only the Born rule probabilities apply > to stable worlds. Read the literature. > > yet you accept the Born rule as a brute fact in your own framework without > any justification beyond "that’s just how quantum mechanics works." > > That's why physics is an empirical science. Some things are derived from > observation, not "self evident axioms". > > > Meanwhile, your single-history approach provides no mechanism for why > probabilities match experimental results. > > Do you think MWI would be true even it didn't match experimental results?? > > You claim that probability "just works" without explaining why the > realized history should respect the Born distribution at all. You dismiss > branch weighting in MWI as unproven, yet you offer no competing explanation > for why a single sequence of events should follow probabilistic predictions. > > Ultimately, your position amounts to: "Random events happen, probabilities > just work, and there’s no deeper reason for anything." > > That seems to the case for QM. It may anger you or disappoint you, but > Nature doesn't care about your demands. > > That’s not an explanation—it’s an assertion that we shouldn’t ask > questions. If you’re satisfied with that, fine, but let’s not pretend it’s > a superior foundation for probability. It’s just giving up on understanding > why reality follows probabilistic laws in the first place. > > I'm fine with asking questions. I'd be fine with finding way to derive > Born's rule without assuming it or something equivalent. What I'm not fine > with is inventing imaginary world's and claiming that's the same thing. > > Brent > > > Quentin > > Le jeu. 6 févr. 2025, 01:09, Bruce Kellett <[email protected]> a > écrit : > >> On Thu, Feb 6, 2025 at 11:49 AM Quentin Anciaux <[email protected]> >> wrote: >> >>> Bruce, >>> >>> You’re making a distinction between single-event probabilities and >>> repeated trials, but you’re not addressing the core issue: in a >>> single-history universe, probability is only ever descriptive, not >>> explanatory. You claim that if an asteroid has an 80% chance of impact but >>> doesn’t hit, then the 20% chance was simply realized. But this explanation >>> is entirely retrospective—it tells us nothing about why this history, >>> rather than any other, is the one that unfolded. >>> >> >> That is the nature of random events. It seems that your real objection is >> to randomness, events that have no simple mechanical explanation. That is >> quantum mechanics, and you just have to get used to it. >> >> You say, "assuming the calculations were accurate, then there certainly >>> was an 80% chance of hitting, and a 20% chance of missing." But in what >>> sense was that 80% ever real? In the only history that exists, the asteroid >>> never had an 80% chance of hitting—it always had a 100% chance of missing >>> because that’s what happened. The probability was just a number assigned >>> before the event, with no actual force in determining the outcome. >>> >> >> I think the calculations are based on prior experience. They are real >> enough, not just empty air. >> >> In a multiverse framework, probabilities are grounded in actual >>> distributions across histories. The 80% means that in 80% of branches, the >>> asteroid hits, and in 20%, it misses. This gives probability an explanatory >>> role—it describes the structure of reality, not just an arbitrary number >>> assigned to something that never had a chance of happening. >>> >> >> Unfortunately, it is well known that branch counting is a failed >> enterprise in quantum mechanics. So the claim that something or the other >> is true in 80% of the branches is just empty rhetoric. >> >> You claim that MWI has no way to connect probabilities to the >>> wavefunction, but that’s false. The structure of the wavefunction naturally >>> assigns measure to branches, >>> >> >> Does it now? And how does it do that? You have a well-developed ability >> to make endless unevidenced assumptions and bend them to your will. Start >> trying to prove some of this! >> >> >> and those measures correspond to the squared amplitudes of the >>> coefficients—the Born rule emerges from this structure. >>> >> >> It does not without many additional assumptions. The attempts to derive >> the Born rule from Everett have all failed. >> >> You keep asserting that probabilities in MWI are meaningless because "all >>> possibilities happen," but that’s only true if you ignore the fact that >>> measure matters. Not all branches are weighted equally, and the frequencies >>> of outcomes reflect those weights. >>> >> >> But you have not shown how these weights arise, or how outcomes depend on >> these 'weights'. >> >> The issue isn’t whether we can calculate probabilities in a >>> single-history world—it’s whether those probabilities have any real >>> ontological meaning. You claim that in a single-history world, >>> probabilities "just work," but that’s not an explanation. It’s just a way >>> of pretending that numbers assigned before an event have some deeper >>> reality when, in truth, they don’t. In the end, the only thing that exists >>> is the one history that happens, and everything else was just an illusion >>> of possibility. >>> >> >> You are still looking for an 'explanation' of random events. You will >> look in vain, because no such explanation can be forthcoming. >> >> Bruce >> -- >> You received this message because you are subscribed to the Google Groups >> "Everything List" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> To view this discussion visit >> https://groups.google.com/d/msgid/everything-list/CAFxXSLSoueSm74SRU0nie_b%3D%3DCDft6rWboAJPMKfyNLUbHAhOA%40mail.gmail.com >> <https://groups.google.com/d/msgid/everything-list/CAFxXSLSoueSm74SRU0nie_b%3D%3DCDft6rWboAJPMKfyNLUbHAhOA%40mail.gmail.com?utm_medium=email&utm_source=footer> >> . >> > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion visit > https://groups.google.com/d/msgid/everything-list/CAMW2kAoUzMipVLp0HwY%2BHXeDKDWhH988t9U6pMWne8%2BZmF3JRg%40mail.gmail.com > <https://groups.google.com/d/msgid/everything-list/CAMW2kAoUzMipVLp0HwY%2BHXeDKDWhH988t9U6pMWne8%2BZmF3JRg%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > -- > You received this message because you are subscribed to the Google Groups > "Everything List" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion visit > https://groups.google.com/d/msgid/everything-list/583254b6-006c-47c6-8c60-b470d94b17f0%40gmail.com > <https://groups.google.com/d/msgid/everything-list/583254b6-006c-47c6-8c60-b470d94b17f0%40gmail.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "Everything List" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/everything-list/CAMW2kAoKrc9o1sW4LXR_NCMvCxGvVQ0fObdXPVQYSQYQJ%2BTPvg%40mail.gmail.com.
