Linguistic Anthropology

The study of language has been part of anthropology since the discipline started in the 1ate 1870s. This site is a place for linguistic anthropologists to post their work and discuss important events and trends in the field.

Monday, April 06, 2009

Respect for B.L. Whorf?

Benjamin Lee Whorf doesn't get much respect from linguists, at least in certain quarters. His notions of linguistic relativism - or just as often, notions attributed to him after the fact - are frequently refuted in popular texts on language and linguistics. Geoffrey Pullum's The Great Eskimo Vocabulary Hoax or Steven Pinker's The Language Instinct, for example, seem to take glee in demolishing both relativism and Mr. Whorf (though Pullum, at least, admits "Whorf has a lasting place in the history of linguistics, a place few of us can aspire to. He is basically responsible for opening up our access to an entire language [classical Mayan] that had previously been inaccessible" (1991:160)). As George Lakoff says, "[Relativism] has become a bête noir, identified with scholarly irresponsibility, fuzzy thinking, lack of rigor, and even immorality. Disavowals and disproofs are de rigeur—even I felt obliged to cite the standard disproof of 'total relativism'," and Whorf is "the most celebrated relativist of this [twentieth] century" (1987:304).

I was therefore pleasantly surprised to come across an empirical treatment of linguistic relativism, and then a kind word for Mr. Whorf himself, in two different outlets this morning.

First, on NPR's Morning Edition, Robert Krulwich talked with psychologist Lera Boroditsky about some apparent effects of grammatical gender. According to the radio piece (I don't know the research it was based on) a group of German speakers, when asked to describe a picture of a bridge, used words such as beautiful, elegant, and slender. A group of Spanish speakers described the same picture using words such as strong, sturdy, and towering. Boroditsky says, "[We] asked German and Spanish speakers to describe objects having opposite gender assignment in those two languages. The descriptions they gave differed in a way predicted by grammatical gender." The feminine die brucke may be seen as more "feminine" than (masculine) el puente. Even though the referent - the object in the world that the words point to - is the same, a grammatical difference may cause speakers of different languages to tend to see the world differently.

Also this morning, I read a piece on Language Log by Mark Liberman called Betting on the poor boy: Whorf strikes back (though it was actually posted yesterday). In it, Liberman challenges a suggestion in The Economist magazine that "the working memories of children who have been raised in poverty have smaller capacities than those of middle-class children." The Economist piece is based on a study by Farah et al. (2006) in which a cohort of African American girls from middle socioeconomic status was compared to a cohort of African American girls from lower socioeconomic status on a number of cognitive tests.

Liberman suggests that the language used in The Economist seems to suggest that one group of girls performed well and the other group performed badly. This actually misstates the facts: the performance of the two groups overlaps extensively, with small but significant differences in some between-group averages.

Here's where Whorf comes into it. Liberman concludes:
[Statistics and effect sizes] are not effectively taught or widely learned, even among quantitatively-minded intellectuals. But I also think that there's a linguistic aspect. If Benjamin Lee Whorf were alive, he might argue that our whole society is intellectually hamstrung by the way that English — like all the other languages of the world — tends to make us think about the evaluation and comparison of the properties of members of groups. And, I think, he might be right.
Our language, like most all languages, makes it relatively easy to talk about "differences between groups" in ways that make us think that the members of the groups differ in the same way, even when the groups overlap.

Labels: ,