Ivan A. Sag (sag@Csli.Stanford.EDU)
Tue, 9 Dec 97 18:39:47 PST
Dear Karel and Tibor, I understand your concerns about reviewing. Gosse Bouma, Rob Malouf and I were all bewildered (not to mention upset) when we received a letter rejecting an abstract sent to a recent conference. We thought that remarks like the following were particularly unfair: >The topics addressed in this abstract are important. Unfortunately, it >does not offer any new empirical consideration, nor does it provide a >precise account of the adjuncts-as-complements approach. Instead, it >offers a bunch of pseudo-formal pretensions (which are partly false >and partly incomprehensible) and terminological confusions. We felt this way particularly because (1) the abstract outlined solutions to defects of the Pollard-Sag 94 extraction analysis that had been carefully documented by Hukari and Levine (and which remain unsolved in other approaches, as far as we are aware) and (2) the solutions were carefully explained, fully implemented, and systematically compared with alternative approaches. But you may be surprised to learn that this was not the Formal Grammar Conference, but rather the HPSG conference held in Ithaca last July. Oh, I should add that Carl Pollard also had an abstract rejected from that HPSG conference... So what does this all this mean? Perhaps it means that Pollard and I and our collaborators have all lost contact with the mainstream of HPSG research or perhaps that Gosse has led Rob and me away from serious work into that dark, dank world of pseudo-formal pretension that he's so famous for (just kidding, Gosse...:-)), but I'm not ready to believe that any of us has devolved to such an extent. The abstract must not have been clear enough or else it struck some nerve on the part of one or more reviewer (I don't know...) And I certainly don't harbor any grudge against the program committee, who as far as I can see worked very hard to put together an outstanding program in Ithaca, even if we weren't on it. I think it's really important to keep in mind that review processes can't ever be perfect. Reviewers sometimes misinterpret abstracts; sometimes they have biases that they can't control; sometimes they see something they like in a paper that turns out to be crap; sometimes they even invert the rating scales by accident and report the wrong evaluation. I was once on a program committee that almost rejected an abstract by Peter Ladefoged because, though they didn't know much about the language discussed in the abstract, they knew that the definitive work on that language had been done by Ladefoged, who wasn't even mentioned in the abstract. There will always be errors. What can you do? You can try to constitute mature and fair program committees (the one for the upcoming FG-HPSG-CG conference looks pretty good to me). You can hope that the program committee has clearly accepted the responsibility of making the decisions, taking the reviewers' comments merely as advisory. You can try to write crystal clear abstracts, using available examples of accepted abstracts as models. But you shouldn't assume that whole communities of reviewers are out to get you. As I've said before in postings to this list, there are people of every theoretical stripe who feel that their abstracts have been treated unfairly, e.g. GBists whose WCCFL abstract wasn't in the 10 percent of abstracts accepted. The Standing Committee engaged in a long discussion about venue for this upcoming HPSG conference (I was on sabbatical and out of the loop for most of it) and there were two sites that had offered to host HPSG-98 (Groningen and Saarbruecken), which takes considerable financial and logistical resources, by the way. The idea of trying to build better communication between the HPSG and Categorial Grammar communities has a long history, going back to the 1985 Categorial Grammar conference in Tucson, continuing with a number of events that Dick Oehrle, Michael Moortgat and I have arranged over the years. I sincerely hope that in Saarbruecken there will be every opportunity for HPSG researchers to communicate with one another (in whatever gory feature structure detail pleases them) as well as with CG people and others, including LFG people, in whatever terms are possible. I see this kind of outreach as critical for the field. As long as all the various constraint-based grammar communities are isolated from one another, they will appear to people entering the field of linguistics as nothing more than splinter groups of minor consequence when compared with the GB/P&P/MP empire which manufactures most of the consent in the world of syntax. Even in Computational Linguistics, a similar point could be made, given the diversity of alternative approaches to NLP. So, I want to encourage you to think more positively about this whole upcoming event. Our positive attitude can really make a difference. Now the issues about democracy in our community are worth addressing independently. In 1995, a group of us started this informal standing committee to manage continuity of the HPSG conference. That committee has people rotating off it soon and it might be reasonable to ask people to suggest nominations for positions on the standing committee. Such suggestions should be sent to John Nerbonne, I believe (correct me if I'm wrong, John). Perhaps the standing committee should consult this list as it considers alternative venues for future meetings (it's time to pin down HPSG-99 now, for example). There is also the possibility of creating an official HPSG society with a `constitution', if people think that would be useful. I originially thought that was overkill, given that not even WCCFL, NELS etc. have such organizations, but maybe we need one. We should certainly do whatever we can to dispel the impression that people are being excluded from decision processes. Reactions? All Best, Ivan ------------------------------------------------------------------------- Ivan A. Sag Professor of Linguistics and Symbolic Systems Email: sag@csli.stanford.edu WWW: http://hpsg.stanford.edu/hpsg/sag.html Dept. of Linguistics CSLI - Ventura Hall Fax: 650-723-5666 Fax: 650-725-2166 Office: MJH 040B Office: Cordura 228 Phone: 650-725-2323 Phone: 650-723-2876 ------------------------------------------------------------------------- Stanford University - Stanford, CA 94305 USA -------------------------------------------------------------------------
This archive was generated by hypermail 2.0b3 on Fri Dec 18 1998 - 20:36:10 PST