Kopfbild


HPSG 2013

Stefan Müller: Some Remarks on the State of the Field and on Ways to Ensure Progress: Description, Analysis, Formalization, Implementation

One problem of current theorizing in mainstream generative grammar (MGG) is the impressionistic use of data (Fanselow, 2009). Another problem is that constructs that are manifested in one language or a language class are taken as evidence for invisible structures in other languages (Chinque, 1999; Meinunger, 2000).

Another problem of the post GB era is the explosion of theoretical variants that make incompatible basic assumtions. This makes it very hard to get a coherent picture of where we are and of how to combine insights from different authors/schools.

In my talk I will briefly comment on the Poverty of the Stimulus argument (Chomsky, 1971, p. 29–33; 2013, p. 39) and show that many of the claims that have been made in the past 60 years cannot be upheld any longer in the light of results from input-based approaches to language acquisition (e.g. Bod, 2009; Freudenthal, et. al., 2006, 2009). The consequence of this is that gramars for particular languages have to be motivated solely by evidence from the language under consideration. I will then discuss the approach taken in the CoreGram project, in which crosslinguistic generalizations are derived in a bottom-up way from grammars that are acquirable in an input-driven way.

I will compare this method to the method used in MGG and address Croft's criticism of the MGG methodology. I will show that the bottom-up approach makes predictions and how Cinque's insights could be integrated into HPSG grammars without stipulating irrelevant empty nodes in all languages of the world.

As a way out of the crises I will suggest a return to the empirical basis (compare the quote of von Neumann's text on the state of the art in mathematics) and I consider results from corpus research, psycholinguistic/neurolinguistic experiments relevant in this respect. I will suggest that every (book) publication should come with a list of covered data so that it becomes possible to see whether more data is covered than in earlier proposals. This would make results comparable and would be a strong incentive not to fall behind what was understood already. As an example I will use again the grammars of the CoreGram project, which are backed up by implementations, which ensures the consistency of the formalization and yields new insights and discoveries which would not be possible without thorough formalization and implementation. The every growing lists of covered phenomena ensures progress and helps with systematic testing. Although computer implementations are very helpful in veryfying complex theories, this approach does not require it and can be used for non-computational work as well.

References

  • Bod, Rens. 2009. From Exemplar to Grammar: Integrating Analogy and Probability in Language Learning. Cognitive Science 33(4), 752–793.
  • Chomsky, Noam. 1971. Problems of Knowledge and Freedom. London: Fontana.
  • Chomsky, Noam. 2013. Problems of Projection. Lingua 130, 33–49.
  • Cinque, Guglielmo. 1999. Adverbs and Functional Heads. A Cross-Linguistic Perspective. New York, Oxford: Oxford University Press.
  • Cinque, Guglielmo und Rizzi, Luigi. 2010. The Cartography of Syntactic Structures. In Bernd Heine und Heiko Narrog (Eds), The Oxford Handbook of Linguistic Analysis, pages 51–65, Oxford: Oxford University Press.
  • Croft, William. 2009. Methods for Finding Language Universals in Syntax. In Sergio Scalise, Elisabetta Magni and Antonietta Bisetto (eds.), Universals of Language Today, volume 76 of Studies in Natural Language and Linguistic Theory , pages 145–164, Springer Netherlands.
  • Fanselow, Gisbert. 2009. Die (generative) Syntax in den Zeiten der Empiriediskussion. Zeitschrift für Sprachwissenschaft 28(1), 133–139.
  • Freudenthal, Daniel, Pine, Julian M. and Gobet, Fernand. 2006. Modeling the Development of Children’s Use of Optional Infinitives in Dutch and English Using MOSAIC. Cognitive Science 30(2), 277–310.
  • Freudenthal, Daniel, Pine, Julian M. and Gobet, Fernand. 2009. Simulating the Referential Properties of Dutch, German, and English Root Infinitives in MOSAIC. Language Learning and Development 5(1), 1–29.
  • Meinunger, André (2000), Syntactic Aspects of Topic and Comment, Amsterdam/Philadelphia: John Benjamins Publishing Co., volume 38 of Linguistik Aktuell/Linguistics Today.
  • Müller, Stefan. 2013a. The CoreGram Project: A Brief Overview and Motivation. In Denys Duchier and Yannick Parmentier (Eds): Proceedings of the Workshop on High-level Methodologies for Grammar Engineering (HMGE 2013), Düsseldorf.
  • Müller, Stefan. 2013b. The CoreGram Project: Theoretical Linguistics, Theory Development and Verification. Ms FU Berlin.
  • Ross, John Robert. 2011. Letter to the MIT Linguistics Department for its 50th Anniversary. http://ling50.mit.edu/replies/haj-ross, 26.08.2013.