[HST spent too long trying to get the astonishing fact of language
 evolution on the table as something which deserves a place in GA]

The inscrutability of the size of Universe is another thing that
defies comprehension.

If it's just chance that we happen to have come to be in a corner of
that Universe where recursive compositional expressive mechanisms are
possible the unlikelihood isn't an argument against it being the case.

Suppose I were to say to a high-school class we agree to have
recursively-specifiable syntactic and semantic language, and have a
brain that produces and consumes its product, and yes, it's unlikely
in the extreme, but so-be-it.

Is there anything to conclude from that.

Imagine the course we're teaching, and a typology of unlikelihoods,
smallness and small-enough-ness.  Very^4 unlikely that we'll gasp for
breath because the oxygen has all left the room.  Do we have to have a
view on whether there's a universe almost like ours where the same
UUID _does_ get minted for two different objects?  

What are we doing?  We want to...  MacIntyre says we need 'real'
language to reflect on/assess our beliefs and desires.  Evolved
corners of the universe that have the requisite properties that [this
process] can be enjoined are astonishingly unlikely, but we are
evidently in one.

What more do we need, in order to have intentionality.

[HST] MacIntyre grants (persistent of) categorisations
[BCS] Sufficient, or Necessary?  [HST] I'll take that as a homework assignment

[BCS] Suppose we have a minimal 'higher' animal: persistent individuals and
categorisation.  Do you have anything yet that reeks of the religious?
My intuition is 'no'.  The critter has to stand up on its hind legs
and _wonder_, in order for there to be... in order for its finitude,
one might say, to lead it to [HST ask Why?][ref. _Child's Christmas in
Wales]

MacIntyre on Davidson? Counterfactuals are crucial to practical
reasoning.  BCS says I could imagine a counterfactual toting rational
argumenter, who still lacks this openness to God and to wonder.

And until an AI system can wonder, I won't be particularly
interested/concerned, it will still be a mere machine.

Haugeland distinguishes the impossible from the possible but not
actual from the actual.

"A blue proton" gets you in to trouble.

With respect of our project, and the opening salvos of GA, there is a
need to characterise the territory that we want to open up, and
something has to be very open to not reduce to boring behaviourism.

The speaker has to be gripped by a certain openness, a certain
possibility, will this cake rise whatever.  How to open the space
enough for wonder to be genuinely gripping, seems important.

We want people to want to read this after the first page or two, even
if they can't say exactly why.