Language, Culture and Thought

Major Questions

A central question of anthropology is whether, and how, different human cultures incline their members toward different understandings of the world. Investigation into the role of language in these understandings is a key project of linguistic anthropology.

Language, Culture, and Thought

For nearly 150 years a theoretical framework known as linguistic relativity dominated linguistics and, in turn, linguistic anthropology. This can be summarized as the idea that human languages are extremely diverse and that different languages may be organized according to locally distinctive principles. Wilhelm von Humboldt argued that these differences produced distinctive worldviews. Boas made significant contributions to this line of thought. His 1889 essay, "On Alternating Sounds," shows how speakers of one language, even when highly trained, often cannot accurately distinguish the sounds of another; their hearing is organized in terms of the sound system of the native tongue. In his 1911 "Introduction" to the Handbook of North American Indian Languages, Boas emphasized that the characteristic constructions of one language might pick out conceptual distinctions that were neglected in others. In a famous and often mischaracterized example, Boas pointed out that four Inuit words, aput ("snow on the ground"), qana ("falling snow"), piqsirpoq ("drifting snow"), and qimuqsuq ("a snowdrift"), share no common root. In English, in contrast, all these ideas require constructions incorporating the element snow. As Boas was not a linguistic determinist, he felt that a language such as that of the Inuit could easily develop an encoding for the more abstract notion of snow if the need arose.

Benjamin Whorf, a student of Sapir who was originally trained as an engineer, determined that relatively covert, yet far-reaching, patterns of co-occurrence between linguistic forms could strongly influence "habitual modes of thought and behavior" (as opposed to careful reflection). In one of his best-known illustrations of this point, he showed how European languages treat terms for units of time in the same way they treat terms for countable physical entities such as stones and dogs, which, unlike days or years, can be experienced by human beings in the form of several simultaneous exemplars. In contrast, in Hopi, a Native American language spoken in Arizona, terms for units of time do not accept plural affixation, and they occur only with ordinal numbers. Thus Hopi permits "the third day" but not "three days." Whorf argued that these patterns resonated in cultural manifestations such as European clocks and calendars, which stand for collections of units of time, and Hopi rituals that aim to regulate temporal succession and cyclicity. Since language, in Whorf's view, was the more intricately organized variable in this system of resonances, he felt that its structure shaped habitual forms of thought and behavior, rather than vice versa. That is, his position was a mild form of linguistic determinism.

Dell Hymes, a student of Carl F. Voegelin, who was a student of Sapir, argued in a 1966 paper for "two types of linguistic relativity": to the differences of form discussed by Whorf should be added differences of function. The functions assigned to varieties of a single language, or to several languages in a multilingual repertoire, may vary cross-culturally. Functional relativity can be illustrated by the question of how a speaker's native language marks community membership. For some speakers of the so-called standard languages of European national units, such as French and Castilian Spanish, competence in the standard is a central manifestation of nationality and, indeed, of patriotism itself. Such speakers may strongly oppose usages that depart from the official standard (such as giving children names in foreign languages or using foreign words on signboards) and may see cultivation of the national language as a logical function of government. In contrast, among highly multilingual Native American groups in the Upper Vaupes basin in Brazil and Colombia, one language in an individual's repertoire, the so-called father language, signals membership in the father's lineage. However, an individual must marry outside the group that claims this father language, and prestige lies not in conformity to it but in good control of a large number of languages within the local system.

Since the early 1960s the theory of relativity of linguistic form has faced sharp challenges. Hymes pointed out that a single language within the multilingual repertoire of a native of the Upper Vaupes would be unlikely to have important influence on that speaker's worldview. Anthropological linguist Joseph Greenberg identified many "language universals," ranging from limits on the possible arrays of distinctive sounds to limits on diversity in word order. Noam Chomsky, a formal theoretical linguist, argued that the language capacity must include a substantial genetic component and that the goal of linguistic science should be to identify the elements of this "universal grammar." Anthropological linguists Brent Berlin and Paul Kay showed that the apparent unwieldy diversity of color-term systems could be organized in terms of a very few fundamental principles, such that once the number of "basic color terms" in a language is known, the color domains that they designate can be largely predicted. For instance, systems with three terms will label a domain of darkness, centered in black; a domain of lightness, centered in white; and a domain of hue, centered in red. Systems with five terms will retain the black and white labels and will label three hue-based domains: one centered in yellow, a second centered in red, and a third domain of cool hues centered in blue or green.

Despite such challenges, research within a modified relativistic framework continues in linguistic anthropology alongside work on universals. Even though scholars now recognize that the diversity of languages is limited, many believe that local realizations of universal principles often yield "Whorfian effects": naive conclusions about the nature of the world suggested by the categories of a language. Research has shown that Whorfian effects appear very early in the speech of children and can be identified even in technical and scientific discourse. For instance, Berlin and Kay's basic color terms theory may depend on a language-specific fact: that these two scholars, as speakers of English, use a word, color, that refers to both brightness and hue. Many languages do not have such a term. Experimental studies conducted in the field show that speakers of English and Yucatec (a Mayan language of Mexico) respond differently to stimuli involving number and substance, in ways that correlate closely with differences in how the two languages treat these dimensions of experience. Speakers of Navajo—a language requiring that mention of more-animate referents (such as adult humans) always precedes the mention of less animate ones (such as animals) in transitive sentences (sentences that express the action of one entity on another)—understand the world to be organized as a complex hierarchy of types of entities that vary in their potential for motion, ranging from holy beings with the power to create the world to inert logs. Furthermore, the pragmatic and rhetorical dimensions of language, as well as poetic and artistic uses, are not constrained by the same kinds of universal principles that limit variation in sentence construction. These can focus and intensify experience, producing locally distinctive forms of cultural knowledge. For instance, the Kuna of Panama, in a chant for the magical control of dangerous snakes, exploit the contrast between the rhyming verbal suffixes -mai ("horizontal position") and -nai ("hanging position") to suggest how the magician controls the serpent by raising it from a horizontal position on the ground into the air, where it "hangs" under his power.