[Robert Evans]

Is it possible to search the BHS-W4 text for all verbs that have, for example, a Strong's number in the 3000s?

I need to include Strong's numbers in some searches because one of my two other co-authors on a book project doesn't read Hebrew and works on a Windows machine using Logos Scholar Silver (or something like that). He indicates that Logos can easily pull up a list of verbs in the Hebrew Bible, in Strong's number order.

The way I've been able to do it so far in Accordance yields only about a dozen or so verbs, but misses quite a few more, according to my Logos-using colleague. He's right at least about some of them (for example, Accordance shows H3001 - not in my search results - as an infinitive construct that is clearly identified as a "VERB" in the Instant Details box, so I must have not entered the search argument correctly). Maybe someone has already done this and can help me find my mistake? Thanks for any help you can give me on this.

Helen has suggested a work-around to this challenge, involving separate searches for all verbs with lemmas starting with yod, kaph, lamedh and mem (e.g., y*@[VERB], etc.). This way, I would just manually look up the first and last Strong's numbers in the 3000s range, do the four searches pe-yod, pe-kaph, pe-lamed, and pe-mem, manually select only those of the the pe-yod and pe-mem verbs that are between the first and last Strong's numbers inclusive, and cobble together a complete list - that Logos can do with one search argument, or so I'm told.

Surely Accordance has a more excellent way?

In His service,
Robert

===================================================
Robert L. Evans,
Assoc. Dean, College of Humanities and International Studies
University of the Nations, mailstop #244
Kailua-Kona, HI 96740 USA

direct voice phone: 808-326-4465
fax phone: 808-326-4406
email: humanities@uofnkona.edu
===================================================

[Robb Brunansky]

How are you currently constructing the search? That might help us see how to refine it. The basic problem I see is that BHS-W4 is not linked to keyed numbers (at least as far as I can tell). Perhaps the Logos Hebrew text has the key numbers built into the database? That might be the difference. At any rate, post your current search argument on the forum for all of us to see when you get a chance.

[Robert Evans]

I've discovered a number of ways that don't work! One that I thought might do it was to search BHS-W4 for all verbs (i.e., Grammatical Tag ... , Verb) and use the Hits command to get word list (User's Guide, p. 8-14, Grammatical Supplement, p. G1-12). If I could isolate just the words (not the rest of the verses) that matched the search criteria (i.e., the verbs in BHS) then I thought maybe I could view that list in a keyed text such as NAS95S, etc., and if I could do that, then maybe Accordance could identify those words by Key number. Doesn't seem to work, though. I even tried the HITS-k command with similar frustrating results. I also thought maybe I could extract the verbs from BHS using HITS=i, but those results had not only verbs highlighted in the full verses, but also highlighted particles such as 'al (upon), which is a preposition, not a verb at all!

So thanks for being willing to check this - I'm looking forward to learning more about Accordance.

In His service,
Robert

This post has been edited by Robert Evans: 21 August 2006 - 04:53 PM

[jpkang]

Robert Evans, on Aug 21 2006, 05:49 PM, said:

I also thought maybe I could extract the verbs from BHS using HITS=i, but those results had not only verbs highlighted in the full verses, but also highlighted particles such as 'al (upon), which is a preposition, not a verb at all!

Can you please copy and paste this search as you entered it? (we can figure out the Hebrew characters if they get garbled). It sounds like in this case you were only looking for words starting with ayin or else you were searching by root (since the preposition 'al is understood to be derived from the verbal 'lh "to go up" by HALOT). It's otherwise hard for me to know or explain your results.

And can you give more details regarding "Logos can easily pull up a list of verbs in the Hebrew Bible, in Strong's number order"? Their website explanation of how their BHS works doesn't include any mention of Strong's key numbers, so I am curious how it could be done, even with more than one step.

I suspect Helen's suggestion is ultimately the only way to do what you're asking for at this point. And will you be needing to search for different ranges of Strong's numbers?

[Tom]

I too, would be interested in hearing how Logos accomplishes this.

Tom

This post has been edited by Tom: 22 August 2006 - 09:31 AM

[Robert Evans]

jpkang, on Aug 21 2006, 08:06 PM, said:

Can you please copy and paste this search as you entered it?

I opened a search window in BHS-W4 (window title "BHS-W4") and did a word search for verbs (search argument: "[VERB]").

Then I opened a second search window also in BHS-W4 (window title "BHS-W4 2"), and did a word search using the HITS standalone command (search argument: "[HITS BHS-W4"), trying to get "the word list" referred to in the User's Guides. But this second search appears to produce the same results as the first search did, i.e., not a word list but the complete text of the verses. The Grammatical Supplement says that "the default HITS search ... in a grammatically tagged text [uses] the list of exact lemmas found by the search in the first window" (emphasis added) - that was what I was trying to get, but evidently didn't.

Then I redid the search in the second window looking for the inflected forms (search argument: "[HITS=i BHS-W4]"), but these new results contained not only the verbs but also various non-verbs such as the preposition 'al, the noun 'or (light), etc.

Changing the results of the first search from BHS-W4 to a Keyed text like KJVS and then searching that first window's results for Key numbers (search argument: "[HITS=k BHS-W4]") doesn't work either, probably because Accordance apparently still thinks the results are in BHS-W4.

jpkang, on Aug 21 2006, 08:06 PM, said:

And will you be needing to search for different ranges of Strong's numbers?

It would be helpful (reduce potential errors, save time, etc.) for this project to be able to utilize a simple search for various ranges of Strong's (or G/K) numbers without resorting to a lengthy workaround.

I'll check with my colleague to find out how he searched Logos and post that info.

Thanks to all for your willingness to help!

Robert

[Robb Brunansky]

I don't know if this will help, but you could search NAS95S or KJVS for [Key H3???] to get a list of all Hebrew words in that begin with the Strong's number H3. Have you been able to get anything useful from that?

[Robert Evans]

Robb Brunansky, on Aug 22 2006, 11:50 AM, said:

I don't know if this will help, but you could search NAS95S or KJVS for [Key H3???] to get a list of all Hebrew words in that begin with the Strong's number H3. Have you been able to get anything useful from that?

Yes, I had tried that - searching KJVS for example for "[Key H3???]", and the results look like all verses that have at least one word with a KEY number in the 3000s. But from there I kept running into more dead ends, or so it seemed. For example, I tried a HITS search on that first window (by opening a second search window, KJVS 2, and searching for first "[HITS KJVS]", and then "[HITS=k KJVS]"). In each of these two searches, the results in the second window contained verses that were not in the first window (e.g., Gen. 1.2) and words with Key numbers not in the 3000s (e.g., "face", H6440).

Aside from the odd results (why would a HITS search on a window turn up verses that were not in that window, and highlight words that were not in the search argument of the first window?), all I had, at least in the first window, was a collection of verses that had at least one word in them with a Key number in the 3000s, while what I needed was a list of words (lemmas) in the OT whose Key numbers were in the 3000s.

Thanks again to all for any ideas on this issue.

In His service,
Robert

[Robb Brunansky]

After you perform the search, you can get a list of hit words and the number of times they occur by clicking details and looking at the analysis tab. Is that helpful?

[Lorinda H. M. Hoover]

I'm not sure why the [Hits=k BHS-W4] search produced the results it did; I tried it myself and got the same issue. But I also couldn't find any discussion of using "=k" with the HITS command in the help.

In reading your post, I'm not sure you understand how the HITS command works. While the help uses the term "word list," using HITS doesn't produce a simple list of hit words (As Robb has suggested, you can get that list by clicking the details button and choosing the analysis tab) Instead, the HITS command uses the hits from the first search window to search another text or another section of the text. For example, during my sermon preparation, I often want to check for vocabulary common to more than one reading. So, I create one window for the LXX1 with the search string * <AND> [RANGE xxx] where xxx is the OT reading for that week. Then I open a GNT Search window and enter the search string [HITS LXX1]<AND>[RANGE yyy] where yyy is the Gospel or Epistle reading for that week. This gives me a list of verses in the specified GNT range that contain vocabulary from the LXX1 range I set. But back to your issue...


I haven't been able to figure out a way to isolate Verbs with a specific key number range. The problem, of course, is that the keyed texts don't include grammatical tagging information, and the grammatically tagged texts don't include key number information. I'm guessing this would be tricky to do, since I don't think the Strongs numbers line up consistently with the lemmas used in the BHS-W4 tagging schema. (I don't know how G/K numbers work, as I don't have one of those texts. Do they by any chance differentiate between parts of speech?)

What you can do, is search for [KEY 3???] in a keyed text (eg KJVS) and then open a BHS search window and use the CONTENTS command [CONTENTS KJVS]. This will give you the same list of verses in the BHS window as in the Strong's window. You can do something similar by opening a BHS pane in the KJVS window, of course. In neither case are any words in the BHS text highlighted as hits. I don't know if this helps you or not.

[Robert Evans]

Robb Brunansky, on Aug 23 2006, 03:36 PM, said:

After you perform the search, you can get a list of hit words and the number of times they occur by clicking details and looking at the analysis tab. Is that helpful?

I also thought that would be a promising direction, and it was one of the first things I tried, because I use the the Details function frequently in preparing material for by Greek and Hebrew students. But in this case, using the analysis info has not yet proved helpful. If from the Analysis window we export an RTF file (the plain text option is grayed out probably because of the Hebrew font), we get a list of the English words in alphabetical order ("abhor" through "youth", in the case of the KJVS), each of these words having one or more Key numbers associated with it. This document is over 6500 lines long and is not formatted for easy isolation of the Key numbers. For example, many of these English words are associated with more than one Key number, so in order to make a consistently tab-delineated database, these Key numbers would need to have the English headword inserted manually in front of each Key number where it is not already there. Then the database could be sorted by Key number, and all key numbers not falling within the 3000 range could be manually deleted. The next step would be to manually delete all non-verbs from the list. This last step alone would be quite time-consuming, and it would be a blessing if there were a simpler way.

So if I'm missing something here, I would be very grateful for any ideas - thanks again to you and others who've taken time to think about this challenge with me.

[Robert Evans]

Lorinda H. M. Hoover, on Aug 23 2006, 11:17 PM, said:

But I also couldn't find any discussion of using "=k" with the HITS command in the help.

You are right - I had to go to the User’s Guide and the Grammatical Supplement to find information about the "=k" (User's Guide, p. 8-16; Grammatical Supplement, p. G1-12).

Lorinda H. M. Hoover, on Aug 23 2006, 11:17 PM, said:

In reading your post, I'm not sure you understand how the HITS command works.

You are also right about my misunderstanding of the HITS command - I see now that it doesn't actually produce a word list, just uses the word list. But I confess I still don't quite understand how that is different from the CONTENTS command? The User's Guide says (p. 8-14): "Unlike the CONTENTS command which uses the list of verses, the HITS command incorporates the actual search results from the other window. This powerful command allows you to compare the vocabulary of different passages, different translations, or even different texts such as the Qumran and the Hebrew Bible." Doing something like this is a significant part of our current project that to my knowledge has never been done before, and this description raised my hopes that Accordance could help.

Lorinda H. M. Hoover, on Aug 23 2006, 11:17 PM, said:

What you can do, is search for [KEY 3???] in a keyed text (eg KJVS) and then open a BHS search window and use the CONTENTS command [CONTENTS KJVS]. This will give you the same list of verses in the BHS window as in the Strong's window. You can do something similar by opening a BHS pane in the KJVS window, of course. In neither case are any words in the BHS text highlighted as hits. I don't know if this helps you or not.

When I try this CONTENTS search (i.e., search in first window "KJVS" for [KEY H3???], then search in second window "BHS-W4" for [CONTENTS KJVS]), the Analysis Details pane is empty - i.e., "Total number of verses = 18204 (total number of verses displayed = 18204) [CONTENTS KJVS] (0 total words)". While it certainly doesn't hurt to have a reference list of every verse in BHS that uses a KJV word whose Key number is in the 3000s, I was just hoping to find a reasonably quick way to produce a simple list of the Hebrew words associated with those Key numbers.

[Lorinda H. M. Hoover]

Quote

I was just hoping to find a reasonably quick way to produce a simple list of the Hebrew words associated with those Key numbers.

What if you did the [KEY 3???] search in the Hebrew Strong's tool? This would provide you with a list of key numbers and Hebrew words, and a gloss. Unfortunately, it won't give you a listing of all the English words used in the KJV to translate each word. This may be a limitation of the e-texts, as I remember the same issue existing in Quickverse's Strongs module several years ago.

It's still not exactly what you are looking for, but is it any closer?

[Ruben Gomez]

Lorinda H. M. Hoover, on Aug 24 2006, 09:51 PM, said:

What if you did the [KEY 3???] search in the Hebrew Strong's tool?

Just to expand on Lorinda's suggestion, you could try the following:

1. If you have the NIV Hebrew module, select the Hebrew Entry search field and type "*?" (without the quotes). Under More options, change the Extra Fields pop-up menu to "2". You will now have two extra search fields below. Leave the default AND in both cases. Proceed to change the first field to G/K Number and type "3???". Then go to the second field, select Part of Speech and simply write "v". Run the search and you should get 582 hits. This will give you the 194 instances where a Hebrew word (*?) is tagged with a number in the 3000's range (3???) and is considered a verb (v). You could then use this data to fine-tune your future Accordance searches.

2. If you have NAS Hebrew you have to follow the same steps, but you'll have to choose Key Number instead of G/K Number. Also, in the third search entry box, select English Gloss and write "to". Since there are no abbreviations for the different parts of speech, we have to resort to the "to" that is used with English infinitives if we want to find verbs.

3. If you are limited to Hebrew Strong's, you will have to follow the same procedure explained above, but choose English Content in the third search entry box.

The first method is the best and most accurate, but you can also get reasonably accurate results if you have to use NAS Hebrew or Hebrew Strong's. It will require some more effort on your part, but it can be done.

I hope this helps some. Unfortunately, unless you have a Hebrew text which is BOTH grammatically tagged and contains Strong's numbers, there is no easy way to get the results you are after.

Best,

[Robert Evans]

Lorinda H. M. Hoover, on Aug 24 2006, 09:51 AM, said:

What if you did the [KEY 3???] search in the Hebrew Strong's tool?

Thank you very much for that suggestion, Lorinda - although we unfortunately can't choose the KEY command from the Search menu when we are in Hebrew Strong's (same is true for NAS Hebrew), I discovered that Hebrew Strong's (but, interestingly, not NAS Hebrew) is formatted consistently enough to be able to be made sortable by Key number.

I ended up using the "Save as Text File…" command to save the whole tool (the "Copy" command is apparently limited to a maximum of 500 verses or "paragraphs in a tool" at a time, so even though we can actually highlight just the range 3000-3999 in the tool, we can't copy all of that range at one time. It would take several passes to just copy that range and paste each 500-paragraph part into a word-processing/database document).

So in a word processor I opened the new document containing the whole tool and deleted everything but the range I wanted (reducing the file from almost 8 MB to less than 900 KB). Then I used Find/Replace to substitute tabs for both the period-space between the Strong's number and the Hebrew word, as well as the semi-colon–space between the pronunciation and the definition. At this point, the entire range could be made into a table.

After deleting unnecessary columns, I ended up with two-column list of Strong's numbers in the first column and Hebrew word/transliteration/pronunciation all in the second column. A few more minutes of "clean-up" formatting and I had a usable 25-page document. The number of pages could be reduced with a few more minutes of formatting (e.g., 9-point font instead of 12, two columns instead of one, narrower margins, etc.), but I'll do that later if necessary. Still don't have just verbs in that range, but at least we know that Accordance can indeed generate data for a list of words from any range of Key numbers from the Hebrew Strong's tool, thanks to your suggestion.

This post has been edited by Robert Evans: 25 August 2006 - 03:16 PM

[Robert Evans]

Ruben Gomez, on Aug 24 2006, 10:59 PM, said:

Just to expand on Lorinda's suggestion, you could try the following:

1. If you have the NIV Hebrew module, select the Hebrew Entry search field and type "*?" (without the quotes). Under More options, change the Extra Fields pop-up menu to "2". You will now have two extra search fields below. Leave the default AND in both cases. Proceed to change the first field to G/K Number and type "3???". Then go to the second field, select Part of Speech and simply write "v". Run the search and you should get 582 hits. This will give you the 194 instances where a Hebrew word (*?) is tagged with a number in the 3000's range (3???) and is considered a verb (v). You could then use this data to fine-tune your future Accordance searches.

2. If you have NAS Hebrew you have to follow the same steps, but you'll have to choose Key Number instead of G/K Number. Also, in the third search entry box, select English Gloss and write "to". Since there are no abbreviations for the different parts of speech, we have to resort to the "to" that is used with English infinitives if we want to find verbs.

3. If you are limited to Hebrew Strong's, you will have to follow the same procedure explained above, but choose English Content in the third search entry box.

The first method is the best and most accurate, but you can also get reasonably accurate results if you have to use NAS Hebrew or Hebrew Strong's. It will require some more effort on your part, but it can be done.

Thank you, Ruben, for another helpful solution. Your idea will isolate verbs from other words in the list. I don't have NIV Hebrew yet, but I worked through your idea in NAS Hebrew and in Hebrew Strong's, and I was able to bookmark each verb found. Is there a way to collect all the bookmarked paragraphs/articles and somehow save or export them separately from the unmarked ones? (The Selection menu commands, i.e., to Add Marked Verses To … a Reference List, are grayed out after I do the search you describe.) If not, then it looks like I can at least Save as Text File … the entire tool, and then open that document and manually delete all the non-marks. In the NAS Hebrew, this search doesn't catch Key numbers that have a letter like "a" or "b" suffixed to them, or verbs that are not identified with the word "to" (such as verbs which are cross-referenced by the phrase "the same as" to another verb outside the range) - but there are relatively few of these.

Ruben Gomez, on Aug 24 2006, 10:59 PM, said:

Unfortunately, unless you have a Hebrew text which is BOTH grammatically tagged and contains Strong's numbers, there is no easy way to get the results you are after.

It would be great, as you imply, if we had available the BHS-W4 (as well as GNT-T, etc.) with Key numbers as part of the tag - the results of a search in such a database would probably be close to 100% accurate. Thanks again for your help.

In His service,
Robert