semantic definitions

[Larry Trask]

Pat Ryan writes:

> Now, I would say that Larry would be the best choice to explain why a
> semantic approach "doesn't work as well" but since I have repeatedly asked,
> and in my opinion have not gotten a responsive answer to the question of
> 'why the "slot-and-filler" approach (Larry's terminology) be characterized
> as working better', perhaps you will jump in to explain with specifics not
> with generalities why you also hold this view.

Well, I don't think it's up to me, or to anybody, to demonstrate *ab initio*
why semantic definitions of parts of speech don't work.  Semantic definitions
have been attempted many times, and all such attempts have been dismal
failures.

If Pat, or anybody, wants to argue that parts of speech *can* be defined
semantically, then it is up to him to make a case, which we can then discuss.
So, Pat, let's have it: where is your semantic definition of 'noun' or 'verb'
or 'adjective' or 'pronoun'?  Put it on the table, and let us have a look at
it.

Perhaps you'd like to consider cases like 'arrive' and 'arrival'.  These appear
to have the same meaning: we can say, equally, 'before she arrived' or 'before
her arrival' -- yet one is a verb while the other is a noun.  So how can the
two be distinguished on semantic grounds?

Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk