1.5 Processes of Change 29intersubjectifi cation (encoding of meanings focused on the addressee), both of
which have been linked crucially to grammaticalization (e.g., Traugott 1995b ,
2003b , 2010 ; Traugott and König 1991 ). For example, the comment clause I
dare say undergoes subjectifi cation of the original meaning ‘have courage to
utter in words’ and becomes epistemic , or a speaker- oriented expression of
tentativeness, while if you will ‘if you are willing to say so’ acquires the more
subjective meaning ‘if you will allow me to say so’ (Brinton 2008 : 93– 97,
177– 180). Intersubjectifi cation is obvious in the case of you see , which no
longer denotes physical vision but claims mutual understanding and avoids a
threat to negative face (see Brinton 2008 : 159).
But in some respects, the development of pragmatic markers diverges
from what is thought to occur during grammaticalization. They do not seem
to acquire the prototypical grammatical qualities that we expect in a fully
grammaticalized infl ection or clitic. They do not typically reduce phonetic-
ally and fuse with the host form but remain independent items.^31 They do not
become an obligatory element in a grammatical paradigm, thus undergoing
neither “paradigmaticization” nor “obligatorifi cation” in Lehmann ’s terms
(2002a). In respect to Lehmann’s ( 2002a ) parameters, the two that have been
felt to be most problematic in respect to pragmatic markers are “fi xation ”
and “condensation.” As noted above, pragmatic markers are movable, not
syntactically fi xed, though they do seem to prefer clause boundaries and are
often found in sentence- initial or fi nal position. As pragmatic markers gram-
maticalize and assume procedural functions, their scope in fact expands, as
they come to have scope over larger segments of discourse ( Traugott 1995a ,
2003a ). This is di rectly contrary to the scope reduction one sees, for example,
in the development from word > infl ection (e.g., main verb have to auxiliary
have to infl ection in French ) (see Lehmann 2002a : 128ff.). However, scope
reduction as a necessary component of grammaticalization has been chal-
lenged. On the one hand, Tabor and Traugott ( 1998 ) argue instead for scope
expansion in grammaticalization under tightly regulated conditions which
they call the “C- Command Scope- Increase Hypothesis.” On the other hand,
Fischer ( 2007b : 280– 297) argues against a simple, direct increase in scope
from clause- internal adverb into pragmatic marker. She proposes a different
path: The adverbial comes to occur in a separate, higher clause (e.g., OE þæt
is sarlic þætte ‘it is sad/ grievous that’) or an independent phrase ( to soðe ‘to
truth/ truthfully’) which occupies topic position and which, by defi nition, has
scope over the rest of the proposition; word order shows that these forms are
independent and outside the main clausal structure. Ellipsis leads to reduction
of this clause and rise of the pragmatic marker.
31 In respect to autonomy, pragmatic markers may be compared to function words (e.g., will , the ,
an , to ), which are nonetheless considered fully grammaticalized.