From: Greg Perkins (gregp@mit.edu)
Date: Tue Mar 03 2009 - 10:36:07 MST
On Tue, Mar 3, 2009 at 10:50 AM, John K Clark wrote:
> On Tue, 3 Mar 2009 "Stuart Armstrong" said:
>
>> if perfect copying or uploading was around, this is correct.
>> But until that happens, we must obsses about the original!
>>
> There can only be one answer, atoms. So you must also obsess about all
> those sacrosanct little atoms you pissed down the toilet over the years
> since you were born.
>
No! There are other answers, see Johnicholas' list.
Bryan Bishop wrote: "Well, not necessarily just the atoms, but also the
state space, which to some extent is unretrievable, i.e. see Boltzmann,
thermodynamics and so on."
The answer that most are actually worried about is time continuity of
the pattern that resides within in the activation states of the brain's
neurons (in the *arrangement* of the matter's state space). That does
not seem to be too terribly dependent on specific atoms. However it does
seem to need to be kept running on the hardware rather continuously to
qualify as the same self. As evidenced by coma victims often having
seriously changed personalities.
And that seems _unbelievably_ hard to do quickly enough. I think if the
transition was essentially transparent between the old brain and the new
brain, to the point that thought processes were indeed continuing in
such a way that their hardware was not distinguishable, then it would be
an acceptable transition. I mean to say that if a powerful enough
simulation of the time evolution of the matter's state space were
ongoing throughout the process, I see no reason this would cause any
alarm. It happens every night and day.
I think the discussion of this hypothetical teleportation is rather
worthless without the caveat that somehow this pattern of brain activity
(and maybe body activity while we're at it, though that's likely a less
pressing concern since it's so much slower) is simulated during the
building of the new body in a different spatial location. I'd be willing
to accept teleportation in this case personally; we have shown that we
are not specifically tied to the underlying atoms themselves, but we do
seem to need time continuity...
Charles Hixson wrote, "I see "me" as basically an artifact of memory."
Yes, Yes, and Yes. This is the key. As we start out at the star-crashing
level, we have frustratingly little evidence of the ability of the
system to remember us. As we approach human society, then our families,
then our body and mind, we remember more and more. This memory is, I
believe, what people associate with their selves, in general.
I believe it is important to consider this on a gradient. In some way,
yes, I feel that my self is my light cone and so my potential self is
little different from the physical universe itself.
Johnicholas Hines wrote:
> 4. Your self after a radical atom-flushing treatment replaces all of
> the atoms in your body.
>
I think this gets very nicely to the question at hand... this seems to
be the end of the societally important human self. 5 is on the border,
in that our society does recognize it as continuous, but most people
would find it a hollow shell of the former. The rest are not societally
protected or at all universally accepted.
Johnicholas Hines wrote:
> 5. Your self after a stroke that causes you to change in knowledge or emotions.
> 6. Your clone, grown from your DNA.
> 7. Your self after a genome-editing virus transforms you into
> (technically) a different species - mutually infertile with unedited
> humans, but otherwise identical.
> 8. Your life's work, a body of text that might be republished and
> reread by college students for centuries.
> 9. Your children.
> 10. A high-quality simulation of you.
> 11. Your self after surgery on your brain to change your personality.
> 12. All of your future light cone.
And the only one that I'm willing to argue with the societally accepted
position on is #10: if our cell states are not tied to specific atoms,
it seems a long shot that our brain state would be somehow tied to the
specific cells (though the internal cell organization and time
continuity do both seem to be important).
The critical caveat is that we cannot be confident of our
self-continuity past any time discontinuity in our brain's patterns:
that is what we must maintain across any hardware transitions, to
guarantee the continuity of memory, the foundation of our self.
- Greg Perkins
ps: Sorry if it's a bit hard to read, lots of quoting, etc. Just
catching up on reading now.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:01:04 MDT