"is the same as"

[Larry Trask]

[ moderator re-formatted ]

Stanley Friesen writes:

[LT]

>> At least in biology, such relations as "can interbreed with" or "cannot
>> readily be distinguished by eye from" need not be transitive -- I agree.
>> But by what right can we identify the relation "is the same as" with one of
>> these?  And what would be the point of doing this?

>  Because:
>
>  A) language is a biological phenomenon, and behaves like other such.

Not really.  Our language faculty, in the view of most linguists, is part
of our biology, but the language faculty is not the subject matter of
historical linguistics: it is the same for everybody.

Historical linguistics deals in the particularities of individual languages,
and these are not biological in nature.  Rather, they are largely social.
And, *pace* Richard Dawkins, social phenomena do not behave like genes.

>  B) language differentiation acts *very* much like biological speciation,
>  except for happening much faster.

"Very much" in some respects, but not in all.  There are important differences.
Biologists recognize a certain amount of gene transmission between species,
but only within limits -- ignoring our own genetic engineering.

But languages allow sideways transmission without limit.

Note, for example, the enormous Latino-Romance influence upon the very
distantly related English and upon the unrelated Basque.  In biological
terms, this is rather as though ostriches had received massive gene
transfusions from tigers or starfish.

>  C) the 'mutual comprehensibility' definition of separate languages is
>  almost exactly equivalent to the biological species definition as a
>  criterion for recognizing species.

Really?  I doubt it.

Mutual comprehensibility is a continuum ranging from 0% to 100%, with
everything in between.  It is also not fixed: with exposure, mutual
comprehensibility can greatly increase.

The same is not true of biological species.  We do not find pairs of
species which can interbreed only at the 86% level, or only at the
32% level.  And chimps and gorillas do not become more inter-fertile
by living alongside each other.

>  D) as others have been pointing out here, the similarities are so close
>  that it is even useful to apply cladistic methodology in the study of
>  historical linguistics.

This has been widely done, and almost every possible parallel has been
noted.  See, for example, Roger Lass's latest book.  But historical
linguistics is still not biological taxonomy.

>  In other words, the two sets of phenomena are so extremely similar that it
>  is ineffective to try and treat them very differently.

Sorry; I disagree strongly.  The differences are large and important.

Just to take an obvious example: what would you say was the linguistic
equivalent of the biological gene?

This strikes me as a pretty big difference.

>  [P.S. the salamander ring I mentioned is formally considered one species
>  for taxonomic purposes].

So it is.  And a dialect continuum is sometimes treated as a single language
for linguistic purposes.  But the decision is largely arbitrary.

Difficulties and all, biological species are a lot more real than are
languages.

[LT on the Romance dialect continuum]

>> Indeed, and this is a common state of affairs.  But how does this constitute
>> an argument for treating "is the same as" as a non-transitive relation?
>> Better, I suggest, to forget about this last relation altogether, and to
>> speak instead of some more appropriate relation, such as "is readily
>> mutually comprehensible with" -- which again I agree is not going to be
>> transitive.

>  That is more or less what I *mean* by "the same as".

Well, if that's all you mean by "is the same as", why not drop the vague
wording and stick to the explicit one?

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk