Probability
I continued reading his "Relational Quantum mechanics" paper and I might as well comment on it here, to see if someone would disagree.
"First of all, one may ask what is the “actual”, “absolute” relation between the description of the world relative to O and the one relative to P. This is a question debated in the context of “perspectival”
interpretations of quantum mechanics. I think that the question is ill-posed. The absolute state of affairs of the world is a meaningless notion; asking about the absolute relation between two descriptions is precisely asking about such an absolute state of affairs of the world. Therefore there is no meaning in the “absolute” relation between the views of different observers. In particular, there is no way of deducing the view of one from the view of the other.
Does this mean that there is no relation whatsoever between views of different observers? Certainly not..."
-- Rovelli, Relational Quantum Mechanics
Sounds great so far!
"There is an important physical reason behind this fact: It is possible to compare different views, but the process of comparison is always a physical interaction"
-- Rovelli, Relational Quantum Mechanics
Also Excellent IMO.
"Suppose a physical quantity q has value with respect to you, as well as with respect to me. Can we compare these values? Yes we can, by communicating among us."
-- Rovelli, Relational Quantum Mechanics
Again, I'm all in with this! Extremely sound reasoning.
"But communication is a physical interaction and therefore is quantum mechanical. In particular, it is intrinsically probabilistic."
-- Rovelli, Relational Quantum Mechanics
Here I start to get a little bit reserved. He is referring to "quantum mechanical interactions", and the notion of "probability", at the same time he in various places avoids discussing the physical meaning of probability. And I wonder if "quantum mechanical interactions" implies some constraining ideas of what an interaction is. I'll put a marker here and keep reading.
Now he pulls out the quantum measurement formalism and here his previously plausible reasoning looses me,
An observer P(observer2), observes S(System)-O(observer1).
"but a quantum property of the quantum S-O system, that can be investigated by P, and whose yes/no answers are, in general, determined only probabilistically."
-- Rovelli, Relational Quantum Mechanics
Again, he uses the probabilistic concept. Where is the physical meaning with probability?
Is the probability itself "observerd"? If so how? If not, then what is it doing in here and what the line of reasoning that brings in that concept?
My issues is not with "
only probabilistically", because a deterministic probability is not just "only" to me. It's a strong statement.
It'd say it's used here as a choice of rule of reasoning more than anything physical. This choice comes out as formally arbitrary to me. I think there must be another way to continue that initiated spirit of Rovelli that I agree with up until this point.
He seems to make a semi-synthesis here, but still in the middle apply the same old QM (more or less).
Here, I personally feel that Rovelli's plausible and excellent reasoning appears somewhat broken?
What do you think, I am missing something? or maybe this is to be seen as the best we can do "so far", and Rovelli himself is still looking for improvements? In that case it makes more sense. Maybe I'm asking too much of this.
I personally do not feel confident in building further onto this platform. I'd prefer to step back to the point where is excellent reasoning seems to get fuzzy, and to me the unclear point is the physical meaning of probability. This is also a common issue for choosing information measures.
My opinion is that there are at least two issue with QM.
First is the objectivity issue. And here I think rovelli's idea of relative information is right on, in the sense of subjective bayesian probability. I also share his view on emergent objectivity. I think it's plain brilliant.
But there is another issues that I at least can't see that he deals with. All of his reasoning is built on the probability concept. Wether bayesian or non-bayesian, the question of it's physical meaning still persists. Here the issues of information capacity also becomes the factors. I have a feeling that he refers to some information concept, which again if we are talking about shannon is more or less an arbitrary measure, still based on an underlying probability measure.
I am not sure I got his spirit with the supposedly plausible information postulates he made, I need to read them again, but my spontaneous reaction was doubt.
So to the point of choosing observables, in this context, what is the status of probability?
I'll keep reading later...
/Fredrik
