Re: X-Bar Theory

I post  it again as I sent it only to Robert. I am really sorry.
>Tule!
>
>        I was reading up on ways to diagram syntax designs, and two different ways of
>doing this came up, BNF (very complicated to look at, i think) and the X-Bar
>Theory.  I searched google for X-Bar theory, and all i could find was (very)
>skeletal information on it. I was wondering if any of you have had experiance
>with this, or happen to know where I can find detailed information on it.
Hi,

I had a class of generative grammar last year : roughly, X-bar theory can be
viewed as a module that gives the underlying structure of the language.The main
idea is that all phrases , at every level, have a structure such as :
X'' = Spec + X'
X' = X + whatever you want (Y'')

X (or X°) is the head of the phrase, its minimal projection : it can be a noun
(N), a verb (V) an adjective (A) or a preposition (P). X' is more or less a
group structure (for instance 'pretty girl' is a N'), and specifier (Spec) is a
position that allows to have a complete phrase ('the pretty girl' is a N'', and
'the' is the Spec of N'). So a N'' is just a NP, a V'' is VP and so on.

All sentences, in every languages, are supposed to be headed by a C'' node, where
C is a position for complementizer (Comp). It is very abstract, and not that
much interesting, but thanks to that node we can have relative and subordinate
structures (this really much complex).
Tat node gives a Spec(C') and a C', and and the C' becomes a C and a I'', where I
is Inflexion (tense, mood aspect and so on. Then the Sped(I') will be a noun
phrase in much cases, and the complement of I will be a V''.
I'll try to make it clearer :

C'' > Spec (C') + C'
C' > C + I''
I'' > Spec (I') + I'
Spec(I') > N'' (the pretty girl)
I' > I + V'' (cry)

I is for instance  imperfective present, and V'' can be complexified.

So we get a sentence such as The pretty girl is crying.

You may wonder what C positions are for. They are actually useful for movement
operations, but it would imply to explain all the theory which is really hard.
My piece of advice would be not to bother with that : it is complex, hard to
understand and hard-reading (especially Chomsky, because you will have to read
it if you want to learn generative grammar) and most linguists have left this
theory, mainly because it doesn't care about semantics (but they understood
they couldn't do without semantics, so they included theta-roles module) . It
would be better to try to read functional approaches (yes, I'll talk again
about Dik's 'The Theory of Functional Grammar", but it's worth and easy
reading) which are much more explanatory about how languages work and differ,
rather than try to merge them in narrow-minded structural type. Chomskyan
generative grammar isn't a good approach to language, is too much "aprioritic"
(??) or "hypothetical" (as opposed to empirical).

Do not hesitate if you want me to be more precise, because what I wrote must be
ununderstandable if you don't know a bit about generative grammar :(.

Reply