Understanding a Syntax Search

Michel Pauw

I have two questions:

1. (easy one) is there already any guide available that helps me understand all the functions of the Syntax Search? I do not mean basic functions, but one that should give an answer to questions like 'how do I set up a good query to find all SVO sentences that has a genitive as part of the subject?'
2. I was reproducing the query that is displayed in the 8th screenshot on https://wiki.logos.com/A_Strategy_for_Syntax_Search, but my query gives 0 results. What do I wrong? Here is my screenshot.

By the way: I am trying to set up the query in my first question, but first tried to get results without the genitive part.

Graham Criddle

Regarding your second question.

The wiki screenshot has dashed lines, indicating "matching skips levels"

Try that.

Sorry can't go into more detail at the moment as not near my computer

Dave Hooton

Michel Pauw said:

(easy one) is there already any guide available that helps me understand all the functions of the Syntax Search? I do not mean basic functions, but one that should give an answer to questions like 'how do I set up a good query to find all SVO sentences that has a genitive as part of the subject?'

The main thing is to select the appropriate Greek Syntax engine and follow the strategy of running a Bible/Morph Search to give you candidate results whose structure you can study/emulate in the Clause Visualization resources in order to provide a suitable syntax query. My preferences for Greek are

1. Cascadia - its structure produces more compact/powerful queries; or
2. OpenText - in some ways the simplest to use; and
3. Lexham Syntactic - its grammatical elements are restricted and difficult to use

When utilising a candidate result for your query understand that all of its elements are not necessary as you can use "Matching skips levels" to provide a more generic query (to match other results). Use "Unordered" to allow for different order of elements.

Michel Pauw

https://community.logos.com/discussion/comment/778490#Comment_778490

Graham Criddle said:

The wiki screenshot has dashed lines, indicating "matching skips levels"

Try that.

That doesn't help. I now made sure that everything is exactly the same as in the example. I have now 4 line arrows and 2 dotted arrows. It gives 0 results in 1.83 sec [:(]

Michel Pauw

https://community.logos.com/discussion/comment/778701#Comment_778701

Dave Hooton said:

When utilising a candidate result for your query understand that all of its elements are not necessary as you can use "Matching skips levels" to provide a more generic query (to match other results). Use "Unordered" to allow for different order of elements.

Thank you, your advise pointed me to the Clause Visualisation resources.

Another question: what is the difference between 'matching skips levels' and 'unordered'?

Dave Hooton

https://community.logos.com/discussion/comment/782697#Comment_782697

Michel Pauw said:

Another question: what is the difference between 'matching skips levels' and 'unordered'?

They are different functions:

• Skipping levels allows you to specify  something like Clause -----> Word without concern for the intermediate levels e.g. instead of Clause ---> Clause Function ----> Word  or  Clause ---> Clause Function ----> Phrase  ---> Word. But if you know that candidate results don't always have a Phrase then Clause ---> Clause Function ----> Word (with 'skipping' on the Word) is more accurate than Clause ---> Word (with 'skipping' on the Word) because you can specify what the Clause Function should be.
  
  
• 'Unordered' is useful if (for example) a Clause contains a Subject Clause Function and a Verbal Clause Function but they can appear in any order

It saves having to repeat the structure with the Verbal function followed by the Subject. There is also an implicit "matching skips levels" between the items on either side of Unordered i.e. between the Clause and the Clause Functions.

Michel Pauw

https://community.logos.com/discussion/comment/782920#Comment_782920

Thank you, Dave. That makes sense. And your example gives me also a nice view of what searches I can do. So far I didn't realize that you can combine case, clause function and part-of-speech in one Clause Function. That's pretty powerful...