[Lexicog] folk taxonomies and universal semantic domains within lexical software

[Ron Moe]

Wayne has raised an issue that I have been concerned about for quite some time. He correctly notes that my list of domains is designed as an etic tool for collecting words. We use the IPA to begin our investigation of the phonetics of a language. But our goal is to describe the phonemic system. In the same way we can start our investigation of the semantics of a language with an etic list of domains. But our goal is to describe the emic system. I wrote an article (available on the DDP website) which explores ways to move from the DDP domains to an emic list that reflects the semantic categories and world view of a particular language.

 

Currently FLEx supports DDP. When importing MDF data, the \is and \sd fields must match DDP version 4 or FLEx will add all your domains to the DDP list inside FLEX. Unfortunately this creates a bit of a mess if your list is very different from the DDP list. You can import your semantic domains to residue and then set up a user defined field to keep your list of domains. But this is less than ideal.

 

FLEx includes the DDP list of domains with all the elicitation questions and sample words in a special ‘Categorized Entry’ tool. So you can use DDP from within FLEx to collect words and efficiently enter them into FLEx. You can also edit the DDP list of domains, adding and deleting domains as necessary. You can print/display all the domains or only those that have words in them. You can provide a vernacular label for each domain. But we still need to develop this area of the program. One current problem is that you cannot change the nine major domains. You can change any sub-domain. We also need to work on help files/tutorials to teach the user how to develop an emic list of domains. We may also need to make it easier to modify the list of domains. There may be an advantage to setting up two systems of domains, one etic (e.g. the DDP list or some future standard) and one emic. This would enable people to use the etic list for cross-linguistic comparisons and the emic list for publication. Another advantage of maintaining an etic classification within your database is to facilitate reference to standardized materials. For instance I am beginning to write materials to help lexicographers do semantic research and write definitions. These materials will be indexed to the DDP list of domains, since they will be written for a particular domain. For instance there will be a module on investigating and describing emotion terms (DDP 3.4). One easy way to maintain both an etic and emic list of domains would be to allow the user to add an emic number for each domain and sort on either the etic number or the emic number. The user could then choose which domains are relevant for his language, organize/renumber them according to his world view, and export the emic list for publication. But the etic numbers would still be in the database for reference purposes. I’ll see if we can’t add this to FLEx. It’s still under development so we have the opportunity to request features that we need.

 

Ron Moe

 

   _____  

From: lexicographylist at yahoogroups.com [mailto:lexicographylist at yahoogroups.com] On Behalf Of Wayne Leman
Sent: Thursday, November 09, 2006 8:19 AM
To: lexicographylist at yahoogroups.com
Subject: [Lexicog] folk taxonomies and universal semantic domains within lexical software

 

  

Following is a slightly revised copy of a message I just sent to a colleague who is on the FieldWorks (FW) team. The issues are relevant to each of us who attempt to discover indigenous semantic domains:

 

___, after you skimmed my database before attempting import to FW FLex, you mentioned that my semantic domains did not exactly fit Ron Moe's list. Here is why: As Ken Pike used to emphasize, there are etic and emic approaches to language data. Ron Moe's approach to lexical research is one of the best I have ever seen. His approach can gather lexical data in a short period of time what many people used to take years to collect. (Of course, one can also benefit greatly from other traditional approaches to lexical data gathering, including text collection and gleaning those texts for new lexical items.) It is appropriate that language software be keyed to the good semantic domains that Ron has developed after careful research of his own, in consultation with others working in lexicography. (I've had helpful exchanges with Ron, whose work I appreciate very much.)

 

But for any dictionary, we need to ask who the audience for it is and whether we want etic or emic categories to be displayed. *IF* a dictionary is intended to be used by the native speakers of a language, it is wise to shift from using any universal etic semantic domain list to the semantic folk taxonomies of those native speakers. As you probably know, different people in different cultures have differents semantic domains. So, for display, such as in Lexique Pro (a fantastic program for use by native speakers for displaying dictionary data), there is wisdom in using native (emic) semantic domains, rather than those of a more universal etic set of domains such as those in Ron's DDP (HYPERLINK "http://www.sil.org/computing/ddp/"http://www.sil.-org/computing/-ddp/).

 

FLex (part of FW) should allow for using Ron's domains for guiding semantic category discovery, but it should also allow for converting semantic domains to those actually used by a language/culture group.

 

We have tried hard over the past 30 years to work toward using Cheyenne semantic domains so that Cheyennes can locate lexical items within the semantic domains that they actually use.

 

Wayne

-----
Wayne Leman
Cheyenne dictionary online:
HYPERLINK "http://www11.asphost4free.com/cheyennedictionary/default.htm"http://www11.-asphost4free.-com/cheyennedict-ionary/default.-htm

 


--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.409 / Virus Database: 268.14.0/524 - Release Date: 11/8/2006



-- 
No virus found in this outgoing message.
Checked by AVG Free Edition.
Version: 7.1.409 / Virus Database: 268.14.0/524 - Release Date: 11/8/2006
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lexicography/attachments/20061108/4cd7a25d/attachment.htm>