Interpretatioo / 142
coding of colunns C and D actually helped In the task by keeping distinct
the function of the utterance fron its position In the dynanic of the
interaction.
In looking for patterns in the sustaining of Interaction, though, it is
necessary to go beyond the level of the utterance. On the basis of types
of discourse suggested by Xishler (1978, see Annex 2 4) and Villians et
al.(1982), the conversations which were more than 10 turns long were
considered (Io.=99). In these, children are unlikely to sustain
conversation for long stretches, and often lose the initial dominance. The
most common pattern was found to be an alternation of lChainlng' , and
'Arching', as in:
CS Ol 05. Ol to 13 Child = TC Moises, M = Mother
in the open-air kitchen while preparing food
Ol M Jfoisds, carry the firewood to GraoDjr AogeliDa (Initiation)
02 Child what for? (Bequest in Reply Move)
03 M for her to cook with (Statement in Confirmation Move)
04 M it's Dot too much (Request in Confirmation Move)
05 Child I'll take all this to her (Response In Reply Move)
06 M was it you who chopped all that? (Request in Confirmat. M.)
07 Child yes (Response in Reply Move)
08 M so get some of that for her! (Request in Confirmation Move)
09 Child I'll get seme for her (Response in Reply Move)
10 M (...) [unclear] (probably Request in Confirmation Move)
11 Child ah! [no] (Response in Reply Move)
12 M are you goiog to carry it? (Request in Confirmation Move)
13 Child yes (Conclusion)
The Tynamic of Interaction' of this conversation has been coded as 1-3-
9-4-6-4-6-4-6-4-6-4-8. The 'Arching' exchange Is in 02 and 03, but does
not question the dominance of the adult.
Compare the following, where the child is in control: