Re: RT list: inference in encoding?

Hi May, Bill, Ernst-August,

I have just received a ScienceDirect alert with a new article in Cognition
that seems relevant:

Arnold, Jennifer E., Loisa Bennetto, Joshua J. Diehl 2009: `Reference
production in young speakers with and without autism: Effects of
discourse status and processing constraints', _Cognition_ 110(2):131-146.

Regards,
Christoph

On Fri, Jan 23, 2009 at 11:38:01AM +0000, Mai Zaki wrote:
> Hi everyone,
>
> Thanks a lot for your comments. What I am thinking now is: it seems that there
> are two kinds of inferences speakers (writers) make at the level of production:
> (a) inferences related to the assumptions they want to make and how they
> correspond to their own abilities and preferences (which is part of the RT
> comprehension procedure); and (b) inferences related to the speaker's
> assumptions about the hearer's cognitive environment (as Thornstein explains)
> which would also include the assumptions Billy talks about ('assumptions not
> only about which assumptions will be manifest to addressees but also about what
> addressees will do when presented with the evidence provided by their
> utterances' )
> The question is, does any one kind of inferences have primacy over the other?
> In other words, in analysing specific examples, can we say that in choosing a
> particular linguistic form (say, a referring expression) the speaker has
> achieved a certain cognitive effect (a certain implicated conclusion for
> example) because he has made a set of inferences of one specific kind? Another
> question is, as far as speakers (or writers) are concerned, do they make a
> conscious decision about which kind of inferences they will let to prevail?
>
> On the other hand, in regards to the chicken-and-egg questions Ernst-August
> asks, isn't it in effect asking Why do people communicate? I think the layman's
> explanation would be "to convey certain assumptions", but the technical
> explanation would add "which extends the mutual cognitive environment"..
>
> Mai
>
>
> On 1/23/09, Billy Clark <B.Clark@mdx.ac.uk> wrote:
>
> Hi,
>
> Just butting in a bit on the discussion which Mai's message started.
> Mai's question, about how communicators choose which linguistic form to
> use, is closely connected to the question I asked on behalf of myself
> and Nicky, about the inferences writers make when choosing what to
> write. Ernst-August makes some useful comments which show he's thinking
> along similar lines to us. We also think that there's a sense in which
> the communicator's inferential task is more complex than the
> interpreter's. Quoting ourselves (in an abstract we just submitted),
> 'Communicators need to make assumptions not only about which assumptions
> will be manifest to addressees but also about what addressees will do
> when presented with the evidence provided by their utterances'.
>
> We're focusing particularly on writing partly because there's quite a
> lot of informal discussion among writers about these inferences (even
> though they don't always recognise this explicitly) and partly because
> of our interest in applying these ideas in teaching.
>
> We're aware of work on production in general, but there seems to be very
> little specifically on the inferences made by communicators, which makes
> us a bit nervous that we're missing something. One source we are
> referring to, of course, is Ernst-August's work on translation, which
> has a lot to say about all of this.
>
> So do let us know if you come across anything else you think we should
> know about.
>
> Thanks and best wishes,
>
> Billy
>
>
>
>
>
>

-- 
Dr. Christoph Unger
SIL International 
Hammerhof 23
67308 Albisheim
Germany
Phone: +49 6355 989939