PDA

View Full Version : How to find different form of identical Hebrew verbs



Michael Burer
09-02-2006, 01:01 AM
A friend asked me today how to distinguish and find Hebrew verbs that are spelled identically but have different meanings, like

lvm, to rule
lvm, to compare.

At this point I'm stumped, especially since both verbs follow the qatal pattern in the Qal. Is there a way to do this from the lexicon entry for the word itself?

Ben Spackman
09-03-2006, 01:48 PM
I'm not at my computer, so no BW is handy, but you can distinguish homonyms in the search line. When you mouse over a word with a homonym, it should tell you if it's homonym I, II, etc., and there's a space for that when doing morphological searches in the WTM. Sorry I can't be more specific at the moment.

Michael Burer
09-04-2006, 10:46 PM
I see that there is a tag for that info, but what would a command line search look like for this? Those codes are so far down the line into the specialized Hebrew codes that I am getting lost pretty easily.

Ben Spackman
09-05-2006, 11:08 AM
Are you using the morphology help? You just type in ? until you get to the right place, as indicated by the pop-up morphology help window. (Again, I don't have BW handy, so I can't just provide an example.)

Charlie
09-05-2006, 01:25 PM
A friend asked me today how to distinguish and find Hebrew verbs that are spelled identically but have different meanings, like

lvm, to rule
lvm, to compare.



Hi Mike,

Your two searches above would look like this on the Command Line with the WTM as the search version:

For lvm, to rule type:
.lvm@v*Hb*

For lvm, to compare type:
.lvm@v*Ha*

See section 44d of the online help for more info on specific codes in the WTM.

Michael Burer
09-06-2006, 08:35 AM
For lvm, to rule type:
.lvm@v*Hb*

For lvm, to compare type:
.lvm@v*Ha*


Thanks, Charlie.

Let me ask a follow up question. I assumed that the asterisk in the command line had to terminate the query. (I've always used the morphological help and asterisk is defined there as "all following codes"). From your examples, it looks like that is not the case: The asterisk can actually stand for any number of codes; I can use it to replace all the ones between the "v" for verb and the "H" for homonym. Do I understand this correctly?

Charlie
09-06-2006, 10:12 AM
Mike,

The asterisk and the question mark are both wildcards. They are not codes. They mean the following:

* stands for any number of characters including zero characters.
? stands for exactly one character.

These wildcards are used the same way no matter what Bible version you're searching: KJV, WTT, WTM, etc.

So the following KJV search:
.*woman
...will find all verses containing, woman, bondwoman, kinswoman, freewoman.

The following WTT search will find all verses containing words begining with the letters yhla (e.g. ~yhla, wnyhla, ~kyhla, etc.):
.*yhla

The tricky part with morphology searches is understanding that the "words" in a morphology version are of the form lemma@parsingcodes. That is, every place the WTT had one of the words found in the search above, the WTM would have ~yhla@parsingcodes, where the parsingcodes would specify the exact form of ~yhla found in that spot in the WTT. The following WTM search will find all verses containing words that have been tagged with the lemma ~yhla:
.~yhla@*

The most important thing to remember with command line searches is that syntax (the use of * ? / . ' ) does not change. Syntax is the same no matter what your search version is.

So the reason I appended an asterisk is just a precaution. I don't have any of the morphology schemes memorized (no one should) so ending morph searches in an asterisk is always a good idea just to be sure you don't loose any valid hits. Remember, unlike the ?, the * doesn't require anything to be in it's place.

I know that may be more info than you asked for, but perhaps it'll help others reading the posts.

See the tutorial posted here for more details: http://www.bibleworks.com/forums/showthread.php?t=1483

Michael Burer
09-06-2006, 01:00 PM
Charlie,

This is very helpful. Let me ask another question which will help me think this through. I believe I am correct in stating that order does matter in the morph codes. For example, in searching the BGM

.*@nnms*

will return all nouns which are nominative masculine singular, but

.*@snmn*

will not.

What I want to understand is the way the system processes the codes. I assume the first code is of supreme importance, as that determines the part of speech and then other variables fall into place. Once you get past the part-of-speech code, though, the system could process the codes in any order since for each code subset for a part of speech there will be no repetition of letters. The command line morph help forces me to put the codes in a certain order and will only allow the * at the end, but in reality I could do things differently. For example, I could find all plural nouns in the BGM with two different searches:

.*@n??p*
.*@n*p*

Either one will work. The first search is the only way to do this with the morph help turned on, i.e., I am forced to do it this way. The second one can be typed in with the morph help turned off.

Am I thinking about this correctly?

Charlie
09-06-2006, 05:30 PM
Yes, this is a helpful discussion. I hope others are listening in.

Keep a firm distinction between regular characters in a word and wildcards (* and ?). The order of characters mattes just as much in morphology versions (WTM, BGM, etc.) as it does in any other Bible version. Obviously the following two searches will not both return hits for verses containing the word woman:

.woman
.owman

For the same reason the codes after the @ also need to be in the same order as they appear in the "word." (Put the WTM on display and look at the words to see what I mean.)

Now, you may be confused by the fact that a search using an asterisk * and a search using a question mark both return the same number of hits. That will often be the case, but you need to undertsand the destinction between the two wildcards as I outlined in my post this morning. The following two searches both return the same hits only because English is very limited on words beginning "wom" and ending with an "n"...

.wom?n
.wom*n

The first search is actually requiring a letter to be between the m and the n (remember the ? means exactly one character). The second search is not requiring anything to be in position between the m and n, but in English the string "womn" is not a word. So both of the searches return the same hits but they are not exactly equivalent searches.

The same thing is true of the two search examples you used. They might return the same hits, but that doesn't make them identical. The first search specifies a "p" in the fourth position after the @. The second search allows the "p" to be in any position other than the first. In the WTM the letter "p" means plural in the fourth position for nouns, but may mean something entirely different in a different position.

.*@n??p*
.*@n*p*