PDA

View Full Version : Hebrew Vocab Search by Root Frequency count



JBonham
11-06-2011, 04:27 AM
While it is not as easy to do in Bible works as in Accordance, I finally figured out a way to do it, though it took some prep work of an .iel file.

A) Make a .iel in Word List Manager where roots are distinguished:

1. In order to prevent the Word List Manager from merging different homonyms stemming from the same consonants (which will badly mess up your frequency counts and prevent filtering), you must first search for each root independently, for each part of speech (to be explained below). There are eight possible roots in BibleWorks (Hx, Ha, Hb, Hc, Hd, He, Hf, Hg), and four parts of speech in which roots are distinguished (nouns, Particles, adjectives, verbs). This means you will have to do 32 searches, but be sure to do it with the following settings...

2. First in the WTM version, search for the a part of speech along with a certain homonym root. For example, nouns + root x would look like *@nc*+*Hx* I put the "c" in because I wanted to avoid proper names in my search, but you could take this out if you wanted. However, if you do, you will not be able to distinguish them later.

3. Under Results, hit Options, and select Send highlighted words to Word List Manager. UNCHECK keep Hebrew vowels, and UNCHECK keep morph codes. If you keep the morph codes, the homonyms will not stack together properly. If you keep the vowels, your .iel based search will not work properly later in the WTM. When you do this right it should give you a very clean, untagged list. But since we know what tags the list should have, we can edit them all together very easily.

4. Under the Select menu, hit Select All. Under the Edit menu, hit Change Morphology on selected words. Type in the same tag you used for your search in step 2 (for example @nc*+*Hx*). Now all roots in your list have both the proper tag and correct frequencies. The reason we need the part of speech in the morphology tag is because we removed the vowels before, and now in order to manipulate the data within a certain verse range we need to know what kind of word it is so roots don't bleed together. For example נפשׁ as a verb, marked homonym x (supposedly no other roots), occurs 3 times. But as a noun, also marked homonym x, it occurs hundreds of times. Ultimately you don't want BibleWorks to mix these two homonym roots together.

5. Save your list as an .iel file that is easily identifiable (for example Hebrew common nouns root 0).

6. Now go back and preform steps 2. through 5. until you have a saved .iel file for the results of each search. There's really only 23 because many parts of speech don't occur in the higher root numbers.

7. Do one more search for pronouns @p* even though they do not exhibit variations in homonyms. This way your list will be more complete.

B. Carefully, merge each of your twenty or so lists into one master list.

1. Load two lists.

2. In the Select menu hit Select all.

3. In the Edit menu hit copy selected words to other list.

4. In the file menu load your next list. Select and copy it to the growing list and so on until every .iel file you have made in the previous steps has been merged. Pay attention to which of the two columns you have selected at the bottom so that you know which direction the copying will be done. You can also keep track of the word counts at the bottom of each column to double check that they are adding together properly.

5. Once you are done, save this larger file as a new master .iel file. When you are done, your master list should have each homonym grouped separately, with a tag that looks something like the ענה@v*+*Ha* (315x), but it will be distinguished from ענה@v*+*Hb* (79x).

C) Edit your master root list to what every desired frequency you are looking to search. For example >100 or <100 is a common one since the Reader's Hebrew Bible basis it's footnotes on this number.

1. To do this in the Sort menu, sort your master list by frequency. Even if it already looks like it is, do it again since when you added the previous lists to it it merely appended them to the bottom (lame!).

2. Select the first word of the frequency you either want to keep or delete, hold the shift button, scroll to where you want to stop, and with the shift key still held down, click your final word. This should highlight all words in between. You can now let go of shift and individually deselect certain words you deem unworthy of your search criteria, if any sneaked in.

3. Now under the edit menu either delete the selected words, or invert the selection and then delete the selected word, depending on what sub-list you want to make. If you make a mistake, relax, you can just reload your master list and try again. Save your sub list as something you'll easily recognize later, Like "Hebrew Vocab 100-250".

D) Finally the search you've all been waiting for.

1. Open a new Advanced Search Engine for the WTM version. Double click on the word box and select Inclusion/exclusion list (IEL). Then click Load IEL. Find one of the master lists that has the root frequency you want and select it, and hit OK in the Word Box Options screen. This is the part that won't work properly if you didn't take out the vowels. For some reason, BibleWorks misreads some (not all) of the Hebrew vowels as gibberish in this kind of search and will only some of your words.

2. Now click on the bottom of the screen where it says Search lim. so you can pick just the chapter you are studying. If you want to find every instance in the whole Hebrew Bible, skip this step, but be warned it will take a while.

3. Click search, and voila. You can search by Hebrew root frequency and within a certain portion of scripture. It was a lot of work to get here, so you may want to permanently highlight the words it finds with a subtle but different font color using the highlight tool.



My two cents after a very un-intuitive two weeks and one long night:

The are some slight differences in word count between Pelt & Pratico's "Vocabulary Guide to Biblical Hebrew" which is based on Accordance searches. Most of these differences are negligible, being off by only one or two occurrences. There are a few that are more sizable, however, so let the record show that there are holes in this method search. Perhaps those duplicating this method will be able to help patch these holes with search coding and merging tips.

In full disclosure, I am still using BibleWorks 6, mostly because I am not confident that coughing up just as much as I already paid for the program will enable it to do what I described here any simpler fashion. Version 9, did attempt to solve this HUGE Hebrew faux pas by allowing you to right click a word and search for Lemma + Homonym, but this does not address how the Word List Manage manipulates this date carelessly. The BibleWorks Hebrew database has to transition to a fundamental ROOT BASED system if it wants to stay a viable contender in the digital Biblical studies field.

I give this to the forum because I found no web site that explained how to do this (and the .iel files is too large at 96.6k to upload here) and I with great pain and labor figured it out on my own with a lot of $$$ motivation. It saved me from buying a Mac and Accordance, which simplifies this process to mere seconds. And it saved me the humiliation of having to ask my Accordance using friends to make lists for me. It did make me think that BibleWorks is way overpriced for how hard I had to man-handle it to get it to do what similarly priced software can do with ease. But let it be stated for the record that Bibleworks is versatile enough to do anything, even things it was not intentionally designed to do. It also showed me how slanted BibleWorks is toward the Greek New Testament rather than the Hebrew Bible.

Joe Bonham