jIjatlhpu' jIH:

> if the bare verb was univalent to begin with (i.e. couldn't normally take

> an object, like {Qong}, {QaQ}), then the derivative with {-'eghmoH}

> probably cannot take an object either. How would one shoehorn an

> explicit object into, say, {bel'eghmoH} "please oneself"?

mujangpu' SuStel, jatlh:

> It's not a question of valency, it's a question of syntactic roles. What is

> having something done to it? That's your object.

I think we're saying the same thing in two different ways. The two agreement slots of a verb (and thus its valency) correspond to the syntactic roles of subject and object in any case. All I'm getting at is that the way {-'egh} and {-moH} together affect the structure of verb agreement (and therefore syntactic roles) should imply that if you can't add an object to a bare verb {X}, you probably can't add one to the verb {X-'eghmoH} either. Do you disagree?

I believe I disagree. If I wanted to say I cause the Klingon to please himself, that could be tlhIngan vIbel'eghmoH jIH. I am doing something. Something is being done to the Klingon. Subject and object. The -'egh tells me that the performer of bel (and not the subject, as TKD says) pleases himself; the -moH tells me that the subject of the sentence causes the action to happen.

These suffixes cause only semantic changes. Whatever meaning the suffixes add to the verb, it still takes one subject that is the performer of the verb and one object that is the performee of the verb. If you want to go through a rigamarole of valences and transitivity and so on, have fun. I feel confident Okrand wasn't thinking about those things; he was thinking of syntax.

You specifically challenged whether the sentences showed anything exceptional or new, and that's what I responded to. It's exceptional in that it explicitly violates two rules about -chuq, whatever the justification. It's new in that no one ever said anything about it before.

-- 
SuStel
http://trimboli.name