Dear Lee,
Are you familiar with any of the experiments that have been performed
regarding "quantum counterfactuals" or "null measurements"? It turns out
that the fact that some particular measure *was not made* counts just as
much, and thus affects the results of a measurement, of an actual
measurement that was made. Thus information of any occurrence or
non-occurrence of a measure of a QM system, coded in an OM, will make a
difference that can not be hand waved away.
This is why I am introducing the notion of a witness.
Interleaving...
----- Original Message -----
From: "Lee Corbin" <lcorbin.domain.name.hidden>
To: <everything-list.domain.name.hidden>
Sent: Tuesday, June 28, 2005 7:43 PM
Subject: RE: Continuity, Observer Moments and Memory of a Past
> Stephen writes
>
>> > [LC]
>> > I'm skeptical of "continuity" requirements. Now I do not believe in
>> > Greg Egan's equations in "Permutation City": according to a premise
>> > of the story, it order to obtain the you of tomorrow, there is a
>> > short-cut alternative to just letting you run. And that is to
>> > determine the solutions of an immense number of differential
>> > equations that do not in fact emulate your intermediary states.
>> > If this were so, then it may be that you could discontinuously
>> > skip past all of tonight and tomorrow's experiences, and just
>> > start living by directly experiencing the day after that.
>>
>> [SPK]
>> Does it not seem that the continuity requirement is such that it only
>> comes up when we consider either an external 3rd person P.o.V. of a OM
>> chaining P.o.V. or an internal 1st person of a memory of having changed
>> one's mind about something is some other OM?
> [LC]
> Yes, it does seem to come up under both points of view. For some
> reason I can't fathom. To me, there is no use of it either way:
[SPK]
Does this "unfathomness" not bother you? It seems to me that we might be
able to navigate our way through this, if we are careful and faithfull to
our findings are we go. ;-)
> [LC]
> Let's take the first-person point of view first. What if in some
> universe it so happened that none of your Monday OMs existed? I
> say that you would have Sunday's OMs followed by Tuesday's OMs,
> and be none-the-wiser. The discontinuity would not be important
> at all.
[SPK]
I agree that the discontinuity is irrelevant if and only if the quantity
of "duration" coded into the OM is important. It can be easily argued that
the duration of coherence of a QM system strongly affects the kinds of
interactions that it may have, therefor if the content of an OM codes
anything having to do with QM system in interaction, the duration of the OM
chaining will matter.
On the other hand, there is obviously, an invariance with respect to
"reparametrizations" involved. For example, for a given set of OMs, so long
as the total content is the same, it does not matter is a week is made up
one one day or seven!
> [LC]
> I considered the third-person view above, but here's more: *If*
> something weird like Greg Egan investigated were true, then one's
> experiences could be discontinuous. However, as actual machines
> we compute all our intermediate states, and so experience is
> continuous (no matter how it may seem via memory erasure or
> brain malfunction).
[SPK]
If we are going to play around with the idea of OMs, then it is
necessary that those OMs contain or code, somehow, all the relevant
experiences that some observer might have. If we are going to seriously
entain this toy model has having something to do with the world we
experience, then somewhere somehow it has to account for the empirical
facts, otherwise... ;-)
If Egan's idea is "true", it does not matter, from a 3rd PoV the OMs can
be implemented in any order, all that matters is the information that they
code.
> [LC]
> (In Repeated Experience, this characteristic is not present. If
> your behavior today is a reenactment of an original calculation
> from billions of years ago, then in principle it would be possible
> for you to experience Sunday, then be terminated, and then a
> recorded state from the original Tuesday billions of years ago
> commenced.)
[SPK]
It seems that closed-time-like-loops are only a problem if:
1) OMs code some kind of trace of all OMs that would be in its causal "past"
and
2) There is no upperbound on the memory space used so that no contradiction
will *ever* obtain.
The repeted experience scenario only causes contradiction IF somehow a
memory of a previous emulation remains; "Ground Hog Day" style.
>
>> [SPK]
>> If we erase our memories of having done the loathsome
>> event after actually going through it, do we not need to
>> also erase the memory of any other witness that might be
>> able to remind us of the event?
> [LC]
> No, witnesses can be bought off, or, where I come from, snuffed
> out. So long as they keep their traps shut, see, I don't
> start to wonder about it. But anyway, having them tell me in
> all sincerity about things I don't remember isn't anything
> *like* actually having had the experiences: suppose that you
> convince me that the Nazis ran a version of me in 1944 and
> tortured the bejesus out of me---well, I'd rather listen to
> your story a thousand times than actually go through the
> alleged experience.
[SPK]
No Lee, you have to account for all possible forms of witnesses, not
just human ones! Please take a look at the following:
http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9512/9512008.pdf
http://arxiv.org/abs/cond-mat/0202082
http://arxiv.org/abs/quant-ph/9704028
http://www.dops.dk/pictures/pdf/archive/1999/99_4_art1.pdf
http://prola.aps.org/abstract/PRA/v41/i5/p2295_1
http://prola.aps.org/abstract/PRL/v84/i1/p59_1
This point is highly counter intuitive, explaining why it might not have
come out in this thread so far, but the null and "nondemolition"
measurements arre very much a reality, having been experimentaly found and
replicated.
>> [SPK]
>> How can a OM encode a trace of other OMs such that it
>> can capture the notion of "remembering something"?
> [LC]
> Nature has given our brains a way, somehow. But remember,
> those do not need to be veridical OMs. In principle you
> can insert the memory of me having been elected president
> into my brain. Doesn't make it so at all.
[SPK]
You are just restateting of the COMP thesis, consiousness' content in
hardware independent. Be careful what you assume.
> [LC]
> (Yes, yes, yes, I know: in some universe I *was* elected
> president. But truly, that's irrelevant.)
[SPK]
And in another, you where tortured to death, we are assuming *all*
possible OMs related to Lee Cordin exist a priori. It seems that we only
disallow OMs that include aspects that would contradict Lee Corbin's
existence. Those would code some other entities OM, maybe. ;-)
>> [SPK]
>>
>> What, exactly makes me "me"? Is it "I am what I
>> remember myself to be"?
> [LC]
> I would say yes. If we got hold of Napoleon's memories and
> inserted them in your brain while discarding SPK's original
> ones, then Napoleon would be alive and you'd be dead.
[SPK]
My wife might not notice a difference but my mother would! Here the role
of entities that have OMs of their own and continuity requirements MUST be
taken into account, thus the idea of witnesses can not be dismissed!
>> [SPK]
>> I don't have any problem believing that I *am* all of my duplicates
>> either, so long as there is some way that the "me" that I remember now
>> could
>> be smoothly continued within the world of a duplicate *and* all witnesses
>> agree that I did all that I claim I did with in that world.
> [LC]
> Like I said, witnesses aren't important, and are expendable. Why
> do you care about witnesses? Besides, the quality of your life
> over the next few minutes is not affected by whether or not it
> is true that God created the world 30 seconds ago (and so you
> never had any experiences, and even your mother and father are
> not reliable witnesses).
[SPK]
I hope that I have dispelled that incorrect assumption. If not, we can
discuss it further. ;-)
snip
Kindest regards,
Stephen
Received on Tue Jun 28 2005 - 23:37:00 PDT