David A. Johns (
4 Feb 1995 20:57:27 GMT

In article <> (Stephen Carlson) writes:

# It seems that there is a similar problem in linguistics, but the
# linguists don't get all bent out of shape about it.
# You can start off in Spain walking from one village to the next
# near the coast of the Mediterranean, go through Catalonia,
# France, and into Italy, and have every village dialect be so
# close to its neighbors that they are mutually intelligible. With
# this dialect continuum, linguists still find it useful to
# discuss, compare and contrast Spanish, Catalan, French, and
# Italian as separate languages.
# Same thing for a trail of villages from Bavaria to Belgium.
# If the concept of race is fundamentally flawed because there are
# 5 billion of them, then the concept of language must be similarly
# flawed because there are 5 billion idiolects. Why are the
# linguists still in business? Are they deluded?

No, you are, I'm afraid, at least in this case.

When linguists discuss "French" and "Spanish", they are normally
talking about the standard languages, which are relatively arbitrary
creations. When they discuss dialects, they use such terms only as
rough guides. What they are really interested in is isoglosses --
boundaries between different features, such as uvular vs trilled <r>,
etc. These isoglosses are very similar in concept to the clines that
population biologists use.

I suppose standard languages are conceptually similar to the various
breeds of dogs -- artificially selected objects which might well
spread and replace all the mutts. But nothing similar to dog breeding
has ever occurred among people, at least to a measurable extent, and
just to keep the analogy going, standard languages tend to develop
local variants too.

So I think there's something to the analogy, but it supports "no
races" rather than "races".

David Johns