Marcus Smith wrote:
> Jörg Rhiemeier wrote:
> >Yoon Ha Lee wrote:
> >
> > > Mine uses active case marking and there *is* no passive (I couldn't think
> > > of any way it would make sense).
> >
> >Same in my lang (Nur-ellen).  I also came to the conclusion that passive
> >doesn't make sense in it.  A passive would move the object into the
> >subject
> >position, but by the logic of active case marking it would have to be
> >put
> >in the objective rather than the agentive case, which would mean no
> >change
> >of the case.
> (These comments are direct at both comments above).
> Yes, the subject of a passive would be marked as a patient/object.  But is
> that all there is to being a subject, case?

No, there isn't.  In fact, the case of a subject can vary, especially in
Nur-ellen with its various degrees of volition.

>       Telek is active, and here is a
> small list of uses for the passive:
> 1.  Forcing same-subject marking on clauses (rather than
> different-subject).  (Great for rhyming, though the Telen don't rhyme all
> that often as far as I can tell).

This seems to have to do something with switch reference, a concept
I not yet understand properly and doesn't exist in Nur-ellen
(at least not in its current stage of design).
Especially I have little clue about how it is used.  Could you perhaps
so kind and give examples?

> 2.  Keeping the focus of the conversation on a single entity, whether or
> not is an agent or patient in any given sentence.  (Notice the different
> between: "I have a friend name John.  John was hit by a car" vs. "I have a
> friend named John.  A car hit John."  They have the same meaning, but the
> focus is different.)

This is done via word order in Nur-ellen.  Basic word order is SVO in
Nur-ellen, but it can vary freely, and the focus can be kept on the same
entity by putting it in front.  So, a sentence like "This book was
by David" could be rendered thus:

Hi    parv      David     tetent.
this [OBJ.]book AGT.David write.PAST

The phrase "this book" could not take any other case because _parv_ is
inanimate.  Your example "John was hit by a car" would also put
"a car" in an instrumental phrase because Nur-ellen does not allow
inanimate agents, and leave the agent slot empty.

(Damn!  It is hard to find appropriate examples when one has so many
in the lexicon!  Must find more words.)

> 3.  Allowing one to say what happened to someone else, when the perpetrator
> is not known without recourse to structures like "Someone hit John".  (ie,
> the focus stays on John in the passive, as in #2).  This is useful for
> cases like "John got crushed" but I don't know if the "crusher" was a car,
> tree, boulder, etc.

You can always leave out the agent.  A zero-agent sentence is perfectly
grammatical.  Something like "Beleg was killed." could be rendered thus:

Veleg     dagnent.
OBJ.Beleg kill-PAST

The patient can of course also be dropped.

I am considering introducing an impersonal pronoun rather than simply
omitting phrases.  Possibly with variations on where the phrase is just
deleted and where an impersonal pronoun is used.

> A lang that does not allow inanimate agents would have a difficult time
> saying something like "A boulder crushed my car."  But "My car was crushed,
> it was a boulder's fault" -- that's simple to do.

The latter trick doesn't work in Nur-ellen either, as "fault" (I haven't
fixed the word yet) requires a possessive genitive phrase (_e_ +
(There are two genitives in Nur-ellen, both marked by the same
_e_, but differing in the case of the noun itself.  The possessive
is _e_ + agentive, it is used to mark alienable possession such as
_bar e Feanor_ "Feanor's house", also as a subjective genitive with
verbal nouns; the partitive genitive is _e_ + objective, it is used to
inalienable possession (of body parts, etc.;
e.g. _dol e Veanor_ "Feanor's head"), also as an objective genitive
with verbal nouns, and to indicate that something is only partly

Martin medent bas.
Martin ate bread.

Martin medent e bas.
Martin ate some bread.

>       This isn't an issue for
> Telek, since it does allow inanimate agents.
> >Elves tend to look upon such language abuse with
> >disgust,
> So do the Telen.
> >and see the lack of a passive voice in their language as a virtue rather
> >than a defect.
> The Telen are quite proud of their ability to make subtle differences in
> pragmatics that would be impossible (or at least very lengthy) without a
> passive.

I haven't worked out the gory details of syntax and verb morphology yet.
There will probably be more to it than word order tricks and phrase
deletion.  Surely, an Elvish language should have quite a number of
interesting syntactic devices, and I am still collecting ideas.

> In fact, they have other operations that can create new subjects and
> objects.  "Subject Possessor Raising" takes the possessor of the subject,
> and makes it the new subject, much like passivization does with an
> object.  "Object Possessor Raising" takes the possessor of an object and
> makes it the new object of the sentence.  They do these when the possessor
> is more important to the conversation than the actual subjects or
> objects.  Say, I'm talking about John, and I'm going to say something about
> John's hat being blown away in the wind.  I would say "John hat blew away"
> where "John" is the subject (so is "hat", but it is inert).  I don't know
> of any active natlang that lacks Object Possessor Raising.  Maybe I ought
> to write this up as a snippet of Telek for everybody to read the details about.

Yes, it sounds interesting, but find it nigh impossible to understand
without seeing any examples.

Thanks for your enlightening comments!  I'll have to think more about
matters, and I'm looking forward to further explanations.