Over the past few years there has
been increasing evidence for the
status of constructions in grammar as emergent phenomena requiring
analysis as holistic entities on their own: as participants in
networks (= morphological paradigms and inventories of phrasal
types)
of similar whole entities, rather than as epiphenomena resulting
from
the simple combination of constitutive parts. Both words and clausal
constructions can be viewed from this perspective as recombinant
gestalts.
As the empirical evidence for constructions expands in different
research domains (linguistics, psycholinguistics, language
acquisition) there has been an accompanying increase in proposals
concerning what counts as construction-theoretic proposals of
grammar, i.e., adapting the notions of Ackerman & Webelhuth 1998 and
Culicover & Jackendoff 2005 what the appropriate ``toolkit'' for
construction-theoretic approaches should consist of. Consider the
following illustrative data from German:
1a.. Active resulative predicate:
Sie hat die Schuhe krumm gelaufen
she has the shoes crooked walked
`She has walked the shoes crooked.'
1b. Passive resulative predicate:
Die Schuhe sind von ihr krumm gelaufen worden
the shoes are by her crooked walked become
`The shoes were walked crooked by her.'
1c. Adjectival derivate:
Die krumm} gelaufenen Schuhe zieht sie nicht mehr an
the crooked walked shoes wears she not more part
`She doesn't wear the shoes any more that she walked crooked.'
As argued in Ackerman and Webelhuth 1998, given certain standard
lexicalist assumptions, the complex predicate krumm laufen `to walk
flat' is best interpreted as a lexical construction, with a
periphrastic surface exponence consisting of two independent
words. We argued, following such construction-theoretic
morphological
proposals as Bochner 1993, that morphology is basically
pattern-based,
that words are realizations of content rather than incremental
compositions of morphemes, and that it is an unwarranted
stipulation (adopted in both lexicalist and non-lexicalist theories)
to restrict the domain of the morphological/lexical component to
only
synthetic (== single) wordforms. Indeed, morphological approaches
based on languages like English that insist on all words being
synthetic do not scale up to phenomena in a host of other languages
where morphological paradigms contain both synthetic and
periphrastic
forms e.g., the papers in Ackerman et. al. 2007.
Simplifying somewhat, we hypothesized certain criteria for
determining
the lexicality of an entity. In particular, if e.g., meaning changes
and valence changes are solely lexical and cannot be effectuated by
syntactic context, then the fact that krumm laufen is transitive
while
laufen is not argues straightfowardly for its lexical nature. What
we
referred to as a lexical combinatorial item, accordingly, may
receive
multiword surface expression in syntactic constructions: since it is
lexical it can serve as the source for passivization, in (1b), or as
the base for category-changing derivational morphology, in (1c).
This
entails the identification of a lexical component and a phrasal
component, as well as the articulation of principles responsible
explaining their interaction. As recognized in that theory, there
are
lexical constructions and there ar phrasal constructions: both are
necessary, they interact, and together they partition the grammar.
In this course we will explore what it means to be a lexical and
(morphological) versus a phrasal construction. What are the
similarities and what are the differences between these distinct,
but
interacting domains? The guiding assumption will be that aspects of
the traditional (pre-structuralist and pre-generative) Word and
Paradigm model of inflectional and derivational morphology (see
Matthews 1972, Aronoff 1994, Anderson 1994, Stump 2001, Booij 2005)
provide crucial insights into the necessary toolkits for grammatical
theory. In line with this we will exam various contributions to a
resurgent word-based perspective on morphological theory. This, of
course, will entail detailed examination of morphological systems in
various languages. Complementing this focus on morphology, we will
examine a particular syntactic phenomenon, specifically, a
prenominal
relative clause construction which presents a challenge to all
syntactic theories: while all present theories can describe the
properties of this type of relative (with more or less stipulation
to
their basic assumptions), none either predict it nor explain it.
Following the analysis and arguments in Ackerman, Nikolaeva (and
Malouf) we will explore the hypothesis only a construction-theoretic
approach with access constructions within networks of multiple
inheritance hierarchies provides hope for a descriptively and
explanatorily adequate analysis of this type of relative clause.
Throughout we will keep in mind issues of how construction-theoretic
approaches fit within the context of current research in cognitive-
neuroscience.
Ackerman, F., I. Nikolaeva (& R. Malouf). Descriptive Typology and
Grammatical Theory: A Construction Theoretic Study in Morphosyntax
CSLI Publication/University of Chicago Press. To appear in 2007.
Ackerman, F., James P. Blevins, & G. P. Stump, editors. Paradigms
and
Periphrasis. CSLI Publications/University of Chicago Press. To
appear
in 2007
Ackerman, F. and G. Webelhuth. 1998. A Theory of Predicates. CSLI
Publications.
Anderson, S.R. 1992. A-morphous Morphology. Cambridge University
Press.
Aronoff, M. 1994. Morphology by itself. MIT Press.
Bochner, H. 1993. Simplicity in Generative Grammar. Mouton de
Gruyter.
Booij, G. 2005. The Grammar of Words: An introduction to linguistic
morphology. Oxford University Press.
Culicover, P. and R. Jackendoff. 2005. Simpler Syntax. Oxford
University Press.
Matthews, P. H. 1972. Inflectional Morphology. Cambridge University
Press.
Stump, G. 2001. Inflectional Morphology. Cambridge University Press.