Writing Dialogues in Signed Languages

Charles Butler chazzer3332000 at YAHOO.COM
Thu Sep 15 16:23:41 UTC 2005


This brings out some questions in my mind.
 
For ASL there are a number of "classifier" signs that cannot easily be translated for use in a gloss-based dictionary, but fit under the "icon" category.  Coming up with a way to catalogue  these in a dictionary retrieval system (such as is used in a gloss-based translation program) will always be a challenge.  
 
Perhaps we need to come up with an I-set or a C-set of concepts that can be retrieved by gloss so that we know "classifier" is needed here, not "single word".  C-vehicle, C-round flat object, C-space locator, C-long thin object.  It's not intuitive for lookup, but linguistically it would be valid for study.
 
Charles Butler

Antônio Carlos da Rocha Costa <rocha at ATLAS.UCPEL.TCHE.BR> wrote:
Barbara Pennacchi wrote:

> When SL research was focused on "single signs" or one-people simple 
> signed sentences, glossing didn't show all its inadequacies. But now 
> the playground in SL research allowed by the use of glossing is 
> starting to feel a bit too small for some researchers.
...
> So this is one of the explanations for the insistence of french SL 
> researchers. They're growing bored of researching the same topics and 
> want to move on different topics, related to studying SL as an "oral 
> language", a language used in real-time between more than one person.

Thanks for making this point. It clearly shows that often science 
evolves not only by the objective requirements of the things being 
studied, but also by the subjective requirements of the scientists. It's 
good to be sincere about that!

On the other hand, that made me remember of a second thing I noticed 
in my conversation with french sign language linguists. They give a 
strong emphasis to the subject of the iconicity of sign languages, much 
more than the american sign language linguists seem to give (as far as I 
can see).

Roughly (if I could grasp the idea), iconicity means that the 
language has the ability to mold expressions to the form of the objects 
about which one is speaking. In oral language this is mainly represented 
by onomatopeia. In sign language, this is very strong in connection to 
the geometrical forms of the objects, but not only with that. But 
onomatopeia has a secondary role in oral language communication, while 
iconicity seems to have an essential role in sign language communication.

Also, in connection to iconicity, comes the ability to rapidly 
generating new expressions according to the immediate needs of a 
conversation, as new configurations of objects are referenced in 
dialogues. This may lead to signs being created by a signer, to be used 
just one single time in a conversation, and to be never used anymore.

What is the problem that this ideas related to iconicity bring to 
SignWriting? In principle, no one. If you knwo SignWriting, you know 
that in principle any sign can be written, no matter if it is a well 
established, traditional sign in a given sign language, or if it is a 
brand new sign, just created and never to be used in the future.

The problem that those linguists seem to see is not with SignWriting 
itself. The problem they seem to see is with the use (and abuse) of 
dictionaries: the over emphasis in the use of electronic dictionaries to 
support the writing of sign languages in computer editors may inhibit 
the creativity of the language, specially in connection with the 
iconicity needs of expressive conversations.

I think this is another important point that french sign linguists 
are making, and that the SignWriting community has to think about:
- should dictionaries be really present in SW text editors?
- should they be easily accessible to support text writing?
- or, should they be provided as a very separate feature, to be used 
just for consultation purposes as conventional printed dictionaries of 
oral languages are?
- should SW text editors should have auto-completion features like 
conventional text editors have?

All the best,

Antônio Carlos

-- 
Antônio Carlos da Rocha Costa
Escola de Informática - UCPel


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/sw-l/attachments/20050915/9f9bb3ce/attachment.html>


More information about the Sw-l mailing list