LINGUIST List 6.872

Mon 26 Jun 1995

Confs: Conceptual Structures (ICCS95), Evolution of Lg

Editor for this issue: <>


  • Gerard Ellis, International Conference on Conceptual Structures ICCS95 (programme)
  • Jim Hurford, Evolution of Language

    Message 1: International Conference on Conceptual Structures ICCS95 (programme)

    Date: Sun, 18 Jun 1995 16:16:46 International Conference on Conceptual Structures ICCS95 (programme)
    From: Gerard Ellis <>
    Subject: International Conference on Conceptual Structures ICCS95 (programme)

    ADVANCE PROGRAMME 3rd International Conference on Conceptual Structures August 14-18, 1995 University of California, Santa Cruz

    Sponsored by: IBM, Santa Teresa Laboratory, San Jose. University of California at Santa Cruz. Royal Melbourne University of Technology, Australia. American Association of Artificial Intelligence AAAI.

    THEME Conceptual structures are a modern treatment of Charles Sanders Peirce's Existential Graphs which are a graphic notation for classical logic with higher order extensions developed in 1896. Peirce viewed existential graphs as ``his luckiest discovery'' and ``a logic of the future''.

    John Sowa showed that conceptual graphs can be mapped to classical predicate calculus or order sorted logic, and are thus seen as a (graphic) notation for logic. However, it is the topological nature of formulas (topology was a field Peirce helped develop) which conceptual graphs make clear, and which can be exploited in reasoning and processing. Conceptual graphs are intuitive because they allow humans to exploit their powerful pattern matching abilities to a larger extent than does the classical notation. Conceptual graphs can be viewed as an attempt to build a unified modelling language and reasoning tool. Conceptual graphs can model data, functional and dynamic aspects of systems. They form a unified diagrammatic tool which can integrate Entity-Relationship diagrams, Finite State Machines, Petri Nets, and Dataflow diagrams.

    ICCS95 home page URL:


    Program Chair Local Arrangements Chair Gerard Ellis Robert Levinson Royal Melbourne Univ of Technology Univ of California, Santa Cruz Australia USA

    Finance Chair Honorary Chair Bill Rich John Sowa IBM San Jose, California State University of New York USA USA


    Hassan Ait-Kaci (Canada) Dickson Lukose (Australia) Harmen van den Berg (Netherlands) Craig McDonald (Australia) Duane Boning (USA) Guy Mineau (Canada) Boris Carbonneill (France) Jens-Uwe Moeller (Germany) Michel Chein (France) Bernard Moulin (Canada) Key Sun Choi (Korea) Marie Laure Mugnier (France) Peter Creasy (Australia) Jonathan Oh (USA) Walling Cyre (USA) Heike Petermann (Germany) Harry Delugach (USA) Heather Pfeiffer (USA) Judy Dick (USA) James Slagle (USA) Peter Eklund (Australia) Bill Tepfenhart (USA) Bruno Emond (Canada) Eileen Way (USA) Norman Foo (Australia) Michel Wermelinger (Portugal) Brian Gaines (Canada) Mark Willems (Netherlands) Adil Kabbaj (Canada) Walter Wilson (USA) Fritz Lehmann (USA) Vilas Wuwongse (Thailand)

    Auxiliary Reviewers Alex Bejan (USA) Tao Lin (Australia) Phil Kime (UK) Maurice Pagnucco (Australia)


    Syntax, Semantics, and Pragmatics of Contexts

    John F. Sowa SUNY Binghamton

    A proposed standard for conceptual graphs is being developed by ANSI Technical Committee X3T2. The standard is based on the common CG core that has been used and implemented in various projects since 1984. Some purists have insisted on perserving a minimal core that is simpler than the 1984 version and closer in spirit to C. S. Peirce's original existential graphs. Others have been trying to increase the expressive power of CGs to match the generalized quantifiers and referents of natural languages, but the various modifications and extensions have introduced incompatibilities. To accommodate both groups, the proposed CG standard will be based on a minimal core that has a simple mapping to predicate calculus and KIF. It will also have an extensibility mechanism for defining generalized quantifiers and referents in a controlled and systematic way. As far as possible, the 1984 syntax and features will be accommodated either by the core or by the extensibility mechanisms. But the extensions will also support some new representations that can be useful for NL semantics, object-oriented systems, and optimized computations.

    Graphical Logic

    Hassan Ait-Kaci Simon Fraser University

    This talk will discuss the graphical nature of structures used in knowledge representation and logic programming. It will draw from my work on the foundations of the LIFE programming language and will emphasize the graph-theoretic underpinnings of the notions of subsumption, approximation, and interpretation.

    Ontology Revision

    Norman Foo University of Sydney

    Knowledge systems are usually static in their ontlogical assumptions. To make them dynamic, there are several levels of theory change that can be considered. Change without the introduction of new concepts is subsumed under existing theories of belief revision. If new concepts have to be introduced, we are in the domain of ontology revision about which very little is known. This talk will briefly outline a widely accepted belief revision meta-logic, then move on to circumstances in which it can be shown that ontology revision is necessary. Connections will be made to classical results in recursion theory and the problem of theoretical terms. Type hierarchies will be used to illustrate some of the main points.

    A Triadic Approach to Formal Concept Analysis

    Fritz Lehmann (GRandAI Software, California) and Rudolf Wille (Technische Hochschule Darmstadt, Germany)

    Formal Concept Analysis, as it has been developed during the last fifteen years, is based on a dyadic understanding of a concept constituted by its extension and its intension. It starts with the primitive notion of a formal (or dyadic) context which combines a set of (formal) objects, a set of (formal) attributes and a binary relation between those sets indicating when an object has a certain attribute. A formal concept of such a dyadic context is constituted by its extension formed by objects of that context and by its intension formed by attributes of that context. The formal concepts of a given formal context always form, with respect to the subconcept- superconcept-relation, a complete lattice which is called the concept lattice of the context. In more than 100 projects of application in numerous areas of interest, concept lattices have been activated for analysing and exploring data and knowledge. Recently, those experiences (but also philosophical considerations based on Peirce's pragmatic philosophy) have suggested a triadic approach to formal concept analysis. It starts with the notion of a triadic context combining (formal) objects, attributes and conditions under which objects may have certain attributes. By the Basic Theorem of Triadic Concept Analysis, it has been clarified what kind of structures are formed by the triadic concepts of triadic contexts. The representation of those structures by triadic diagrams has also been studied. Examples of the conceptual analysis of triadic data sets may demonstrate the usefulness of the new approach.

    Tutorial: Introduction to Conceptual Graphs Gerard Ellis, RMIT

    In this tutorial we will introduce conceptual graphs as a graphic notation for first order logic. We will illustrate how to define concepts and relationships, and construct concept and relation hierarchies. We will introduce the generalization hierarchy over conceptual graphs and the canonical formation rules which are the foundation of conceptual graph theory. We will examine advanced uses of conceptual graphs for object modelling and object systems. We will show how Peirce's inference rules can be used for reasoning in conceptual graphs.

    Full details of the ICCS95 programme and the latest information regarding ICCS95 can be found on the World Wide Web under

    Message 2: Evolution of Language

    Date: Mon, 19 Jun 95 16:13:18 BSEvolution of Language
    From: Jim Hurford <>
    Subject: Evolution of Language

    Content-Length: 2423


    University of Edinburgh, April 1st - 4th 1996

    Organizing Committee: Prof. Jean Aitchison (Oxford University), Dr Chris Knight (University of East London), Prof. James R Hurford (University of Edinburgh).

    We are planning a conference rather tightly focussed around the following two issues (and their interrelationship);

    * Chronology of the spread of mankind over the planet, and its relationship to language. * The continuity/discontinuity of the language faculty with other human and nonhuman systems.

    Preliminary programme: This is far from settled, but the outline below is probably a good indication of what will be on offer. The titles below are all tentative, and based on preliminary negotiations with the speakers. There will be other speakers who submit abstracts, to be refereed by the organizing committee.

    * Chris Stringer, Natural History Museum, London, `Paleontological orientation'

    * Bjorn Lindblom, U. Stockholm, `Evolution of the human vocal tract'

    *Andrew Carstairs-McCarthy, U.Canterbury, N.Z., `The ``double articulation'' or ``duality of patterning'' in human language as an adaptation'

    * Frederick Newmeyer, U.Washington, `The drift from parsing principles to innate principles of grammar -- ``phylogenetic grammaticalization'' '

    * Derek Bickerton, U.Hawaii, `(Dis)continuity between ``protolanguage'' and language'

    * Ray Jackendoff, Brandeis U., `Formal parallels between language and other faculties', OR `Intermediate stages betwen ``protolanguage'' and language'

    * Robert Boyd, UCLA, `The interaction of biological and cultural evolution, with specific reference to language'

    * Johanna Nichols, U.C.Berkeley, `The origin and dispersal of human language'

    * Merritt Ruhlen, ex Stanford, `The linguistic and human family tree'

    * Leon Stassen, U.Nijmegen, `A language typologist's reflections on the debate on the global diffusion of languages'

    * Michael Studdert-Kennedy (Haskins Labs), Dan Dennett (Tufts) and other participants -- review of the conference and summing up.

    For further information contact: Professor James R Hurford, Department of Linguistics, University of Edinburgh, Adam Ferguson Building, 40 George Square, Edinburgh EH8 9LL, Scotland, UK.


    Further information will be sent out in late August or early September, 1995.