Generating a Word List of Greek Words with Frequency by Lemma?

[jstrong]

As part of a big project I'm looking into the distribution of words in the Gospel of Luke and in the New Testament. I want a list that shows the words according to the number of times they appear in the text. The Word List tool looked like it was just the sort of thing I needed to crunch these numbers, but what it's spitting out isn't useful at all. It's showing a different entry for every word in each form, e.g adelphos gets one entry and adelphon is listed separately so that it looks like the latter is a rare word. Is there a way to generate a frequency word list so that entries of the same lemma will be counted as a single entry?

[Lee]

You need to use a morphology version (e.g., BYM or BNM). And the easiest way to get the list is to open a morphology version and then open the "Context" tab in the analysis window. The "Book Context" in the upper right corner will list the lemmas in Luke by order of frequency.

[jstrong]

You need to use a morphology version (e.g., BYM or BNM). And the easiest way to get the list is to open a morphology version and then open the "Context" tab in the analysis window. The "Book Context" in the upper right corner will list the lemmas in Luke by order of frequency.

This does the same thing as before as far as I can tell, listing a bunch of identical words rather than lemma. If you scroll down to words appearing just once you'll see a bunch of entries of the same word in a different case.

[Jim Wert]

If you use a morphology version, you have to tell BW to show you no morph codes.

In the context tab, after having made a morphology version your search version, right click in the context tab, select "Context/Word Tab Options", check "Hide Morphology Codes".

In the Word List Manager, when you generate your word list, make sure that a morph version is specified at the top of the "Create Word List" window, and uncheck "Keep 99 morph codes".

--Jim

[jstrong]

If you use a morphology version, you have to tell BW to show you no morph codes.

In the context tab, after having made a morphology version your search version, right click in the context tab, select "Context/Word Tab Options", check "Hide Morphology Codes".

In the Word List Manager, when you generate your word list, make sure that a morph version is specified at the top of the "Create Word List" window, and uncheck "Keep 99 morph codes".

--Jim

Maybe I'm not seeing what you are, but I've unchecked the "hide morphology codes" and the "keep 99 morph codes" and this doesn't seem to make any difference in either the context tab or the word list... In the entries for lemmas appearing once it lists for example nukta followed by nuktas, and mikron, mikros mikrwn all as discrete entries.

[Jim Wert]

Maybe I'm not seeing what you are, but I've unchecked the "hide morphology codes" and the "keep 99 morph codes" and this doesn't seem to make any difference in either the context tab or the word list... In the entries for lemmas appearing once it lists for example nukta followed by nuktas, and mikron, mikros mikrwn all as discrete entries.

Looks like you are using a text version rather than a morphological version. Your are showing forms, not lemmas.
To get a list of lemmas you need to use (make your search version), for example, BNM, not BNT; BYM, not BZT, etc.

If you are using the "Context" tab, "Hide Morphology Codes" must be checked; perhaps my post was unclear.

--Jim

[DavidR]

jstrong, I'm seeing the same thing in the Context tab that you are. I don't use that tool much, but testing just now with BNM set as the search version and morphology codes turned off in the Context tab options, and searching on νύξ, I get listings for each of its declined forms. It shows up more clearly if I right-click in the list and set it to sort alphabetically. Screenshot attached. It's obviously not hard to add up all the forms for this relatively uncommon word; ἀδελφός requires a little more work. It would surprise me, though, if there wouldn't be some way to get BW to show you a count of lemmas rather than forms; I just don't know what it is!

[Donald Cobb]

Maybe I'm not seeing what you are, but I've unchecked the "hide morphology codes" and the "keep 99 morph codes" and this doesn't seem to make any difference in either the context tab or the word list... In the entries for lemmas appearing once it lists for example nukta followed by nuktas, and mikron, mikros mikrwn all as discrete entries.

Hi JStrong,

There is one detail that may be throwing you. Once you've done your query and opened the Word List Manager, uncheck everything (except, perhaps, for keeping the Greek accents), then make sure that in the version box at the top of the WLM, you have selected BGM, GNM or another morphology version. This will not automatically conform to your search version, so you have to set it manually (once you've done that, though, it will retain the version you chose in the previous search).

I've forgotten that on occasion, and have unexpectedly run into the same problem you describe.

I hope that helps!

Regards,

Donald Cobb
Aix-en-Provence, France

[jstrong]

Ok I'm pretty sure I got it to work in the word list finally, or at least as best as it can work. It still separates things I wish it didn't like πρεσβυτέριον from πρεσβύτερος, I'm assuming because it thinks they are different parts of speech. Is there any way to get it to join nouns and verbs from the same stem?

Is there any way to export this somewhere with verse numbers attached?

[Donald Cobb]

Ok I'm pretty sure I got it to work in the word list finally, or at least as best as it can work. It still separates things I wish it didn't like πρεσβυτέριον from πρεσβύτερος, I'm assuming because it thinks they are different parts of speech. Is there any way to get it to join nouns and verbs from the same stem?

Is there any way to export this somewhere with verse numbers attached?

Hi again. Two quick comments:
- πρεσβυτήριον and πρεσβύτερος actually are different parts of speech. One is a neuter noun (the -τηριον in Greek is a diminutive that is always neuter, generally designating the place where the noun happens), the other is an adjective.
- I would venture to say that, by definition, the Word list Manager can't give you verses. It functions on the level of words. I don't know of any way you can 1) do a query to turn up every word in every verse and then 2) assign each of all those words to the verses in which they are found. Perhaps it can be done, but someone else will have to chime in here.

It would be easy, however, and perhaps more helpful to focus on certain words and spot-search those (say, perhaps, those words that occur only once or twice in the chapters you are working on). Do you really want to know all the verses in which you can find, say, καί or δέ?

Hope that can help a little.

Regards,

Donald Cobb
Aix-en-Provence, France