Habitual and perfect marking and statives

I'm having a lot of trouble deciding on the aspect system in one of
my conlangs. One of the central points of contention is how to mark
habituals and "doer" nouns. (First of all, habituals and doer nouns
will probably have the same form, since the language doesn't really
distinguish nouns from verbs.)

On one hand, I could let habituals be expressed by the bare verb
stem, and mark verbs occurring at a specific time specially. I
somewhat like this idea, because the "nouns" in this language are
mostly built up of verbal roots, and if I decided to have an affix
for habituality, way too many nouns would end up with this affix as
well. (Of course, this could just be an accepted part of nominal
morphology, just as there are a huge number of o- and a-stem nouns in
Indo-European languages.)

On the other hand, this seems way too much like English to me, in
which the habitual is plain (well, except for some verbs like verbs
of perception, opinion, and knowledge) while the present tense is
generally marked with a periphrastic "progressive." But when I think
about it, most Indo-European languages seem to leave habitual verbs
unmarked - even in languages where the true present tense is also
unmarked. So one basic question I have is: what are some different
ways of marking habitual events? And how common is it to have *no*
special marking for it?

Another issue concerns conflation of aspect with stativity*. I have
seen a few conlangs, and at least one reconstructed natlang (PIE
according to one analysis), where the perfect aspect of eventive
verbs is conflated with the stative verb derived from the eventive.
I.e., stative verbs and eventive perfects have the same form. I like
this idea quite a bit, but again I would like to ask how common a
pattern it is.

However, another idea I've had is for the *habitual* of eventive
verbs to have the same form as stative verbs. Are there any languages
that do this? This way, a noun which is eventive at its core, e.g.
"speaker = one who habitually speaks," would be formed parallel to a
noun which is stative at its core, like "red thing", or nominal at
its core, such as "person."

Another thing that occurs to me is to mark ability rather than, or in
addition to, habituality. The two concepts overlap quite a bit, I
think; a "writer" is someone who *can* write. What are some ways this
is grammaticalized in languages? (In fact, as I've read recently, a
verb originally meaning "to know" can evolve into an auxiliary to
mark habituality.)

*sigh* Since it's hard to decide, maybe I could just have a stativish
derivation which can apply to all verbs, but whose actual semantics
are determined by the particular verb. Perhaps "kill[+STAT]" would be
"killer = one who has killed" (certainly in English one only needs to
have killed once to be called a killer!), while "speak[+STAT]" would
be "speaker = one who can speak", and "write[+STAT]" would be "writer
= someone who writes habitually." I just have to be careful not to
let English semantics have too much influence.

Also, while I'm thinking about it, have there been any
classifications or analyses of verbs which posit a kind of verb
intermediate between eventive and stative? I've been thinking about
the verb "to rule," which doesn't seem to quite fit into either
category to me (except in the case of specific decisions made by a
court).

* Is there a better word than "stativity"?

Reply