No single ancestor - breath of fresh air

[Larry Trask]

Lloyd Anderson writes:

[on the family-tree model]

> Is there a better model, which captures all of the virtues of the family-tree
> model without limiting us to that model when it is clearly not applicable?
> Perhaps dialect-network and family-tree superimposed in some way
> (perhaps what was referred to in another recent communication)?

There are many models on offer.  In my forthcoming dictionary of historical and
comparative linguistics (out in March over here and in March or April in the
US), I group the various proposals under seven headings:

1. family-tree
2. wave
3. rake
4. rhizotic
5. crystallization
6. social-network
7. punctuated-equilibrium

The first three emphasize divergence, while the last four focus more on
convergence, or at least regard convergence and divergence as equally
important.  Only the last four have anything to say about non-genetic
languages, such as creoles and mixed languages.

I don't think any of these can be regarded as the ultimate model.  Each is good
at handling some things, bad at handling others.  They are, I think, best
regarded as complementary, rather than as competing.  As always, linguistic
reality is too complex to be captured adequately within a single model.

However, I want to stress that the family-tree model remains our standard
model.  This is just too successful in too many cases to be brushed aside.

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk