X-Message-Number: 16113
Date: Mon, 23 Apr 2001 20:48:01 -0700
From: Lee Corbin <>
Subject: Re: space and time

Robert Ettinger in #16100 once again rakes those of us over
the coals who subscribe to the information theory of identity
about our over-reliance upon the concept of "isomorphism".
He gleefully points out that we lack an explanation as to why
isomorphism over time appears to be able to support conscious
activity, but isomorphism across space appears not to.  (Hint
for what follows: so does he.)

To bring these lofty ideas down to where it's easy for me to
think about them, we can translate the foregoing as follows:
In terms of physics, your experience over the last thirty
seconds is described as a fantastic number of state transitions
of an object built of molecules (namely you).  That is, we
think of your brain and body as a physical assemblege of atoms
all obeying physical law.  The isomorphism comes in when we
insist that the later instants of you (say approaching the
thirty-second boundary) are extremely similar to earlier
instances of you, say near the first few of the thirty seconds.
In fact, your whole development during that thirty seconds is
understood, physically speaking, as an almost continuous process.

But whether this process really is continuous or not, or 
whether it must be described as abrupt transitions from 
one of the 10^10^45 possible states that any human could
be in to another of one of the 10^10^45 possible states
that a human could be in, is, I believe, from the point
of view of wanting to understand personal identity, moot.
In fact, I think that it was a tactical error to have 
ever used quantum mechanics to try to explain to anybody
anything about personal identity.

Okay, now in the previous paragraph I described what isomorphism
over time is, more or less, and why you and your last thirty
seconds are a good example of isomorphism over time being able
to support a conscious process.  (You were conscious, weren't
you, or has this thread actually knocked you out?)

But if we try to achieve the same thing across space---rather
than time---we get what I have called "a succession of frozen
states".  It may be that we undergo about 10^54 state transitions
per several seconds (the number that Tipler gives as a maximum),
but he also allows that it may be quite a bit smaller.  For the
sake of concreteness, suppose that you're in a trillion different
states over the course of the last five minutes.  Then to display
a space-isomorphism, we line up, in a row, in deep space one
trillion plaster cast molds of your head that each correspond
in total detail to one of those states that you were in.  Ah,
good, frozen heads---it's getting graphic, so wake up!

Well, as we move past this assemblege of heads at the proper
velocity v, we have a striking re-enactment of your last five
minutes!  We see your lips move, your eyes dart from side to
side, and various expressions cross your face, at the incredible
resolution of a million frames per second.  Not only that, but
the neuron potentials of your (clay) neurons undergo exactly
the same changes that your real neurons did a while ago.

Yet we must regard the frozen heads as completely inanimate.
(In cryonics, we rightly regard frozen heads as inanimate,

and hope that we shall in the future be able to restore life
and movement to them.  But presently, they are not living.)
No matter how vividly they portray your last few minutes' 
experiences, it's only a portrayal---nothing more.
No one is having any actual experiences.

Now I have for twenty years been captivated by this Problem
of the Succession of Frozen States.  It demands an explanation
as to why!  I posted the following on the Fountain_Society
(Uploading Discussion Group) on February 13:

  I do know of two "refutations" of the notion that this SFS should
  be regarded as a conscious emulation.  One is reductio ad absurdum:
  they might as well be scattered throughout space as in a row, and
  then might as well overlap, or be spread very thin.  (This idea is
  exploited in Greg Egan's novel "Permutation City" to great effect.)
  Since we are not accustomed to thinking of patterns of dust in the
  universe as conscious, therefore the SFS should not be regarded as
  conscious.  But the problem with this solution, as eminently practical
  as it is, is that the replacement of the time direction by a space
  direction doesn't on the face of it seem to be an incorrect step;
  an explanation is still lacking for why it is wrong, exactly.

  The second "refutation" is that the SFS doesn't achieve information
  flow.  Normally, a brain evolves from one state to another by means
  of ordinary causal processes that shift information from prior states
  to later states.  Apparently in the SFS, such information flow is
  lacking, since the frozen states are not causally connected.  (As
  an aside, one can usefully cast this problem in terms of conscious
  states on a Life board, i.e., generations of patterns as defined in
  Conway's Game of Life.  Here also the information flow from one
  generation to another is unmistakeable.)   The problem with this
  explanation, though, is that it relies upon the concepts of 
  "causality" and "flow";  the former is highly suspicious in terms

  of our best physical theories, and the latter is just a mask for
  the concept of "time", and so has the same flaw as the first 
  explanation I gave above.

  Does anyone have a good explanation for why time should support
  consciousness, but another spatial dimension doesn't appear to?

There were no responses.

Okay, so Robert Ettinger thinks that this is the Achilles Heel
of the information theory of identity.  He somehow thinks that
just because no one has (to our knowledge) explained this must
imply that all bets are off as far as you and your duplicates
go.  It's not even fair to lay this paradox at the feel of
functionalists in general, to say nothing about identity
theorists.

There are a number of thought experiments that are persuasive
that one ought to regard one's duplicates as one's self.  None
of them rely on any claims about isomorphism!

When driving your car, you brake when you want to slow down.
Because of isomorphism?  Because you think that the quantum
mechanical version of you that finally arrives at the stop-
sign is isomorphic (time-isomorphism) with you the way that
you are when you're putting on the brak?  (It's talk like
this that gives philosophy a bad name.)  No!  It's because
you anticipate waiting quietly at the stop sign for traffic
to clear, if you do brake, and getting creamed by a truck
if you don't!

The other ongoing threads about identity at this time, and in
previous debates, mostly focus on what we should anticipate,
upon being uploaded or upon being duplicated.  I don't think
that the time-isomorphism vs. space-isomorphism paradox is
all that relevant.


Lee Corbin

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=16113