possessive [form of a] pronoun

[ECOLING at aol.com]

It has taken some time to get to the bottom of the
differences in the use of terminology here,
but I think we have arrived:

To summarize into its most succinct form:

The relation:

he :: his

is from syntactic and semantic points of view
essentially the same as the relation

the man we met yesterday :: the man we met yesterday's

and this parallel extends to many different uses of the
apostrophe-s form, whatever one wishes to call it,
whether the syntactic relation marks
possession, part-whole relation,
or agency or object-status relative to a verbal noun.

*

Therefore the form "his" can quite properly be referred to as
the possessive or genitive form of "he",
just as much as the form "the man we met yesterday's "
can quite properly be referred to as
the possessive or genitive form of
"the man we met yesterday".

*EVEN* using "pronoun" in a purely distributional meaning,
the reading "possessive [form of a] pronoun" for "possessive pronoun"
is fully legitimate.  Possessive pronouns, so meant,
were originally case-forms of (nominative) pronouns,
and are still to a considerable extent so analyzable --

(or substitute whatever term one wishes instead of "case" --
if the English "  's  " is not considered a case form,
but is considered for example something of a clitic,
since it can follow entire NPs including a relative clause,
as illustrated above,
it need not follow only simple noun stems).

*

At the *same* time (and neither of these excludes the other),
on the basis of a distributional analysis,
just as Trask says, the possessive forms of NP's
do not substitute for nouns, they substitute for NP's
(whether pronoun, or whether complex NP including relative clause).

*

Therefore, there are legitimate syntactic and semantic
reasons to use the terminology "possessive pronoun" in this way,
where "possessive pronoun" can quite plausibly be taken
as meaning "possessive form of a pronoun",
and that can be meant primarily based on syntatic and semantic
grounds rather than on morphological or slot-filler analysis.
This is true whether or not it is someone else's standard terminology.
We all know what Pat Ryan has been referring to,
and we all are capable of understanding his terminology
and giving it the best interpretation we can,
to proceed to discuss content.

End of argument, as far as I am concerned.
In the future, I will simply refer back to this message.

Lloyd Anderson