Some hints on getting started- Once a few forms have been elicited, re-check them, and look for forms with identical melodies and group them.
- It is very helpful to have a lexical database, to know about phonotactics, likely morpheme boundary positions.
Thoughts on some common questionsWhat are good notations for transcribing tone? A variety of options are available, but the most transparent seems to be stylised contours, one per syllable, inside brackets, e.g. [ -_ ]
Should you use elicitation frames? The value of using elicitation frames is: (a) we get rid of the phrase-final effect; (b) we may see distinctions that would not appear otherwise; (c) gives a reference point for relative pitch. However, note that the isolation context is still a valid context for elicitation.
Should you repeat words back to the speaker for confirmation? If they correct you, is it for reason other than the tone? You could whistle or hum the melody back to the speaker for confirmation. Yet, speakers don't necessarily perceive such melodies in the same way when they are hummed vs spoken with segmental content.
When should you enter everything into a database? A database lets you quickly sort and categorise materials, but requires time to set up. Entering transcriptions could be slow. Its probably best to transcribe into a notebook first, then keyboard the transcriptions.
|
|