NA-28 λέγω vs LXX εἶπον tagging

[Susan]

Does it make sense to others that in the NA-28 the aorist forms of λέγω are tagged as lexeme λέγω, whereas in the LXX Rahlfs module they're tagged as lexeme εἶπον? I have no principled opinion on the matter (BDAG has a separate entry for εἶπον), but it's mildly annoying when attempting to use the same search string for both texts, and it's a pervasive false positive if one runs contrast searches between the two text using HITS such as the one described here. 

 

(If anybody wants to run the search at that link, note that the tag is now [noun proper] rather than [noun propername].)

 

(One happy consequence of the LXX scheme is that the search [verb aorist] @λέγω in the LXX finds a couple really interesting forms. Prov. 25:7 ῥηθῆναι does not appear to belong among them; aorist passives *ῥηθ are otherwise tagged εἶπον, as expected.)

 

(I do realize Accordance isn't responsible for the tagging, and LXX and NT appear to have been done by different people. Maybe that's the only reason they're different, I'm not sure.)

 

 

 

[Λύχνις Δαν]

I guess it makes sense in that Mounce (and Kovisto) followed the theory he (Mounce) presents in MBG for the GNT. It would appear just quickly looking over LSJ, BDAG and MBG that the use of επω as a first singular active ind. died out early and then you have a full verb represent in two or even three parts (ερω for future). LSJ reports :

 

εἶπον (pres. ἔπω is used by Nic. Al. 429, 490, etc., but the pres. in use is φημί, λέγω, ἀγορεύω

“εἶπον,” LSJ, 489.

 

Now, if one for a moment accepts that that in LSJ was enough for a tagger of the LXX to consider that the roots were different and thus tag them that way then that two makes sense. If you look in Muraoka, A Greek-English Lexicon of the Septuagint he two notes both forms but begins by "(aor.act.ind.) and goes on to explain its future and other forms noting at the end of the entry's introduction "For the pres. and impf., λεγω is used.".  So if one accepts that authority you might tag the LXX the same way and the GNT.

 

So the real question I think is what is the aim of your tagging. Is it for diachronic analysis of morphology ? If so the same tagging system recognising the distinct 'original' stems and roots in ALL the texts in the corpus under study makes sense, for the detailed analysis you'll likely want to do. If your aim is to tie together the words that form such a composite verb as λεγω so that it is simple to find the right forms according to usage in the works in question then combining them makes sense.

 

I have thought for a while that one wants the ability to layer multiple tagging layers over a single text. Acc does this but not for differing morphologies against a single text. This is a concrete example of one case where you might indeed want that. It is even possible to imagine a tagging layer composed of the tagging elements of a lower layer tagging. This would be more complex and while it might suit this case here - being able to say that pres, impf from λεγω, aor. from ειπον and future from ερω (via ειπον by the look of if - the exact case here is not clear to me) - but it's questionable that it would serve other cases well. Oh sparse layers - cool - interesting - hmmm.... more on that later. That would resolve the issue of use and maintenance. Sorry that last bit might not make sense, just thinking out loud.

 

Given your use case though, it would be nice if they treated them the same way.

 

I wonder if the Prov 25:7 is actually a tagging bug.

 

Thx

D

[Susan]

Thanks for your thoughts, Daniel, that makes a lot of sense. There are others that are handled differently in LXX tagging too, mostly issues of "deponency", e.g.  ἀντιλαμβάνομαι in the LXX vs. ἀντιλαμβάνω in the NA-28 (although the latter also only contains middle forms). Similarly ἐντέλλομαι in the LXX vs. ἐντέλλω in the NA-28; κοιμάομαι in LXX vs. κοιμάω in NA-28,  etc. But also, conversely, ἐκλέγω in the LXX vs. ἐκλέγομαι in the NA-28. Go figure. 

 

There's also ὄμνυμι in the LXX vs. ὀμνύω in the NA-28. Possibly an actual diachronic shift, although the LXX forms mostly look to me like they could be derived from ὀμνύω. The morphology here is beyond me...

 

Agreed about Prov 25:7 -- is there a place to report these things?

[R. Mansfield]

Agreed about Prov 25:7 -- is there a place to report these things?

Just select the text, bring up a contextual menu, and choose "Report a correction."

[Susan]

Just select the text, bring up a contextual menu, and choose "Report a correction."

 

Thanks, I guess I was assuming the tagging itself was not Accordance's problem, but maybe you all feed back to the authors? 

Edited July 10, 2016 by Susan

[Ken Simpson]

Susan, the tagging of the Greek text is our own, the tagging of the Hebrew is from Westminster-Groves.

 

Always worth reporting.

Edited July 10, 2016 by Ken Simpson

[Susan]

I see, thanks, will do!

 

If that's the case I am surprised about all of the differences between LXX and NT tagging mentioned above. Any chance of getting these things smoothed out, or they are intentional?

[Michael J. Bolesta]

Susan

 

According to the "About this text..." and the Accordance Read Me module (for LXX1 and LXX2), different scholars were involved in creating the morphological databases. For Rahlfs, the work was by Kraft, Taylor, and Wheeler. NA28 was by Mounce and Koivisto. The differences you observe may reflect the scholarly opinion of the respective teams.

 

This is my speculation concerning the divergences noted.

[Ken Simpson]

You know what they say, if you get 5 greek scholars in a room, you will get 6 grammars!

[Michael J. Bolesta]

You know what they say, if you get 5 greek scholars in a room, you will get 6 grammars!

 

Only 6? 

[Ken Simpson]

Only 6? 

They're young. Give them time.

[Fabian]

Susan, the tagging of the Greek text is our own, the tagging of the Hebrew is from Westminster-Groves.

 

Always worth reporting.

Own? Or Mounce-Koivisto which works for Accordance?

 

Greetings

 

Fabian

[mgvh]

Just the every 5 year reminder...

I was just doing some searches on speaking parts, and like the OP @Susan so long ago, it took me a while to realize that if I tried to run the same search strings on NA28 and on LXX Rahlf's looking for forms of λεγω, I could not do it. LXX Rahlf's distinguishes between λεγω and ειπον as lexemes. (Oddly, they both pick up the future, perfect, and aorist passive forms of ερω.) Grrr... took me a while to figure out what was going on, and then I found this old thread.

BUT...

I found a work around! The combined Greek Bible in Accordance uses Swete's tagged LXX and the GNT, and that tagging follows the GNT / NA28. Of course, Swete's LXX is not identical to Rahlf's, but it's close enough to get me what I need for now.

,