IE versus *PIE

[Larry Trask]

--On Saturday, May 12, 2001 2:51 pm -0400, Robert Orr,
<colkitto at sprint.ca>, wrote:

> I think part of the problem is that family tree diagrams tend to be
> oversimplified.  A close reading of the relevant passages of Dixon's book
> shows that he seems to arguing for some sort of recognition  that family
> tree diagrams should be far more complex than normally presented.

In a manner of speaking, yes.  But Dixon's central point is that *apparent*
language families can arise by long-term convergence, by the steady
diffusion of features across language boundaries.  He argues that family
trees of the familiar type only come about in exceptional circumstances --
what he calls 'punctuations'.

> One analogy that we should consider is bushes which can sprout branches
> which later can fuse again, or can fuse with branches from other bushes.

And exactly this has been proposed by Malcolm Ross.  See especially this:

  M. Ross. 1997. 'Speech networks and kinds of speech-community event'. In
  Roger Blench and Matthew Spriggs (eds), Archaeology and Language I:
  Theoretical and Methodological Orientations, London: Routledge, pp.
  209-261.

> Of course, it may be impossible to reduce such complexity to a page, and
> linguists may have to discuss family trees in the same way that
> cartographers discuss Mercator's Projection (even Peter's may not entirely
> free of such problems.)

In fact, there has been quite a bit of lively discussion of alternatives to
the traditional family-tree model in the last few years.  I've put together
a little talk on the subject, which I hope to write up for publication one
day.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk

Tel: (01273)-678693 (from UK); +44-1273-678693 (from abroad)
Fax: (01273)-671320 (from UK); +44-1273-671320 (from abroad)