Quoting Peter Bleackley <[log in to unmask]>:

> Has anyone evolved an inflecting conlang from an agglutinating one? If
> so,
> how did you go about it?

My mailer went crashy on me, and I think that my first reply did not get thru.
In any case, I wish to preemptorily apologize if this turns out to be, in
effect, a double post.

Tersnuvu, one of my sketchier conlangs, passed thru a such transition early in
it's development. It's ancestor, Old Azainic, was isolating, with number,
tense, mood, etc optionally specified by adverbs and other particles. In the
branch leading to Tersnuvu, many such markers became first mandatory, and then
coalesced with the head word to become inflectional endings. Eg, _ibh_ "many"
became a plural ending; _sabhar ibh_ "ray many">_savariv_ "rays". Because of
such siffixification, Old Tersnuvu came to be an agglutinating language with a
largish number of cases, tenses, etc, in addition to this sg-pl distinction.

Further sound changes ruined this system, leading to the fusional modern
Tersnuvu dialects. In the prestigeous literary Taxte dialect, _savariv_
mutated into _sevr_, the singular being _savr_. Most of the cases and verbal
forms went the way of the dodo in the process, leaving the modern variants
with relatively small numbers of more or less irregularly formed forms.

It may be noted that some modern dialects, including literary Taxte, persist
in using the singular form of nouns with cardinals. I dread to see the
anadewism for this one ...