In the second of a series of interviews, Barry Smith discusses his COSIT'95 paper, and provides a personal and thought-provoking perspective on COSIT. You can download a copy of Barry's COSIT'95 paper, now one of the most often cited COSIT papers, using the link on the right of this page.
Can you summarize the key ideas/results in the paper?
Barry Smith: The paper introduces the opposition between fiat and bona fide boundaries. The former are human-demarcation-induced boundaries -- for example between Montana and the rest of the American continent, or between your arm and the rest of your body. The latter are boundaries corresponding to genuine physical discontinuities, which would exist even in the absence of all drawing of boundaries by human beings or other cognitive agents. This distinction between two types of boundaries generates a categorization of different sorts of objects (fiat objects, bona fide objects, and various mixed types). The fiat/bona fide opposition is applied first of all in the areas of geography and of administrative and property law and it grew out of my initial work in these areas. However, once the distinction is made, then it can be used to suggest solutions to a number of philosophical problems, for example pertaining to the distinction between natural categories such as dog or cat, and human demarcation-induced categories such as red and green.
What was the main motivation behind writing this particular paper?
Barry Smith: There were two motivations. One was the desire to do justice to the ways in which, in using language, we demarcate (carve out) portions of reality in complicated ways. For many years I had been attempting to use ideas derived from mereology and topology to do justice to this phenomenon (with the ultimate goal of producing a semantics for natural language based on mereotopology rather than on set theory). My thinking in this respect was influenced by Len Talmy, and also by a philosopher by the name of Johannes Daubert, an early follower of Edmund Husserl, who classified the different sorts of demarcation phenomena effected by different sorts of use of language, and thereby anticipated some elements of what later came to be called the theory of speech acts. The second motivation for this work were the various geopolitical events taking place around that time: the collapse of Yugoslavia, the unification of Germany, the first Gulf War, and the idea of Sir Percy Craddock and other heroes of the British Empire sitting in London who, by drawing lines on maps, were able to bring entire countries into being by sheer fiat.
Why did you submit this paper to COSIT (rather than another conference)?
Barry Smith: Since around 1994 I had been aware of the existence of David Mark in Buffalo as someone who was interested in the interplay of mereotopology and cognition; David and I began to collaborate intensively only later, but he had already in 1994 awakened my curiosity as regards the COSIT community. Given the fact that the meeting that year was to be held in Austria (one of my main philosophical interests had for a long time been in the area of Austrian philosophy) this seemed like a good opportunity to get to know the people involved.
What are your memories of the initial response to the paper when presented at the COSIT conference?
Barry Smith: Actually I did not read the paper at the meeting, relying on the fact that it had been available to all the participants in the published proceedings. Instead, I presented an off-the-cuff lecture entitled "Performative Maps," which consisted in no small part in an attack on the philosophical ideas of George Lakoff. George has of course made many important contributions to linguistics, but over and over again he and his followers have advanced philosophical claims (for example about the existence of mind-independent universals, or about the nature of mind-independent reality) which they support with arguments which have long ago been refuted by philosophers.
What feedback did you get during the conference?
Barry Smith: Am I allowed to say that I won the prize, that year, for best presentation? I think this was in part because I was the first (and still almost the only) philosopher to have become involved in the work of COSIT. But it was also because many people in the audience hated what I said -- but this meant also that it stuck in their minds, and started them thinking (so that I hope that I can claim that my campaign against the influence of Lakoff has been successful).
How did the feedback you received at the COSIT conference affect the development of the work?
Barry Smith: It got me involved, first of all, with various efforts in cadastral and legal ontology. In the following year I presented at another conference in Austria a paper entitled "The Cognitive Geometry of War", which attempted to demonstrate that many wars and ethnic conflicts through the ages have come about because people think that there are such things as natural borders, i.e. because they do not take sufficiently seriously the role of fiat boundaries in structuring geopolitical reality. (The first Gulf War is an example of this phenomenon, as also are the Israel-Palestine and the Irish conflicts; not however the second Gulf War.)
What collaborations developed with other COSIT participants as a result of the presentation?
Barry Smith: David Mark and I began to collaborate intensively in 1997 and we have published a series of papers on geographic ontology since then (see http://www.geog.buffalo.edu/ncgia/ontology/). Many other collaborations have ensued, including current cooperation between IFOMIS, the Medical Ontology Institute which I have founded in Leipzig, and the Qualitative Spatial Reasoning Group in Leeds.
How has the research presented in the COSIT paper developed since the COSIT conference?
Barry Smith: David Mark and I received NSF funding for our work in ontology, and my collaboration with David and with the wider COSIT community played an important role in my receiving the Wolfgang Paul Award from the Humboldt Foundation in 2001. In a number of papers since 1995 I have tried to show how the basis fiat/bona fide idea can be applied in a variety of fields where physical and fiat boundaries are at work, including the field of cognitive linguistics, ecology, and ecological psychology. I also developed with Achille Varzi a new way of approaching mereotopology, according to which two distinct axiomatizations are required to deal with the two different sorts of boundaries phenomena. The mereotopology needed by fiat boundaries is non-standard -- it does not recognize the opposition between closed and open objects. Originally we had assumed that the mereotopology appropriate for bona fide boundaries would be structurally equivalent to classical point-set topology, and that it would be independent of fiat mereotopology; we discovered, however, that bona fide mereotopology presupposes fiat mereotopology in complicated ways, so that there is no way of developing the two theories independently of each other. See http://ontology.buffalo.edu/smith/articles/fiatvs.html
What is the most important feature of COSIT conference series for you?
Barry Smith: My main impression of the meeting -- which was the first meeting I attended in which geographers played an important role -- was how nice the participants were. Philosophers, of course, adopt a much more critical (not to say savage) attitude to what they hear at philosophical meetings. (And when one looks at what has happened to geography because so few people were willing to criticize the deconstructionist acid which is eating away at the discipline, then one cannot help but conclude that philosophers have a point.)