Provides an in-depth look at the issues of spatial relationships and how to accurately describe them in natural language.

In January of 1989, the National Center for Geographic Information and Analysis (NCGIA) organized and hosted a workshopentitled "Languages of Spatial Relations". This meeting was the "Specialist Meeting" for NCGIA Research Initiative #2, which has the same title as the workshop. The research mission of the NCGIA involves removing impediments to the analysis and use of geographic information in theoretical and applied research; one group of conceptual and technical impediments centers around language andlanguages.

The Specialist Meeting brought together geographers, cognitive linguists, computer scientists, engineers, and others in threedays of interdisciplinary discussions on the representation of geographic space-- representation in language, in thought, and in computers. One theme of the meeting was that geographic information systems (GIS) must have at least some capabilities to deal with natural language.

One of the objectives of the Specialist Meeting was to compile a list of "researchable questions" related to the topic, to be included in the report on the meeting. These questions were to emphasize problems that could be tackled within a 12-24 month period, by Center personnel, by other Specialist Meeting participants, and/or by the research community at large.

This report describes the results of NCGIA Research Initiative 2 "Languages of Spatial Relations". It compares the background and objectives of the initiative as originally planned with the understanding of the problem following the work carried out during the initiative. It then assesses the work in detail against five criteria: 1. Research Accomplished; 2. Research Agenda Development; 3. Contribution to GIS Education; 4. Implications for Science Policy; and 5. Comments on the Research Initiative Process. These criteria are applied to the initiative's research objectives to:

1)  Identify formal cognitive/semantic models of spatial concepts and relations in natural language;

2) Develop reliable methods for determining reference frames for spatial language;

3) Construct formal mathematical/logical models of spatial concepts and relations based on topology and geometry; and

4) Integrate the two kinds of formal models into a general theory of spatial relations.

This paper is a part of Research Initiative 10, "Spatio-Temporal Reasoning in GIS," of the U.S. National Center for Geographic Information and Analysis (NCGIA). 

This report contains two papers, plus supplementary material. The first paper develops and presents the formal mathematical definitions of the 9-intersection; it is under consideration by a journal at the time of this writing. The second paper reports of cognitive testing, based on the mathematical model presented in the first paper. It contains the complete set of stimuli used in Mark and Egenhofer’s experimental work up to January 1994.

The great majority of existing geographic information systems have been designed by English or German speakers. Since human natural languages impose structure on the cognition and perception of space, time, and other concepts, GIS data models, and especially GIS query languages and human interfaces, can be expected to contain artifacts of the language spoken by their designers, most commonly English. At a practical level, natural language studies, particularly those based on cognitive linguistics, are important issues for the designers of geographical information systems, because GISs of the future should be able to handle natural language in a number of situations. We propose that the design of any natural language modules for GIS should be based on language universals and language primitives derived from cross-linguistic studies of geographic concepts and relations. Otherwise, it will be even more difficult to transfer English-designed systems into the many other languages that can be expected to make up an ever-increasing proportion of GIS markets in the future. This paper expands on the general principles of cognitive linguistics mentioned above, with emphasis on cross-linguistic issues. Then, we will present a more detailed study of the primitive geographic relations represented in most Indo-European languages by prepositions. We will concentrate on differences between English and Spanish, but also draw on other languages.

La gran mayoriá de los sistemas de información geográficos existentes han sido diseñbados por hablantes de lengua inglesa o alemana. Como los lenguaJes humanos naturales imponen una estructura en la cognición y percepción del espacio, tiempo y otros conceptos, cabe esperar que los modelos de datos y especialmente los lenguajes de consulta de los SIGs y el interface del usuario, contengan artefactos de la lengua hablada por sus disefiadores, el inglés en la mayoriá de los casos. A nivel practico los estudios sobre lenguajes naturales, particularmente esos basados en linguística cognitiva, son temas importantes para los diseñadores de sistemas de información geográficos porque los SIGs del futuro deberian de ser capaces de manejar lenguajes naturales en diversas situaciones. El diseño de cualquier módulo de lenguaje natural para SIGs deberia de estar basado en lenguas universales y lenguas primitivas derivadas de estudios interlinguísticos sobre conceptos y relaciones geográficos. De otro modo, será incluso más dificil transferir los sistemas diseflados en inglos a las muchas lenguas en las que se espera una proporción creciente de mercados para los SIGs en el futuro. Este trabajo desarrolla los principios generales de la linguística cognitiva mencionada anteriormente, con énfasis en temas interlinguísticos. A continuación presentaremos un estudio más detallado de las relaciones geográficas primitivas representada por preposiciones en la mayoría de las lenguas europeas. Nos concentraremos en las diferencias entre el inglés y el español, pero tambien usaremos otras lenguas.

This report describes the outcome of the Specialist Meeting of the NCGIA Research Initiative 21 on "Formal Models of Commonsense Geographic Worlds". The meeting was held in San Marcos, TX on October 30- November 3, 1996.

Research Initiative 21 is concerned with the development of formal models of commonsense geographic worlds. Discussions at the Specialist Meeting focused on the commonsense or naive geographic reasoning that people perform and whose outcome makes intuitive sense to most people. The Specialist Meeting brought together specialists from geographic information science, artificial intelligence, computer science, geography, developmental psychology, and behavioral science to foster discussions leading towards a better understanding of the nature of naive geographic reasoning and how better to incorporate naive geographic knowledge and reasoning into GIS.

This Report serves to document the discussions held during the meeting and, most importantly, delivers a set of researchable questions that forms the basis for future research in this area. Participants collected close to 50 questions worthy of further consideration and research on topics relating to the fundamentals of Naive Geography, developmental influences on Naive Geography, and the impact of Naive Geography on GIS.