Jump to content

Search perplexity


Rod Decker

Recommended Posts

I'm stumped. I've tried multiple variations and can't figure this one out. It's probably simple, but any pointers would be appreciated.

 

I want to find all anarthrous, oblique case participles (i.e., gen., dat., acc) that do not have a substantive in the same clause which agrees in gen, number, or case.

 

That should provide a good preliminary list for identifying examples of anarthrous substantival participles (which is my goal). And yes, I know, they could be nom. also, but the potential list will have less chaff if I omit nominatives (which are more likely going to be adverbial).

Link to comment
Share on other sites

Rod,

 

Here's the best I could come up with:

 

1. Drag an Article item into the first column, then drag a NOT item on top of it.

2. Drag a Verb item into the second column. In the dialog box, select Participle from the Mood pop-up menu and Nominative from the Case pop-up menu. Also check the NOT checkbox next to the Case pop-up.

3. Drag the following items into the third column: Noun, Pronoun, Adjective, and Verb. Constrain the Mood of the Verb item to participle. (If I've left out any possible substantives, let me know.)

4. Drag a NOT item over each of the items in the third column.

5. Drag a WITHIN above the first two columns and give it a reasonable proximity. I used 5.

6. Drag an AGREE above the first two columns and check gender, number, case.

7. Drag a WITHIN above the second and third columns and give it a wide proximity. I used 50.

8. Drag an AGREE above the second and third columns and check gender, number, case.

 

This search finds anarthrous nominative participles and excludes the possibility of a following substantive agreeing in gender, number, and case. Unfortunately, it does not exclude the possibility of a preceding substantive agreeing in gender, number, and case. When I attempted to exclude that possibility I kept running into rules against having consecutive negative columns and the like.

 

I hope this helps.

Link to comment
Share on other sites

Unfortunately, it does not exclude the possibility of a preceding substantive agreeing in gender, number, and case. When I attempted to exclude that possibility I kept running into rules against having consecutive negative columns and the like.

 

The consecutive negative cols. was one of the problems I had also.

 

This is a good start. Can we tweak it? What's the possibility of doing two constructs and linking them to handle both the preceding substantive and the following one? I've not tried that before, but I've attached my attempt. Your search is in the first construct window (that alone results in 531 hits), then my attempt at excluding the preceding agreements is in the 2d. When I link the two together with an AND, I get 225 hits, but the very first one is invalid (see the two forms I circled in green).

 

Related question: instead of listing N, P, A, V/p in 3d col., could an ANY tag be used? Does it make any difference?

 

I'll play with this a bit more when I get to my study, but this is my first round this morning.

 

Thanks for your help.

post-14490-126925367744_thumb.jpg

Link to comment
Share on other sites

Rod,

 

I had considered trying to use multiple constructs, but the problem is that there is no way to specify agreement (or lack thereof) between items in two different Constructs. That's why you're getting the false hits.

Link to comment
Share on other sites

Hmmm. I see your point. Does this mean that I've hit a limitation in Acordance search ability? Perhaps this ought to go into the "capabilities expansion" hopper for Roy to work on. There really ought to be a way to do what seems like a not unreasonable nor rare sort of search request.

 

Is there any way to use an ANY in the first column that would exclude either an article or a noun, etc. that agrees with the participle int he 2d col.? I tried a couple variations without finding the right one. (Negative parameters always cause me trouble!)

Link to comment
Share on other sites

I had considered trying to use multiple constructs, but the problem is that there is no way to specify agreement (or lack thereof) between items in two different Constructs. That's why you're getting the false hits.

 

Could you not use the HITS command here? Do the first construct search, open a new tab, put in the HITS command and the AND command and then make a new construct for the other direction. If I'm thinking through the logic correctly, the second construct would be searching only within the hit verses of the first construct, yes? Or am I missing something?

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...