Towards a speaker model of FG 351SENTENCE
1 4 9
SUBJ VC SAT
2 5 7 10 12
PRO AUX PRED PREP N
3 6 8 11 13
/he/ /would/ /come/ /into/ /class/
Figure 4. Partial expression tree; stage 1
After (12) – previously (11a) – has been articulated, working memory
will contain this tree, or at least such a substructure of it as to support the
rest of the syntactic entity under construction. The same will be assumed
for the IL and RL structures of (14) and (15). There is a short pause now, in
which new pragmatic and semantic material is being processed, corre-
sponding to (20).
(20) ... a-t .. uh- you know three or f
This unit is complicated and particularly interesting. It starts with a
preposition, which is an indicator of a temporal satellite, as we know on
hindsight. This preposition is articulated, but in a hesitant, prolonged way,
apparently because the speaker is not sure about the temporal details of the
satellite. Therefore, extensions will be made at the IL and RL levels. We
will represent these as follows. First, an update will take place on the struc-
ture in (13). An extra referential act R 3 will be added at the IL level, being
the cognitive version of the temporal satellite under creation. This leads to
(21); the new part is printed in bold.
(21) (M 6 : [INFRM [(A 6 : [DECL (P 1 )S (P 2 )A
(C 6 : [R 1 (x 1 )TOP R 2 (x 8 :CLASS)FOC T 1 (f 1 : GO INTO)
R 3 (x 9 :TIME)FOC])])]])