Arabic and IE

[Larry Trask]

----------------------------Original message----------------------------
Not for the first time, Alexis M R has raised a troubling question of
terminology.

When we say that languages A and B are related, we mean that a genetic
link between them has been established to general satisfaction -- though
I am, of course, begging the question of what constitutes "general
satisfaction".

But, when this state of affairs does not obtain, things get awkward.
We commonly say that A and B are not related, but Alexis has, quite
properly, pointed out that this wording can easily give outsiders the
wrong impression, by implying that a link between A and B has been
disproved -- an impossible conclusion.

So, as Alexis argues, we really ought to be more careful and say
something like "no link between A and B has at present been shown to
exist".  But this precise wording is long and cumbersome, and it's just
too much of a mouthful to use over and over again.

When I talk to linguists, I have no hesitation in saying "A and B are
not related", since I assume that any linguist will know what I mean --
though Alexis has objected to this, too, in an earlier posting.  When
talking to non-linguists or to beginning students, I try to be more
careful, but even so I often retreat to the wording "not discoverably
related".  Others use "not relatable".  But both of these, if taken
literally, would appear to imply that no relationship can ever be
discovered -- exactly the thing that's bothering Alexis.

So what should we say?  Damned if I know.  I can think of no wording
which is brief enough to be used constantly and accurate enough to
satisfy the most fastidious among us.  Maybe we ought to stage a
competition to find a suitable form of words.

As for Semitic and IE, I am puzzled that this issue continues to attract
such attention.  Of Greenberg's four African families, Afro-Asiatic is
the only one that appears to command universal assent -- or, at least, I
can't name a single linguist who queries it or rejects it, in spite of
the fact that no generally accepted reconstruction of Proto-AA exists at
present.  That being so, and Semitic being accepted by all as a branch
of AA, why should people be expending so much energy on trying to link
Semitic to IE?  In my view, if a feature of Semitic cannot be shown to
have been present in Proto-AA, then it is simply not available for any
external comparisons -- except by someone who denies the Semitic-AA
link, of course.

Of course, I did confess that I was a hard-liner on these matters. ;-)

Over to you, Alexis.


Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk