8 replies to this topic

[Harvey]

(Sorry to start a new topic, but I couldn't figure out how to insert an image, or attach one, to a reply to an existing topic.)

Besides the remaining programming work to be done on syntax searches, is there a reason why the attached search would not pick up אשר מתחת לשמים in Gen. 1:7 (and many other similar phrases)?  Screen shot 2012-07-04 at 2.53.10 PM.png   60.72K   35 downloads

[Robert Holmstedt]

It is because what follows the אשׁר is not a clause; rather, the אשׁר as the function word introducing the relative clause stands as the first word within the clause. Thus, there will be *no* examples of the construction you are looking for with the search above.

[Ken Simpson]

Hi Robert, I think I know what you mean, but could you clarify this phrase for me please?

"as the function word introducing the relative clause stands as the first word within the clause"

it just doesn't quite seem to make English sense to me, though I think I can figure it out, I'd rather be sure of what you mean.

Thanks
Ken

[Robert Holmstedt]

Interesting ... an American/Canadian English versus Australian English processing difference? That would be fun to unpack.

Anyway, I can hardly say it more clearly and succinctly. The syntactic position of the אשׁר (which is a function word) is at the beginning of the subordinate clause it introduces. In some older Hebrew grammars, the אשׁר is described as being part of the main clause rather than the subordinate clause. This is inaccurate (it reflected both an Indo-European framework for looking at Hebrew and an over-emphasis on the nominal etymological of אשׁר). Our syntactic tagging positions the אשׁר accurately -- within the subordinate clause. It is no different that causal כי, which as the function word introducing a causal clause stands as the first word within the clause.

So, the search would have to look like this:

 Screen Shot 2012-07-05 at 7.58.59 AM.png   19.83K   23 downloads

Note that the currently this search does not work -- in the latest iteration, the Null appears to have been restricted to Subjects and so Null Predicates within the clausal structure are currently impossible to find (at least, in this search -- I can find them in other contexts, which suggests this is a rather tricky little bug). Please be patient. I have been told that the programming is a bit like trying to stuff a suitcase full -- stuff one more item in and some small one pops out. But this is the search as it will eventually be done properly. The Preposition is identified as the Complement within the Complement phrase and no PLACE feature should be necessary in this hierarchical structure.

Edited by Robert Holmstedt, 05 July 2012 - 08:05 AM.

[Ken Simpson]

Thanks Robert. We speak funny down here, but nonetheless this explanation made sense to me (even if it is less succinct!). And knowing this, the original sentence makes sense to me too!

Much appreciated.

Ken

[Robert Holmstedt]

Dialects are fun, eh? ;-)

Happy to be of help.

[Harvey]

Thanks, Robert. What threw (and throws) me off is that when I ran this search, it picked up an instance of what I believe is the structure in question---in Judg. 21:19. I now see that there is a second clause shown there, under the relative clause, as in my search; I didn't notice that this was not the case elsewhere. Is there any reason for the difference in the analysis in Judg. 21:19, or is this just an inconsistency?

And, by the way: how do you insert a screenshot in a reply to a post?

Edited by Harvey, 08 July 2012 - 08:59 PM.

[Robert Holmstedt]

Harvey,

To post a screenshot you have to choose the "More Reply Options" button and then choose the image, upload it, and insert it into the post (it will insert at the cursor point).

The case in Judg 21.19 was caught in your search only because it is a relatively rare example of a single אשׁר that governs more than one relative clause. The way we tag (I won't go into details here) means that clausal status is signalled both for the clause starting with the אשׁר and the two clauses that the אשׁר introduces. It's neither an inconsistency nor an error. And it will also be caught by the normal (proper) search when it works fully.

[Harvey]

Thanks, again, Robert.