From: Keith Henson (email@example.com)
Date: Tue 22 Mar 2005 - 05:14:33 GMT
At 11:28 AM 21/03/05 +0000, you wrote:
>I'd like to thank Scott for his cogent follow ups -- he gets this. It is
>maybe too subtle for you Keith, as your from-the-hip (in sooo may ways)
>opening comment betrays.
> > Consider the mechanism you used for this post Chris.
>For propagating _email_ not for replicating, in your head, the firing
>patterns of groups of neurons in my head!
Please show me where I have *ever* made a claim that replicating memes
requires replicating the firing pattern of neurons.
>Anyway, allow me to unpack this. Point is that there are no memes in the
>way that there are genes -- genes are (more or less) discrete units that
>exist independently in the world. I could track particular sequences
>through generations (even, to a point, in terms of the actual individual
>physical nucleotides, give or take a few exchanged hydrogens). There is no
>direct analogue in memetics. Not one idea anyone has stored in their head
>is _exactly_ the same (in all but the physical stuff) as anyone else's;
>whereas that is not true of genes -- there is sequence variation (within
>known, measurable limits) but there is also a lot of invariant sequence
>which is literally, exactly, demonstrably the same.
No gene copy (in base pairs) of substantial length is exactly the same
either. Not counting variable methylation, a copy is going to differ in
ways you could measure with nanotech tools in the distribution of carbon
isotopes (13 and 14).
Further in the sense of a gene being the information source for a protean,
lots of base pair substitution leads to the same outcome because of the
degeneracy of the DNA coding.
But this is the same kind of nitpicking you are trying to do on memes.
"In Xanadu did Kubla Khan
a stately pleasure dome decree
where Alph the sacred river ran
through caverns meaasureless to man
down to a sunless sea."
"In Xanadu did Kubla Khan
A stately pleasure-dome decree:
Where Alph, the sacred river, ran
Through caverns measureless to man
Down to a sunless sea."
Is the first, from memory and missing bits of punctuation, close enough to
the original Coleridge (which I just looked up) to be "the same"? If
"sameness" is doing a bit by bit comparison of the ASCII, no. If someone were reciting, yes. An observer who knew the poem would without any doubt recognize this as the start of Coleridge's famous poem.
>The reason I want to zoom in on this issue is to avoid lazy thinking; as I
>said metaphors are incredibly useful, but only up to a point. When are we
>going to get to grips with _what is actually happening_ and stop faffing
>about with black boxism? What sort of brain might evolve to facilitate
>this 'life of the mind' (cos to be sure even in a world with as many bad
>as good ideas learning still goes to fixation as a trait, as at least one
>paper has shown [in phil.trans.roy.soc.B iirc]).
>What I want to see is how we go from a (literal) muscle head, to an
>automaton, to an automaton with some conditioning, to a learning beast, to
>a 'thinking' beast.
Are you familiar with William Calvin? He might not have the correct model,
but I think his is the best on the subject to date. See
>I also want to address (for the umpteenth time) the linked fallacies (1)
>that 'we' have free will and (2) that there is any true randomness.
I don't have the slightest clue as to why you would drag this subject
in. But if you insist, try Google. You will get 1,250 hits for "Marvin
Minsky" "free will".
Typical (ranting against Minsky an early and major figure in AI)
"A third position recently taken against naturalism, and one of
the most interesting, is found in Marvin Minsky's The Society of Mind.
Minsky first admits that free will is an illusion, that chance and
causality alone account for our actions. But then, in a startling
display of double-think, he advises us against using this knowledge in
our everyday lives. Why? Because he feels that the concept of free
will is too deeply ingrained and too functionally irreplaceable to let
"'No matter that the physical world provides no room for freedom
of will: that concept is essential to our models of the mental realm.
Too much of our psychology is based on it for us to ever give it up.
We're virtually forced to maintain that belief, even though we know
>Maybe alt.memetics provide more comfort for a hollow para-paradigm?
>Scott was bang on on the examples anyway so I won't do more than precis
>that here; I can make a surface copy of a behaviour; I can try to assign
>intent; that is all. Your meme being created/lost whatever is
>inappropriate at any but the coarsest level; it may be the case that,
>given my internal state, I interpret an artefact in an _apparently_
>similar way (in terms of observables), but no 'informational soul' existed
>in that thing to be preserved or lost. What if I only thought I saw
>something that wasn't really there -- what then? Why should someone else's
>intent govern the role of an artefact anyway (PC = doorstop)?
>We all learn to use our hands by having hands that are connected in the
>same sort of generic way, and develop discrete behaviours around our hands
>(artefact makes 'meme' -- see Karl Sims' sims), but even then they are not
>going to be _exactly_ the same. Surface similarity is all that matters at
>the end of the day. It is once we get around the idea that we _literally_
>share stuff when 'copying', rather than converging on a perceived
>phenotype, that we can for example very simply explain the vertical
>transmission of familial abuse, or the difference between flaky free
>thinkers (including me -- although as a long-time cat owner I may just
>have toxoplasmosis) and bookish dullards with total recall, or why a
>'small' loss of ability in Asperger's seems so often to result in such
>disproportionate gains in specific areas.
>Show me the money.
Sheesh. Keith Henson
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
This archive was generated by hypermail 2.1.5 : Tue 22 Mar 2005 - 05:33:31 GMT