Vnouns in סמיכות פרודה state

[Hilla-Merhav]

@amir-zeldes

Following our correspondence in the Guidelines doc, I am coping your last instruction for the sentence: החוק מעגן את היותה של מדינת ישראל מדינה יהודית so we can resume this discussion: "et heyota medina yehudit" is an object clause introduced by "et", so et/SCONJ/mark is a child of "medina" (note it is not ADP/case, so we can tell it's not "et ha-medina")" Thank you very much, I have followed this instruction. But due to the fact the copula can't govern anything (as I understand) I still am not sure how should I analyze the סמיכות פרודה.

My current (and wrong, I am sure) analysis is attached. I want to change it so the ישראל would be the nsubj of מדינה יהודית, but I am not sure how should I treat של, and the PRON ה too.

[amir-zeldes]

I hadn't thought about this case... First of all medina could not be obj under this analysis, since it's a clause, so it would have been ccomp if anything. But you're right about it being a hard constraint that cop can't have children, so the only way to do this would be to make yisrael and the possessive "h" be children of medina somehow. If we did that, probably the best I can come up with would be either:

1. "h" is nsubj to medina, and yisrael is nmod:poss to "h"
2. "h" (or "yisrael") is nsubj to medina, and "h" is dislocated to medina

But both of these are so ugly I think it's better to give up and just treat "heyot" as a NOUN and medina as xcomp to it.

[Hilla-Merhav]

Thank you very much, this anlysis makes a lot of sense. Is it OK if I add it to the Guidelines?

[amir-zeldes]

Yes, I think this is the best thing we can do here. Language is just really complex... :)