>>>Smyth notes that a relative in the nominative or dative is *very
rarely attracted* and cites Thucydides 7.67 as an example of the
former. It wouldn't surprise me to find a few more if we could do an
efficient search on the TLG. A Thucydidian passage carries a great
deal of weight.
Thanks, Don. This sounds like exactly what I was looking for as a
counterexample to falsify what I have always assumed. I'll have to
hot-foot it to the library now, to see whether this passage from
Thycydides really refutes me. If so, Carl Conrad will have maintained
his point without having to type one word in its defense! (Actually,
I think he did defend it when I raised the question before, but I
don't recall any citation of a specific example, and I remained
unconvinced. Carl is certainly justified in not bothering to repeat
himself when I raised the point again.)
Smyth's comment raises another question. As I was taught direct
attraction, it is limited to relatives that would have been written
in the accusative. Smyth points out that the nominative and dative
are only rarely attracted to another case. What about the genitive?
Am I correct in assuming that it is common for a genitive relative
pronoun to be attracted to the case of its antecedent? This seems
more likely to me than the other alternative that Smyth's statement
could imply: that the genitive is never attracted to another case.
I guess, while I'm at the library, I'd better check Symth, too. I
suspect he'll have the answer to my question without my having to
impose upon the list members more than I already have.
This has been helpful to me. Thanks again, Don.
Oh. One more point. Don mentions a TLG search. I don't know what
sorts of searches can be done in TLG, but this is a great example of
a problem where the ability to search electronically for certain
basic syntactical relationships would be extremely valuable. The idea
would be to search for relative pronouns in an oblique case
functioning as subject of a finite verb (it's hard for me to imagine
an relative pronoun functioning as predicate nominative). Finding
relatives in an oblique case is no problem with existing databases,
but I'm not aware of one that allows searching for words with a
particular syntactical function. If TLG can do this, I'd sure like to
know about it. I've always assumed that the difficulty of tagging
passages where the syntax allows a variety of analyses is a major
hindrance that has not yet been overcome, and may never be. For now
I'm content with picking others' brains and books for help.
In Love to God and Neighbor,
Bob Jones University