Re: physics of uploading minds.

From: Phillip Huggan (
Date: Sun Oct 30 2005 - 13:47:26 MST

One of the links Damian posted summarizes my position on consciousness nicely:
Minute imperfections shouldn't matter. But if consciousness is experienced as a field travelling through time, any disconnect or degradation of the field impinges on personal identity. I don't understand your second sentence. There is no such thing as a static mind, objectively or subjectively in character. Self identity requires the arrow of time and the is facilitated by quantum mechanics.
  An upload diverges from the original as soon as it is created, even if we assume a perfect level of copying resolution. The uploaded mind, once activated, will exert an EM field different from the original. That was the point of my (horribly vague) Rubik's cube analogy. Even if the upload appears to have a physically identical configuration, its field will be unique. Two seemingly identical photons exert separate fields, it is a fact of physics.
  MWI creates uploads in a way that preserves the continuity of one's identity in a way that is severed by upload attempts described to date. I would define identity as the continuity of the fields generated by one's brain areas responsible for memories, personality, etc. The continuity doesn't lie with one's past, but with one's future as some actions lie within one's ability to exercise free-will to control.

Michael Wilson <> wrote:
But this is beside the point; as several people have been saying, the
question isn't 'how can we perfectly copy physical objects, specifically
neurons', it's 'why should we /care/ if the imperfections are minute, far
below the threshold of a systematic effect on behaviour or indeed minute
in comparison to the general noise and chaos that biological brains
experience?'. Why should our standard of personhood, including self
identity, rely on perfect physical continuity when this continuity seems
completely seperable from every objective and subjective character of
a mind? This is why I was criticising your assumptions, not your
conclusions, and so far you haven't offered anything to strengthen those

> the uploaded mind's brain substrate will not be in any way attached
> to the original mind's brain.

...such as why an 'attachment' is necessary.

> The upload can still be sentient and his memories will cause him to
> think he is the original,

Until the two copies begin to diverge, why does substrate continuity
make one any more the 'original' than the other? It's physically
measureable, but why does it /matter/? Once they have diverged, why
does one have any more claim to the prior history and identity than
the other?

I would note that if the many-worlds interpretation is correct then
there are no 'originals'; you're not being copied constantly, you're
being /split/ constantly, at a level of detail many orders of magnitude
below anything that could be considered neural computation. This
cosmology may or may not be correct, but if you can't get your
philosophy of unique individuals to work under many-worlds then you're
fighting a losing battle to get it to work under any other form of

> The process describes how to maintain continuity of consciousness,
> not how to preserve identity... But personal identity emerges
> from matter

You keep talking about 'identity', but normally that means memories,
personality and assorted external trappings. Since you've accepted
that this things can be duplicated to beyond the threshold of
objective or subjective measurability, you must mean something else,
and it would help if you defined the manner of this 'emergence' (a
term used here in its typical application).

> like a Rubik's Cube can be solved if it were attached via tiny
> strings to billions of other points in the universe.

I'm afraid I don't understand this analogy; how does this work

* Michael Wilson

How much free photo storage do you get? Store your holiday
snaps for FREE with Yahoo! Photos

 Yahoo! FareChase - Search multiple travel sites in one click.

This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:52 MDT