Chapter 7 - Complementiser PhrasesThe fronted negative obviously follows the complementiser and so cannot be in
specifier of CP. However, note that the fronted negation is still accompanied by
auxiliary inversion and therefore there must be a head position to accommodate the
auxiliary. This would suggest the following structure:
(120) CP
C'
C XP
NegP X'
X IP
Under these assumptions the negative phrase is fronted to the specifier of some phrase
that comes between the complementiser and the IP. The head of this phrase is then
where the inverted auxiliary sits.
What is the nature of XP and how does it interact with the other movements we
have reviewed? If we are to maintain our view that functional categories select for a
very limited selection of complements, then as XP is the complement of a
complementiser this argues that it is something like an IP. However, presumably X is a
functional element itself and it takes an IP complement, which makes it more like a
complementiser. Given that complementisers are categorised as [+F, –N, –V] and
inflections are [+F, –N, +V] categories, a category which shares properties of them
both would be [+F, –N], with an undefined V feature. Let us refer to this category a
little ‘i’, reflecting the use of ‘v’ to represent a verbal element with an undefined F
feature. This element heads an iP and so the structure can be represented as:
(121) CP
C'
C iP
NegP i'
i IP
Note that the fronted negative is like the focus in its interaction with the topic: the
topic precedes the fronted negative:
(122) a I said that, in this town, never have I been so embarrassed
b *I said that never have, in this town, I been so embarrassed
We can account for the distribution of the topic if we suggest that it adjoins to the
highest phrase that it can. In main clauses the topic can adjoin to the CP and therefore
as this is the highest phrase, this is where the topic will adjoin. In embedded contexts,
something prevents the topic from adjoining so high up. Perhaps there has to be a