[ExI] Probability is in the Mind

Lee Corbin lcorbin at rawbw.com
Sat Mar 15 03:39:38 UTC 2008


Jef writes

----- Original Message ----- 
From: "Jef Allbright" <jef at jefallbright.net>
Sent: Wednesday, March 12, 2008 11:21 AM

> Lee wrote:
> 
>>    One ultimate purpose of philosophy, and I argue the most important
>>    one, is be prescriptive. Philosophy most vitally---for me and for many
>>    others---should instruct us about what actions to take and what
>>    decisions to make.
>>    ...
>>    Ultimately, again, I want to know what actions I should take and
>>    what I can expect the different outcomes to be like.
> 
> My interest in philosophy is pragmatic, but our key difference is that
> I emphasize the importance of awareness of an always broader (and
> ultimately uncertain) context, while you characteristically use terms
> like "the simple truth" 

That's almost totally false. I emphasize---and have emphasized ever
since 1972  Eternal Truth No. 1:  "Nothing is simple".  If I have
used that phrase, "the simple truth is..." and I probably have, it is
as a figure of speech, nothing more.  E.g., I wanted suggest a simpler
conjecture than the proposed one or the one under discussion.

> and strive to convey models in "objective" terms.

You're dead on there!  :-)   Please criticize this further, if you'd like to,
and have the time.

> I emphasize the importance of increasing coherence over increasing context.
> You demonstrate that coherence tends naturally to increase inversely
> with context.

Interesting. I admit that coherence *does* naturally tend to decrease with
increasing context. (That's one reason for Eternal Truth No. 1.)  But your
goal---if that's not too strong a word---of increasing coherence along with
(or "over") context sounds praiseworthy.

> Note that coherence applies only to the map, not the territory.

I think I understand you.  I agree with this: "the territory" is reality
and can't properly be deemed coherent or incoherent, though I
do steadfastly maintain that the "the territory" (i.e. reality) is not
simple.  I concede that "simple" here does drag in epistemological
concerns, part of your main point here, I think.

> Within a fully defined context (Peano arithmetic, predicate logic,
> basic game theory...), your approach is beyond reproach.  However,
> within an increasingly uncertain context -- as we find ourselves to be
> as necessarily subjective agents within a world of accelerating change
> -- our emphasis should shift away from hopes of an increasingly
> accurate model of expected consequences, toward increasing effective
> modeling in principle (to be applied at each moment of decision.)

A provocative claim. I admit to striving vigorously---so long as it is possible
---for increasingly accurate models. I think that that's more or less the
paradigm of science.  Think so?  Anyway, I'm not really sure of the fine
point you appear to be making about the difference, i.e., striving towards
increasingly effective modeling instead of striving for increasingly accurate
models.  (And pardon me, sincerely, if I have misrepresented you there.)

[Pseudo-nasty ponder:  Could Jef continue to write consistently,
coherently, and understandably if he were forbidden use of the
word "increasingly"?  Would, or could he ever find paraphrases
or synonyms?] 

> The game itself is evolving (it always was, but we hadn't really
> noticed) and it's becoming increasingly important not only to act
> effectively, but to adapt increasingly effectively in order to act
> increasingly effectively.  Wash, rinse, repeat.

I.e., to adapt with increasing effectiveness.  I agree, if you mean
that things are changing more and more quickly culturally,
scientifically, technologically and so on.  Yes.

>>  And so I prefer to pose situations that call for decisions, for
>>  actions.
> 
> The first several years of my professional life I was a successful,
> even gifted, troubleshooter of scientific instruments....
> 
> ...And I found that some of my most talented engineers were blind
> to the importance of context.  Some would brilliantly address
> electro-mechanical problems within the equipment but fail to perceive
> the importance of the environment of the lab (temperature, humidity,
> vibration, fluctuating power or ground loops...)  Most did recognize
> the importance of the *physical* environment but then failed to
> recognize that the customer doesn't really care about the equipment,
> but rather the technical solutions they provide. Very few realized
> that in the larger context both equipment and solutions were
> subordinate to the customer's (mere) perception.

Thanks, a good anecdote with an important point. 

> The point here is that the hard-core engineer was typically set firmly
> in his belief that he was acting most decisively in optimizing
> objective performance of the instrument.  Anything else was
> wishy-washy, vague and abstract.  But I would much rather have (and
> pay) engineers whose primary value was to achieve customer
> satisfaction, and would then act (including optimizing technical
> solutions and equipment performance) as appropriate.

Yes, just so.  I do find some resonance between what you are saying
with the way that people tend to apply abstract principles at the
expense of recognizing the requirements of real, practical situations.
The Libertarians are a beautiful case in point. Like Ayn Rand before
them (but gratefully not quite so extreme, most of them) they would
settle on certain *principles* and completely ignore context. As I
have said before, their writings often seem to assume that every
human being is a white, middle-class individualist in the American
tradition. In exasperation, I would sometimes wonder just how
much they'd read about other people, other races, and other
cultures. (Sorry for the digression.)

>>  Otherwise, the terminological confusions and irreconcilable
>>  philosophies may be simply moot.
> 
> <snipped examples of specific thought experiments>
> 
>>  But unless you delineate very exact, sharp *decisions* and
>>  *actions* in scenarios that serve to distinguish your beliefs
>>  from everyone else---and put less emphasis on the abstract
>>  hard-to-follow verbiage---then you won't be facing up to
>>  the true challenge, namely, prescribing courses of action.
> 
> Lee, you discovered many years ago that, contrary to popular
> individual and cultural biases, it is incoherent to conceive of
> personal identity as dependent on any particular physical embodiment,
> nor any "discrete trajectory" of its constituent particles or
> processes, nor any requirement of a discrete self, tangible or not.

Well, yes, but that's putting it a bit too strongly. Some people
indeed can consistently and ingeniously defend identity as dependent
on a particular physical embodiment, and other defend other variations
we oppose. I surmise that the cost is really only a certain amount of
"awkwardness", or perhaps having to defend sentences whose words
contrast starkly with the words' usual meanings.

> In short, you arrived at a functionalist account of personal identity
> defined in terms of a particular (dynamic) pattern of information.
> 
> Your model is an improvement, coherent and extensible over a context
> broader than common-sense, and all the more satisfying and seductive
> since it appears to offer newfound potential for extended survival of
> the self.

Thank you.  :-)

> Your view posits exact identity of duplicates at T=0, with identity
> then diminishing as some function of diverging pattern/functional
> similarity due to accumulating change brought about by separate paths
> of interaction with the world.  This is delightful, because it means
> (with logical certainty) that sufficiently (?) similar copies,
> regardless of displacement in place or time, must be considered
> (existing/surviving) identical persons.

Ahem, I thought it was nice too  :-)

> But just as the popular common-sense view fails when extended while
> yours succeeds, your view fails when extended only to be superseded
> by a model coherent within an even broader context.

That could be possible!  Although, again, "fails" is too strong a word.
At most, I can be forced to support certain awkward conclusions.
For example, since 1981 I've been at loggerheads with myself over
whether a succession of frozen states can be conscious. A friend and
I thought that we solved it in 1988. But the solution was, and is,
costly:  there are several questions which are exceedingly difficult
for me to field. But all I can say, even after my pathetic plea at
the end of www.leecorbin.com/FSF.html is "can anyone help?". 

> Because personal identity is not a property of any object.
> Personal identity is entirely a function of the observer. Even
> when the subject *is* the observer.

Naturally, I hotly dispute both those claims, as you know.

> And personal identity based on pattern/function similarity is merely a
> narrow special case of personal identity based on perceived agency.

Which I also think---at this time---to be not far short of nonsense.

> Personal identity based on perceived agency encompasses and
> accommodates the preservation of identity over the lifetime of a
> person, regardless of any degree of physical/functional change
> due to aging, disfigurement, body-morphing or enhancement,
> memory-alteration,...to the extent that the agent is perceived as
> acting on behalf of a particular entity.

I do remember our earlier discussions concerning this point, though
not as well as I would like. 

> Personal identity based on perceived agency allows for special-purpose
> replicants, each a valid instance of "self" to the extent that they
> act as an agent of a particular entity, regardless of physical (or
> virtual) form or function.
> 
> Personal  identity based on perceived agency resolves the obvious and
> immediate paradox of self in direct conflict with self, for instance as a
> consequence of duplicates stepping out of the duplication machine
> with both intent on control/enjoyment of the same property, spouse,
> etc.
> 
> Personal identity based on perceived agency accommodates the full
> range of moral/legal/social responsibility and is extensible, without
> the obvious contradictions of a pattern/functional similarity where,
> for instance Lee1 is held responsible for the violent actions of his
> near duplicate Lee2 who went a little crazy while working out in the
> asteroid mines.  [Of course, I retort that our laws must evolve to
> rationally deal with the reality here---we easily see in your not-
> very-good example that there was an explanation for Lee2's 
> behavior, and that it seems very questionable to try to "blame"
> Lee1 for any of it.]
> 
> Of course this has all been discussed before and in more detail,
> available in the archives.

Yes it has.  But that review was good. That's why I didn't snip it
in this reply.

> So Lee, a theory of personal identity more coherent over greater
> context is available to you.  But as the believer in the popular
> common-sense view must give up an attachment to belief in a discrete
> self, you must give up an attachment to belief in an ontological self.

Never!  I'll die first.  (Literally.)  But of course  :-) , I really am open
to criticism and even crave it, and so my mind is open, (or at least
as open as most of us are capable of so-imagining our minds to be).

> And everything will be just the same, but with a more coherent model,
> supporting more effective action.

But what I still find crucially missing is that "your philosophy" perhaps
makes no difference. What I mean is, what if there is no actual difference
between two given philosophical theories?  That, unsuspected by their
proponents, the two are isomorphic given mere terminological and even
slightly semantic substitutions?  How does one know that doctrine A
is not really just another way of saying doctrine B?  As you know, the
history of philosophy is littered with quite a number of entirely
consequential-less arguments. Changing a coordinate system doesn't
affect "the territory" at all.

You seem to shy away from describing how your views would make
any *practical* difference. In other words, just how does someone 
act if they subscribe to your view and not mine? Not coincidentally,
I fear, you also shy away in many instances of giving sharp answers
to sharp scenarios.  And no, it won't do (as I'm sure you realize) that
as a consequence people will merely judge 80 year old Aunt Mary to
be a "different person" than 40 year old Aunt Mary. Who really cares
what "different person" actually means unless it has consequences for
how we would act, and decisions we would make?

Lee

P.S.  Jef, thank you for the clearest post I can ever remember reading
from you. It was a pleasure to follow. I never had to struggle with a
sentence (well, maybe one or two) trying to conjecture what you
meant.  And in the cases where ambiguity did persist, you followed
it up with another sentence which resolved it. Finally, you provided
a very good example to *illustrate* what you were saying, giving 
not only me, but any reader a "warm fuzzy" that he was understanding
you.




More information about the extropy-chat mailing list