Syntax search fails with agreement of not present term

I am using a syntax search to find participles without an article. The following generally works well (although it picks up John 6:12, which surprised me, but I think I can see why):

But if I add the requirement that the article not agree with the participle, then the search fails and returns nothing.

Agreement between the article and participle is inevitable if the article is a specifier with the participle, so I cannot see why this narrowing down of the search should cease to return any results. Does anyone have insight into what is happening?

Find more posts tagged with

Comments

NetworkGeek

This is strange. I do not think John 6:12 should show up, as it seems to clearly have a specifier and an article. In your first query, John 6:12 shows up whether or not the Modifier 1 is marked as present or not present. I played around with some other approaches, including adding Matching skips levels where necessary for that example, and no matter what John 6:12 always seems to show up. I even tried just two words, "no" article and participle/verb, and John 6:12 shows up.

I am not completely sure I understand what you are looking for. In the second example, if you were saying just exclude certain articles where the case/gender/number match, maybe that word (Word 2) should be the one marked "not present". But that didn't seem to work either.

I am interested now too, to see if someone has an idea. I haven't done a lot with the "not present", I am always trying to find stuff that is present :-) Let's see what others say I hope!

Graham Criddle

https://community.logos.com/discussion/comment/182890#Comment_182890

In terms of Jn 6:14 in the first case I assume it is matching because of their being two head terms, the first of which (green oval) has a specifier and the second (red oval) which doesn't.

No insight on the second question yet!

Graham

Graham Criddle

https://community.logos.com/discussion/comment/182895#Comment_182895

I think there must be some error in the matching logic!

I restricted the search range to Matthew 1 (just to save time and reduce the number of hits) and explicitly specified the case, gender and number of Word1.

When I forced Word2 to have the same values (as below) I got four hits (as expected):

If I changed the definition of Word2 so that it used a rule to match case, gender and number I got no hits:

As far as I can see these two searches are functionally identical which suggests that there is some error in logic in processing the matching rules (as opposed to morphology) - maybe in the case where the term in question is forced to be not present.

If I have missed something obvious I would love to know.

Hope this helps someone to narrow it down a little further!

Graham

Harry Hahne

I also cannot figure out why a similar search would not match Rom 15:6 on πατέρα, which does not have an article.

Rom 15:6 is matched on στόματι, which also has no article. In other instances, it highlights multiple matches in the same verse. Here is how it is tagged in opentext.

You can see that στόματι has a specifier, but not an article, which is fine. But πατέρα has no specifier of any kind, so it seems like it should match.

Graham Criddle

https://community.logos.com/discussion/comment/182903#Comment_182903

Hi Harry

When I run that search I do get a match on πατέρα as shown below (I restricted the search to the verse in question):

The last four matches are all of the last head term in the verse and matching against πατέρα, κυρίου, Ἰησοῦ and Χριστοῦ as expected.

If I receive the "matching skip levels" option on Word1 then I just get the two hits - one against στόματι and one against πατέρα.

So it seems to be working correctly here unless I am missing what you mean.

Graham

Dave Hooton

The following generally works well (although it picks up John 6:12, which surprised me, but I think I can see why):

It gives 4971 results. If you specify the Head Term with NO article (and "Matching skips levels") you get 4968 results. I haven't figured the difference.

Agreement between the article and participle is inevitable if the article is a specifier with the participle, so I cannot see why this narrowing down of the search should cease to return any results. Does anyone have insight into what is happening?

Here you want the specifier to be present NOT absent. And the article to disagree in one of more ways.

This (case disagrees) gives one result in Phil 3:6.

Change it and you find Gender disagrees in Phil 3:6 and Rev 11:4

Number always agrees!

There are 1671 results of participle with Article, and 1669 in which Case/Gender/Number agree.