X99Lynx at aol.com X99Lynx at aol.com
Tue Jan 26 16:44:28 UTC 1999

In a message dated 1/26/99 8:54:14 AM, DLW wrote:

<<I must, with some reluctance, support Mr. Hubey on this one.
There are reasons to think that biological evolution is relevant to
linguistic evolution. >>

Yes, but it then becomes very important to understand evolution science and
the key difference between the two kinds of studies.  Here's an example:

<<For example, it has recently been determined that the three kinds
of zebras (there I go with zebras again) are not in fact closely related
to each other.>>

Modern conclusions can be misread if we don't take them in historical context.

One of the big breakthroughs of Darwinian theory was that it freed biological
history from total morphological relationships.  Lamarck and popular thinking,
reinforced by the system of phylum and genera, assumed that common
morphological traits meant common descent or relation.  There was really
nothing else to go by than the idea that if two types of animals looked alike
they were more closely related than animals that looked different from one
The popular version was that you will look more like your father than your
distant cousin and extending that to bigger realm of animal relationship, etc.

What the theory of natural selection did in this case was shift focus from
strict inheritance to the environment as dictating morphology.  So that the
closeness of body structure or the appearance of a unique structure (like
zebra stripes) does not necessarily tell us the history of descent.  In the
extreme case, concepts like parallel evolution or independent occurence can
actually mean that a trait shared in common (like wings in birds and bats) do
not demonstrate common ancestry but instead common adaptation.  (Even modern
genetics, without the confirmed Darwinian hypothesis of random mutation, would
not have been able to predict this little trick of Mother Nature.)

The big example of evolution theory at work isn't zebras.  It's the recent
conclusion, against all common sense, that birds (Avis) are the descendants of
dinosaurs.  There was no DNA ("genetic") evidence for this, it is based on
paleobiological taxonomic evidence.

Compare this to historical linguistics.  The breakthrough here was to
establish a genuine morphology (by which I mean also for these purposes
phonology) of descent.  The breakthrough with Grimm's Law was that form and
traits could give you a predictable pattern of descent.

This is not parallel to the science of evolution so much as it is the science
of genetics.  Both genetics and linguistics focus on internal structure.
Natural selection focuses on external forces.

Linguistics is a powerful internally consistent tool.  But once you try to
apply external forces to its analysis, you are diluting or misapplying it.

Just as random mutation breaks the laws of genetic inheritance, non-linguistic
factors disrupt the unity assumption of linguistics.  Of course, this in no
way says that either genetics or linguistics are flawed.  It just means you
have to know when they apply and when to use them and when something has
intervened - like mutation or a counter-linguistic historical event.

As far as language structure being subject to natural selection, there is an
important difference between biology and language structure.  Language
structure - like genetic structure - must be conservative about forms.  This
is because language's function is communication.  Communication demands
predictability.  I must know by your words you meant the same thing today that
you meant yesterday .  Or communication fails.

Biological evolution, on the other hand, is anti-conservative.  The diversity
of forms is totally dependent on "innovations" and nothing else.  Genetics is
conservative like language structure.  It conserves forms.  Evolution is the

Steve Long

More information about the Indo-european mailing list