Search perplexity, noncontiguous elements

[Rod Decker]

In Baruch 6:25 there is what I think is an uncommon syntax:

 

ἄνευ ποδῶν ἐπ᾿ ὤμοις φέρονται ἐνδεικνύμενοι τὴν ἑαυτῶν ἀτιμίαν τοῖς ἀνθρώποις, αἰσχύνονταί τε καὶ οἱ θεραπεύοντες αὐτὰ

διὰ τό,

μήποτε ἐπὶ τὴν γῆν πέσῃ, δι᾿ αὐτῶν

ἀνίστασθαι·

 

Without feet they are carried on shoulders, showing their disgrace to the people, so even those who serve them are ashamed

because it

[the pagan idol]

is caused to stand

by them [i.e., by the attendants], lest it fall to the ground.

 

The δια το is followed by an embedded clause, set off by commas, complete with its own finite verb, and even another instance of the same preposition δι' which is not part of the δια το + infinitive construction.

 

In trying to find other similar situations I've experimented with several searches. The following one does return the Baruch 6:25 example, but none other.

 

"διὰ το" <AND> <WITHIN 0 Words> ., <FOLLOWED BY> <WITHIN 4-10 Words> [verb infin]

 

I have a couple of questions. First, is this a valid search for the construction I'm looking for? If so, then should I assume that this is the only example of this construction in the LXX (& NT since the same search there returns zero hits)? Second, is there a better way to do this, one that might not require an actual comma following δια το (which I specified with the ".,"), but which would still identify the preposition + article separated from its infinitival object by an embedded clause? Is there a syntax search that would be more appropriate or more accurate to describe this pattern?

[David Lang]

Rod,

 

First, you don't need to combine the AND and WITHIN commands. The WITHIN by itself is simply an AND with a proximity constraint.

 

Second, this looks like the only place in ANY Greek text currently available in Accordance where διὰ το is followed immediately by a comma. Doing just the first half of this search in the Search All window shows that to be the case (although it also shows some false hits we need to look into).

 

As far as doing the search without relying on the comma, I did a construct search with διὰ in the first column, το in the second, and an infinitive VERB in the third. I used a WITHIN of 1 between the first and second columns, and a WITHIN of 4-10 between the second and third. I then added an INTER between the second and third columns, placed a VERB infinitive within the INTER, and negated it with the NOT.

 

That search will find any case where 4-10 words appear between διὰ το and the infinitive. It returned 17 hits in LXX and 4 in the GNT. I'll let you sort through the results to see if they are at all similar to the Baruch example.

 

Note that I didn't use any syntax characteristics. I had thought about trying it, but decided against it when I remembered that we don't have syntax tagging of the LXX. That would have meant that any syntax search I did for GNT would have to be reworked for LXX, so I just stuck with grammatical tags.

 

Hope this helps.

[Rod Decker]

Thanks Dave. Due note of the AND taken. So the basic search is just

 

"διὰ το" <WITHIN 0 Words> .,

 

As for the construct, all the hits show infinitives that are in the same clause as the δια το with no intervening clause, so they aren't parallel with Baruch. It may be that the basic search is adequate, but the accuracy of the conclusion would depend entirely on the editors' punctuation.

 

I tried a variation: making the INTER positive instead of negative and specifying VERB (INDICATIVE, SUBJUNCTIVE, IMPERATIVE, OPTATIVE)

but this doesn't even find the Baruch example, so I must be doing something wrong. I would think that this search would find a construction in which there *was* a finite verb intervening before the infinitive, but it doesn't.

[David Lang]

Hmmm. I'm not sure why the search with the positive INTER is failing. It looks like a bug to me. We'll check into it.

[Helen Brown]

The positive INTER always means that only this item can intervene between the columns, or nothing at all.

[Joe Weaks]

Hmmm. I'm not sure why the search with the positive INTER is failing.

If you're going to use the INTER command, you have to include every possible option for the words that can be found. Like in the screen shot, which returns the 17 hits in the LXX.

What would be cool is if you could use the ANY tag, and just do an [iNTER] with [ANY -infinitive]. But alas, no.