Problem with Greek searching

[Donald Cobb]

Perhaps @Silas Marrs or @David Lang can help here. Either there's something I don't understand or there's a truly bothersome aspect to the search function:

 

- I was wanting to do a search on πέμπω in the fourth Gospel, specifically trying to find those passages where God/the Father sends the Son. So I construct this search: πεμπω@ [VERB singular aorist] [RANGE John] . It only returns one hit, i.e., John 13,20. Strange behavior since, although there are two occurrences in this verse alone, it only turns up one of them (I've bolded the first and underscored the second, which was not picked up): John 13,20 ἀμὴν ἀμὴν λέγω ὑμῖν, ὁ λαμβάνων ἄν τινα πέμψω ἐμὲ λαμβάνει, ὁ δὲ ἐμὲ λαμβάνων λαμβάνει  ⸆ τὸν πέμψαντά με.

 

It's actually worse than this, because a search specifically on the the active aorist participial form (πεμπω [VERB aorist active participle singular] [RANGE John]) turns up 26 hits. I don't understand the difference here because Acc. parses participles as verbal forms (of course). So, by any normal standards, my initial search should turn up 27 occurrences. 

 

- Looking further into this I see that, when I select "participle," it grays out "number." There is a second "number" category that comes up that has to be checked. I'm not sure why this is so, but it means that, in order to do a complete search, I either have to do two separate searches, one on verbal forms and the other one participial forms, or combine the two categories [OR]. This is not at all intuitive and shouldn't be necessary. The Father's "sending" the Son is is a major theological theme in the fourth Gospel, so when my search only turned up one occurrence, I immediately knew there was something wrong. But I shouldn't have to remind myself of this idiosyncrasy every time I want to search one something similar elsewhere. 

 

Can this issue please be addressed? Thanks!

 

 

[c. stirling bartholomew]

This isn't new. Get same results in Acc 9.6.8. You can safely ignore number if that is giving you grief. πέμψω plural isn't common.  The hierarchy of tagging in Accord has some interesting features. Long ago given up wondering why. I Just find a work around. 

 

Edited July 2, 2023 by c. stirling bartholomew

[Donald Cobb]

39 minutes ago, c. stirling bartholomew said:

This isn't new. Get same results in Acc 9.6.8. You can safely ignore number if that is giving you grief. πέμψω plural isn't common.  The hierarchy of tagging in Accord has some interesting features. Long ago given up wondering why. I Just find a work around. 

 

 

Thanks Stirling. I'm not too concerned about the specific example. I've taught on John's Gospel for practically twenty years, so it was just to verify some statistics. What bothers me is realizing that in other cases, it could be easy to blithely ignore the fact that, without what is a fundamentally unnecessary two-step process, significant data is not being taken into account.

[Tim Planche]

same happens for me on 14 using a mac

 

I reprise the issue water I do a construct search or typed search and across different bibles.

 

There does seem to be an issue with how number is tagged in participles and other moods.

 

interesting error message about searching for singular in both participles and other moods if you try in  a construct search  

 

seems number is tagged separately for participles and other moods of verbs - which kind of makes sense given he way it may search in different hierarchies 

 

as a solution - the following seems to work and finds the required 27 instances of both participles and a subjunctive 

 

"πεμπω @ [VERB aorist -plural  ] [RANGE john] "

 

will be goods to hear from those who know more than I  - but is looks like you have to treat participles and other moods differently when searching with number 

 

[jlm]

23 hours ago, Donald Cobb said:

Perhaps @Silas Marrs or @David Lang can help here. Either there's something I don't understand or there's a truly bothersome aspect to the search function:

 

- I was wanting to do a search on πέμπω in the fourth Gospel, specifically trying to find those passages where God/the Father sends the Son. So I construct this search: πεμπω@ [VERB singular aorist] [RANGE John] . It only returns one hit, i.e., John 13,20. Strange behavior since, although there are two occurrences in this verse alone, it only turns up one of them (I've bolded the first and underscored the second, which was not picked up): John 13,20 ἀμὴν ἀμὴν λέγω ὑμῖν, ὁ λαμβάνων ἄν τινα πέμψω ἐμὲ λαμβάνει, ὁ δὲ ἐμὲ λαμβάνων λαμβάνει  ⸆ τὸν πέμψαντά με.

 

It's actually worse than this, because a search specifically on the the active aorist participial form (πεμπω [VERB aorist active participle singular] [RANGE John]) turns up 26 hits. I don't understand the difference here because Acc. parses participles as verbal forms (of course). So, by any normal standards, my initial search should turn up 27 occurrences. 

 

- Looking further into this I see that, when I select "participle," it grays out "number." There is a second "number" category that comes up that has to be checked. I'm not sure why this is so, but it means that, in order to do a complete search, I either have to do two separate searches, one on verbal forms and the other one participial forms, or combine the two categories [OR]. This is not at all intuitive and shouldn't be necessary. The Father's "sending" the Son is is a major theological theme in the fourth Gospel, so when my search only turned up one occurrence, I immediately knew there was something wrong. But I shouldn't have to remind myself of this idiosyncrasy every time I want to search one something similar elsewhere. 

 

Can this issue please be addressed? Thanks!

 

 

I think the logic here is that a participle is not singular in the same sense that a finite verb is singular. So the first search is looking for verbs that have a singular subject, and excludes participles, because they have no subject. If you want to find verbs that are singular in either sense, you need an OR.

Edited July 3, 2023 by jlm
Fix typo

[Dr. Nathan Parker]

I'll monitor this. If there's a bug report or feature request I need to file anytime, just let me know.

[Donald Cobb]

1 hour ago, Nathan Parker said:

I'll monitor this. If there's a bug report or feature request I need to file anytime, just let me know.

 

Thanks Nathan. I started the post as a bug report but as I wrote and tried to work out the problem a little more clearly, I realized it was more a design problem. So does that make it a bug report or a feature request? Perhaps @Silas Marrs or @David Lang would be best positioned to say.

[Donald Cobb]

6 hours ago, jlm said:

I think the logic here is that a participle is not singular in the same sense that a finite verb is singular. So the first search is looking for verbs that have a singular subject, and excludes participles, because they have no subject. If you want to find verbs that are singular in either sense, you need an OR.

 

From a grammatical point of view, I suppose that could make sense. But from a practical point of view, if I'm looking for a singular form of a verb and participles are included as a form of the verb, I really don't understand the usefulness of separating the two. If I wanted to separate the participles, it would make better sense to use the "not" function.

[darrylmy]

1 hour ago, Donald Cobb said:

from a practical point of view, if I'm looking for a singular form of a verb and participles are included as a form of the verb, I really don't understand the usefulness of separating the two

 

Until David os Silas jump in, perhaps the relevant info from the Help files can help shed light on the situation. Regarding this issue, the Help files under "Tag Code" it states:

 

Quote

The participle number is a separate tag from the regular verb number. Thus to find all plural verbs search for [VERB plural] <OR> [VERB participle plural].

 

So from a practical point of view this actually makes sense because there is unique tagging. And there has to be. Grammatically, the number for verbs is connected to their person (1st, 2nd, 3rd). Grammatically, participles are not just verbs, but they are verbal-adjectives, and the number for participles is connected to their adjectival characteristic (rather than to number regarding person which the participle does not have).

 

Regarding the <NOT> function, if you think about it, when you enter πεμπω@[VERB singular aorist] [RANGE John] (i.e., specifying the the number for person), Accordance could be considered (for all intents and purposes) to be effectively applying a <NOT> command implicitly (in that it is not searching for participles because of the person parameter).

Not sure if that makes sense, but it's making sense in my head, haha. Hope that helps!

[Dr. Nathan Parker]

For now, file it as a Feature Request. I can always move it to a bug report if need be.