I wrote:

> >Even in the sense of being "bound" by not occuring alone,
> >"the" is just a form of "this/that", and "a/an" equal "one",
> >which do stand alone. So in my dialect, they are liberated.

then, Nik wrote:

> I disagree. It's not uncommon for clitics to have "free" forms. For
> instance, in French, _je_ is a clitic, but has the free form _moi_.
> Also, I don't see "this/that" as being equivalent to "the". They are
> bound. The forms "the", "a", and "an" can NEVER occur outside the noun
> phrase. They are incontestably clitics. "A/an" has a free form "one",
> but "the" has no free form.

I admit this at least partially; clitic may be the right term.
But "the" is to "this/that" as "a/an" are to "one", so if one/an
has a free form, that/the does too -- this is getting messy.

I'd like to hear what you thought of Trask's story, which seems
unassailable. It may be that we need a different term for
articles that glom (Basque), vs. those that do not (English).
Maybe it is just "clitic" vs. "affix"? No, you said clitics
have bound and free forms; this can never be pinned down.

then, Kristian Jensen wrote:

> Wow! That's interesting! You mean to say that in your dialect you
> still consider "the" a demonstrative, and you still consider "a/an"
> a numeral? So say I ask you a question like "Which one is yours?",
> you could answer "The!"? That sounds really weird to me.

I would say "This!" or "That!" or "One!", very carefully henceforth.
Now that the water is muddy and I am thoroughly confused about
the most common words in my native language, I surrender the field.

Would you use "bound" for anything other than determiners?
A verb like "is" certainly does not stand alone, but isn't bound.
Terminology not only confuses, it mutates too quickly to remember.
I would never get through a modern linguistics course.