Le ven. 10 janv. 2025, 07:58, Bruce Kellett <bhkellet...@gmail.com> a écrit :
> On Fri, Jan 10, 2025 at 5:43 PM Quentin Anciaux <allco...@gmail.com> > wrote: > >> Le ven. 10 janv. 2025, 07:31, Bruce Kellett <bhkellet...@gmail.com> a >> écrit : >> >>> When you roll a die, the probability of a four is 1/6. Do the other >>> possibilities have to exist? Even if they do, they have no influence on the >>> outcome you actually observe. Probabilities are useful for predicting >>> possible outcomes; so I can, in advance, predict that the possibility that >>> I shall get a four is 1/6. That is all there is to it. That is what >>> probabilities do. There is no need for the other possible outcomes to >>> exist, either in advance or after the fact. They play no role, outside of >>> your imagination. >>> >>> Bruce >>> >> >> The difference with your die example is that in a single-history >> framework, the probabilities we assign are only meaningful if the ensemble >> of possible outcomes has some kind of reality. When you roll a die, the >> probability of a four being 1/6 relies on the existence of a mechanism >> where all six sides are genuinely possible outcomes before the roll. If we >> later observe that, in the single history of the universe, a four never >> occurs, then its actual probability was not 1/6—it was zero. (It’s an >> example) >> >> In a single-world view, the other possibilities never had any reality at >> all—they were not actualized, not even hypothetically. So, the act of >> assigning probabilities becomes a purely formal exercise, disconnected from >> what can or cannot happen. The other sides of the die are reduced to >> abstractions, with no causal role in the observed outcome. >> >> Probabilities are indeed tools for predicting outcomes, but their >> usefulness depends on the assumption that the possibilities they describe >> have some grounding in reality. Without that grounding, they become empty >> numbers. In frameworks like many-worlds, those possibilities exist as >> actualized outcomes in other branches, giving substance to the >> probabilities. But in a single-history framework, the unobserved >> possibilities never existed, making the probabilities feel like a game of >> imagination rather than a reflection of the world. >> >> If the "other possible outcomes" have no reality, even as hypothetical >> constructs, then the act of predicting probabilities loses its explanatory >> value. They become detached from the reality they aim to describe. That’s >> the fundamental issue I’m pointing to. Without some ontological weight for >> the ensemble of possibilities, probabilities are just formal tools with no >> connection to what actually happens. >> > > That is just patent nonsense. Formal tools are quite capable of giving the > right answer for the realized world; (and the right answer is what actually > happens.) > > Bruce > Bruce, I agree that formal tools can provide correct predictions for the realized world. However, in a single-history framework, the "right answer" they provide—the actual outcome—reduces probabilities to mere retrospective descriptors. Probabilities only have meaning if the ensemble of possibilities they refer to has some grounding, even if not directly realized. Otherwise, we are attributing significance to calculations about scenarios that never existed and had no potential to exist within the single-history paradigm. For example, if in this single-history universe, a particular outcome (e.g., rolling a four on a die) never occurs, then retroactively, its probability was effectively zero. The formal tools used to calculate probabilities would still "predict" the possibilities, but in a framework where only one history is ever realized, those unrealized possibilities had no causal or explanatory connection to the outcome. The tools become exercises in abstraction, detached from the realized world. This is the crux of the issue: probabilities, in a single-history view, do not reflect anything about what could happen—they only describe what did happen after the fact. Without a substantive ensemble of possibilities, probabilities lose their role as meaningful descriptors of potential and become post-hoc justifications for a single outcome. In frameworks where possibilities are real (e.g., many-worlds), probabilities gain coherence because they describe distributions over actualized outcomes, not abstract, non-existent alternatives. The point is not that formal tools fail to provide correct answers—they do. The issue is that, in a single-history framework, the connection between those tools and the nature of reality becomes tenuous, leaving the probabilities they calculate feeling arbitrary and conceptually empty. Quentin > -- > 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 everything-list+unsubscr...@googlegroups.com. > To view this discussion visit > https://groups.google.com/d/msgid/everything-list/CAFxXSLSk-3qcCW_-0eNAxk5GhK8TLr5DZpYtM%3D_sEMZAEvVd-A%40mail.gmail.com > <https://groups.google.com/d/msgid/everything-list/CAFxXSLSk-3qcCW_-0eNAxk5GhK8TLr5DZpYtM%3D_sEMZAEvVd-A%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 everything-list+unsubscr...@googlegroups.com. To view this discussion visit https://groups.google.com/d/msgid/everything-list/CAMW2kAqa%3Due3a7Ji1a1HZn101ZZsERgBgVrygjM96y2%3D2PCcQQ%40mail.gmail.com.
