Shakespeare, or his Juliet, might have thought that a rose by any name would smell as sweet. But, for years, social scientists have been engaged in lively debate about whether different cultures, and their different languages, typically name objects and ideas in ways that are universal among peoples, or whether, on the other hand, language reflects each culture's uniquely constructed perspective on the world.

A compelling series of studies have begun to lend support to the more universal view - that despite differences in cultures, languages indeed seem to reflect a closely related set of core perceptions that are strikingly similar across cultural boundaries. Recent studies suggest that this similarity occurs even with words that reflect such abstract concepts as emotions.

These studies were possible thanks to the collaboration between two NSF sponsored research projects being carried out at the University of California, Irvine. The first, directed by Carmella C. Moore, is focused on the substantive question of the universality of semantic structure and funded by the Cultural Anthropology Program. The second, co-directed by A. Kimball Romney and William H. Batchelder, is focused on the development of statistical tools and models for social measurement and funded by the Methodology, Measurement and Statistics Program.

Past studies by Moore and Romney, and their colleagues, have shown that various cultures and their languages appear to perceive concrete items, such as colors or animals, in similar ways. More recently, the study of emotions provided a team of social scientists led by Moore and Romney with an ideal springboard for looking at less concrete words as cognitive representations across cultures.

The research team's central question was this: to what degree are perceptions about emotions similar among widely divergent cultures? Or, contrarily, do each culture's beliefs, practices, and experiences shape its perceptions and language so uniquely that many words are not actually describing similar experiences? A rose may be a rose in America or Japan, but do the words describing what an American experiences as joy, or shame, have direct corollaries in Japanese or Chinese, or other languages?

To compare semantic representations of emotions, researchers first assembled a list of 15 common, frequently used emotional terms in both English and Japanese: anger-haragaatsu, envy-uraymashi, happy-ureshii, and others. Separate groups of exclusively English or exclusively Japanese speaking college students were then asked to judge the similarity in meaning among the words. Using new multidimensional scaling methods the researchers constructed a spatial representation of the semantic structure of the words. In this representation the emotion terms that were judged similar in meaning are closer to each other than those that were judged dissimilar. The first dimension that emerged from the analysis was pleasant-unpleasant; for example, happy is pleasant while sad or hate is unpleasant. The second dimension was active versus passive; for example, angry would be a more active term, bored would be a more passive term.

After plotting each individual's response onto a two-dimensional graph, the researchers were able to compare the group's responses statistically. The results suggested that, in general, emotional terms appear to be perceived in remarkably similar ways across the two cultures. Most terms ended up plotted in strikingly similar patterns. Still, there were some notable exceptions. For instance, wrote the researchers, "the measured location of the English word shame is more extreme on the "unpleasant" dimension than that of the Japanese counterpart hazukashii. We think that this difference reflects a genuine difference in the meanings of the two words rather than being an accident of measurement or sampling variability." Two other terms, anxious and bored, also showed quite large differences.

If, as this research suggests, semantic domains are more universal than relative, it might mean that specific regions of the human brain are "hard wired" for the perception and expression of commonly shared concepts. Among other issues, learning more about this could lead to an improved understanding of how an individual's cognition fails in such diseases as Alzheimer's.

Importantly, the Moore and Romney team, which includes Dr. Craig D. Rusch and research assistant Ti-Lien Hsia, have provided tools for measuring, with known accuracy, the extent to which "pictures" or cognitive representations in the mind of one person correspond to those in the mind of another. These new methods go beyond a polarized approach to examining cultural universals vs. culture relativity (e.g. "there are no similarities in the semantic structure of said languages" or "all languages have the same semantic structure,") and make it possible to measure the extent of the similarities and differences in the semantic structure of different languages.