Nik Taylor wrote:

> > These languages evolved at times
> > when the concept of "life" was defined in a more intuitive way than what
> > modern science tells us what it is.
>
> Which is why plants are rarely in "animate" categories.

Are they so rarely?  I know that they sometimes aren't, but I think it
is
pretty intuitive to include them into animate as they are very obviously
alive, and that's why plants are animate in Nur-ellen.  Verbs such as
"grow" and "blossom" are in fact active verbs, requiring an agentive
argument.
(I notice that this is somewhat at odds with the volitional hierarchy
I have posted here earlier, and I am considering reworking this.)

I am also considering having the animacy status of some words being
volatile, such that they are animate in certain contexts (e.g. poetry)
and inanimate in others.
This would prompt double objective forms (the poetic agentive being the
same as the common-usage objective, and the poetic objective being a
lenited form of the former); however, what I consider is simply that
these "semi-animate" nouns are  technically animate, but the agentive
being only very rarely used outside poetry,
and their use outside of these special contexts would seem almost as
bizarre as if someone was to invent an agentive of a
well-known-to-be-inanimate noun (such as **_tring_ as an agentive of
_dring_ "hammer" - if there really was such an animate noun, its
objective would indeed be _dring_) in order to force it into an agentive
slot.  Some of these "poetic agentives" might be irregular or even
suppletive.

> > Actually, I have allowed myself some freedom in Nur-ellen.  Celestial
> > bodies, for example, are animate
>
> Same here, they're in gender 6, which I call "pseudo-animate", that is,
> non-animal animate.

Is this gender also used for plants?

>       In an earlier stage, what is now a
> sentient/nonsentient distinction in the verbal inflection was an
> animate/inanimate distinction, so G1-6 took one agreement, and G7 took
> another.  Even now, there's still a trace of that animate/inanimate
> distinction - inanimate nouns *rarely* precede animate nouns, even if
> the inanimate is the subject - "The stone squashed John" would be
> "squashed John-abs stone-erg".  But, I'm uncomfortable with G7 nouns
> taking ergative, they might take instrumental.

I assume that W is a VSO language, with the exception that inanimate
subjects are moved to the end of the clause, thus in a sense labelling
them as not-really-subjects.  What you are considering to do - mark them
with instrumental instead of ergative - would fit, and is pretty much
the same as in Nur-ellen.
Or you could consider your ergative case as you have it now polysemic,
marking a subject if it precedes the object, and an instrument if it
follows, as in

broke John-ERG window-ABS hammer-ERG

"John broke the window with a hammer."

Nur-ellen translation, just for the fun of it:

Dzhon    ristent    i   hent   ni   dring.
AGT.John smash-PAST the window INST hammer

(The cases of _hent_ and _dring_ are not tagged in the interlinear as
these two nouns, being inanimate, cannot take any other case than
objective.)

> > the
> > nouns _byrokrat_ "bureaucrat" and _natsi_ "Nazi" are both inanimate,
> > because bureaucrats and Nazis tend to behave like automata and demons,
> > respectively.
>
> Fascinating!  W doesn't have anything like that.  Sentient beings are in
> genders 1-3 (1 = feminine, 2 = masculine, 3 = epicene),

The only place where Nur-ellen has different gender forms (besides the
animate/inanimte distinction) are 3rd person singular pronouns, which
occur in masculine, feminine and epicene forms.  The masculine and
feminine forms are only used for biological males and females (with some
mythologically motivated exceptions: _An`r_ (Sun) and _Kem_ (Earth) are
considered female, while _It`l_ (Moon) and _Men`l_ (Sky) are male),
while epicene can be used with all animate  beings.  The epicene pronoun
is also the one to use with collective entities such as nations,
families etc.

There is of course no gender distinction within inanimate pronouns.
(There are also, for obvious reasons, no inanimate 1st and 2nd person
pronouns.)

>       but "sentient"
> is defined as "speaking", so an infant is lanabási, with la-,

Looks like a class prefix system as found in Bantu languages.

>       the gender
> 4 (domestic animal) prefix.

So, as for now I can identify:

1 feminine speaking
2 masculine speaking
3 epicene speaking
4 infant/domestic animal
5 feral animal???
6 "pseudo-animate"
7 inanimate

I guess gender 5 are feral animals.  Are there even more genders,
or is the list complete?

Syld,

Joerg.