Jump to content

What am I missing (Hebrew Syntax Search)


Martin Z

Recommended Posts

Hello,

 

I tried to reproduce what Dr. J demonstrates in the tutorial podcast: Understanding Syntax (around 16:07). But Accordance cannot find anything.

What am I missing? (I use ANY clause instead of independent clause, which I expected to find more results if there are).

 

post-32138-0-70992600-1513139878_thumb.png

 

Thanks!

Link to comment
Share on other sites

You should either...

1. increase the search depth to at least 1, or

2. insert a SUBJECT PHRASE and PREDICATE PHRASE after the CLAUSE item.

 

In the syntax tree, a subject word is always part of a subject phrase. Same goes for predicate, complement, and adjunct.

 

post-31158-0-85748500-1513168005_thumb.png

Link to comment
Share on other sites

The syntax searching has been refined since Dr. J's podcast. The manual for syntax searching would be helpful for such searches.

Link to comment
Share on other sites

Thanks, Jordan.

Something must have happened to the database.

 

Here is a screenshot of Dr. J's tutorial (Depth = 0):

post-32138-0-81057300-1513169010_thumb.png

Here is my search (depth = 1):

post-32138-0-01456800-1513169038_thumb.png

 

These results seem to be identical.

 

Yours has 11 hits.

There are differences between the searches. I need to dig a little bit more. But if someone can explain mine and Dr. J's searches, it would be helpful!


The syntax searching has been refined since Dr. J's podcast. The manual for syntax searching would be helpful for such searches.

Ah. Here's the official answer! Thanks!

Link to comment
Share on other sites

I sound gave added that Jordan's search is set up correctly. If the phrase levels are left out (I.e., Dr. J's search but depth =1), the results are bound to be "fuzzy"

Link to comment
Share on other sites

Thanks, Dr. Holmstedt for taking your time to answer my question and giving comments.

The 11 hits do look correct to me.

The only thing to refine it is to remove the "Predicate" and "Subject" items. They look redundant. And dropping these two items does not change the search result.

Link to comment
Share on other sites

I may need more help on it.

I changed the number of the predicate into "singular", expecting to see a lot of hits. But there are only 91 hits.

post-32138-0-48317900-1513173198_thumb.png

 

I was expecting at least Gen 1:1 (and a bunch of others) should show up.

 

When I search Elohim as Subject, there are 417 hits.

In theory, the verb is "singular" + the verb is "plural" = 417.

This is what I wanted. How to set up a Construct search that will find all the 417 occurrences where Elohim is the subject (separating the singular verbs and plural verbs search)?

 

Can anyone help?

 

Thanks!

Link to comment
Share on other sites

You have set up the search to look for subject-predicate order. Check the "search both directions" box right below the Hebrew construct and you'll get many more hits.

Link to comment
Share on other sites

Thanks again, Dr. Holmstedt.

 

I did check the search both direction box already.

 

post-32138-0-47736500-1513174461_thumb.png

Link to comment
Share on other sites

Martin,

This is a bug. That search should have grabbed Gen 1:1 (among others).

I will report it.

 

Also, keep in mind that there are instances where אלהים is the subject of a null predicate. 

 

Of course, this got me thinking. So, in the second image below there are a few infinitive constructs tagged as the predicate. 

 

So, your equation should be: 

the verb is "singular" + the verb is "plural" +  other predicates  = 417

 

post-31158-0-73199000-1513176796_thumb.png

 

post-31158-0-69251500-1513177277_thumb.png

Link to comment
Share on other sites

Correction. Not a bug (as Joel pointed out to me). Since the predicate phrase is split in Gen 1:1 (among others), the predicate phrase structure should be set to "one segment" as opposed to "all segments." 

 

Per the Help Files: This option will not find embedded phrases on the same level, such as a subject with parts of a predicate on both sides.

 

In fact, the help files cite Gen 1:1 as an example....sorry for that.

 

post-31158-0-25566800-1513181130_thumb.png

Link to comment
Share on other sites

Hello Jordon,

 

Shouldn't "All segment" include "One segment" search results?

 

2. Phrase Structure—Due to the discontinuity of many Hebrew constituents (see below, §3.2), it is possible to specify if a search should find all elements of constituents, regardless whether they are split or not (“All segment”) or force the search to look for discontinuous constituents (“One segment”—finding split phrases) or non-discontinuous constituents (“Contiguous”—finding phrases without any splits). [A Brief User’s Guide for the Accordance Hebrew Syntax Database, pp. 17-18]

Link to comment
Share on other sites

Martin,

 

Yes, an All segment search will include all segments, *except* for the case where you are looking for elements in between the different segments of a phrase.  Since the predicate phrase in Gen 1:1 is being split by the subject, it is neither before nor after the phrase.  Thus, specifying that "all segments" be before the subject or "all segments" be after the subject will never find a subject in the middle.  This is precisely why we implemented the 'One segment' option.

  • Like 2
Link to comment
Share on other sites

Martin,

 

Yes, an All segment search will include all segments, *except* for the case where you are looking for elements in between the different segments of a phrase.  Since the predicate phrase in Gen 1:1 is being split by the subject, it is neither before nor after the phrase.  Thus, specifying that "all segments" be before the subject or "all segments" be after the subject will never find a subject in the middle.  This is precisely why we implemented the 'One segment' option.

Apparently I have misunderstood the expression "find all elements of constituents, regardless whether they are split or not (“All segment”)."

I thought search for All segment will include those phrases that are "split" as well. I thought this means "split": we will also "find a subject in the middle."

Link to comment
Share on other sites

Joel,

This is an interesting issue -- what, then, distinguishes continguous from all segments in a search for types like Gen 1:1? If all segments restricts the predicates in Martin's search to include only those in which the entire predicate is on one side or the other of the subject, isn't this redundant with specifying the predicate as contiguous?

  • Like 1
Link to comment
Share on other sites

Its important to consider the search as a whole, or try it out yourself!  If you just take the standard predicate phrase "all segments" and search for it in Genesis, you'll see it finds split and not split predicates.  This is exactly as the documentation described.  If you set the phrase to contiguous it will not find any of the split predicates.  This is also exactly as the documentation described.

 

*However*, and I think this is where you are getting tripped up, we aren't just searching for a predicate phrase in this case - we've searched for a predicate phrase preceded by a subject (or followed by with Reverse Directions).  In the case of verses like Gen 1:1, the subject does not precede "all segments" nor does it follow "all segments".  So since the search specifies a particular positional relationship of the subject, that is why it does not find that particular split phrase.

  • Like 1
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...