ISCL Proseminar (Sommersemester 2023)

Grammar Formalisms in Computational Linguistics

Abstract:

Given that natural languages cannot be characterized by simply listing all possible sentences and their meaning, a range of grammar formalisms have been developed to characterize form and meaning in a general and compact way. The approaches differ in terms of their focus, empirical coverage, formal foundations, expressive power, conceptualization of generalizations, and the processing regimes that have been developed for those formalisms.

After a general overview of grammar types in the Chomsky Hierarchy, we will discuss plain context-free grammars as a baseline on which we will introduce and compare several current grammar formalisms. The plan is to include a discussion of Head-Driven Phrase Structure Grammar (HPSG), Tree Adjoining Grammars (TAG), Lexical Functional Grammar (LFG), and Combinatoric Categorial Grammars (CCG).

The focus will be on obtaining a sound working knowledge of how different formalisms capture some of the fundamental phenomena of natural language syntax: argument and adjunct realization, agreement and government, middle-distance phenomena (e.g., equi, raising), long-distance phenomena (e.g., fronting).

While the more recent Chomskyan frameworks (Minimalism, Bare Phrase Structure) are not covered in the course given their lack of relevance for computational (or empirical) linguistic work, the grammar formalisms course essentially addresses a desideratum clearly expressed in Chomsky (1981, pp. 335-6): “It is an open question whether full-scale formalization is a worthwhile endeavor at the moment ... My personal feeling is that the point has been reached where these further steps should be undertaken, that there is sufficient depth and complexity of argument so that formalization will not merely be a pointless technical exercise but may bring to light errors or gaps and hidden assumptions, and may yield new theoretical insights and suggest new empirical problems for investigation.”

Instructor: Detmar Meurers

Tutor: Valentin Pickard

Course meets:

Note: Active attendance for both lectures and one lab session per week is obligatory.

Credits: BA 9 CP, MA 6 CP
For other degree programs, contact me for requirements and credits.

Moodle: We will be using the university Moodle site for the course, primarily for the discussing forum and to access course materials. Our course is accessible under Moodle at https://moodle.zdv.uni-tuebingen.de/course/view.php?id=3304.

To log into this specific Moodle site, you use your general ZDV university account id and password. Moodle and privacy: Note that Moodle generally keeps detailed logs of your interaction with the system, e.g., when you log in, etc.

Email: In the Moodle system everyone in the course can send messages to other participants in the class, and I will use this to contact you for class related matters. Such email gets sent to your regular ZDV account (@student.uni-tuebingen.de). So register in the Moodle during the first week of the semester, and read your university email regularly.

Nature of course and my expectations: This is a core course in ISCL and I expect each participant to take an active role in the class: i) regularly participate in the class sessions, with your camera on, ii) read any reading assignments before class, iii) actively participate in the practical sessions and carefully complete all assignments and implementation activities and homeworks.

Grading: Successful completion of the course will be based on participation, homeworks, and the final exam held in the last week of the semester, on Tuesday July 25.

Academic conduct and misconduct: Learning and research are driven by discussion and free exchange of ideas, motivations, and perspectives. So you are encouraged to work in groups, discuss, and exchange ideas. At the same time, the foundation of the free exchange of ideas is that everyone is open about where they obtained which information. Concretely, this means you are expected to always make explicit when you’ve worked on something as a team – and keep in mind that being part of a team means sharing the work! For text you write, you always have to provide explicit references for any ideas or passages you reuse from somewhere else. This includes text “found” on the web, where you should cite the url of the web site in case no more official publication is available. Failure to follow these important guidelines is academic misconduct, which will be sanctioned by failing you on the assignment, exam, or the entire class depending on the severity of the violation.

Topics we plan to cover