Ask-A-Linguist Message Details
|Subject:||Consistency-checking manual transcriptions|
For languages with no writing system or no very standardized
writing system, are there any tools that support consistency-
checking of manually created transcriptions? For example, I
was reading a text in a grammar for an endangered language,
and the writer had glossed and translated both uwE and iwE as
''2SG/you''; and I wasn't sure if that was intentional or a typo
(iwE was much more frequent in the text than uwE).
If not, how do linguists working with languages without
standardized orthographies achieve transcription consistency,
for approximately phonemic transcriptions?
Previous responses have answered some parts of the question(s)
you may have been asking. I'll try a slightly different tack, and
assume you are trying to write in the language and/or analyze it.
In that case, you may well need to do a bit more analysis of the
language. The safest thing to do is take the differences you
encounter (which others have pointed out *could* be either
correct or random, inconsistent 'errors') and assume they
indicate something and try to find out what that is. If that
doesn't work out, you could assume that it is 'free variation'
but this is usually a last gasp, rather desperate measure which
indicates the analyst giving up rather than a real phenomenon. If
the conditioning phenomenon is pragmatic, for example, this may
not be evident from the data available, and you may just have to
immerse yourself more in the culture where the language is spoken
in order to tease out the conditioning factor(s). Even rather
basic issues, like getting the best orthography in place to make
literacy materials, may depend on the best phonological analysis
of the language, which a good analyst may sometimes need decades
to arrive at; literacy may be urgent enough in some cases to
require a 'best guess' approach, but always in consultation with
intelligent native speakers (these, one always finds, are
available insofar as the language is still spoken, and may have
their own ideas about linguistic matters).
I hope this has been of some use to you. Feel free to write us
back if you need further help or advice.
James L. Fidelholtz
Graduate Program in Language Sciences
Instituto de Ciencias Sociales y Humanidades
Benem'erita Universidad Aut'onoma de Puebla, M'EXICO
|Reply From:||James L Fidelholtz click here to access email|