Overt movement of negated arguments to Spec NegP in later Middle English was proposed by Ingham (2000, 2002), in which
sentential negator not and displaced negated objects were in complementary distribution. However, the earliest data
considered in that research (cf. also van der Wurff (1999)), are dated to the
1450s, and most examples came from the Paston letters, e.g.
(1) I wol [NegP non answeri [VP yeve hym
ti ] ]
Paston D 600, 34 (1460)
Evidence of the phenomenon is predicted to become available earlier than this; already in the 14th
century not had become generally grammaticised in Spec NegP (Frisch 1997, van Kemenade 2000, Ingham 2004). In the present
study earlier prose data are therefore analysed, corresponding to the Helsinki Corpus period M3 (1360-1420), and from a wider
range of geographical areas. Displaced negated object cases such as (1) were found to occur in about 20% of contexts. OV with
ordinary objects was no longer found. Co-occurrence of not with an in
situ negated object in S-Aux-V-O structures also occurred quite widely, e.g.:
(2) …žat wolt not don no lecherie
Mirror of St Edmund, Vernon ms., Horstman 248.346
(3) 3ou ne shalt nat eek make no lesynges Chaucer, Pars. 1018
(4) žat he ne wyl no3te lere na science
Northern prose Rule of St Benet 39,32
Although the rate of usage was fairly low (about 12-20% of contexts depending on the region), it contrasts with the
complete absence in the same texts of not in clauses with displaced objects such as (1). It also contrasts with the
absence of not in S-Aux-V clauses with negated subjects, e.g.:
(5) …žat no creature may attenden hit parfiteliche
Mirror of St Edmund, Vernon ms., Horstman 260,2
(6) No man ne sholde know his owne engendrure
Chaucer, Pars. 923
(7) Na man wyl prayse žaim
Northern prose Rule of St Benet 7,8
Where a negated subject and a negated object co-occurred, not was never found, as expected if the negated subject
had traversed Spec NegP leaving a trace,
These findings are taken as further support
for the claim in Ingham (2000) that not in Spec NegP would block movement of a negated subject out of VP passing through
Spec NegP, whereas negated objects remain in VP, and therefore allow the optional appearance of not in Spec NegP. The
region with the lowest rate of negated OV, the South-West also had the lowest rate of not accompanying a negated object
in S-Aux-V-O configurations.
It has been argued that Spec NegP became a movement site in English when the grammar switched from an interpretable
head feature to an interpretable spec feature on NegP (Ingham 2004). So far the earliest evidence for this has come from 14th
century expletive negative subjects (Ingham 2003). Displaced objects can now be seen to have behaved in accordance with expectations
at an earlier date than had previously been established.
