On 11/05/2016 2:31 am, Bruno Marchal wrote:
On 10 May 2016, at 15:37, 'scerir' via Everything List wrote:
Following the above reasoning MWI (if it is a truly deterministic
theory)
should violate the locality condition.
I doubt this, but if you find a proof, in the literature (or not), I
am interested. As I explained, and also give references, it seems to
me that the MWI restores both 3p determinacy and 3p locality, making
both the indeterminacy and non-locality only first person plural
phenomenological happening. That is also Everett's position, and I
would say the position of most Everettian (I still don't find any
Everettian claiming that the MWI remains non-local, except the
beginners who often think at first that the entire universe split
instantaneously, but this does not deserve to be commented as nobody
believes in this anymore).
I came across the following brief statement by Goldstein et al:
*Many-worlds and relational interpretations of quantum theory**
*
Strictly speaking, there is yet another assumption, besides locality and
the "no conspiracy" condition that is necessary for the proof of Bell's
theorem: one has to assume that, after the experiment on one given side
is performed, its ±1-valued outcome is a well-defined element of
physical reality. (Recall that in Section 6, in order to apply Bell's
definition of locality to the type of experiment considered in Section
5, we assumed that the outcomes A1 and A2 were functions of the local
beables in regions 1 and 2, respectively.) Now one might wonder how
anyone could deny that assumption. After all, the outcome of the
experiment is recorded by the configuration of a macroscopic object
(say, a pointer position, ink on a piece of paper, etc.) that can be
directly inspected by a human experimenter. However, there exists one
fairly popular interpretation of quantum theory that does deny that one
has (after the experiments are concluded) a well-defined physically real
±1-valued outcome on each side: the many-worlds interpretation90. More
precisely, according to the many-worlds interpretation, both outcomes
are equally real on each side, so that it doesn't make sense to talk
about "the one ±1-valued outcome that actually occurs". Certain
"relational" interpretations of quantum theory91 also deny that a
completed experiment has a well-defined physically real outcome. It is
possible that this type of strategy could succeed in evading the
consequences of Bell's theorem, allowing for the possibility of a
universe governed by a local theory such that conscious observers living
in that universe attest to the validity of the quantum predictions.
However, it is not clear how to actually do the trick. There are many
difficulties and the subject is rather subtle. To begin with, there are
controversies around the problem of finding an appropriate formulation
of a many-worlds (or relational) interpretation. Moreover, it is not
clear whether such an appropriate formulation can be made local, given
that the wave function — which seems to be all there is in standard
formulations of many-worlds theories — is not a localized object; in the
terminology of Bell, it is not a local beable. (Indeed, if a theory has
no local beables, it is certainly not meaningful to ask whether it is
local or not in the relevant sense.) A formulation of a version of the
many-worlds interpretation which includes, in addition to the wave
function, some local beables, was presented in a recent paper92, but it
was found by the authors to be non-local. The question of whether a
many-worlds (or relational) approach can be taken advantage of to create
a local (and empirically viable) theory thus remains open — as does the
question of how seriously one should take a theory of this type, should
it be successfully constructed.
http://www.scholarpedia.org/article/Bell%27s_theorem#Manyworlds_and_relational_interpretations_of_quantum_theory
This Scholarpedia article (*6*, 8378) is available on-line and has many
additional references.
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 post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/everything-list.
For more options, visit https://groups.google.com/d/optout.
