Discourse -y Stuff
I have started writing up the ALS 2006 paper I did on 'det' ( a determiner in Kriol), and then furthered at the New Mexico 'High Desert Linguistics Society' conference. One thing I would never have thought of to do before that I have found has interesting results, is to look at things from a discourse analysis perspective. In this case, I had already looked at and noted an unusually high frequency of occurance in discourse (naturally occuring conversation), but another interesting thing came up, which is to look at its distribution when it is a 'core argument' (subject or direct object) of the verb- (no keep reading!). The interesting thing is that at first glance at least the 'det' Noun construction ( I mean like det fish, det Sophie etc) seems to occur the majority of the time as a subject of a transitive verb, and almost never in the other positions... but this needs a closer investigation.
I am really enjoying learning about new ways of analysing language in its functional context, though I am not a complete convert, I think there is a lot to be said for grammar 'emerging' from , or at least being heavily influenced by how the language is used when people speak it. By this I mean that there are other constraints on language, dyamic ones, other than the grammatical ones, such as how much information a person likes to process at a given time, how much breath you have to say each part etc etc.
One thing I had never thought of before is breaking down conversations into 'Intonation Units' (IU's) as you write them down. This solves my prior problems of 'what is a clause' ( try look at a conversation transcript and find one!), which becomes less relevant as IU's can be a standardised way of quantifying (e.g. it is 350 IU's) and analysing the language data. IU's are relatively easy to identify (a single intonation contour..), though I did take a semester long course to try and get the hang of it...but the interesting thing is is that they appear to be 'universal' across languages (something to do with length of breath? 'thinking for speaking?').
I have learnt so many things since I arrived here and I want to tell anyone who is interested about it all. So if you have any questions (baba?) I will do my best to mine out some references and tell you..
Unfortunately the last ALS (2006) was a bit difficult as N died the day before it started, and it was all a bit of a blur, but I just submitted an abstract to ALS 2007 titled ' Reference in Spoken Discourse in the Ngukurr Aboriginal Community: The Status of Proper Names'- which seems a hot topic these days (cf. Garde, et. al), and I am really enjoying researching it, I think NJ would have too.
I am really enjoying learning about new ways of analysing language in its functional context, though I am not a complete convert, I think there is a lot to be said for grammar 'emerging' from , or at least being heavily influenced by how the language is used when people speak it. By this I mean that there are other constraints on language, dyamic ones, other than the grammatical ones, such as how much information a person likes to process at a given time, how much breath you have to say each part etc etc.
One thing I had never thought of before is breaking down conversations into 'Intonation Units' (IU's) as you write them down. This solves my prior problems of 'what is a clause' ( try look at a conversation transcript and find one!), which becomes less relevant as IU's can be a standardised way of quantifying (e.g. it is 350 IU's) and analysing the language data. IU's are relatively easy to identify (a single intonation contour..), though I did take a semester long course to try and get the hang of it...but the interesting thing is is that they appear to be 'universal' across languages (something to do with length of breath? 'thinking for speaking?').
I have learnt so many things since I arrived here and I want to tell anyone who is interested about it all. So if you have any questions (baba?) I will do my best to mine out some references and tell you..
Unfortunately the last ALS (2006) was a bit difficult as N died the day before it started, and it was all a bit of a blur, but I just submitted an abstract to ALS 2007 titled ' Reference in Spoken Discourse in the Ngukurr Aboriginal Community: The Status of Proper Names'- which seems a hot topic these days (cf. Garde, et. al), and I am really enjoying researching it, I think NJ would have too.
posted by Sophie at 10:45 pm
3 Comments:
Yay! Oh Yay!
She's posting on discourse!!
I have so much to also write up after my recent Blackwood paper, and will try to post it soon. In trying to describe the discourse functions of the demonstratives in Dalabon, I'm trying to get beyond what Garde (2002) identifies for the BKW demonstratives, namely that they have 'various emphatic indexical and recognitional functions.'
What are these 'various emphatic indexical functions'? So far, I'm lumping them in the 'now' function... (as in 'that man now', or perhaps better: 'that very man')
Looking at the Dalaban data and trying to use discourse analysis tools, what I felt it boiled down to was: 'How on earth do you 'measure' referent accessibility in an Aboriginal language (given Garde's (2002) 'Be brief, Be oblique' principle)?' (=what on earth do you consider to be a 'last mention'?!) and 'Is it even worth trying to 'measure' accessibility?'
Nick Evans made the probably very useful comment that one needs to be mindful of Walsh's 'broadcast model' of Aboriginal communication, and that the speaker can't assume that the hearer is even tuned in the whole time, so the emphatic markers alert the hearers to what they really need to be paying attention to.
So, maybe this whole 'accessibility' in terms of how many IUs since last mention or persistence isn't relevant at all. Where to from here? If NE is right, then speakers are only marking referents which are important in narrative structure. Narrative analysis a la Labov and Waletsky (1968)?
Oh man! I wish I was there to talk to you baba- do you have skype? Are there demonstratives that can occur with proper names? with generics? (in the det N sense). One of the things that I am trying to show in my paper is that the data kind of shows that referents often though they are specific may not be 'identifiable', and therefor are not definite- i think it has more to do with not 'overstating' referents, rather than minimising them. As in, it isnt at all unusual to actually guess specifically who the person is talking about... actually I am not sure where I am going with this! Is your data mainly narrative? Just one speaker? Elicited? Skype me! I would love to hear more! I can be found under my gmail address.
can you email me your gmail address? I don't seem to have it...
Post a Comment
<< Home