The language­geography connection

By BRENT CUNNINGHAM
Reporter Staff

When is a "hill" a "mountain"? When does a road "cross" a park? When is a "pond" in English not a "pond" in French?

They may sound like riddles, but for David Mark these questions point to a potentially fertile area for geographic studies. A professor in the Department of Geography and director of the National Center for Geographic Information and Analysis, Mark has spent the last 10 years investigating how language and meaning relate to our understanding of geographic features and spatial relations.

Often thought of as the study of rivers, capitals, resources and population, "descriptive" geography of this sort, Mark said, is only one approach among many. He noted that, due in part to the rise of computers in the 1960s, geographic conceptions of space and information have changed and expanded, crossing into computer science, social theory and the cognitive sciences. At the same time, these fields have brought their own concerns to geography.

"It doesn't make sense," said Mark, "to study the physical environment or societies without considering the geographic perspective. For example, almost all social and environmental processes are attenuated by distance-the further apart things are, the less effect they can have on each other."

But geography and language? Geography and philosophy?

"Not many geographers," Mark admitted, "are using the grammar, vocabulary and usage of language as a tool for teasing apart the factors that influence people's spatial behavior and spatial reasoning. But I have trouble coming up with experiments that don't use language."

According to Mark, comparative linguistics can reveal a great deal about how people think about space and geographic categories. In French, for example, the word for "pond" is "etang." "It's defined exactly the way 'pond' is defined in English," said Mark, "and most etang are ponds, but there are things that are etang in French that would not be ponds in English. 'Etang' seems to focus more on the water quality."

While his findings remain tenuous, Mark said early experiments point to the possibility that geographic categories are not "natural" but "constructed" out of the relative position of the people who name them. But this raises another question: do people speaking different languages perceive the feature differently or do they only define it differently? The simple geographic question-"what is a pond?"-spreads rapidly into philosophy, social theory and the cognitive sciences.

"To what extent is the geographic reality socially constructed? asked Mark. "To what extent is it cognitively constructed, and to what extent is it objectively, physically 'out there'?"

At the same time that Mark is finding evidence that the "what" of geography may be relative, other experiments with the "where" of geography seem to be yielding somewhat different results. Early research into spatial relations seems to indicate that concepts of relative position may be less subject to cultural and linguistic variation, more universal, than geographic categories like "pond" or "sea."

Using simple drawings of an area (a "park") and a line (a "road") Mark and Max J. Egenhofer from the University of Maine tried to pinpoint when subjects might describe the road as "crossing" the park, when they might describe the road as going "into" the park, and so on.

Working with native speakers of English, French, Spanish, German, Chinese and Hindi, Mark said he was "surprised at how similar the concepts seem to be across the languages, which then raises the question of where these concepts came from. I would presume somehow from experience with the world. The similarities are so strong that it seems it must be-I think-somehow prelinguistic, or somehow related to perception.

"The evidence right now," he added, "is that there are six spatial relations between lines and regions: entirely inside, entirely outside, entering, crossing, tangent but not touching at an end and tangent touching at an end.

"But there's a lot of work that needs to be done," he cautioned.

If research discovers whether spatial relations do or do not have connections to a prelinguistic universality, what sort of applications would it have? "I got into some of this from wondering about software design," Mark said. "Can you just take geographic technologies that have been developed in an English-speaking culture, translate the command names and have it make sense to people in other cultures?

"It seems like cultural imperialism to just assume that, of course, everyone can use the same tools," Mark noted. "I thought we should assume that the spatial relations might be different. I was kind of expecting that I would find differences, but...it's starting to suggest that maybe we can just change the commands, at least for the languages we've looked at."

While these kinds of issues did exist in the "analog" world, when geographers and cartographers used overlays and other methods to represent spatial relations, Mark said they have become more central to geography since the rise of computer graphics in the 1960s. Geographic Information Systems (GIS), which generally refers to all "digital" systems for presenting and interpreting geographic information, now is considered a distinct, if amorphous, field. While it continues to have strong links to geography, its concerns cut across the disciplines, pulling heavily from psychology and computer science.

Currently, Mark is helping lead a team of geographers studying the history of GIS. "We want to look at how, where, when and why the actual software systems were developed," he said. "We also want to investigate how much of GIS is a legacy of cartography. The thought experiment is: what would happen if some culture invented GIS but had never made graphic maps? Would GIS be different? Is GIS mostly still automated mapping?"

The history of GIS, Mark noted, is contentious, even today. Some people, for example, argue that GIS was "invented," others that it was "discovered."

"One of the GIS innovators thinks that he just discovered it," said Mark, "that it was already there in the nature of mathematics and the world." Some, he said, would contend that mathematics is itself a human creation devoid of objective "truth."

It is a debate that can easily get caught up in a "sort of humanist, postmodern" political critique, Mark noted. If scientists say they "discover" something already there, is it a denial of their own cultural biases and political agendas? Or, could the assertion that scientists "invent" be a way for them to ignore disagreeable data or politics?

While Mark said he considers these questions valuable for GIS and geography in general, he said his background and instincts are better suited to the scientific method than to broad theorizing. "I really enjoy speculating and introspection," he said, "but before I'm really going to claim something, I'm typically going to want to do an experiment."