Date: Sun, 15 Jun 1997 10:58:06 +1000 (EST)
From: Chris Cleirigh <cleirig@speech.usyd.edu.au>
Message-Id: <199706150058.KAA27415@fortis.speech.usyd.edu.au>
To: memetics@mmu.ac.uk
Subject: Memes & Language
Mark,
Thanks for your useful postings of late.
I have a few comments.
Mark Mills wrote:
>Chris,
>
>bbenzon> its the physical stuff... which is the memes..
>
>cleirig>>Does this view stand up when applied to language?
>>Ink patterns on paper are memes, and their cultural phenotypes
>>are "the mental things in the brain"?
>>How does language emerge on this model...
>>phenotype first, for example?
>
>There is only a problem if one assumes 'only memetic' phenotypes exist in
>the brain. I see no reason they can't exist in both the brain and the
>environment. The important point is they need to be 'physical stuff' in
>both places.
>
Comment:
In applying the meme hypothesis to language, I find it useful to distinguish
between brain-situated memotypes and externalised meme phenotyes. No
argument from me about the need to be 'physical stuff' in both places,
with the rider that, in using Edelman's Theory of Neuronal Group Selection
as my brain model, I equate memes (as replicators/replicateds, pace Mario)
with the categorising *activity* of global mappings of locally selected
neuronal groups.
Mark Mills wrote:
>In my model, 'brain memes' are the patterns of memory proteins deposited in
>brain synapses. I suspect they provide a service analogous to 'linkage
>weights' in computational neural networks. 'Artifact memes' are the
>patterns of ink spots on paper.
Comment:
In my Dawkins-Edelman model of language, linguistic artefacts such as
patterns of ink spots on paper, like speech sounds, are extended phenotypes
of linguistic memes. Unextended phenotypes are perceivable behaviours
that express linguistic memes, such as signing in sign languages.
Mark Mills wrote:
>Both 'brain' and 'artifact' memes are much older than human history. The
>key thing is that in both cases, they are made of physical stuff. The
>vibrations in the air we call 'speech' are not memes, but phenotypes. They
>have no physical form. Examples of 'artifact' memes in the animal kingdom
>are well described by Bonner (though he doesn't use the term meme himself).
Comment:
But you would agree that linguistic memes and their artefacts are not older
than human *pre*history? I agree that vibrations in the air we call 'speech'
are (extended) phenotypes of memes, but would add that they can be described
by physics, whether or not they are regarded as physical form.
Since you have twice stressed 'physical stuff', you give me the impression
that you ascribe a 'nonphysical' position on memes to me. Perhaps, I can
clarify my position by pointing to Tim Perper's misinterpretation of a
previous posting of mine.
Remember that I said that I thought that memes (as memotypes) only existed in
neural systems, rather than anywhere else, because the "information" in meme
phenotypes is not information unless there are brains around to categorise it.
Tim Perper then wrote (inter alia) "But that is from the recipient's view.
and then went on to imply that I was advocating "solipsism of the Bishop
Berkeley variety" (see the log for the full text).
Tim's miscontrual revolved around two main confusions. One was that he thought
I was only talking about the recipients of information and not the sender.
But of course, the sender (here we are talking of humans) has a brain
and categorises phenomena with it before sending information. In cases of
information flow between non-human systems, this way of categorising
phenomena (as information flow) is itself information created and distributed
by the activity of brains.
His second confusion was to equate the categorisable phenomena (recognisable
events in one domain) with the human categorisations of them (recognition
events in neural domains -- which may be externalised as meme phenotypes).
This led him to think that I was saying that phenomena only exist in brains,
when what I was saying that categorisations of them only exist because of
brains.
My position is just that making sense of the world, and making sense of
making sense of the world are adaptive activities of organisms, that
knowledge is embodied. Phenomena, of course, exist independent of our
knowledge of them...that is why it is to our adaptive advantage to
categorise them reliably and pass that (memetic) information around
using sensorimotor systems.
Mark Mills wrote:
>Thus, I can explain language development out of 'mimetic' abilities made
>possible by 'brain memes.' These 'brain memes' record sound and provide a
>model for sound reproduction (phenotypes). A conceptual leap (unaddressed
>here) is required to explain the explosion of linguistic expression in
>'artifact memes' beginning 20,000 years ago, but that is a rate of change
>problem, not a framework problem. Both 'brain' and 'artifact' memes
>existed millions of years ago.
Comment:
I would say rather that the 'mimetic' abilities involved in the evolution
of language are made possible (inter alia) by brain organisation, especially
the correlation of conceptualising activity with sensorimotor repertoires,
which provides copying fidelity for concepts and allows linguistic memes
to be expressed as sound (extended phenotypes) and be replicated in other
brains where they may be mutated, recombined and selected in the construction
of future texts.
Thanks again Mark for all the thoughtful comments you have supplied.
Chris
===============================================================
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)
see: http://www.cpm.mmu.ac.uk/jom-emit