Biblical Hebrew Forum

["Dave Washburn" <dwashbur AT nyx.net>]

Peter wrote:
> Dear Moon,
> 
> I think I may have confused you by bringing in the comparison with 
> English. Hebrew may be no more like English than Korean. In fact I 
> think a closer English equivalent to ):A$ER might be "the following" 
> followed by a colon. For it seems to me that the word ):A$ER is 
> related syntactically to the main clause rather than to the 
> subordinate clause. Azerbaijani has a similar conjunction "ki" (though 
> it cannot take prepositions) in which this relation is made clear in 
> that (according to the pronunciation) a comma is always written after 
> "ki", making it part of the main clause rather than the subordinate 
> clause. I think you are confused because you are trying to parse 
> ):A$ER as part of the subordinate clause. On my view HABBAYIT ):A$ER 
> HF)IY$ BOW could be glossed "the following house: the man (is) in it". 
> But HABBAYIT BA):A$ER HF)IY$ would not make sense: it would mean 
> something like "the house in the following (one): the man is". Thus 
> KA):A$ER would be "like the following one:" or simply "as follows:".

This discussion has sort of drifted away from my original question, 
which is my fault because I haven't been in it (due to illness).  
Recall that my question had to do with the legitimacy of TG trace 
theory as regards the Hebrew construction.  Now, if we assume 
that it's a legitimate theory, then Moon's two examples

--
It also seems that (5) sounds better than (6), though both are
not grammatical.

(5)  the house that the man is in it.
(6)  the house which the man is in it.
--

Are not grammatical because the relative has been moved out of 
the object-of-preposition slot, to wit:

The house __ the man is in [that/which]
                    __________________|
                    |
The house which/that the man is in (t)

where (t) is the "trace" left behind by the moved item.  By 
definition, a trace cannot be filled by another element, so putting 
"it" in that slot produces an ungrammatical clause.  However, in 
Hebrew it doesn't work that way.  In Hebrew, if we leave the 
preposition where it is then it has to attach to something, hence 

(7) HABBAYIT ):A$ER HF)IY$ BOW (the house which the man (is) 
in it)

but not

(8)  * HABBAYIT ):A$ER HF)IY$ B:

However, it seems to me that if trace theory is correct, we should 
never see constructions like HABBAYIT ):A$ER HF)IY$ BOW 
because they would be inherently ungrammatical.  Yet it's a 
perfectly good structure in BH.  

Of course, a side-effect of all this is that Peter's idea of ):$ER being 
part of the main clause is unlikely because either way it's a moved 
constituent from the "lower" clause; it seems to me that even in the 
case of constructions like 

(9) HABBAYIT BA):A$ER HF)IY$ (the house in which the man (is) )

this is the case, hence using his format we should gloss (7) as 
"the following house: the one which the man is in (it)" and (9) as 
"the following house: (the one) in which the man (is)."

I want to come back to Peter's earlier idea about whether the 
):A$ER "goes with" the main clause in the moved-preposition 
construction and with the "lower" clause in the resumptive-pronoun 
construction, but I'll do that separately once I actually locate that 
post again...

In any case, the resumptive phenomenon in Hebrew makes me 
question whether trace theory is on the right track.

Dave Washburn
http://www.nyx.net/~dwashbur
A Bible that's falling apart means a life that isn't.