Trouble figuring out the syntax commands

[Donald Cobb]

Hello all,

 

Can anyone help me with this? I'm looking for the occurrences of πιστεύω taking its object with a dative. I can do this with the construct pane: πιστεύω + (noun, participle) in the dative case. But that also gives me several false positives. With the syntax module, I should logically put the second element as the verbal complement but that's too broad. I'm thinking I should also be able to specify that I want the nous within the same clause but I'm not sure how to do that. Any help on this would be appreciated!

[Lawrence]

I'm not sure how to restrict all components of a search to a single clause, but if the dative followed the verb, you can use this search string:

• πιστευω ([noun dative], [verb participle dative])

The comma within the parentheses serves as a disjunction: it will match a dative noun or a dative participle.

 

If you wanted a bit more leeway to allow intervening words, you can specify that the two parts are with, say, 5 words of each other:

• πιστευω <within 5 words> ([noun dative], [verb participle dative])

[Donald Cobb]

22 minutes ago, Lawrence said:

I'm not sure how to restrict all components of a search to a single clause, but if the dative followed the verb, you can use this search string:

• πιστευω ([noun dative], [verb participle dative])

The comma within the parentheses serves as a disjunction: it will match a dative noun or a dative participle.

 

If you wanted a bit more leeway to allow intervening words, you can specify that the two parts are with, say, 5 words of each other:

• πιστευω <within 5 words> ([noun dative], [verb participle dative])

 

Thank you Lawrence. Yes, I can do that, the problem is the false positives which I'd like to avoid, e.g.: 

 

καὶ μακαρία ἡ πιστεύσασα ὅτι ἔσται τελείωσις τοῖς λελαλημένοις αὐτῇ παρὰ κυρίου. (Luke 1,45)

 

⸀εὐθὺς κράξας ὁ πατὴρ τοῦ παιδίου ⸆ ἔλεγεν· πιστεύω· βοήθει μου τῇ ἀπιστίᾳ. (Mark 9,24)

 

This is where a syntax tree should be helpful, to keep the dative noun as the object of the verb. I'm sure it can be done, I'm just not familiar enough with the logic of the syntax module.

[darrylmy]

Hi @Donald Cobb, I was able to build a construct search that seems to find mostly what you are looking for (32 instances, although Rom 10:19 does show up with πιστεύσῃς modified by the prepositional phrase, ἐν τῇ καρδίᾳ σου; 1 Th. 1:7 also seems to be a false positive as well).

 

Here is the construct I used:

 

The first thing you have to change is the phrase parameters (double click the heading "PHRASE"):

 

 

Be sure to deselect "Allow hits within sub-classes". From the help: "If this option is selected, elements in other clauses will be included in the search." This will result in false positives.

 

 

Change the depth for each column. With the depth of the noun being infinite, it gives 32 hits. But oddly enough, there are only 28 occurrences (with the hits in Rom 10:19 and 1 Th 1:7 totaling two hits). I think this is because Rom 10:9 is being modified by a prepositional phrase (the hit) and has a direct object ὅτι clause (which accounts for an extra hit which isn't highlighted because its not a dative noun but still syntactically related). Additionally, in 1 Th 1:7, πιστεύουσιν is being modified by two prepositional phrases, but only accounts for one (of the objects of the preposition), which accounts for the additional hit.

 

All that said, I adjusted the depth (from the help, "how closely related the searched terms must be relative to this phrase") of the noun column to "2," and I correctly get 28 hits (without Rom 10:9 and 1 Th 1:7).

 

Hopefully that was clear, I was kind of learning as I was going.

 

 

 

 

[Kristin]

Hi @Donald Cobb,

I had been waiting to see how others would respond, but at this point I would just like to add my two cents that I have never found constructs to be 100% accurate. They will often produce a few false hits or not pick up on a few hits. This is not really an issue with the construct (I don't think), but rather because the construct is based on the assumption that everything has been tagged correctly. And as shown by a few threads going around with the issues with the "Word Study", sometimes accurate tagging is an issue.

As a result, I find the construct to be VERY helpful to get a general idea, or in situations where 90% is good enough. However, if I really need accurate results, I frankly do it by hand.

Kristin

[mgvh]

I also do not have full confidence in a construct search. Partly it's a matter of my uncertainty about how it all works with the various depth levels, etc.

But, I also think that we are counting on the accuracy of the syntax analysis that the construct search uses. There are times where there can reasonably be multiple ways of analyzing a sentence structure, and I have found instances which, imo, are inconsistencies on how things analyzed.

Still, I'm from back in the day when I had to use the Moulton & Geden Concordance and plow through things one at a time...

[Donald Cobb]

On 12/3/2022 at 7:01 PM, darrylmy said:

Hi @Donald Cobb, I was able to build a construct search that seems to find mostly what you are looking for (32 instances, although Rom 10:19 does show up with πιστεύσῃς modified by the prepositional phrase, ἐν τῇ καρδίᾳ σου; 1 Th. 1:7 also seems to be a false positive as well).

 

Here is the construct I used:

 

The first thing you have to change is the phrase parameters (double click the heading "PHRASE"):

 

 

Be sure to deselect "Allow hits within sub-classes". From the help: "If this option is selected, elements in other clauses will be included in the search." This will result in false positives.

 

 

Change the depth for each column. With the depth of the noun being infinite, it gives 32 hits. But oddly enough, there are only 28 occurrences (with the hits in Rom 10:19 and 1 Th 1:7 totaling two hits). I think this is because Rom 10:9 is being modified by a prepositional phrase (the hit) and has a direct object ὅτι clause (which accounts for an extra hit which isn't highlighted because its not a dative noun but still syntactically related). Additionally, in 1 Th 1:7, πιστεύουσιν is being modified by two prepositional phrases, but only accounts for one (of the objects of the preposition), which accounts for the additional hit.

 

All that said, I adjusted the depth (from the help, "how closely related the searched terms must be relative to this phrase") of the noun column to "2," and I correctly get 28 hits (without Rom 10:9 and 1 Th 1:7).

 

Hopefully that was clear, I was kind of learning as I was going.

 

 

 

 

 

Thank you Darryl, this is helpful. Over the weekend I tried a couple things using "predicate" and "complement" categories but that left out several passages that should have been included. With the search you suggest, there are a few false positives (Rm 8,18; 1 Tm 3,16; Rm 10,10). Interestingly, when I ran this, it didn't pick up either Rm 10,9 or 1 Th 1,7. I'm not sure why. 

 

Other anomaly: in all but one result, the search picks up πιστεύω + noun with no intervening preposition (έν). However, it does pick up Mc 1,15 (πιστεύετε ἐν τῷ εὐαγγελίῳ) but not Jn 3,15 (ἵνα πᾶς ὁ πιστεύων ἐν αὐτῷ ἔχῃ ζωὴν αἰώνιον). Although this verse can be translated "that anyone who believes might have eternal life in him"), the syntax tree doesn't seem to have analyzed it it that way. 

 

Thanks also to @Kristin and @mgvh for your comments. I agree that the syntax modules don't inspire 100% confidence. I asked the question in this post, in part, to try to get a handle on this module that I've had for a while but have yet to really get a handle on. I don't know if there is an in-depth explanation somewhere. I'd love to see it if there is. Obviously, to make sure the search was complete would mean doing more thorough research. This is a good start, though. 

 

[mgvh]

After a LOT of playing around w/ parameters, this is the construct I used that returned the best results. There are still some false hits, and I don't know how many missed hits.

Note:

Predicate phrase: one segment; do not allow hits w/in subclauses

Complement phrase: all segments; do not allow hits w/in subclauses

So much depends on depth.

 

[darrylmy]

After trying to discover the logic of how mgvh's got to his search, I ended up at the same search by selecting a positive hit and right-clicking to choose the "construct" > "clause" option and then modifying it to exclude specific grammatical parameters (to generalize them). This technique might be a helpful begging point in the future.

 

Also, when modifying the depth from mgvh's search, at least two missing hits turned up. Here is the search I used (the parameters look slightly different because of how I got to the search):

 

 

In all honestly, even with playing around with this search, I still haven't gotten my head around syntax searching.

[mgvh]

So.... I went into my Logos program to see what I could find there. I spent way too much trying to figure out how to do a Syntax search and couldn't generate any results at all. I ended figuring out how to do it with their Clause search, but it wasn't immediately intuitive. In fact, I got no results at all until I discovered that in their coding scheme, a dative object for the verb pisteuw was labeled as an indirect object.

Below is the list that it generated, and it looks very good. I compared it to the list I got with Accordance, but I still can't get rid of instances with datives in prepositional phrases.

Logos also found 3 hits that Accordance missed: Mark 16.14; John 5.24; Acts 24.14. Those are in somewhat complicated constructions, so something may be going on with that. In any case, if you @darrylmyare still working on it, see if you can generate a list to match this.

Matthew 21:25
Matthew 21:32
Mark 11:31
Mark 16:13
Mark 16:14
Luke 1:20
Luke 16:11
Luke 20:5
John 2:22
John 2:24
John 4:21
John 4:50
John 5:24
John 5:38
John 5:46
John 5:47
John 6:30
John 8:31
John 8:45
John 8:46
John 10:37
John 10:38
John 12:38
John 14:11
Acts 5:14
Acts 8:12
Acts 16:34
Acts 18:8
Acts 24:14
Acts 26:27
Acts 27:25
Romans 4:3
Romans 10:16
Galatians 3:6
2 Thessalonians 2:11
2 Thessalonians 2:12
2 Timothy 1:12
Titus 3:8
James 2:23
1 John 3:23
1 John 4:1
1 John 5:10

 

 

[Donald Cobb]

Many thanks for this Mark and Darryl. I'm having trouble reproducing the search. I'll have to come back to it a little later. The presence of a preposition didn't necessarily bother me. What surprised me was more that some results didn't exclude the ἐν whereas others were clearly left out because of it. 

[darrylmy]

After spending some time on this search I was only able to get close to the results you @mgvh posted, but with the following differences:

 

- John 2:24
+ John 16:30
- Acts 18:8
+ Rom 4:18
+ 1 Cor. 3:5

 

Here is the construct that got those results: