NT Greek with Morph data?

[sungkhum]

Hello,

I currently serve as a missionary in Cambodia and I am working with a Cambodian brother on creating some Greek tools for pastors. One tool we would like to see in the Khmer language is a NT Greek/Khmer interlinear. I've used Fieldworks, and started doing some tests, but coming across this group I wanted to ask if anyone has a complete NT Greek database that also includes morphological data that we could base our work on with instead of beginning from scratch.  I have come across this: https://github.com/morphgnt/sblgnt and other such resources, but ideally something already in Fieldworks would be the quickest.

Any ideas/help would be much appreciated!

Thanks,

Nathan Wells

http://www.actioncambodia.org

[csta...@gmail.com]

If you come across an answer to this question, I would be interested in having a copy of the FW database.

Coleen Starwalt

> --
>
> You received this message because you are subscribed to the discussion group "FLEx list". This group is hosted by Google Groups and is open for anyone to browse.
>
> To post to this group, send email to flex...@googlegroups.com
>
> To unsubscribe from this group, send email to flex-list+...@googlegroups.com
>
> For more options, visit this group at http://groups.google.com/group/flex-list
>
>

[Ronald Moe]

Hi Nathan,

I have a NT Greek FieldWorks project that I have been working on for many years in my spare time. I am hoping to change assignments in the near future and start working on it full-time.

 

As Eric noted, the primary problem is copyright issues. I began work a couple of decades ago using Shoebox. At some point I got permission to adapt Louw and Nida’s “Greek-English Lexicon of the New Testament” for use in LinguaLinks. I began adding part of speech, lexeme/stem form (vs. citation form), inflection features, affixes, allomorphs, etymology, frequency, and other kinds of information that Louw and Nida’s original database was lacking. I also began doing word studies on various words, using the NT as a text corpus. But I found that I often disagreed with Louw and Nida’s definitions. The dictionary seemed to be written as an aid to translators, which meant that the definitions were more concerned about translation and interpretation issues than in describing the semantics and function of the Greek word. So I eventually came to the decision to start over. I’ve written 1,793 definitions so far, covering 1,449 lexemes out of a total of 8,670. That last figure inludes roots, affixes, variants, and irregularly inflected forms. Most of those only need a gloss.

 

I also began work on parsing the NT. But LinguaLinks at that time did not have a functioning parser, so I did the parsing in Shoebox. It was a huge eye-opener for me, because the traditional morphological analysis that I had been taught in school turned out to be wrong. So I had to start analyzing Greek morphology using modern linguistics. It turns out that many of the difficulties in traditional Greek grammars are nothing more than simple morphophonemics. Although the parsers in FLEx are not perfect, they are wonderfully helpful in arriving at a good understanding of the morphology of a language.

 

My hope is to form a team to work on developing the database into a full FW project with the NT as the text corpus. Quite a few people have expressed interest in working on the project, but we have lacked two things--the ability to work simultaneously from a distance and the organizational structure to keep things moving. The FW and WeSay developers have been getting closer and closer to the ability to put an entire project online, just as we can do now with a lexical database via LanguageDepot and Chorus. (Perhaps it is already working and I just need to learn how.) We would need to put the entire project on LanguageDepot, grant people editorial priviledges, and determine who should work on what.

 

We also need to incorporate a version of the NT that is in the public domain. We have the SBL NT text and someone worked on putting it into TE or Paratext. But I never incorporated it into my project. So between the SBL text and my dictionary we have the two essential components for a complete FW project that could be distributed freely. But there may be opposition to using the SBL text instead of the UBS text. Unfortunately the UBS text is copyrighted and cannot be distributed. One solution is to do all the foundational work on the SBL text and then later apply it to the UBS text for those people and applications that have permission to use the UBS text.

 

This will give you an idea where we are. We can communicate further off-line about specifics of sharing the file, etc.

 

Ron Moe

 

---

--

You received this message because you are subscribed to the discussion group "FLEx list". This group is hosted by Google Groups and is open for anyone to browse.
To post to this group, send email to flex...@googlegroups.com
To unsubscribe from this group, send email to flex-list+...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/flex-list

---

No virus found in this message.
Checked by AVG - www.avg.com
Version: 2012.0.2197 / Virus Database: 2437/5259 - Release Date: 09/09/12

[Randy Regnier]

On 9/10/2012 12:29 PM, Ronald Moe wrote:

My hope is to form a team to work on developing the database into a full FW project with the NT as the text corpus. Quite a few people have expressed interest in working on the project, but we have lacked two things--the ability to work simultaneously from a distance and the organizational structure to keep things moving. The FW and WeSay developers have been getting closer and closer to the ability to put an entire project online, just as we can do now with a lexical database via LanguageDepot and Chorus. (Perhaps it is already working and I just need to learn how.) We would need to put the entire project on LanguageDepot, grant people editorial priviledges, and determine who should work on what.

FLEX Bridge aims to support sharing an entire FW language project. It has had a couple rounds of alpha testing to date, and I expect another version (another alpha or perhaps beta) to come out before too much longer. The FW developers are switching systems for how they store source code, and that is taking some time.

Randy Regnier

[Ronald Moe]

Hi Randy,

This is good news. Do you need any more testers? I would be happy to be one. I’ve done some checking and I have the entire SBL NT imported into TE, although the formatting needs some fixing up. So I have a large dictionary and text corpus that could be used to do the testing. Kevin Warfel has expressed a desire to work with me on the parser. So we would need the ability to share the entire project, including lexicon, parser information, and texts.

Ron Moe

 

---

No virus found in this message.
Checked by AVG - www.avg.com

Version: 2012.0.2221 / Virus Database: 2437/5260 - Release Date: 09/10/12

[Randy Regnier]

FLEx Bridge (uses the same S/R 'engine' [Chorus] that is used by WeSay) allows sharing of:

1. Entire contents of the main xml data file, with the "fwdata" extension, (includes the lexicon, parser information, and texts you mentioned)
2. Writing system definitions,
3. Layout configuration information (next version),
4. Media files, (some limitations apply)

Susanna Imrie sends out the requests/invitations for testers on this list. The last one was in mid-July. So, stay tuned and monitor the list for the next round.

Randy Regnier

[Nathan Wells]

Thanks Ronald,

Much work still to be done, but exciting to hear of the possibilities! Your work sound very interesting (it would be cool to see some examples), maybe I will look around for an automated way to import the SBL data in the mean time (like you said, UBS would be nice, but copyright issues prevent most people from using that). If anyone has ideas on how to do the conversion, let me know, I'm not all that experienced with FLEx.

It is great to see more collaboration happening, and having used the WeSay dictionary collaboration personally, wonderful to hear more is coming for FW.

Thanks again for your time and expertise!
Nathan

Ronald Moe <Ron...@sil.org> wrote:

Hi Nathan,

I have a NT Greek FieldWorks project that I have been working on for many years in my spare time. I am hoping to change assignments in the near future and start working on it full-time.

 

As Eric noted, the primary problem is copyright issues. I began work a couple of decades ago using Shoebox. At some point I got permission to adapt Louw and Nida’s “Greek-English Lexicon of the New Testament” for use in LinguaLinks. I began adding part of speech, lexeme/stem form (vs. citation form), inflection features, affixes, allomorphs, etymology, frequency, and other kinds of information that Louw and Nida’s original database was lacking. I also began doing word studies on various words, using the NT as a text corpus. But I found that I often disagreed with Louw and Nida’s definitions. The dictionary seemed to be written as an aid to translators, which meant that the definitions were more concerned about translation and interpretation issues than in describing the semantics and function of the Greek word. So I eventually came to the decision to start over. I’ve written 1,793 definitions so far, covering 1,449 lexemes out of a total of 8,670. That last figure inludes roots, affixes, variants, and irregularly inflected forms. Most of those only need a gloss.

 

I also began work on parsing the NT. But LinguaLinks at that time did not have a functioning parser, so I did the parsing in Shoebox. It was a huge eye-opener for me, because the traditional morphological analysis that I had been taught in school turned out to be wrong. So I had to start analyzing Greek morphology using modern linguistics. It turns out that many of the difficulties in traditional Greek grammars are nothing more than simple morphophonemics. Although the parsers in FLEx are not perfect, they are wonderfully helpful in arriving at a good understanding of the morphology of a language.

 

My hope is to form a team to work on developing the database into a full FW project with the NT as the text corpus. Quite a few people have expressed interest in working on the project, but we have lacked two things--the ability to work simultaneously from a distance and the organizational structure to keep things moving. The FW and WeSay developers have been getting closer and closer to the ability to put an entire project online, just as we can do now with a lexical database via LanguageDepot and Chorus. (Perhaps it is already working and I just need to learn how.) We would need to put the entire project on LanguageDepot, grant people editorial priviledges, and determine who should work on what.

 

We also need to incorporate a version of the NT that is in the public domain. We have the SBL NT text and someone worked on putting it into TE or Paratext. But I never incorporated it into my project. So between the SBL text and my dictionary we have the two essential components for a complete FW project that could be distributed freely. But there may be opposition to using the SBL text instead of the UBS text. Unfortunately the UBS text is copyrighted and cannot be distributed. One solution is to do all the foundational work on the SBL text and then later apply it to the UBS text for those people and applications that have permission to use the UBS text.

 

This will give you an idea where we are. We can communicate further off-line about specifics of sharing the file, etc.

 

Ron Moe

 

---

From: flex...@googlegroups.com [mailto:flex...@googlegroups.com] On Behalf Of sungkhum

Sent: Monday, September 10, 2012 2:24 AM
To: flex...@googlegroups.com
Subject: [FLEx] NT Greek with Morph data?

 

Hello,

 

I currently serve as a missionary in Cambodia and I am working with a Cambodian brother on creating some Greek tools for pastors. One tool we would like to see in the Khmer language is a NT Greek/Khmer interlinear. I've used Fieldworks, and started doing some tests, but coming across this group I wanted to ask if anyone has a complete NT Greek database that also includes morphological data that we could base our work on with instead of beginning from scratch.  I have come across this: https://github.com/morphgnt/sblgnt and other such resources, but ideally something already in Fieldworks would be the quickest.

Any ideas/help would be much appreciated!

 

Thanks,

Nathan Wells

http://www.actioncambodia.org

--

You received this message because you are subscribed to the discussion group "FLEx list". This group is hosted by Google Groups and is open for anyone to browse.
To post to this group, send email to flex...@googlegroups.com
To unsubscribe from this group, send email to flex-list+...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/flex-list

No virus found in this message.
Checked by AVG - www.avg.com

[Roger]

Ron,

   I am also interested in what you are doing. I have dreamed of doing something similar and have even done some experiments on my own in FLEx but haven't gotten nearly as far as you. Would be very interested to have access to your file whenever you are ready to share it.

Roger Stone 

[Jim Henderson]

Dear Nathan,

 

I have just made for you a backup of a Translation Editor project that contains the free SBL Greek text. (Translation Editor is part of the BTE version of FieldWorks, and is available from the usual FieldWorks download site. Or if you are a member of the LangTran software distribution system, you may have the installer on your computer already. You can join LangTran by going to

https://www.wuala.com/LangTran/Documentation/?key=JdnNdIKqrorL)

 

The Greek text was prepared by SBL, and is free for anyone to use. It was downloaded from http://sblgnt.com/

then converted to SFM. I made the SFM files into a Paratext project. The text contains brackets that are not present in all Greek fonts, so you should download the SBL Greek font, from

http://www.sbl-site.org/educational/biblicalfonts.aspx

and install it.

 

I have put this TE project backup in my DropBox, and you can download it from

https://www.dropbox.com/sh/z5wv644ow2dr56b/tvT-1zkZU8

 

You might want to download it and open it in TE, and see if you can start adding interlinear glosses in Khmer.

 

I hope you find that helpful,

    Jim

[Nathan Wells]

Thanks so much Jim,
That is very helpful! Thanks for taking the time to help!

Do you know of a way to automatically import the morphological tagging data from these files: https://github.com/morphgnt/sblgnt

Having the text in FW is a wonderful start, but we would also like to be able to include the parsing data in the interlinear (which would also make the lexicon data more useful in the future) if possible.  We could manually paste the data in, but since it already exists, having it automatically imported would obviously be ideal.  But my expertise is limited, so I thought I would ask and see if you knew of a way.

Thanks again,
Nathan

[Michael Aubrey]

Hi Nathan,

The morphology in the MorphGNT would require re-writing. Currently, James Tauber, Ulrik Sandborg-Petersen and I have been on revising it so that it deals with certain morphological issues better--though none of us have had time to do anything with it for at least a year and a half. All three of us have other projects, work, and school that keeps us busy, though I did talk to Ulrik about it a little when I was in the Netherlands back in February. However, the format of morphology in FLEX and the tagging in MorphGNT are so dramatically different, the effort to convert it to a FLEX database is so great that it would probably be easier to start from scratch.

For that reason, I'd say the project of Ron Moe and others is probably the best shot. I've been doing the same independently from them. My goals aren't quite the same as their, since I'm aiming toward something more useful for parsing a larger corpus of Koine Greek texts beyond the NT. You can see some of what I've done on my website back in 2009 here:

http://evepheso.wordpress.com/2009/01/16/the-rough-beginnings-of-greek-morphology/

There's a link to a PDF on that page that show successful FLEX parsing of much of the nominal morphology system.

I've gotten a lot done since then and will likely have a relatively complete morphological parser up and running within the next year with a general complete lexicon of affixes. The roots and stems is growing progressively as I've worked through the text of Josephus (Niese's critical text) and portions of the NT (SBLGNT).

Mike Aubrey

PS - Ron mentioned that some people may not be comfortable with the use of the SBLGNT over the UBS text, but its a rather unreasonable discomfort. There is solid scholarship behind its editing and text-critical decisions.

---

From: Nathan Wells <sung...@gmail.com>
To: Jim_He...@sil.org
Cc: flex...@googlegroups.com
Sent: Tuesday, September 11, 2012 8:47 PM
Subject: Re: [FLEx] NT Greek with Morph data?

[Jim Henderson]

Nathan, you asked:

Do you know of a way to automatically import the morphological tagging data from these files: https://github.com/morphgnt/sblgnt

I have looked at the layout of the work that Michael Aubrey and his colleagues have done, and the format is consistent, like this:

040101 P- -------- Ἐν Ἐν ἐν ἐν
040101 N- ----DSF- ἀρχῇ ἀρχῇ ἀρχῇ ἀρχή
040101 V- 3IAI-S-- ἦν ἦν ἦν εἰμί
040101 RA ----NSM- ὁ ὁ ὁ ὁ
040101 N- ----NSM- λόγος, λόγος λόγος λόγος
040101 C- -------- καὶ καὶ καί καί
040101 RA ----NSM- ὁ ὁ ὁ ὁ
040101 N- ----NSM- λόγος λόγος λόγος λόγος
040101 V- 3IAI-S-- ἦν ἦν ἦν εἰμί
040101 P- -------- πρὸς πρὸς πρός πρός
040101 RA ----ASM- τὸν τὸν τόν ὁ
040101 N- ----ASM- θεόν, θεόν θεόν θεός
040101 C- -------- καὶ καὶ καί καί
040101 N- ----NSM- θεὸς θεὸς θεός θεός
040101 V- 3IAI-S-- ἦν ἦν ἦν εἰμί
040101 RA ----NSM- ὁ ὁ ὁ ὁ
040101 N- ----NSM- λόγος. λόγος λόγος λόγος
 

Michael has also documented what the components are, in the PDF file that he referred to. So it would be possible for someone to develop a program that would adapt their work for import into a FieldWorks project. (If I had time to do it, I would use Perl or Python.)

 

But Ron said that he has come up with new ways of parsing the Greek words, based on a better understanding of morphophonemics, so it might be better, as Michael said, to use Ron's project.

 

Ron, would you have a short book ready to release? I think you could use the Send/Receive LIFT item that is available in the File menu of FLEx to replicate your project to one at Language Depot. Nathan and others could become members of the project and experiment with what you have done.

 

Nathan, I have just added two more versions of the SBL project that I made, to my DropBox:

https://www.dropbox.com/sh/z5wv644ow2dr56b/tvT-1zkZU8

One has the SBL text in FLEx with no modifications, the other has simple glossing in English of John 1.1. You could adapt that so that the Writing System for the gloss line is Khmer, and start experimenting with that part of the job.

[Michael Aubrey]

Just to clarify a couple more things...

The analysis in MorphGNT is a simplified tagging scheme currently based on traditional grammatical categories that may or may not fit what's actually in the language. Issues with things like clitics, for example, are not dealt with, but are incredibly important for dealing with any number of issues. The layout you see below is James Tauber's work from the early 1990s (I take no credit for that). What we have been doing since then is not represented there. We're working on a more consistent categorization and analysis based on better models of morphology, but at this point, I'm not sure when it will be completed. What is available on github isn't easily convertible. It's probably impossible. Greek is too complex and the tagging of the MorphGNT is too simple. As I said in the web post with my PDF of FLEx work, even what you see there is a simplified version of noun morphology. There maybe what we can term three declensions, but there are multiple sub-declension classes that are not provided there that no simple morphological tagging scheme represents. In fact, as far as I'm aware, no morphological tagging scheme represents declension classes of any kind. Other issues are the pronominal system which is a complete mess in all morphological tagging databases and the verb will be a complete nightmare, if you have a copy of P. H. Matthew's second edition of Morphology handy, you can see what I'm on that front on pages 168-84 (chapter 9). The tagging in MorphGNT and other similar databases, by definition, cannot deal sufficiently with the very complex issues of extended exponence that the Greek verb demonstrates. To get the verb into FLEx from such tagging would involve a lot of creating something out of nothing.

All that to say, in answer to Nathan's statement:

We could manually paste the data in, but since it already exists, having it automatically imported would obviously be ideal

*IF* there's a way to do it, it wouldn't be automatic and it wouldn't be ideal, which is why I never even tried. I started from scratch instead.

Mike Aubrey

---

From: Jim Henderson <Jim_He...@SIL.org>
To: flex...@googlegroups.com
Sent: Wednesday, September 12, 2012 12:50 AM

[Ronald Moe]

To those of you who have expressed interest in my Greek project,

 

Thanks for your patience. I’ve done some cleanup and put the project backup file in my DropBox folder. I have also requested a new Language Depot project, which is probably a better way for all of us to get the dictionary file and contribute to it. Thanks to Jim Henderson and Kevin Warfel, I was able to import the SBL Greek NT text into TE. So it is now part of the project. As soon as FLEx Bridge is working, we will be able to collaborate on the entire project. If you want the entire project now, I can send you a link to the DropBox folder and you can download it from there. It is over 38 Mbs.

 

Thanks,

Ron Moe

 

---

Version: 2012.0.2221 / Virus Database: 2437/5263 - Release Date: 09/11/12

[Jeff and Peg Shrum]

What could be more fun than taming the tiger of Greek morphology. But are there plans to include Levinsohn’s and others insights into Greek discourse?  This is where the big gains are to be made in pragmatics and real understanding of the meanings of these ancient texts.

 

 

Jeff Shrum

Language Technology Consultant

SIL Mozambique

[Nathan Wells]

Thanks everyone for your time on this! I'm away from my office, so I can't do any direct testing with Khmer, but Jim was kind enough to make a sample and it looks promising! Thanks Ron for making your work available.

I had a question regarding morphology: a few times it has been stated that current morphology data is flawed.
So my question is, how flawed? Meaning is the morphology data in programs like Logos, Bibleworks etc. also flawed (meaning the projects like Ron's are on the cutting edge, working toward correcting what is currently the norm for most), or is the free morphology data (as contained specifically in MorphGNT) flawed?

Are there some articles you can recommend to help me better understand the situation?


Thanks,
Nathan

[Ronald Moe]

Hi Jeff,

 

I don’t have a lot of time right now, but I’m hoping to start sharing Word documents that describe my analysis of particular morphological issues.

 

I have Levinsohn’s book. I’m curious how you see his insights impacting a project to interlinearize  the NT or develop a dictionary. Or were you merely encouraging me to apply his insights to exegesis? I would agree that his work merits attention.

 

Ron

 

---

Version: 2012.0.2221 / Virus Database: 2437/5266 - Release Date: 09/13/12

[Ronald Moe]

We should probably take this discussion of Greek off the FLEx-List. I’ve set up a new Google Group called NTGreekGrammar for those of us who wish to discuss these issues in more depth. Please let me know if you would like to be a part. I’ll start sharing articles using the new group.

Thanks,

Ron Moe

 

---

[Jim Henderson]

Levinsohn's insights have been included in a text available in the Logos system, the Lexham Discourse Greek New Testament. It is available as a component in the Logos software system and also in print. It is described here:

http://www.logos.com/product/3888/lexham-discourse-greek-new-testament-bundle

----- Original Message -----

From: Jeff and Peg Shrum

To: flex...@googlegroups.com

Sent: Thursday, September 13, 2012 2:55 PM

Subject: RE: [FLEx] NT Greek with Morph data?

What could be more fun than taming the tiger of Greek morphology. But are there plans to include Levinsohn’s and others insights into Greek discourse?  This is where the big gains are to be made in pragmatics and real understanding of the meanings of these ancient texts.

 

 

Jeff Shrum

Language Technology Consultant

SIL Mozambique

 

 <snip>

[Kari Valkama]

I would like to be a part of the NTGreekGrammar group. 

I tried to find it, but Google groups search gave 0 hits. 

Yours, 

Kari 

[Griz]

Mike, looks beautiful! Cheers, Ken

On 9/12/2012 9:18 AM, Michael Aubrey wrote:

Hi Nathan,

The morphology in the MorphGNT would require re-writing. Currently, James Tauber, Ulrik Sandborg-Petersen and I have been on revising it so that it deals with certain morphological issues better--though none of us have had time to do anything with it for at least a year and a half. All three of us have other projects, work, and school that keeps us busy, though I did talk to Ulrik about it a little when I was in the Netherlands back in February. However, the format of morphology in FLEX and the tagging in MorphGNT are so dramatically different, the effort to convert it to a FLEX database is so great that it would probably be easier to start from scratch.

For that reason, I'd say the project of Ron Moe and others is probably the best shot. I've been doing the same independently from them. My goals aren't quite the same as their, since I'm aiming toward something more useful for parsing a larger corpus of Koine Greek texts beyond the NT. You can see some of what I've done on my website back in 2009 here:

http://evepheso.wordpress.com/2009/01/16/the-rough-beginnings-of-greek-morphology/

There's a link to a PDF on that page that show successful FLEX parsing of much of the nominal morphology system.

I've gotten a lot done since then and will likely have a relatively complete morphological parser up and running within the next year with a general complete lexicon of affixes. The roots and stems is growing progressively as I've worked through the text of Josephus (Niese's critical text) and portions of the NT (SBLGNT).

Mike Aubrey

PS - Ron mentioned that some people may not be comfortable with the use of the SBLGNT over the UBS text, but its a rather unreasonable discomfort. There is solid scholarship behind its editing and text-critical decisions.

---

From: Nathan Wells <sung...@gmail.com>
To: Jim_He...@sil.org
Cc: flex...@googlegroups.com
Sent: Tuesday, September 11, 2012 8:47 PM
Subject: Re: [FLEx] NT Greek with Morph data?

Thanks so much Jim,
That is very helpful! Thanks for taking the time to help!

Do you know of a way to automatically import the morphological tagging data from these files: https://github.com/morphgnt/sblgnt

Having the text in FW is a wonderful start, but we would also like to be able to include the parsing data in the interlinear (which would also make the lexicon data more useful in the future) if possible.� We could manually paste the data in, but since it already exists, having it automatically imported would obviously be ideal.� But my expertise is limited, so I thought I would ask and see if you knew of a way.

Thanks again,
Nathan

On Wed, Sep 12, 2012 at 9:42 AM, Jim Henderson <Jim_He...@sil.org> wrote:

Dear Nathan,

�

I have just made for you a backup of a Translation Editor project that contains the free SBL Greek text. (Translation Editor is part of the BTE version of FieldWorks, and is available from the usual FieldWorks download site. Or if you are a member of the LangTran software distribution system, you may have the installer on your computer already. You can join LangTran by going to

https://www.wuala.com/LangTran/Documentation/?key=JdnNdIKqrorL)

�

The Greek text was prepared by SBL, and is free for anyone to use. It�was downloaded from http://sblgnt.com/

then converted to SFM. I made the SFM files into a Paratext project. The text contains brackets that are not present in all Greek fonts, so you should download the SBL Greek font, from

http://www.sbl-site.org/educational/biblicalfonts.aspx

and install it.

�

I have put this TE project backup in my DropBox, and you can download it from

https://www.dropbox.com/sh/z5wv644ow2dr56b/tvT-1zkZU8

�

You might want to download it and open it in TE, and see if you can start adding interlinear glosses in Khmer.

�

I hope you find that helpful,

��� Jim

�

----- Original Message -----

From: Nathan Wells

To: flex...@googlegroups.com

Sent: Tuesday, September 11, 2012 12:02 PM

Subject: RE: [FLEx] NT Greek with Morph data?

Thanks Ronald,

Much work still to be done, but exciting to hear of the possibilities! Your work sound very interesting (it would be cool to see some examples), maybe I will look around for an automated way to import the SBL data in the mean time (like you said, UBS would be nice, but copyright issues prevent most people from using that). If anyone has ideas on how to do the conversion, let me know, I'm not all that experienced with FLEx.

It is great to see more collaboration happening, and having used the WeSay dictionary collaboration personally, wonderful to hear more is coming for FW.

Thanks again for your time and expertise!
Nathan

Ronald Moe <Ron...@sil.org> wrote:

Hi Nathan,

I have a NT Greek FieldWorks project that I have been working on for many years in my spare time. I am hoping to change assignments in the near future and start working on it full-time.

As Eric noted, the primary problem is copyright issues. I began work a couple of decades ago using Shoebox. At some point I got permission to adapt Louw and Nida�s �Greek-English Lexicon of the New Testament� for use in LinguaLinks. I began adding part of speech, lexeme/stem form (vs. citation form), inflection features, affixes, allomorphs, etymology, frequency, and other kinds of information that Louw and Nida�s original database was lacking. I also began doing word studies on various words, using the NT as a text corpus. But I found that I often disagreed with Louw and Nida�s definitions. The dictionary seemed to be written as an aid to translators, which meant that the definitions were more concerned about translation and interpretation issues than in describing the semantics and function of the Greek word. So I eventually came to the decision to start over. I�ve written 1,793 definitions so far, covering 1,449 lexemes out of a total of 8,670. That last figure inludes roots, affixes, variants, and irregularly inflected forms. Most of those only need a gloss.

I also began work on parsing the NT. But LinguaLinks at that time did not have a functioning parser, so I did the parsing in Shoebox. It was a huge eye-opener for me, because the traditional morphological analysis that I had been taught in school turned out to be wrong. So I had to start analyzing Greek morphology using modern linguistics. It turns out that many of the difficulties in traditional Greek grammars are nothing more than simple morphophonemics. Although the parsers in FLEx are not perfect, they are wonderfully helpful in arriving at a good understanding of the morphology of a language.

My hope is to form a team to work on developing the database into a full FW project with the NT as the text corpus. Quite a few people have expressed interest in working on the project, but we have lacked two things--the ability to work simultaneously from a distance and the organizational structure to keep things moving. The FW and WeSay developers have been getting closer and closer to the ability to put an entire project online, just as we can do now with a lexical database via LanguageDepot and Chorus. (Perhaps it is already working and I just need to learn how.) We would need to put the entire project on LanguageDepot, grant people editorial priviledges, and determine who should work on what.

We also need to incorporate a version of the NT that is in the public domain. We have the SBL NT text and someone worked on putting it into TE or Paratext. But I never incorporated it into my project. So between the SBL text and my dictionary we have the two essential components for a complete FW project that could be distributed freely. But there may be opposition to using the SBL text instead of the UBS text. Unfortunately the UBS text is copyrighted and cannot be distributed. One solution is to do all the foundational work on the SBL text and then later apply it to the UBS text for those people and applications that have permission to use the UBS text.

This will give you an idea where we are. We can communicate further off-line about specifics of sharing the file, etc.

Ron Moe

---

From: flex...@googlegroups.com [mailto:flex...@googlegroups.com] On Behalf Of sungkhum
Sent: Monday, September 10, 2012 2:24 AM
To: flex...@googlegroups.com
Subject: [FLEx] NT Greek with Morph data?

Hello,

I currently serve as a missionary in Cambodia and I am working with a Cambodian brother on creating some Greek tools for pastors. One tool we would like to see in the Khmer language is a NT Greek/Khmer interlinear. I've used Fieldworks, and started doing some tests, but coming across this group I wanted to ask if anyone has a complete NT Greek database that also includes�morphological�data that we could base our work on with instead of beginning from scratch. �I have come across this:�https://github.com/morphgnt/sblgnt�and other such resources, but ideally something already in Fieldworks would be the quickest.

[Mike Maxwell]

On 9/12/2012 11:34 AM, Michael Aubrey wrote:
> ...as far as I'm aware, no morphological tagging scheme represents declension classes of any
> kind.

I think that's correct. Most (perhaps all other) tagging schemes are intended to represent
syntactic and semantically meaningful characteristics. Declension classes by definition are
neither; they only perform selection of affix allomorphs.

Which leads me to ask: why are they represented in this tagging scheme? Why not omit them?
--
Mike Maxwell
max...@umiacs.umd.edu
"My definition of an interesting universe is
one that has the capacity to study itself."
--Stephen Eastmond

[Michael Aubrey]

That's the question, isn't it.

Personally, I'm not yet convinced that classes aren't meaningful at some level, but we cannot check for sure until we have sufficient nouns tagged.  There is no published study that has ever sought to look for linguistic patterns in either declension classes or gender classes in any era of Ancient Greek. That fact, in of itself, makes me want to include them in a morphological analysis, since FLEx provides the tools, where no other morphological database does, to search and sort for such patterns. The fact of the matter is that most linguists, translators, and biblical scholars generally aren't as interested in treating Greek as a real language in the same way we treat Bella Coola, Tagalog, Hausa, or Lengo. There is no modern linguistic grammar of Koine Greek and no existing reference grammar answers all the questions that contemporary linguists ask about language (There will be soon, however, I've been working on one, typological, functional, and cognitive for the past seven years and maybe in another five...). Generally, we're depressingly content with assuming that everything in Greek has already been done.

Mike Aubrey

---

From: Mike Maxwell <max...@umiacs.umd.edu>
To: flex...@googlegroups.com
Cc: Michael Aubrey <mga...@yahoo.com>
Sent: Tuesday, September 18, 2012 3:37 PM

Subject: Re: [FLEx] NT Greek with Morph data?

-- You received this message because you are subscribed to the discussion group "FLEx list". This group is hosted by Google Groups and is open for anyone to browse.
To post to this group, send email to flex...@googlegroups.com

To unsubscribe from this group, send email to flex-list+unsub...@googlegroups.com

[Kevin Warfel]

Mike,

 

What you’re describing sounds very much like what Ron Moe has done a lot of work on – a linguistic analysis of the base forms of the morphemes in Koiné Greek, along with the morphophonological processes that operate when they come together.  If the two of you aren’t collaborating yet, now is a good time to begin.

 

Blessings,

Kevin

To unsubscribe from this group, send email to flex-list+...@googlegroups.com

[Beth (work) Bryson]

Hi, folks.

This topic departed from being FLEx-specific a really long time ago. Could we move the discussion somewhere else? I know Ron Moe tried. This list really needs to focus on FLEx-related things.

Thanks.

-Beth

[Marcus Vanhountenmeyer]

I realize this is a very old thread.  I'm just wondering if anything has advanced in 5 years?  I'm very interested in having a fully parced Greek New Testament in FLEx.  Has anyone made progress on this?

Thanks!

[Paul Nelson]

You should look at Paratext's interlinear text. The text for Khmer and Greek should be there and you can then use the Interlinear tool there to get a head start on what you want.

Paul

--
You are subscribed to the publicly accessible group "FLEx list".
Only members can post but anyone can view messages on the website.
To change your status, please write to flex_d...@sil.org.
You can join this group by going to http://groups.google.com/group/flex-list.
---
You received this message because you are subscribed to the Google Groups "FLEx list" group.
To unsubscribe from this group and stop receiving emails from it, send an email to flex-list+unsubscribe@googlegroups.com.
To post to this group, send email to flex...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/flex-list/150ed6b4-f8c0-4507-b22a-8c53fb811db8%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ron Moe]

I have a NT Greek dictionary with at least some morphological data (part of speech, irregular forms, allomorphs, roots, entries for affixes, etc.). I'm currently working on refining the senses and adding definitions. Kevin Warfel and I made a small start on the parsing of the NT. I have the SBL Greek NT text loaded. The FLEx project is available via Language Depot. I would be happy to work with you on the parsing. I've done a lot of analysis of the morphology and can help to make the analysis more linguistically valid than what you would find in the older grammars.

Ron Moe

--