Searching in Accordance - a more fully worked example

[Λύχνις Δαν]

Hi ya,

 

  Some time back this query https://www.accordancebible.com/forums/topic/25490-complex-search-in-greek-nt was posted in the forums. At that time I wanted to give a fuller answer than I did in the posts. It has taken a while but I have now posted three videos showing how one can go about this search and the pros and cons of various techniques.

 

  They are :

 

  Part I: Search box based query: https://youtu.be/3owYZgoiNQ0

  Part II: Construct query: https://youtu.be/r5DexXn7uBE

  Part III: Syntax search: https://youtu.be/juRd5usxZNg

 

  I've never actually tried to do this before, and having done more reruns over each than I care to count I am more impressed than ever by the quality of Dr. J's videos. It is really hard to do this well. There are many things one might add to what I've said but I tried to be good and stick to the script  

 

  Anyhow, if you have questions or corrections let me know. Hopefully these are helpful.

 

Thx

D

[accord]

Hi ya,

 

  Some time back this query https://www.accordancebible.com/forums/topic/25490-complex-search-in-greek-nt was posted in the forums. At that time I wanted to give a fuller answer than I did in the posts. It has taken a while but I have now posted three videos showing how one can go about this search and the pros and cons of various techniques.

 

  They are :

 

  Part I: Search box based query: https://youtu.be/3owYZgoiNQ0

  Part II: Construct query: https://youtu.be/r5DexXn7uBE

  Part III: Syntax search: https://youtu.be/juRd5usxZNg

 

Well done, and much appreciated Daniel.

[jlm]

I appreciated these too, especially the last video. The only thing that seemed to be missing is an explanation of why you chose not to express certain characteristics, specifically the fact that there are singular nouns.

[Λύχνις Δαν]

I thought I explained this but perhaps I missed it. I omitted two features from the query. They were a) whether the nouns were articular or not, and b ) whether they were singular or not. Both were excluded for the same reason: the construction occurs with and without the article and with both singular and plural nouns.

 

 

Thx

D

Edited June 9, 2019 by דָנִיאֶל

[GYDOF1729]

Thanks so much.  

[Boris Repschinski]

I was wondering why, in the syntax search, you did not check the "search both directions" box - that gives plenty more positives.

[Mark Allison]

I was wondering why, in the syntax search, you did not check the "search both directions" box - that gives plenty more positives.

 

Right. When searching in Greek, and even more in Hebrew, "search both directions" should almost always be checked.

Edited December 11, 2019 by Mark Allison

[Λύχνις Δαν]

Hi Boris,

 

  Thanx for viewing the videos and commenting.

 

  I had to go back to the other post (the query about the construction) and then to my notes to see what happened there. It is true that this phenomenon occurs in pre- and post-verbal forms.I suspect I was trying to keep the three searches broadly comparable and wanted to roughly match the search box based query which I think was basically one directional.

 

  I have a note all in bold letters in my script notes for the syntax query case to mention in the video using search both directions if you are interested in both forms. I recorded each video several times, as I'm rubbish at this and kept making errors. I know that there were small differences between my rough script notes and what I actually said in some versions. If I forgot to say this in the version that I uploaded, ooops. I clearly did intend to based on my notes. I see actually I had a note that I should have mentioned this in video 2 also.

 

  I do not always use search both directions actually. It really depends what I am trying to find.

 

Thx

D

Edited December 11, 2019 by דָנִיאֶל

[mgvh]

So... I thought I would try to run the search that Daniel was conducting using the new highlight > construct feature in Acc 13.

I went to his example, Mark 4.41, highlighted  ὁ ἄνεμος καὶ ἡ θάλασσα ὑπακούει, right-clicked and chose Construct > Clause

This automatically creates the construct search.

BUT, without changing anything, I ran the search, and it returned no results. ???

I tried it again, but this time I accidentally included the και before ὁ ἄνεμος, and this time it did give me Mark 4.41.

Am I misunderstanding something about clause construct searches?

 

So, I repeated above (without the και), but this time chose Construct > Phrase.

No results. Same deal if I use Sentence.

 

So, I repeated above, but this time chose Construct > Word.

This seems to work.

 

Now, my point here is that I can then simply start removing the specifics.

If I do so with the Word construct, I get a lot of false hits, since it finds many nominatives linked with a conjunction and a singular verb in there somewhere. (E.g., the genealogy in Matthew 1)

 

It would seem to me that the Construct > Clause search should be the way to go, but I can't make it work.

[EDIT: I just went back and looked again at the video with the Syntax search, and you point out the issue with that και in Mark 4.41.]

 

BTW, thanks to Daniel for starting this and sharing the videos. Those are a lot of work!

Edited December 11, 2019 by mgvh

[John Fidel]

I tried the same thing as you and returned no results without changing anything.

 

As a suggestion for improving this it would be helpful to allow us to choose what elements to include such as syntax etc. before constructing the search.

[Λύχνις Δαν]

BTW, thanks to Daniel for starting this and sharing the videos. Those are a lot of work!

 

Welcome. i certainly got a new appreciation of the amount time Dr. J had to spend to produce the videos he did !

 

 

It would seem to me that the Construct > Clause search should be the way to go, but I can't make it work.

[EDIT: I just went back and looked again at the video with the Syntax search, and you point out the issue with that και in Mark 4.41.]

 

I did ? What did I say ? I don't recall and cannot really view the vids right now. (These were done before 13's new feature)

 

thx

D

Edited December 11, 2019 by דָנִיאֶל

[Λύχνις Δαν]

So... I thought I would try to run the search that Daniel was conducting using the new highlight > construct feature in Acc 13.

I went to his example, Mark 4.41, highlighted  ὁ ἄνεμος καὶ ἡ θάλασσα ὑπακούει, right-clicked and chose Construct > Clause

This automatically creates the construct search.

BUT, without changing anything, I ran the search, and it returned no results. ???

I tried it again, but this time I accidentally included the και before ὁ ἄνεμος, and this time it did give me Mark 4.41.

Am I misunderstanding something about clause construct searches?

 

So, I repeated above (without the και), but this time chose Construct > Phrase.

No results. Same deal if I use Sentence.

 

So, I repeated above, but this time chose Construct > Word.

This seems to work.

 

Now, my point here is that I can then simply start removing the specifics.

If I do so with the Word construct, I get a lot of false hits, since it finds many nominatives linked with a conjunction and a singular verb in there somewhere. (E.g., the genealogy in Matthew 1)

 

It would seem to me that the Construct > Clause search should be the way to go, but I can't make it work.

[EDIT: I just went back and looked again at the video with the Syntax search, and you point out the issue with that και in Mark 4.41.]

 

BTW, thanks to Daniel for starting this and sharing the videos. Those are a lot of work!

 

 

The problem appears to be that the the amplify is producing a Subject Phrase - Subject Phrase hierarchy when it is modelled as a Subject Phrase - Compound Subject in the syntax. If you remove the lower Subject Phrase and increase the depth of the remaining one to 1 (from 0) then 4:41 is found.

 

Now I could not enter a compound subject as a phrase so it cannot be created this way in a normal search, if you want to specify the word level syntax for the compound subject.

 

But that suggested another test - turn off word-level syntax and grammatical details in the amplify support. But that fails to find it also because the amplify is inserting a Subject Phrase - Subject Phrase hierarchy. But replacing the lower Subject Phrase with a Compound Subject does find it.

 

Basically, you need a way to have a Compound Subject Phrase. And it doesn't exist.

 

Thx

D

[Boris Repschinski]

 

  I've never actually tried to do this before, and having done more reruns over each than I care to count I am more impressed than ever by the quality of Dr. J's videos. It is really hard to do this well. There are many things one might add to what I've said but I tried to be good and stick to the script

 

I can only second this - the videos are really helpful, yours as well as those of Dr. J.

 

Thinking about your videos again, and about how I usually go about the searches I do, I would make a further suggestion on planning a search with respect to the lists you provide. I would add  the question whether I am actually searching for a syntactical construction or not. That will help in deciding which search method I use. With the example you provided I would have gone straight to the third method. And your videos made clear to me why that is a good idea.

Edited December 13, 2019 by Boris Repschinski