The Complexity Principle and lexical complexity in the English and Dutch
dative alternation
This study investigates the effects of lexical complexity on the
choice of dative alternants in English and Dutch. The lexical complexity of
a given word is operationalized as being proportional to how quickly
speakers can retrieve it from their mental lexicon, for which I consult the
databases of recent megastudies (Keuleers, Diependaele, and Brysbaert 2010: 1). Following the
Complexity Principle (Rohdenburg 1996), which states that cognitively
complex environments favour the grammatically more explicit variant in
linguistic alternations, it could be expected that lexically complex
environments favour prepositional datives. However, the models suggest that
speakers’ choices are not particularly sensitive to the complexity of larger
linguistic environments. Instead speakers aim to place the lexically
easier constituent before the more complex one. This turns out to be one of
the strongest predictors in both languages.
Article outline
- 1.Introduction
- 2.Background
- 2.1Lexical complexity
- 2.2The Complexity Principle
- 2.3The dative alternation
- 3.Methods and data
- 3.1Operationalizing lexical complexity
- 3.2The corpora
- 3.3Delimiting dative contexts
- 3.4Annotation procedure
- 3.4.1Theme and recipient lexical complexity
- 3.4.2Lexical complexity difference
- 3.4.3Preceding word(s) lexical complexity
- 3.4.4Complexity of the constituents
- 3.4.5Complexity of the constituents and their preceding
word(s)
- 3.4.6Filtering of the data and supplementary predictor
variables
- 3.4.7Weight difference
- 3.4.8Discourse status
- 3.4.9Pronominality
- 3.4.10Syntactic complexity
- 3.4.11Definiteness
- 3.4.12Recipient Animacy
- 3.4.13Verb position
- 3.4.14Verb complexity
- 3.4.15Middle field
- 3.5Statistical analysis: Conditional random forest analysis
- 4.Results
- 4.1CRF analysis of the English dataset
- 4.2CRF analysis of the Dutch-language dataset
- 5.Discussion
- 6.Conclusion
-
Acknowledgements
-
Notes
-
References