Le jeu. 9 janv. 2025, 22:16, Jesse Mazer <[email protected]> a écrit :
On Thu, Jan 9, 2025 at 5:38 AM Quentin Anciaux
<[email protected]> wrote:
Brent,
The core disagreement here seems to rest on the role and
status of possibilities. In a single-world framework, the
unrealized possibilities you refer to have no actual existence
or causal link to the realized world. They are simply
conceptual tools to calculate probabilities. But this is
precisely what strikes me as incoherent: why invoke these
possibilities as part of the explanation if they play no real
role in shaping the outcome?
I don't see a fundamental problem here, you can interpret it in
terms of the notion of "hypothetical frequentism" where you are
just talking about the frequencies that would obtain if an
experiment were (hypothetically) repeated an infinite number of
times, even if such repetitions don't occur in reality (assuming
some sort of ontological difference between possible worlds and
the 'real world', a difference which some views like Tegmark's MUH
or David Lewis' modal realism might deny--personally I'm
philosophically inclined to a sort of monism that denies a
distinction between possible and real worlds, as well as denying a
distinction between mathematical forms and the physical universe,
but I don't think the idea of making such distinctions is incoherent).
To me there are other reasons for seeing it implausible that
"collapse of the wavefunction on measurement" should be treated as
real rather than just a useful approximation, though. One is just
that I expect all physical phenomenon should be described by some
unified set of physical laws, applying to small collections of
particles and large measuring instruments alike; those Copenhagen
advocates who treat the collapse as objective don't have any sort
of mathematical model of the laws governing measurement
instrument/quantum system interaction to determine when a collapse
occurs, they just have to put in the notion by hand in an ad hoc
way. There are also "objective collapse theories" which do try to
give a theory in terms of some idea like a collapse happening
spontaneously whenever a collection of entangled particles exceeds
a certain mass, but this would actually give predictions different
from standard QM and seems implausible to me, it is an idea worth
testing of course.
The other big reason to see collapse as not ultimately real is
point made by von Neumann that it's actually arbitrary where you
place the collapse in a series of interactions, it doesn't matter
in terms of predictions whether it happens when the measuring
instrument interacts with the quantum system or only when the
information about that interaction enters a human observer's
brain. See the paper at https://www.jstor.org/stable/3541837
(readable if you sign up for a free jstor membership) which talks
starting on p. 123 about von Neumann's principle of
"psycho-physical parallelism" and on p. 125 quotes von Neumann
that this principle requires us to be able to show "that the
boundary between the observed system and the observer can be
displaced arbitrarily" and that "this boundary can be pushed
arbitrarily into the interior of the body of the actual observer
is the content of the principle of the psycho-physical
parallelism" (p. 126 also quotes him giving an example involving
the measurement of temperature).
I believe one could extend this further and imagine a Wigner's
friend style thought-experiment where a human experimenter is
making a bunch of measurements in a box which is perfectly sealed
off from interactions with the outside world (no decoherence
between the contents of the box and the outside environment) from
some time t=0 until we open it at a later time t=T. The person in
the box could be doing a series of measurements in the
electron-double slit experiment for example, in some cases putting
measuring devices at the slits to see which one the electron went
through, in other cases not, and recording the outcome of all
experiments. If we assumed each such individual measurement
collapsed the wavefunction, we'd get a prediction about the
statistics in cases where the electron was observed, and how they
differed from the statistics when it wasn't observed. If on the
other hand we assumed everything in the box was evolving according
to the Schrodinger equation with no collapse until we opened the
box at t=T, we would get exactly the same prediction about the
statistics seen in the experimenter's records! Except in this case
the different statistics when a measuring device was present at
the slits would be explained in terms of decoherence when the
electron became entangled with the measuring instrument and
records, plus the final collapse of those records at t=T. (it
seems to me that this is a further reason to be dubious of
objective collapse theories--it would make this agreement into
just a 'weird coincidence')
From what I understand the only way we might get different
predictions in the "every measurement causes collapse" picture and
the "collapse of records doesn't happen until box is opened at
t=T" picture is if there's some possibility the records of a
measurement could be thoroughly erased, with no possibility of
reconstructing it from the measured state at t=T. This is the type
of thought-experiment Deutsch suggested to test MWI against
"consciousness causes collapse" interpretations, see discussion of
"Experiment #3" proposed by Deutch, involving a quantum artificial
intelligence which makes measurements and then has its memory
erased, starting on p. 15 at
http://www.columbia.edu/~jpp2139/IssuesInQuantumComputingFD.pdf
Jesse
Jesse,
The issue with invoking "hypothetical frequentism" in a single-world
framework is that the supposed ensemble of possibilities has no
substance. It’s a purely abstract construct, with no ontological
status or causal influence on the realized history. If only one
history exists for all eternity, this ensemble is not just
unrealized—it’s irrelevant. It has no bearing on the single outcome,
no mechanism to "shape" probabilities, and no connection to reality
beyond being a mathematical abstraction.
Quentin
You argue that they reduce the probability of occurrence for
other events. But this reduction is purely formal—an artifact
of the mathematical framework, not something grounded in
reality. In a single-world view, only the realized history
exists, so the "ensemble" of possibilities is entirely
abstract. It doesn’t exist as part of reality, which means it
cannot influence or interact with it. The "reduction of
probability" you mention is simply a mathematical convenience,
not a causal mechanism.
When you calculated probabilities during the Vietnam War,
those calculations had predictive utility, but the unrealized
scenarios were not part of reality—they didn’t shape or
explain the actual outcomes beyond being conceptual
constructs. That’s fine for practical purposes, but when
applied to the nature of existence itself, it becomes
unsatisfying. The single-world framework uses the language of
possibility and probability without giving these notions any
real grounding.
In a multiverse framework, possibilities are not just
mathematical tools—they are ontologically real. Each possible
outcome exists within the structure of the multiverse, and the
probabilities describe the distribution of outcomes across
this ensemble. This provides explanatory depth that the
single-world framework lacks because it doesn’t need to rely
on nonexistent possibilities to "explain" realized outcomes.
Your bridge tournament analogy highlights the difference: in
the single-world view, only one table exists, and all the
other possible hands are irrelevant—they have no role in
shaping the one game that’s played. In the multiverse view,
all tables exist, and the hand you’re dealt corresponds to
your position in the tournament. The probabilities describe
the relative frequency of hands across the tables, not just
the abstract chance of receiving one hand.
The single-world framework asks us to accept that
possibilities exist only in the abstract, with no causal or
explanatory role in the realized world. This reliance on
nonexistent entities to justify outcomes feels like an
incomplete explanation, one that collapses into "it just
happened this way." That’s the heart of my issue with it.
Le jeu. 9 janv. 2025, 07:33, Brent Meeker
<[email protected]> a écrit :
On 1/8/2025 9:42 PM, Quentin Anciaux wrote:
Brent,
The issue I see with a single-world framework is the
reliance on possibilities that have no existence or
causal link to the realized world. In this view,
possibilities are entirely notional, they don’t exist
ontologically, and they have no impact on the single
realized history. This makes their invocation seem
unnecessary, even absurd, because they don’t contribute
to the reality we observe in any meaningful way.
But they do. They reduce its probability of occurence.
If the only thing that exists is the realized world, why
appeal to a theoretical ensemble of possibilities?
Because that's what the equations of quantum mechanics
produces. We're not "appealing to them" where taking them
into account as things that might occur. That's why the
Born rule assigns probabilities less than one to them.
It’s as if the single-world view borrows the language and
tools of probability to describe outcomes but discards
the explanatory depth provided by an actual ensemble.
I'd say that's looking at it exactly backwards, as though
the "tools of probability" on applied to cases that were
really deterministic (had explanatory depth) and what work
is done by the word "actual" in "actual ensemble".
Usually it is an ensemble of possibilities. When you're
dealt a bridge hand no one supposes that all other
possible hands are dealt somewhere else; it is enough that
they merely possible. During the Viet Nam was I
calculated the probability of dropping a bridge with a
Walleye, I calculated the probability of a missile failure
causing it to hit the launching aircraft, I calculated the
probability of a wayward missile going out of the range
safety boundaries, and dozens of other probabilities. I
was always considering a range of instances and their
contrary; but I never needed to suppose the instances were
actually anything more than possibilities. They didn't
have to happen anywhere in any world.
Without the existence of unrealized possibilities, the
concept of "randomness" seems like a placeholder for "it
just happened this way," offering no real insight into
why this one history unfolded.
No, it's a "placeholder" for it could have happened these
other ways but didn't.
In contrast, in a multiverse framework, the ensemble is
not merely theoretical, it has ontological status.
Yes, it's like a bridge tournament in which all possible
hands are dealt at different tables and then you pick one
to sit */at random/*. But wait, that's absurd, we must
sit down at every table. And then we must play every
possible card in every possible order. Otherwise we
cannot speak of the probability of making our bid.
The possibilities exist and have causal relationships
within the broader structure. This provides coherence to
the use of probability, as it describes the distribution
of outcomes across the ensemble, not just within a
single, isolated history.
The single-world framework effectively asks us to accept
a universe where unrealized possibilities are invoked to
explain outcomes, yet they have no actual role in shaping
reality.
Sure they do. If they are things that might possibily
happen then they reduce the probability of something else
happening.
Brent
This reliance on something that neither exists nor
affects the realized world strikes me as deeply incoherent.
Quentin
Le jeu. 9 janv. 2025, 03:14, Brent Meeker
<[email protected]> a écrit :
On 1/8/2025 4:11 AM, Quentin Anciaux wrote:
Brent,
The core of my argument is that in a single-world
framework, the ensemble of possibilities described
by Schrödinger’s equation is only conceptual. If
only one history is realized, then those
"possibilities" don’t exist in any meaningful
way—they’re theoretical abstractions. In the absence
of an actual ensemble from which a selection occurs,
the notion of randomness is a metaphor, not a mechanism.
In the many-worlds framework, every possibility is
realized, so the "selection" is an emergent
phenomenon from within the structure of the
totality. In the single-world view, however, there’s
no actual ensemble. Probabilities merely describe
the likelihood of the one realized outcome, but
there’s no underlying framework where those
possibilities are instantiated. Randomness then
becomes a label for the lack of explanation rather
than a true process.
To say "the single history simply is" and call that
random doesn’t resolve the issue—it just restates
it. Without an ensemble that exists ontologically
(even probabilistically), the idea of selection
collapses because there’s nothing to select from.
The photon emission you mentioned is described by
probabilities in QM, but those probabilities don’t
correspond to real, alternate outcomes in a
single-world framework. The realized outcome is the
only one that exists, and all other "possibilities"
are simply unrealized ideas.
In contrast, in the many-worlds interpretation, the
photon’s emission in one state is one thread of the
total structure, and alternate emissions exist along
other threads. This gives explanatory power to the
probabilities, as they correspond to real structures
within the ensemble.
But small probabilities explain why things */don't/*
exist.
Regarding your point that probabilities lose meaning
in MWI because all possibilities are realized—that’s
not the case. Probabilities in MWI are understood as
the measure of the branching structure relative to
the observer's perspective. They still hold meaning
because they reflect the structure of the
multiverse, not a singular outcome.
What about the one's for which P=0, you could as well
say that reflect the structure of the multiverse.
Will you make an ensemble of them?
The single-world view still strikes me as incoherent
because it leans on the language of probability and
possibility but denies their actual realization.
Without an ensemble, it’s hard to see what
randomness truly means.
In every other application of probability theory (and
for years I headed the Reliability Division at Pt.
Mugu) the ensemble is only notional. It is a the set
of possibilities without assuming that they exist, in
which case they would be actualities. With an
ensemble of which every member exists, randomness
becomes incoherent.
Brent
--
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/207aada7-2a8d-4c9e-8490-a25f23eff83a%40gmail.com
<https://groups.google.com/d/msgid/everything-list/207aada7-2a8d-4c9e-8490-a25f23eff83a%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/CAMW2kAoJKWzYSjXeRQJCNBiXo2Xbk8VseggxV%3D6rgWL4y6aMFQ%40mail.gmail.com
<https://groups.google.com/d/msgid/everything-list/CAMW2kAoJKWzYSjXeRQJCNBiXo2Xbk8VseggxV%3D6rgWL4y6aMFQ%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/2f860f70-c8af-477a-add2-c7192ab54ed2%40gmail.com
<https://groups.google.com/d/msgid/everything-list/2f860f70-c8af-477a-add2-c7192ab54ed2%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/CAMW2kAox73KEyv%3DqWUWsCd7pd9wHgm0jwyAPAfv4mkY%3DZkJH7Q%40mail.gmail.com
<https://groups.google.com/d/msgid/everything-list/CAMW2kAox73KEyv%3DqWUWsCd7pd9wHgm0jwyAPAfv4mkY%3DZkJH7Q%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/CAPCWU3KP1ygXnajTScOFOrH7zCMJ%2BozNpZRuwsWcpR6qM_UUYQ%40mail.gmail.com
<https://groups.google.com/d/msgid/everything-list/CAPCWU3KP1ygXnajTScOFOrH7zCMJ%2BozNpZRuwsWcpR6qM_UUYQ%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/CAMW2kAqjPA6Y1BgnBNzr2mXB8fD4A2Pz9cpeH4Ctok%3DuxLv%2Bkg%40mail.gmail.com
<https://groups.google.com/d/msgid/everything-list/CAMW2kAqjPA6Y1BgnBNzr2mXB8fD4A2Pz9cpeH4Ctok%3DuxLv%2Bkg%40mail.gmail.com?utm_medium=email&utm_source=footer>.
