EACL 2009 Workshop on the Interaction between Linguistics and Computational Linguistics: Virtuous, Vicious or Vacuous?

March 30, 2009

Athens, Greece

[11/5/2009] For those who weren't able to attend in person, a video of the workshop has generously been made available for download by the local organisers

Background

In its infancy, computational linguistics drew heavily on theoretical linguistics. There have been numerous examples of co-development successes between computational and theoretical linguistics over the years (e.g. syntactic theories, discourse processing and language resource development), and significant crossover with other areas of linguistics such as psycholinguistics and corpus linguistics.

Throughout the history of the field, however, there has always been a subset of computational linguistics which has openly distanced itself from theoretical linguistics, perhaps most famously in the field of machine translation (MT) where there is relatively little in the majority of "successful" MT systems that a core linguist would identify with. In the current climate of hard-core empiricism within computational linguistics it is appropriate to reflect on where we have come from and where we are headed relative to the various other fields of linguistics. As part of this reflection, it is timely to look beyond theoretical linguistics to the various other fields of linguistics which have traditionally received less exposure in computational linguistics, including sociolinguistics, historical linguistics, neurolinguistics and evolutionary linguistics.

Aims

This workshop is an attempt to bring together linguists and computational linguists across the broad spectrum of the two fields who operate across or near the computational "divide", to reflect on the relationship between the two fields, including the following questions:

On the basis of exploring answers to these and other questions, the workshop aims to explore possible trajectories for linguistics and computational linguistics, in terms of both concrete low-level tasks and high-level aspirations/synergies.

Target Audience

The workshop is intended to be of interest to both the large numbers of people interested in deep linguistic processing (e.g. grammar developers, computational syntacticians, computational semanticists, researchers working on parsing and generation, and researchers applying deep linguistic processing in various application areas), but also those who have perhaps explicitly distanced themselves from linguistics, or who come from a linguistic background but have moved away from it in their computational linguist research. We also strongly encourage (pure) linguists to come along.

Format

The workshop will adopt the unconventional format of a series of invited talks, followed by an extended panel and discussion session, with no open CFP or submission-based talks. In this sense, it is a hybrid of a workshop, tutorial and special conference session.

Speakers

Invited Speakers:

Mark Johnson (Brown University, USA)
Frank Keller (University of Edinburgh, UK)
Mark Liberman (University of Pennsylvania, USA)
Stelios Piperidis (Institute for Language and Speech Processing, Greece)
Geoffrey Pullum (University of Edinburgh, UK)

Panelists:

Emily M. Bender (University of Washington, USA)
Gregor Erbach (European Union)
Bob Moore (Microsoft Research, USA)
Gertjan van Noord (University of Groningen, Netherlands)
Hans Uszkoreit (Saarland University, Germany)

Programme

8:55–9:00Opening Remarks
09:00-09:45Machine Translation and its Philosophical Accounts
Stelios Piperidis
09:45-10:30The Annotation Conundrum
Mark Liberman
10:30-11:00Coffee Break
11:00-11:45How the Statistical Revolution Changes (Computational) Linguistics
Mark Johnson
11:45-12:30Discussion
12:30-14:00Lunch Break
14:00-14:45Computational Linguistics and Generative Linguistics: The Triumph of Hope over Experience
Geoffrey Pullum
14:45-16:00Panel and Discussion
 Linguistics in Computational Linguistics: Observations and Predictions
Hans Uszkoreit
 Linguistically Naïve != Language Independent: Why NLP Needs Linguistic Typology
Emily M. Bender
 Parsed Corpora for Linguistics
Gertjan van Noord
 Computational Linguistics and Linguistics: What Keeps Them together, What Sets Them apart?
Gregor Erbach
 What Do Computational Linguists Need to Know about Linguistics?
Robert C. Moore
16:00-16:30Coffee Break
16:30-17:15The Interaction of Syntactic Theory and Computational Psycholinguistics
Frank Keller
17:15-17:55Discussion
17:55-18:00Closing Remarks

Workshop Organisers

Timothy Baldwin
University of Melbourne
Valia Kordoni
German Research Centre for Artificial Intelligence (DFKI) and Saarland University

The workshop is endorsed by the Erasmus Mundus European Masters Program in Language and Communication Technologies (LCT).

Last modified: Sat Aug 22 13:10:52 EST 2009