CROSS-MODULAR APPROACHES TO ELLIPSIS

The area of ellipsis resolution and generation has long been neglected in work on natural language processing, and there are few examples of working systems or computational algorithms. However, the misuse or non-use of ellipsis in highly preferred contexts can make a dialogue difficult to understand similar to the way inappropriate referential expressions can impede comprehension. Furthermore, we think ellipsis has many more functions than are currently recognized, an idea discussed in our own contribution to this volume. We are grateful that the ESSLLI organizers have provided a forum for researchers to present new data and analyses. The accepted contributions in our opinion give new insights into the nature and function of ellipsis. The work of Ericsson, Alcántara & Bertomeu and Hoeksema all present results based on corpus study, helping to fill the gap in our knowledge about naturally occurring examples of ellipsis. The work of Repp and Fernandez show in particular how discourse structure and effect the interpretation of ellipsis, and Hielkema, Theune & Hendriks use discourse coherence relations to guide the ellipsis generation process. Like Hielkema, Theune & Hendriks, Ericsson also discusses issues in the appropriate generation of ellipsis, an area where more work needs to be done. Finally, both Nielsen and Kreutz discuss VP-ellipsis, one of the more common forms of ellipsis in English. Nielsen shows an actual implementation to resolve VP- ellipsis, while Kreutz emphasizes the need to incorporate more than traditional knowledge sources. We are certain that the contributions in this proceedings, and the talks in Edinburgh will inspire others to look at the fascinating and challenging problem of incorporating ellipsis into natural language applications. Because without ellipsis, natural language applications will not have the naturalness and functionality present in actual natural language.