What's the best way to do this kind of a search?

[ISalzman]

All,

This past weekend, I was invited to speak at a bible conference whose theme was
לְד֥וֹר וָד֖וֹר, "from generation to generation." In preparing for this conference, I thought I would run a search of the phrase and see in what contexts it was used in the Hebrew Bible. I ran the search on the exact phrase and found the grand total of two hits. "That's strange," I thought. I thought it occurred a whole lot many more times than that. Intuitively, I thought, "let me run this search with the word "dowr" spelled defectively (= dor). Sure enough, the phrase turned up a bunch more times, but this time, the two occurrences with the full spelling (see above) were absent. There's something not right about this picture. It would strike me that if you run a search on a phrase or a word, every occurrence of that word should be a hit, regardless of whether the word is spelled fully or defectively. Apparently, that is not the default search preference in BW. At any rate, given this fact, how would you all run a search on a phrase such as the one above that would guarantee I'm going to get all the results?

I should add that there were times the phrase occurred as "midor ledor," "from generation to generation." And sometimes it occurred just as "ledor dor," without the intervening vav (or waw, if you must!). So any insights on how best to run a search of this nature would greatly be appreciated.

[Glenn Weaver]

Hi Irving,

For this search, the most important thing is to search using the WTM morphology version. I think that you are probably trying to run the search using the WTT text version. If you search using the morphology version, the search results will find every occurence no matter if the defective form is used or not.

So, your search will be a phrase search, with two occurences of the word DWR. Since sometimes the vav appears on the second word (and who knows what other prefixes could be there), we need to account for the possible presence of an intervening word between the two occurences of DWR. (Prefixes in the WTM are treated as complete words, even though they are always joined to a word in the printed text.)

Therefore, on the Command Line, using the WTM, search for the following (using the appropriate Hebrew characters, of course, instead of transliteration):

'dwr *1 dwr

These letters are exact ones that you will include in your search in the WTM. Do not include any accents or vowel points for this search. This search allows for 1 intervening word, so it allows you to find hits that both do and do not include the vav. (If you increase the intervening words to 3 you will gain an additional two hits, though these might not be useful fo you.) The use of the single quote mark instead of a period at the beginning of the search phrase makes this a PHRASE search instead of an AND search.

Please let me know if this doesn't work for you.

Blessings,
Glenn

[ISalzman]

Hey Glenn, thanks, that works great!!

Just a question and a remark. Question: When doing a Hebrew phrase search, am I correct in understanding that the Command Line search phrase should be inputted from right to left just like Hebrew and to reflect the actual Hebrew order of the phrase? (Obviously, dwr + dwr doesn't really matter too much, since the words, and thus, word order, are the same.)

Remark: I found, when inputting the above search, that when you insert 'intervening word syntax' - in this case, *1 - it defaults to left-to-right order. This made it necessary for me to use my left arrow key to revert back to normal Hebrew order. Does this sound right to you? ( I guess this comment turned into a question!)

Thanks Glenn.

Irving

[Glenn Weaver]

Hi Irving,

I'm glad that the search is working for you.

You are correct that the typing order will change automatically, depending upon what you are typing. If what you are typing is actually a Hebrew character, then the Command Lines knows to keep the order right-to-left. But if you type something that is not a Hebrew character, then the order changes to left-to-right.

This is useful in many ways. You do not have to change the typing direction, since the Command Line already does it for you. And when you enter morphology codes, once you add the @ sign the Command Line shifts automatically to enter the codes in left-to-right order.

After you type a Hebrew word, the cursor is at the left side of the word. To add morphology codes you move the cursor to the right side of the word, then enter the @ sign. Then you can enter the desired morphology codes.

When you search on a phrase in Hebrew, the words should be entered in right-to-left order. For example, if you wanted to search on the exact phrase of the first words in Genesis 1:1 (using the WTT), you would enter, from right-to-left, the following characters:


'~yhla arb tyvarb


To run the same phrase search on the WTM, with the @ sign and and asterisk, you would use the following search. Notice the order of the characters. The Hebrew letters are right-to-left, while the @* is left-to-right, and is to the right of the Hebrew word.


'~yhla@* arb@* tyvar@* b@*



Blessings,
Glenn

[ISalzman]

Glenn,

Thanks again for the helpful reply. I really appreciate it.

Blessings,

Irving