13 replies to this topic

[JonathanHuber]

I'm trying to find the NT clauses introduced by ωστε where the main verb in the clause is in the indicative mood, such as in John 3:16. Searching by morphology is insufficient, especially since some clauses include other verbs between ωστε and the main verb (look at Phil. 2:12, where ωστε is the first word of the sentence and the main verb is very last!). Thus, I'm trying to set up a syntax construct but am uncertain how to define this search. For example, how would you define the relationship between ωστε and εδωκεν in John 3:16? The best I've come up with is shown in the attached screenshot, but it's finding some false hits. Does anyone have an idea how to better define this?

Thanks!
Jonathan

Attached Files

•  Screen shot 2011-11-18 at 12.03.59 PM.jpg   18.64K   18 downloads

[James Tucker]

I'm trying to find the NT clauses introduced by ωστε where the main verb in the clause is in the indicative mood, such as in John 3:16. Searching by morphology is insufficient, especially since some clauses include other verbs between ωστε and the main verb (look at Phil. 2:12, where ωστε is the first word of the sentence and the main verb is very last!). Thus, I'm trying to set up a syntax construct but am uncertain how to define this search. For example, how would you define the relationship between ωστε and εδωκεν in John 3:16? The best I've come up with is shown in the attached screenshot, but it's finding some false hits. Does anyone have an idea how to better define this?

Thanks!
Jonathan

Jonathan,
I dont have access on my MacBook Pro to post a screen shot, so I will do so as soon as I can get access.
Here is how to get your results:

• Open GNT-T
• Add Syntax
• Open Construct
• Drag in a Phrase, and designate it as a Predicate Phrase.
• Drag in Adjunct in the first column beneath the Predicate Phrase
• In column under the Predicate, yet adjacent to the Adjunct, drag in the Clause and designate it as a depedent clause.
• The drag in below the Depedent Clause a Predicate
• Then, drag in below the Predicate Verb, and specify that you want the mood to be indicative.

This will find similar syntactical structures as you've witnessed in John 3:16.

 hWSTE w: Indicative.png   31.57K   16 downloads

[JonathanHuber]

Jonathan,
I dont have access on my MacBook Pro to post a screen shot, so I will do so as soon as I can get access.
Here is how to get your results:

• Open GNT-T
• Add Syntax
• Open Construct
• Drag in a Phrase, and designate it as a Predicate Phrase.
• Drag in Adjunct in the first column beneath the Predicate Phrase
• In column under the Predicate, yet adjacent to the Adjunct, drag in the Clause and designate it as a depedent clause.
• The drag in below the Depedent Clause a Predicate
• Then, drag in below the Predicate Verb, and specify that you want the mood to be indicative.

This will find similar syntactical structures as you've witnessed in John 3:16.

 hWSTE w: Indicative.png   31.57K   16 downloads

Thanks for the response. This gets a little closer, but it's still not quite right. For example, in Mark 1:27, the main verb of the hwste clause in an infinitive. The indicate verb in the clause is buried in a sub-clause (is that a term?). This is sort of a test case for me learning the syntax module better: how do I limit the construct search to the higher levels of the syntax tree? Is that possible?

Jonathan

[Robert Holmstedt]

Thanks for the response. This gets a little closer, but it's still not quite right. For example, in Mark 1:27, the main verb of the hwste clause in an infinitive. The indicate verb in the clause is buried in a sub-clause (is that a term?). This is sort of a test case for me learning the syntax module better: how do I limit the construct search to the higher levels of the syntax tree? Is that possible?

Jonathan

Jonathan,

That kind of level specificity is inherent to the tagging scheme and will most certainly be accessible with further refinement of the search programming. We are getting there . . .

Good question and thank you for continuing to work with the syntax modules. More questions like these help us to either explain or improve the modules, or both.

Edited by Robert Holmstedt, 18 November 2011 - 07:48 PM.

[James Tucker]

1321661878[/url]' post='30020']
Thanks for the response. This gets a little closer, but it's still not quite right. For example, in Mark 1:27, the main verb of the hwste clause in an infinitive. The indicate verb in the clause is buried in a sub-clause (is that a term?). This is sort of a test case for me learning the syntax module better: how do I limit the construct search to the higher levels of the syntax tree? Is that possible?

Jonathan

Jonathan,

I am confused. I thought you wanted to find instances of this syntactical construction with the indicative. If you wanted to find any verb, then delete the indicative specification from the Verb Tag. This will include verbal moods such as infinitives.

**Disregard. I see what you are saying. I apologize. I misunderstood that Mk. 1:27 is included within the indicative search when in fact it should not be.

Edited by James Tucker, 18 November 2011 - 08:11 PM.

[JonathanHuber]

Thank you both for your replies. I ultimately got my answer on hwste by the time-tested method of looking up every reference one by one. (This is still much easier in Accordance than if I had to use only print resources.) I'm happy to have such powerful tools and look forward to the continued improvements.

Jonathan

[Marco V. Fabbri]

Jonathan,

perhaps we may go further even using the syntax as it already is.

It is true, as Robert says, that at the moment the search engine doesn't know whether a verb is part of Clause A, or it is part of Clause B that is subordinate to clause A. As in Greek there are many nested subordinates, this problem is a frequent one.

However, we may go around it and restrict the search results.

We can tell Accordance that the Adjunct Clause that includes the indicative (that is, the result Clause where ὥστε takes the indicative) must be one that doesn't also include an infinitive.
I will also tell Accordance that the Dependent Clause must be an Adjunct dependent Clause, as result Clauses are Adjunct Clauses.

This would look like this:

 AdjDepClause.PNG   16.4K   14 downloads

This result will exclude independent Clauses that start with ὥστε after a fullstop. There are some in the GNT. If we want to catch them, too, then we will need to change the Adjucnt dependent Clause to Any Clause.

It will look like this:

 anyClause.PNG   16.31K   14 downloads

If we look at the results, they will now include such verses as Matthew 19:6 or 23:31.

[JonathanHuber]

Oh, this was really helpful. Thank you!

Jonathan

[Marco V. Fabbri]

Not at all.

Perhaps I need to add that the search that I suggested wouldn't catch Phil 2:12, as the verb is in the Imperative.

You would find it, though, with the search for Any Clause, if you changed the verb from Indicative to Imperative

[JonathanHuber]

I realized from your previous post that it's also possible to find some of these results based on morphology alone. Search by verse, clause, or sentence, and make a construct search for wste followed by an indicative with no infinitive between them. That isn't the most precise and of course won't distinguish different types of clauses, but it will find some of the occurrences in books not yet covered by the syntax module.

[Marco V. Fabbri]

I agree that this works. However, it is not a search based on morphology alone. The construct knows about sentences and clauses because they are tagged as such in syntax add-on.
An important part of preparing the syntax consisted of marking where every clause starts and where it ends.
And it is true, as you suggest, that this is enough to prepare a lot of interesting searches that formerly were not possible.

[Helen Brown]

The search with a clause or sentence field is based on the punctuation of the GNT, as it always has been. Of course, the syntax is more accurate, and may produce different results.

[Marco V. Fabbri]

Helen:

I think that you refer to the Search Text window. Then you are right, of course.

Perhaps I was confused when I read that Jonathan talked about a construct search. I thought of the Greek Construct window, where I still expect that Clause means the L element in the syntax.

However, when I re-read Jonathan, I see that he mentions the books that are not covered in the syntax add-on, where he would get no results if the syntax were actually involved, as it is in the Greek Construct. So Search Text it was.

[JonathanHuber]

Yes, I meant the search text window. Sorry, I should have been clearer. An example of this search is below for anyone interested in this.

 Screen shot 2011-11-25 at 12.24.28 PM.jpg   210.2K   10 downloads

Jonathan