Re: empirical "memetics"

From: John Wilkins (wilkins@wehi.EDU.AU)
Date: Wed Sep 20 2000 - 04:01:26 BST

  • Next message: Scott Chase: "Re: empirical "memetics""

    Received: by alpheratz.cpm.aca.mmu.ac.uk id EAA08067 (8.6.9/5.3[ref pg@gmsl.co.uk] for cpm.aca.mmu.ac.uk from fmb-majordomo@mmu.ac.uk); Wed, 20 Sep 2000 04:03:57 +0100
    Date: Wed, 20 Sep 2000 14:01:26 +1100
    From: John Wilkins <wilkins@wehi.EDU.AU>
    Subject: Re: empirical "memetics"
    To: memetics@mmu.ac.uk
    In-Reply-To: <B5ED7EB3.49AF%bbenzon@mindspring.com>
    Message-ID: <MailDrop1.2d7j-PPC.1000920140126@mac463.wehi.edu.au>
    X-Authenticated: <wilkins@wehiz.wehi.edu.au>
    Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
    Sender: fmb-majordomo@mmu.ac.uk
    Precedence: bulk
    Reply-To: memetics@mmu.ac.uk
    

    On Tue, 19 Sep 2000 21:33:43 -0400 bbenzon@mindspring.com (William
    Benzon) wrote:

    >> From: John Wilkins <wilkins@wehi.EDU.AU>
    >> Reply-To: memetics@mmu.ac.uk
    >> Date: Wed, 20 Sep 2000 09:52:32 +1100
    >> To: memetics@mmu.ac.uk
    >> Subject: Re: empirical "memetics"
    >>
    >[snip
    >
    >>>
    >>> And I ask this because, dammit, I don't see it. Improvements and
    >>> alterations are not necessarily evolutions, IMHO.
    >>
    >> This is the reason why I am doing my PhD on species concepts rather
    >than
    >> on cultural concepts - and my answer is that culture speciates when
    >> traditions split. When the making of automobiles is no longer taught
    >as
    >> part of horse carriage making, then automotive engineering has become
    >a
    >> separate tradition and thus must be classified as a distinct lineage.
    >>
    >> As Hull has argued, the fundamental ontological category in evolution
    >is
    >> the lineage.
    >
    >Very interesting. That certainly makes sense in the biological case.
    >Though what does Hull do about hybrids? I've read, here and there, for
    >example, that a quarter of all naturally-occurring plant species are
    >hybrids.
    >
    >I'm not quite sure what it implies for culture. The fact is, cultural
    >lineages are very leaky. In the extreme, we have creolization in
    >language,
    >where a new language emerges from two or more parent languages within
    >two
    >human generations. Assigning the new language to any of the parent
    >lineages
    >is somewhat arbitrary and obscures significant relationships.
    >[Comparative
    >and historical linguistics pretends that linguistics lineages only
    >split,
    >but they deliberately adopt methods that are designed to produce this
    >result.]
    >
    >The same thing happens in other cultural domains. I'm particularly
    >interested in jazz music, for example. It doesn't make sense to assign
    >it
    >either to a lineage originating in Africa or one originating in Europe,
    >but
    >it has ancestors in both places.

    Indeed, I thought it would be the major disanalogy between memetics
    (particularly Hull's work - BTW: has everybody seen the June 2000 issue
    of Biology and Philosophy, with a retrospect on Hull's work?) and
    biology. Then I started to investigate it.

    Reticulation, as it is called, is a major problem for all phylogenetic
    systematics. There is a pretty simple reason for this - when a lineage
    joins rather than splits to form a hybrid, information is lost. Unless
    either the original lineages remain intact or there are splitting
    (articulating) characters (such as LINES or terminal repeats), the
    history is just lost, and you cannot tell the difference between
    convergence (homoplasy), noise, and reticulation.

    It transpires that around 40% of the average mammalian genome is
    reticulate, in the form of endogenous retroviral insertions. Usually
    these are indels and not expressed, but they are there. Somewhere
    between 45% for angiosperms and 95% for gymnosperms (ferns and fern
    allies) are hybrids, and this has casued specialists in that field to
    introduce a range of species concepts to deal with them, the most common
    being "nothospecies", due to WH Wagner (of Wagner Groundplan Divergence
    fame). Another concept is the "compliospecies" where genes from one
    species are "plundered" by another, even though they remain distinct
    lineages.

    Then you have the work of Carl Woese and Ford Doolittle, who are both
    arguing, following the original work of Lyn Margulis and Dorion Sagan,
    that major cross-phylum organelle and gene exchange is responsible for a
    lor of phylogenetic history. An early horizontal transferist is Mike
    Syvanen.

    So it begins to look like the disanalogy is not so great after all. In
    fact, comparing bacterial evolution to cultural, culture is *more*
    articulate than biology. Bacteria (not a monophyletic group, by the way,
    but let it pass, please) exchange genes in the form of small rings of
    DNA called plasmids. When a bacterium dies, it disintegrates in a
    proccess called lysation, and its plasmids float into the medium and are
    often taken up by other bacteria (which need not even be of the same
    family). There are often functional in the new host.

    So the major problem for horizontal transfer is one of diagnosis, and
    there I have no magic wand. If we are to recover history we have to make
    use of articulate patterns in datasets (that is, in hierarchically
    arrayed patterns - groups under groups, in Darwin's words) but if the
    history is reticulate, then we lose that information. The best we can
    hope for is that most change over time involves articulate trees. In
    biology *or* culture.

    --
    

    John Wilkins, Head, Graphic Production The Walter and Eliza Hall Institute of Medical Research Melbourne, Australia <mailto:wilkins@WEHI.EDU.AU> <http://www.users.bigpond.com/thewilkins/darwiniana.html> Homo homini aut deus aut lupus - Erasmus of Rotterdam

    =============================================================== 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



    This archive was generated by hypermail 2b29 : Wed Sep 20 2000 - 04:05:13 BST