X-Message-Number: 7905
Date:  Thu, 20 Mar 97 12:44:20 
From: Mike Perry <>
Subject: Re #7889-7899

On Olaf Henny's problem:

He seems to think that there are two alternatives only: (1) you 
preserve only your CNS or brain, and you come back without your 
endocrine system, and you NEVER recover that, so you are deficient 
thereafter in emotions, etc. or (2) your endocrine system is 
preserved all along and restored along with the rest of you. It 
doesn't seem to occur that there is a third alternative, which is, 
that your endocrine system is not preserved, but is *recreated* from 
information (such as that found in DNA) that *is* preserved. Dolly the 
sheep has a complete (sheep) endocrine system--yet she certainly 
didn't start with one--and it is (presumably) the same, within close 
limits, as that of her "mother" from which she was cloned. So
why would it be essential to preserve the human endocrine system 
in cryopreservation? Why not just recreate it at the appropriate time?


Bob Ettinger says,

>Suppose feeling &
>subjectivity depend on a specific feature of brain anatomy/physiology,
>say (oversimplifying) a particular kind of resonance or standing wave
>in the electrochemistry of some part or aspect of the brain (whether
>local or distributed). Variations in this PHYSICAL  condition may
>CONSTITUTE our feelings of pleasure/pain etc. Certain things have to
>happen or exist in a small region of spacetime. It is entirely
>conceivable (and I suspect probably true) that the possible substrates
>for such are extremely limited--maybe even limited to organic
>systems....And finally, for the umpteenth time, to say that an
>emulation or model of the self circuit would be in some sense the same
>as, or as good as, the original is just an empty assertion, not a
>logical conclusion or demonstrated fact."

To me an ongoing computation that fully *describes* an ongoing
process is "as good" in some reasonable sense, as the
actual process. If a person 
were emulated in this way (ignoring for the moment
that this might have to be a *very* complex program, require a lot of 
computer time, space, etc., and be far beyond our present capability) 
I could converse with the entity being emulated through an 
appropriate hookup, and, done appropriately, it would seem in all 
respects like a person, so I would accept it as a person. (And note 
that being an emulation would open certain possibilities; it would no 
longer be necessary that certain elements of the process be in close 
spacetime proximity, so long as appropriate communication links were 
maintained.) But that wouldn't "prove" it was a person, or even
remotely conscious (whatever that should mean). Perhaps my
claim that this was a person, with feeling, consciousness, etc., would
forever stand as an "empty assertion, not a logical conclusion or
determined fact" to the dedicated doubter. 

One is reminded of the "day person" hypothesis, 
attributed to Thomas Nagel, that basically, we die 
each time we lose consciousness: the person that wakes up, though 
similar in so many ways to the one that fell asleep, is still a 
different being entirely! This would be handy for certain people: 
e.g. a bank robber could snooze, and say "*I* didn't do it!" And 
I suppose there are people who really feel this way, that you could 
never convince otherwise by any amount of argument. "The day person 
hypothesis," they might say, "fits all the experimental evidence you 
can muster, and it feels right to me, so I adopt it." As for Ockham's 
razor (where we try to minimize the amount of unnecessary hypotheses 
in our explanations of things) they might say, "Why make the extra 
assumption that there is this special 'identity' link between me when 
I just woke up and started my life, and this other being I may happen 
to know a lot about?" For me, on the other hand, there are certain 
hypotheses that seem to fit reality and also seem right at the gut 
level, that I adopt, one being that "an emulation, treated 
appropriately, is as good as the 'real' thing." (I could argue at greater 
length here, but let me go on before I doze off!)

Thomas Donaldson writes:

>For all practical purposes, Tipler is wrong. We do not even know if
>the universe is finite. A finite object, acted on by an infinite
>variety of inputs, will behave in an infinite variety of ways. Tipler
>has said some highly percipient things about the universe, but that
>just isn't one of them."

The question I've tried to address in referring to Tipler
is whether finite domains in the universe are, in 
effect, finite state machines. (I didn't try to address the 
question of whether the universe is finite or infinite. In fact 
Tipler argues it is inifinite--in total energy content though not 
spatial extent. That, exactly as he has stated,  may be questionable,
but I think the conclusions about finite subsets of the universe are on
firmer footing.) If a person, in effect is a finite-state machine, as
quantum theory seems to say, the machine itself might have to be
very large, but it can be that way if needed. We don't worry overmuch about
"practical" considerations here! And, if we can be immortal, as we 
hope, someday things will be"practical" to us that are not at all thay
way now!

As for the 
assertion that "A finite object, acted on by an infinite
variety of inputs, will behave in an infinite variety of ways," 
with a finite-state machine, there can't be an infinite variety of 
inputs--those are limited just like the number of states. The only 
way to violate this would be to invoke immortality--over infinite 
time, a person who is truly immortal cannot be described as a 
finite-state machine. (Such a machine exhibits the "Eternal Return" 
in which some finite state space is occupied forever--not true
immortality.) But this is over infinite time. So overall you could 
have a universe that is infinite but still "digital" in the sense 
that its finite subprocesses (bounded in energy content and spatial 
extent) behave as finite-state machines.

But the thrust of Thomas's arguments in his posting (not just what 
I've quoted) seems to be that the world is nondigital in some 
fundamental way--and when you confine yourself to digital processes, you 
lose some fundamental "essence"--so such processes could never 
exhibit feeling and awareness, no matter how complex they are. 
(Somebody correct me if this assessment is wrong!) It looks in 
fact as if Thomas is saying that digital process could not exhibit 
*any* feeling or awareness--qua digital processes at least--only
nondigital ones can do that. (Again 
correct me if I'm wrong.) On the other hand, if everything is digital,
even if it has to be very complex-digital, this argument is pretty
thoroughly overturned.

Peter Merel writes,

>Mike Perry writes,
>
>>Well, according to physics, every system bounded in spatial extent
>>and energy content is a finite-state machine (for an extended
>>discussion see Tipler, *The Physics of Immortality*, esp. pp. 20-44).
>>
>
>This was one of the parts of Tipler's book that I found most
>unconvincing. Every brand of quantum theory I've ever heard about
>involves universe-wide non-locality; perhaps I'm missing something
>obvious, but I can't see how non-locality and finite-state machines go
>together.

Many-worlds upholds locality. Effects that require a non-local 
explanation with other theories do not require such an explanation 
with many worlds. This depends on the fact that the correlations that 
seem to happen at superlight speed cannot be *verified* at superlight 
speed--they have to be verified by observers moving in more usual 
ways. If worlds can split, the propagating splits (also not traveling 
faster than light) can travel ahead of the verifying observers (who
also split of course) and "arrange" things
approprately in advance--so the apparently non-local correlations
are seen. (It might be objected 
too that many-worlds, even if it does preserve locality, introduces 
an additional complication--the splitting iself. So we have what are 
called "non-deterministic finite state machines"--still finite-state 
machines however, and also still "deterministic" in the sense that we 
always know what is going to happen.)

>Also, did I miss seeing it here, or am I the only one to have read in
>this week's New Scientist about the world's first functioning (3 Qbit)
>quantum computer?

I haven't seen this article yet, but as I've said, progress with the 
quantum computer tends to support many-worlds. 

Mike Perry

http://www.alcor.org

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=7905