"Syntactic Amalgams Re-revisited. Apokoinou Sentences in Spoken English"
Knud Lambrecht (UT Austin)
In a 1988 paper entitled "There was a farmer had a dog.
Syntactic amalgams revisited" I proposed an analysis of
a non-standard English construction I called the
'Presentational Amalgam Construction', illustrated by such
sentences as There was a farmer had a dog or
There was a ball of fire shot up through the seats in
front of me. To account for the peculiar syntactic
structure of such sentences I proposed an analysis whereby
the post-copular NP (a farmer, a ball of fire)
functions simultaneously as the object of one clause
('There was a farmer') and the subject of another ('A
farmer had a dog'), rather than analyzing the second
clause as a relative with an empty pronominal subject.
In the present paper, I extend my earlier analysis to a
number of little-known English constructions which I refer
to as Apokoinou constructions, borrowing a term from
classical Greek rhetoric. These constructions have one
basic syntactic property in common: they are complex
sentence-level structures consisting of two matrix
predications that overlap in a shared argument, the koinon.
They can be schematically represented as [A<B>C], where
B is the koinon and where both AB and BC, are clausal
constructions. The koinon is always a subject in BC.
The Apokoinou constructions I analyze can be divided into
three major subcategories: clefts, cleftoids, and non-clefts.
In clefts (e.g. I have <a friend of mine> called me),
A is semantically empty and serves as a focus marker for B,
witness the semantic anomaly of AB 'I have a friend of mine'.
In cleftoids, A is non-empty but semantically redundant, being
lexically identical or similar to C (e.g. I see her <at
Central Market> is where I see her). In non-clefts,
A and C are semantically more or less unrelated (e.g. You
can take <an anti-inflammatory medicine> usually is the
best thing). I will show that each Apokoinou category is
paired with a (more or less specific) discourse function.
The existence and common occurrence of Apokoinou constructions
in spontaneous spoken English discourse corroborates my earlier
analysis of the Presentational Amalgam Construction as a
pragmatically motivated syntactic structure which cannot be
reduced to, or derived from, a subtype of relative-clause
construction. In the Apokoinou types I analyze in this talk it is
generally impossible to reduce the sentential [A<B>C] schema to a
canonical bi-clausal pattern. Apokoinous constitute strong
empirical evidence for the validity of the theoretical construct
of 'grammatical construction' as a ready-made unit of grammar whose
constituent structure may be syntactically idiosyncratic and whose
distribution in larger units cannot be captured by general rules
or principles of syntactic composition.
|
