Is this syntax search possible in the GNT?

[Norman Meyer]

I would like to find all instances of "εἰ" which are in an independent clause but not also hierarchically in a dependent clause. If anyone knows how to perform such a search, please let me know. The usefulness of this search is to investigate the use of "εἰ" outside of the apodosis protasis construction. Below is a pictorial representation of what I would like to do, but Accordance doesn't seem to like negations in the first column. Can this be accomplished with the search window? Or a combination of the search window and construct searches? Thanks in advance for any help!

 

[Tim Planche]

Dear Norman 

Posited this on the other link - Does this work?

 

I had a quick look and this return 228 words in 207 verses (I left the "show hit antecedent or elision" option unchecked)

seems to be getting what you were looking for - but i may be wrong - you may also need to tweak the search depth options

 

apologies if Ihave got the wrong end of the stick - I find these syntax searches a bit challenging

[mgvh]

@Norman Meyer: Could you give a verse that is an example of what you are looking for?
Aren't all instances of ει (=if) part of dependent clauses?

[Norman Meyer]

@mgvh Great question. I was under the impression that all instances of ει are part of dependent clauses, but I had a Greek instructor asserting otherwise so I'm trying to do my own investigation. Mk 8.12 may be an example (see definition #4 in BDAG); but it occurs in direct speech so I'm not sure this is a good example. The discussion I was involved in pertained to ει γε in Eph 3.2. The syntax of Eph 3.1-2 seems to follow a pretty standard conditional clause with the semantic meaning of Evidence/Inference; but most English translations don't see it that way. I have found some commentaries that do (e.g. Heinrich Meyer) and interestingly the syntactic diagrams in Accordance and Logos all seem to take it as a standard apodosis protasis construction (in vss Eph 3:1-2). Anyway, if ει can be used outside of a conditional construction, I thought it would probably match the search criteria suggested above. 

 

Also, as a relatively new user of Accordance, I'm hoping to find a way to construct searches of the structure above (i.e. a negated sub-term) as I think it will be very useful in other circumstances.

Edited October 18, 2023 by Norman Meyer

[Norman Meyer]

@Tim Planche Thanks greatly for the suggestion. Unfortunately, the search you show does not work. I had tried that construction myself! Basically any sentence that has just one ει in a dependent clause would still be returned. The search you provide just guarantees that there won't be two ει in two dependent clauses of the same independent clause.

[mgvh]

@Norman Meyer Ok, I see what you are getting at now, and I do see BDAG 4 (and 5). I still think that in all these instances there is at least an implied apodosis which would be the independent clause.  Mark 9.23 is another sort of example: τὸ εἰ δύνῃ. (I translate with, "What's this, 'If you can'?" Here too, what's implied is the "Help us" that was provided in the context.

Even in instances where it gets translated with "whether" (e.g., Mark 3.2) or "if not / except" (εἰ μὴ), there is some bigger idea behind it.

I.e., I think ει can be used in sentences on its own (and thus looking like an independent clause), especially in oral communication, but it is still part of a larger thought context.

[Tim Planche]

I think with the examples understanding a bit better...

 

to understand a bit more - I realise "Εἰ μἠ" is a bit different and often not used in continual sentences - but frequently rather as an adjunct in a clause - 

 

but would this be an example ?

1Cor. 7:17 ¶     Εἰ μὴ ἑκάστῳ ὡς ἐμέρισεν ὁ κύριος, ἕκαστον ὡς κέκληκεν ὁ θεός, οὕτως περιπατείτω. καὶ οὕτως ἐν ταῖς ἐκκλησίαις πάσαις διατάσσομαι.

 

or are you only looking in the context of a conditional sentences?

[Norman Meyer]

@Tim Planche Yep! That is the kind of construction for which I am looking. I have been able to find verses such as these by merely searching for ει in an independent clause and limiting the search depth to 1. Then, I'm trying to discern, as @mgvh suggested, is there an implied protasis or is this a "non-conditional" use of ει. I tend to think he is right in that there is always some implied protasis if not explicitly stated.

 

HOWEVER, I am just as interested, maybe even more, if there is a way to get Accordance to logically do the kind of search I first suggested: i.e. structurally is it possible to search for a lemma in a larger syntactically tagged clause while negating it's existence in certain kinds of syntactically tagged sub-clauses within that larger clause? If anyone is able to demonstrate how to do this, I'd greatly appreciate it! Note that this would be possible if the HITS command could provide a list of either the verses (which would in a way be a specific RANGE for the search) or the actual instances of the words found. The current 3 options (lexeme, inflected form, and tag) all redo the search without the syntactical restrictions of the first search. 

[mgvh]

I don't have time to work on this, but I think that even 1Cor 7.17 really is a conditional: IF not to each person as the Lord assigned, (i.e.,) each person as the Lord has called them, THEN let them walk in this manner.

There two things that might get closer to what you want in a search.

1) As @Norman Meyer suggests, changing the depth will get different results

2) Note the option to allow or not hits within sub-clauses.

[Norman Meyer]

@mgvh If you're short on time, no problem, but whenever I check or uncheck "Allow hits within sub-clauses" the results stay the same. I don't understand what this is supposed to affect and the documentation isn't clear either. If you have time for a quick sentence or two explanation it would help; but again, no worries if you are short on time. Thanks!

[Norman Meyer]

For anyone interested, I did find out the "Contents" command can be used to achieve the search structure I was desiring above. To do so 1) perform a construct search to find all instances of ει in an independent clause, then 2) perform a construct search to find all the instances of ει in a dependent clause, then 3) use the "Contents" command in a new search window (see screenshot below) to negate the second search. It's not perfect, but it's close (i.e. if there are two instances of ει in a verse this could affect the accuracy of the results; thus it would still be nice if Accordance had a different standalone command that allowed one to use all the actual hits—i.e. instances of the found words—in a search).

 

 

[mgvh]

Quick example...

Mark 2.2 has a complex syntax

For a quick exercise, let's look for instances of συναγω (gather) and θυρα (door). You can see in the syntax that both words are embedded in the same independent clause, but θυρα is buried quite a few levels deep.

First, I'm going to search for an independent clause and allow hits within sub-clauses.

If I start out with a depth of two for both, I won't find Mark 2.2. The verb is actually at a depth of 1, but if I keep changing depth on "door," eventually I get Mark 2.2 w/ a depth of 7.

So, that's how depth works.

 

Now, if I change the depth to infinity BUT I do not allow hits within sub-clauses, I will not get Mark 2.2 since "door" is not at the same level as the "gather."

 

Looking at the syntax chart, I can see that πολλοι, an adjective, is at the same depth (specifically, a depth of 1) as the verb. If I do not allow hits w/in sub-clauses but search for συναγω (gather) and an adjective with at least a depth of one, I will get Mark 2.2.

Hope that helps.