OT: Two language change questions

I've been wondering about two things wrt language change lately:

1. Does it ever happen that a language which differentiates definite
and indefinite (e.g. by articles) loses that distinction?

My hunch is yes -- after all, there are lots of languages in the
world with no definiteness distinction; so, in any such language, it
is either the case that a) the language and its ancestors NEVER had
the distinction, or b) the language or its ancestors at some point
had it, but lost it. It seems to me that in the millennia of
evolution of each such language, it is unlikely that the feature
"does/does not distinguish definiteness" could remain completely stable.

However, I am having trouble imagining the gradual steps involved in
the loss of the distinction. I know that sometimes English uses the
definite article to refer to things which aren't all that definite,
especially in idioms (but then again, they're idioms!), so I can sort
of imagine it, but not quite.

(E.g. "doing the dishes" and "doing the laundry", where IMO the exact
identity of said dishes or laundry  isn't really pertinent; "that
takes the cake", where the cake is strictly metaphorical; "play/act
the fool", which seems to me would more "logically" be "play/act
[like] a fool".)


2. In some languages (e.g. Japanese) you can juxtapose a noun and a
verb (or verb-like thing such as an adjective) in such a way that the
meaning is "the [noun] that <verbs>" or "the <noun> that is <verb>ed"
-- without having to derive or inflect the verb into a verbal
adjective. E.g. Japanese _yonda hon_ "a read book, a book which has
been read". Is it ever possible in such languages to use a pronoun
instead of a noun? E.g. Japanese (hypothetical) *_tabeta kare_ "he
who ate".

Replies