Greetings; this reply has taken some time...
I don't quite agree with your point of view, and the reason is maybe similar to our disagreement in my statement: "It is not useful to talk about 1st person experiences in 3rd person terms, since when we do that we lose the very thing that we want to study."
I think you're right; this is the central disagreement. To spell it out: I do not believe the 1st person/3rd person distinction is fundamental. That is, I think that once you've explained what a system does (and how it does it), you've explained everything. Consciousness is simply a complicated set of behavioral dispositions, which can in principle be explained from an objective, 3rd-person standpoint; and the "1st person" viewpoint is just a description, a way that it naturally occurs to us to put things, because of our psychology (biology).
When you said earlier that:
"In a materialistic framework, ' I ' am a bunch of atoms. These atoms
happen to constitute a system that has self-referential qualities that
we call consciousness."
I took you to be agreeing with me on my view of consciousness. But I think a difference is highlighted here:
I don't need to point out who ' I ' am. I am
concerned with my first-person experiences, and that is easy to
determine without even gaining acces to the other entities who look like
What, exactly, are these "first-person experiences"? Are they really so easy to determine?
I would say that whatever they are, they contain no information on whether you are Eric01 or Eric02. We know that these completely specified entities do exist, but since Eric01 has no way of knowing he is in fact Eric01, it does not matter what he experiences; the difference between Eric01 and Eric02 is not a difference for Eric.
Hal Finney's thought experiment about the 2 identical computers is right on, I think. You have 2 AI programs running in lockstep. Nothing in the programs' experiences can tell them which program they are (let's say each computer has a serial number on the outside; but this is completely inaccessible to the programs running on them). Now I tell both programs that I'm going to turn off the machine labeled "01" and leave "02" running. Our intuitions now point in different directions here.. I do not think it's meaningful here for a program to consider which computer it's running on.
You would ask the next question:
"If you were one of the programs, what would you expect to happen?"
whereas I would ask it:
"If you were the set of these 2 programs, what would you expect to happen?"
So I think it all comes down to which conception of "you" you decide to use.
You are trying to identify ' me ' by somehowpointing it out from the pool
of similar entities in a God's perspective. That may be even impossible,
if there is no God, but that is another discussion.
You're right, it is impossible. (even if "you" were in one of Jürgen's or Bruno's programs, I would think it would be difficult and, to a certain degree, _arbitrary_ to completely isolate the part of the program that is "you" from the part that is everything else.) I think the fact that it is impossible to determine who "you" are is precisely one of the things that makes QTI seem plausible.
' I ' am what I experience myself to be, and that is a completely specified entity since there are no interactions between universes.
I think that your word "that" in this sentence does _not_ actually point to a completely specified entity. Your experiences are not specifying what needs to be specified here. You appear to be saying, "Well, at any rate, I know that _I am me_ and no one else!". But you do not experience yourself as a completely specified entity. And it's still true that a vast number of (almost completely) identical bunches of atoms are saying the same thing as "you", and invoking the same experiences in support of their assertion.
This was my point about Wittgenstein's man who says he knows how tall he is. The answer to the question "How tall are you?" is not "This tall!", accompanied by a gesture with your hand. It seems like an answer, but it won't do - because the question is asking for a measurement (given an agreed-upon metric, etc.). For the purposes of the QTI, the question "who are you?" cannot be answered by doing, as you essentially are doing, "I'm this person!", accompanied by pointing to your chest.
If two universes have any distinction, even in the state of a particle, as
I understand the MWI, they are distinct universes. That is an objective
statement, which has nothing to do with my lack of knowledge about
the small differences. One particular universe (the one ' I ' am in) may
have distinct outcomes in a dangerous situation, so that it splits in, say,
50% universes where I am dead, and 50% where I am alive.
We both agree with this setup of MWI and the outcomes of dangerous events, etc.
You might say, "We agree on all the 3rd person facts". For instance, we both agree that, say, Eric01 will die and Eric02 will live, and this has nothing to do with what either of them know. The disagreement comes when you say, "Yes, but what will *I* experience?". I have been trying to argue that this question and its answer depends on the particular sense of "I" being used.