What am I doing wrong with these search results? (Or is it me?)

[Abram K-J]

I just found this "HITS Odyssey" post from a while back, which was really helpful to me.

 

When I try the search David/Helen mention (with a modification or two to make it work in Accordance 10 with the modules I have, I have a recurring hit I don't understand.

 

If you don't want to go back and read the original post, I'm just doing what's noted there:

 

• Open a GNT tab (whether GNT-T, GNT28-T, etc.). Search for all words with *
• Use these hits results in a new tab for the LXX with this string: "*@- [HITS GNT28-T] @- [NOUN proper] @ [COUNT 50-6500]"
• As the original post notes, after opening Analytics from the second (LXX) tab, "my Analysis now lists every word in the Septuagint which does not appear in the Greek New Testament, which is not a proper noun, and which is used 50 or more times."

First of all, I can't get over how cool it is to be able to do this so quickly and easily.

 

But, secondly, I'm showing εἶπεν highlighted as a hit in the LXX window after the query above. This is odd, since εἶπεν does appear in the GNT28-T, and my search should be showing words that do not appear in the GNT28-T.

 

My LXX (using LXX1 text, most recent Accordance, MacBook late 2008) Analysis window shows this at the top of the Word Count Totals analysis:

 

εἶπον (λέγω) •to say = 4190

 

When I "search this resource" from LXX1 with εἶπεν highlighted, I get 4190 results, which matches the number of results for εἶπον in the Analysis.

 

But why the single letter discrepancy here? Is this an error? I realize both of these inflections are valid, but shouldn't they be identical in terms of what I see in the text and what's showing in the Analysis? The former is εἶπεν; the latter is εἶπον.

 

And, the bigger question, why am I getting in my results a word that actually does appear in the GNT28-T, when I'm searching for words that do not?

 

Or, at least, I think I am... my current learning curve ("growing edge"?) in Accordance is how to do more complex searches, and how to use the argument entry bar more effectively, so it's entirely possible I'm just missing something fairly obvious here.

 

Thanks in advance for any ideas.

 

 

[Joel Brown]

This seems to be a result of different tagging philosophies. Looking at the Instant Details, εἶπεν is the inflected form, with λέγω as the lemma according to GNT-T. But, on LXX1, the lemma is εἶπον, with the root being λέγω.

 

Since your search is based on the lemma, they are considered different words. You could potentially account for this by doing the search based on inflect (use "*" for both tabs) rather than lemma, or perhaps other tweaks.

 

Remember in the instant details, it is "inflect lemma (root) partofspeech tag english gloss".

[Abram K-J]

Thanks, Joel! That explains it all perfectly.

 

Since what was highlighted as hit results in red in my GNT (with the * search) were actual inflected words (i.e., the text), I didn't realize the search was pulling lemmas instead. But it sounds like that's the right/expected Accordance behavior here with the * search? I may have even known this at one point, and lack of use of the info made it fall out of mind....

[Joel Brown]

Right, a plain * search means to find every lemma. This also means words without a lemma won't be found (this applies more to Hebrew, I believe). If you are more interested in the printed word, use "*" instead of *.

[Abram K-J]

Thanks!

[davidmedina]

Very interesting and very cool… I just wish I knew what you guys were talking about lol

 

Well, I understood what Abram is looking for but way over my pay grade, lol. Hopefully one day…

Edited December 16, 2013 by davidmedina