On Thu, 17 Oct 2002, Nick Bostrom wrote:
> I'm want to chicken out on this and postpone getting into the issue in
> depth until we possibly get to it later in the FAQ. There is a vast
> academic literature on what constitutes a person and the conditions for
> preservation of personal identity. For example, one could start by reading
I think we're facing much larger issues with the conservation of identity
during the upload process. Once we're past that, and define idenity as a
process (deterministic, spatiotemporal pattern of the execution of an
image) things are very much clear. Identical input, synchronous execution,
same pattern, same person. (Ditto for not identical input, and trajectory
forcing). Differing input results in bifurcation and exponential
amplification of initial differences, two persons.
However, when we're abstracting select features out of a large, complex
biological system (destroying it in the process) and destill it into a new
representation, the result is a lot more difficult to swallow. Turing-like
testing is not really convincing to the external observers, and one
typically can't compare the behaviour even if the original has not been
destroyed, because in absence of trajectory forcing it would give rise to
two persons.
So this boils down to a question of faith, applied to a gedanken. It's
pretty much impossible to achieve consensus here. I'm definitely voting
for listing all dissenting views, with commentaries maybe.
> Derek Parfit on personal identity (he draws a distinction between
> preservation of personal identity and survival), and Harry Frankfurt on the
> concept of a person, and then read their critics and the debates that have
> arisen around their work. (I'm not aware of any peer-reviewed publication
> by Kurzweil on this topic, so he might not be the best place to start.)
Received on Fri Oct 18 03:40:00 2002
This archive was generated by hypermail 2.1.8 : Mon Sep 08 2003 - 11:37:39 CEST