Linguistics 221A Class Schedule and Links

Linguistics 221A --- Foundations of English Grammar

Instructor: Ivan A. Sag (sag@turing.stanford.edu)

Schedule and Links -- Winter, 2002:

Tu. 1.8 , Th. 1.10 , Tu. 1.15 , Th. 1.17 , Tu. 1.22 , Th. 1.24 , Tu. 1.29 , Th. 1.31
Tu. 2.5 , Th. 2.7 , Tu. 2.12 , Th. 2.14 , Tu. 2.19 , Th. 2.21 , Tu. 2.26 , Th. 2.28
Tu. 3.5 , Th. 3.7 , Tu. 3.12 , Th. 3.14

If you want a very brief introduction to any area of linguistics, you might check out the Linguistic Society of America's homepage at http://www.lsadc.org/web2/fldfr.htm

Tu, 1.8: A. Overview: Framework versus theory. What is HPSG? The common practice of the constraint-based grammar community: methodology. Assumptions behind the framework. Sign-Based Grammar. The basics of feature structures.

Basics of feature structures. Attribute-Value Matrices (AVMs). Rooted Directed Acyclic Graphs (DAGS). Unification. Structure Sharing. This material is covered in Shieber 1986. Read this immediately.

Th, 1.10: We discussed competence vs. performance, the issue of whether grammar should include probabilistic information, and related issues that came up in Chris Manning's class on 1.09.

We began to go over the leading linguistic ideas of HPSG, most notably:

the lexical encoding of linguistic information

the theory of headed phrases; the Head Feature Principle and its historical roots.

the theory of valence and argument `saturation', adapted from Categorial Grammar. For a brief introduction to Categorial Grammar, I recommend:

Mark Steedman. 1999. Categorial Grammar. A short encyclopedia entry MIT Encyclopedia of Cognitive Sciences, ed. by R. Wilson and F. Keil ( .ps file).

See also the Categorial Grammar Homepage at: http://www.cs.man.ac.uk/ai/CG/

Tu, 1.15: By now you should definitely have read Pollard & Sag 94, Pages 1--46, the pages from Shieber 1986, and Pollard 1996. You should also have at least skimmed some of the other introductory stuff linked to the introductory section (sec. 1) of the L221A home page. I'll give out a HW problem today (on the formal stuff) that will be due in a week.

Locality and the motivation for HPSG 2. The feature geometry of HPSG 2 and the inventory of constructions. Nonconfigurational languages. The modularity of linear order constraints.

Th, 1.17: Phrases as feature structures. How different formulations of grammar can suggest different generalizations. Once a CFG is recast in terms of sign-recursion, for example, then concatenation can be seen as a limiting case. Bach's notion of wrapping can be seen in this light. It was developed in a number of papers, including:

Bach, Emmon. 1979. Control in Montague Grammar. Linguistic Inquiry (10):513--531.

The notion of head wrapping introduced in Carl Pollard's (1984) Stanford Dissertation: Generalized Phrase Structure Grammars, Head Grammars, and Natural Language (available in the Linguistics Department or in the Stanford Library) is another interesting generalization of CFG in a sign-based architecture. Kelly Roach proved an equivalence between Pollard's Head Grammars and the Tree-Adjoining Grammars (TAGs) developed by Aravind Joshi and his colleagues. A good starting place for learning about TAGS is the homepage of the XTAG Project at the University of Pennsylvania (http://www.cis.upenn.edu/~xtag/). Perhaps the latest stuff on TAGs can be found in:

Anne Abeillé and Owen Rambow, eds. 2000. Tree Adjoining Grammars: Formalisms, Linguistic Analysis and Processing Stanford: CSLI Publications (blurb)

The wrapping kind of extension of CFG led to another, the `shuffling' extension. Mike Reape was the first to explore this within HPSG:

Reape, Michael. 1993. A Formal Theory of Word Order Doctoral dissertation, University of Edinburgh.

Reape, Michael. 1994. Domain union and word order variation in German. In J. Nerbonne, K. Netter, and C. Pollard, eds., German in Head-Driven Phrase Structure Grammar. Lecture Notes Series No. 46. Pp. 151--197. Stanford University: CSLI Publications (blurb).

An excellent elementary presentation of Head Grammars and `Sequence Union' (Reape's term for the shuffle operation) by Gertjan van Noord is available HERE
Reape's notion of shuffling with Domain Union was refined by Andreas Kathol in his dissertation and a number of papers (some with Pollard). Kathol's theory integrates the traditional idea of topological `fields' (Vorfeld, Mittelfeld, etc.) into the word order theory. An accessible starting point for this work might be these two papers:

Kathol, A. 1995. Verb-`movement' in German and topological fields. In Papers from the 31st Regional Meeting of the Chicago Linguistic Society Vol. 31. Chicago: CLS. ( .ps file)

Kathol, A., and C. Pollard. 1995. On the left periphery of German subordinate clauses. In West Coast Conference on Formal Linguistics, Vol. 14, Stanford University. CSLI Publications/SLA. ( .ps file)

See also Kathol's homepage: http://www.linguistics.berkeley.edu/~kathol/ and maybe also Stefan Mueller's homepage: http://www.dfki.de/~stefan/e_index.html

Tu, 1.22: First problem set (pset-formal.ps) due.

Chapter 3 of Pollard and Sag (1994). Lexical specfication for selection of HEAD, valence, content, features. Passive structures. Control and Raising. Expletives (dummies). Interactions of Expletives, raising, control and passive.

Th, 1.24: The evolution of valence theory. Argument Realization.

Tu, 1.29: 2nd Problem Set Due. Lexical Organization. Classificatory Relations. Alternatives to lexical rules for analyzing argument structure alternations. Feeding relations among lexical alternations. In addition to the Bouma et al. piece, please read:

Jean-Pierre Koenig. 1999a. Two Kinds of Lexical Relations ( .ps file, .pdf file). In J.-P. Koenig. 1999b. Lexical Relations. Stanford: CSLI Publications.

Also, please take a look at:

Beth Levin. 1993.English Verb Classes and Alternations. Chicago: University of Chicago Press. Pages 21-107.

Th, 1.31: More on diathesis alternations. Inflectional Relations. You also might want to take a look at some of:

Jean-Pierre Koenig. 1999c. On-Line Type Construction ( .ps file , .pdf file). In J.-P. Koenig. 1999b. Lexical Relations. Stanford: CSLI Publications.

Another prominent HPSG Lexical theorist is Detmar Meurers at Ohio State. His recent paper -- On Expressing Lexical Generalizations in HPSG, which presents a view interestingly different from that of Koenig, is available HERE ( .ps file). Related work by Meurers on lexical relations can be found in these works:

W. Detmar Meurers and Guido Minnen. 1997. A Computational Treatment of Lexical Rules in HPSG as Covariation in Lexical Entries. Computational Linguistics 23(4). Downloadable HERE.

W. Detmar Meurers. 2000. Lexical Generalizations in the Syntax of German Non-Finite Constructions. Ph.D. thesis. Seminar für Sprachwissenschaft, Universität Tübingen, 1999. Available as Volume 145 in Arbeitspapiere des Sonderforschungsbereichs 340. Downloadable HERE.

Some things to think about:

What is at stake in lexical theory?
How much should be done with inheritance hierarchies?
Are lexical rules better or worse than type-based analyses?
Has Koenig presented convincing arguments that on-line type construction is better?
Is the redundancy that Meurers talks about real?

Tu, 2.5: 3rd Problem Set Due. Start reading Ginzburg and Sag 2000, Chapter 2. You can just skim the first part about the lexicon. Final discussion of lexical representation. Parts of speech. Verbs and complementizers. VFORM values and their justification. CONTENT values. Realis SOAs and Irrealis SOAs. Messages.

Th, 2.7: Issues in the semantics of messages. Literal meaning vs. conveyed meaning. Constraints on clause content. Simple declarative clauses. Clauses ground the message recursion.

Tu, 2.12: Finish Chapter 2 of Ginzburg and Sag. Start discussing the auxiliary system. Please look at:

Rules and Exceptions in the English Auxiliary System.

Th, 2.14: Problem Set 4 is due. More on the English auxiliary system.

Tu, 2.19: Finish the auxiliary system. Some semantic bits from Chapter 3.

Th, 2.21: Demo: John Beavers will demonstrate the LKB implementation of the Ginzburg and Sag grammar.

Tu, 2.26: Chapter 5 of Ginzburg and Sag.

Th, 2.28: Dan Flickinger will give a demo of the English Resource grammar with some related background about the LinGO project.

Tu, 3.5:

Th, 3.7:

Tu, 3.12:

Th, 3.14: