Dear Stephen,
When you say:

##
Advertising

> [...]
>We might not be able to know "what it is like to be a bat"
>but surely we could "know what it is like to be an ameoba"!
It is amusing because I describe often---for exemple my thesis
or http://www.escribe.com/science/theory/m3651.html--- my whole
work as an attempt to know what it is like to be an amoeba.
In my thesis I express myself exactly like that.
I am thinking for sure to a self-dividing amoeba, and that's what
has lead me to the comp indeterminacy.
Frankly if you know what it does look like to be an amoeba,
even in between self-divisions, you should try to describe it!
>Bruno, I am still not convinced that the statements that "If
>we are consistent machine we cannot know which machine we are"
>and "Godel's and Lob's incompleteness prevent us to identify
>any intuitive first person knowledge with objective third person
>communicable statements" mutes my question since it seems that
>it makes my predicament much worse! It seems that your idea
>prevents me from "knowing what it is like to be a bat" by not
>allowing me to have any 1-person certainty at all.
I don't see why. You can still have a lot of 1-person certainties.
It is just that 'knowing which machine you are' is not among them.
But you can keep the 1-certainty that 1+1 = 2, or that there is no
integers p and q such that p/q = sqrt(2), etc.
You can *bet* that you are well defined at such or such level of
description, but you cannot consistently assert you can prove
being well-defined at those levels. (A non-computationalist just
pretend that there are no such levels, not even the quantum one
because the quantum level is emulable classicaly).
All what follows from Godel in this setting is that you cannot
consistently ascribe univocally a well defined machine to your
1-person experience. Once you bet on a level, you can ascribe
to your experience an infinite vague sets of machines, though.
Those machines are going through an infinite vague set of
histories. I should have said perhaps that
you cannot know *precisely* which machine you are.
(Z1* should provide, by construction, the geometry of that vagueness).
Best Regards,
Bruno