FG and the dynamics of discourse 221lient in the speech setting or because it is ‘inalienably possessed’ or other-
wise anchored in the individuality of one of the interlocutors (i.e. through
deictic reference, e.g. those ugly pictures or the woman in the green hat
over there, your left leg, or my sister's second ex-husband). All instances of
identifiability and their expression by a single grammatical category as-
sume the existence of a cognitive schema, or frame.
Activation, on the other hand, evokes Chafe’s (1987: 22) idea “that our
minds contain very large amounts of knowledge or information, and that
only a very small amount of this information can be focused on, or be ‘ac-
tive’ at any one time”. Accordingly, three following degrees of activation
are posited:
(a) active, designating “currently lit up” information that is, as Chafe
has it, “in a person's focus of consciousness at a particular moment”;
(b) semiactive (or accessible), designating information that is in a per-
son’s peripheral consciousness or in their background awareness as a
result of: (i) deactivation from an earlier state (i.e. textual accessibil-
ity); (ii) inference from a cognitive schema or frame, that is, from
some other active or accessible element in the universe of discourse
(i.e. inferential accessibility); and (iii) presence in the text-external
world referent (i.e. situational accessibility).
(c) unused, designating inactive information that is “currently in a per-
son's long-term memory, neither focally nor peripherally active”.
Note that, although D-Topics normally entail some degree of identifi-
ability and/or accessibility, not all identifiable or accessible items become
D-Topics, for the latter must also show a prospective dimension (see note
13). The problem arises how to identify these cognitively oriented D-
Topics. For this sketchy presentation it suffices to assume that D-Topics
generally profile relationships normally anchored in the situational and lin-
guistic co(n)text, in conformity with natural paths of mental accessing (see
notes 5 and 14). This grounding process occurs in successive attentional
frames (intonation units with corresponding Focus spans), whose arrange-
ment is basically linear and affected by the order of presentation (in
principle what is presented first will profile what follows, see Section 2.3),
and which are superimposed on structural elements, whose arrangement is
both linear and hierarchical. There is a natural tendency for attentional
frames to coincide with grammatical constituents, but it is only a tendency,
not an inviolable principle. Conceptual organization need not be directly
mirrored in grammatical constituency, for grammar is a tool for building up