How much spatial information is represented in language? To what extent do languages differ in the expression of geometrical and functional object structure? Is there variation in the role the human and animal body plays as a conceptual model of the structure of objects across languages? Do speakers of all languages employ the same conceptual processes in mapping the structure of the body into that of objects? Does the way the geometrical and functional structure of objects is conceptualized in different languages influence the way spatial relations are identified in these languages? And does the way speakers of different languages talk about spatial relations influence the way they memorize them? This project attempts to find answers to these and similar questions, based on an investigation of the representation of space in 13 indigenous languages of Guatemala and Mexico.
A first in kind, this set of studies seeks to apply methods of semantic typology to elucidate the correlations between two typologically unusual traits of spatial language in Mesoamerica and their possible diffusion through language contact. These are the highly productive use of meronymic (part-whole) terminologies for object parts and spatial regions based primarily on object geometry, and the striking preference for allocentric (non-observer-based, i.e. intrinsic or absolute) over egocentric (observer-based) frames of reference. The recent demonstration of a close alignment in frame-of-reference selection for linguistic encoding, recall memory, and spatial reasoning has triggered a debate over the relative importance of linguistic and cultural factors in frame-of-reference choice. This study aims to advance this debate, capitalizing on the ecological diversity of the Mesoamerican area, cultural factors including modes of production, bilingualism, and education, and the possibility of a linguistic predictor: highly productive meronym systems.
We investigated the role of language and culture in cognition in an area of thought that until the 1970s was considered to be a matter purely of biology and individual variation across people: spatial orientation. When Americans, Europeans, or Mainland Japanese people think or talk about objects in small-scale space - say, the objects on the desk in front of them - they customarily (though unconsciously) use a kind of mental coordinate system - a so-called 'frame of reference' - that is derived from their own body or that of another person. For example, I might say that my teacup is right of my laptop, and by that I mean most likely that it is on that side of my computer that is closest to the right side of my body. Such a reference frame is called 'relative'. We also use 'geocentric' frames, which are based on the environment, for example, when working with cardinal directions. But we use geocentric frames only for large-scale space - say, whem talking about different parts of a city. Only in the last few decades has it become known that there are human populations that make little or no use of relative frames, and some of these in fact use predominantly geocentric frames even at the small scale ('The teacup is east of the laptop'). And members of such populations use geocentric frames not only for talking about objects in space, but also for thinking about them. For example, they will memorize the relative positions of an assemblage of objects vis-a-vis the environment and, when asked to reproduce it, try to preverve its orientation vis-a-vis the environment. In contrast, Europeans and Americans will memorize the configuration with respect to their own body. Two competing interpretations of this finding have been proposed. A nativist view holds that everyone is innately able to use any reference frame, and the observed variation is the result of high levels of literacy and education boosting the use of relative frames, whereas the presence of salient landmarks in the local environment boosts the use of geocentric frames especially in small, tight-nit communities. Proponents assume these factors to affect each person individually, so members of the same communities use similar reference frames only because they tend to share similar levels of literacy and education and the same local environment. The alternative view holds that reference frame use is something we primarily learn from our peers, by observing their behavior. And the way our peers talk and, for example, gesture about space is thought to provide us with important cues. The use of relative or geocentric frames becomes inculcated in young children when they observe adults and older children regularly use these frames in their speech. These proposed factors are not independent of one another. Different languages are often spoken in different environments and by speakers with different average levels of education and literacy. However, sophisticated statistical tools that have only recently become available to cognitice scientists can help break the impasse. We investigated how speakers of eight indigenous languages of Mexico and Nicaragua talk about the location and orientation of objects in space and memorize this information. Six of the languages belong to the Mesoamerican linguistic and cultural area, the decendents of the ancient civilizations of the Olmecs, Mayas, and Aztecs. These languages have influenced one another for millennia. The dominant language of the region is Spanish, and most speakers of the indigenous languages are bilingual in it. We assessed how frequently the indigenous participants use Spanish in their everyday lives and also collected data from monolingual Spanish speakers of Mexico, Nicaragua, and (for comparison) Europe. We found that all the proposed factors indeed influence the use of reference frames, with the apparent exception of education. However, our statistical models show that the role the first language plays cannot be reduced to any combination of the other factors. We were also able to show that the indigenous participants were the more likely to use relative frames in their native languages the more frequently they use Spanish as a second language. We also found significant differences between the Mesoamericans and the non-Mesoamerican indigenous participants. All of these findings suggest that language can indeed play the role of a vehicle in the cultural transmission of particular styles of thinking - here, the use of particlar reference frames. Intriguingly, we found a preference for geocentric memory in populations who prefered geocentric frames for talking about space, but also in populations who preferred a third strategy or showed no clear linguistic preference. Previous research with nonhuman primates has produced evidence of a geocentric bias. We think that this bias may be innate in humans as well, but may become overridden by a culturally learned preference for relative frames in those populations that have it.
