Linguists at the soup kitchen; are there objective physical definitions?

Do linguists have a responsibility for “advocacy” and “empowerment” as Rickford argues? He claims that African-Americans, for example, have contributed greatly to linguistics research on analysis of narratives and other speech events, diachronic issues, and analysis of social class, ethnicity, network, and style in African-American vernacular English (AAVE). In return, linguists have a responsibility to use linguistics to improve people's lives: to improve elementary education by pushing for use of textbooks like the Bridge textbooks which use ebonics along with standard English and have correlated to higher English performance among speakers of AAVE than traditional English textbooks, or to train and employ more African-American linguists. He also thinks linguists should become involved in service for service's sake, such as tutoring, running a book drive, helping teenagers apply for college, etc.

I don't think it makes sense for linguists to focus attention on traditional community service efforts; if they want to effect change they should research best educational practices for learning languages, most effective methods for speakers of vernacular like AAVE to overcome bias in hiring or education, and other linguistic questions. Leave the community service to others; linguists have a comparative advantage in studying linguistics so should focus their efforts there. It could be more effective for for students to do things like community service in linguistics classes (as Rickford recommends); trained linguists should probably focus more on specialized and technical efforts. A counterargument is that linguists should participate in community service for the African-American community because their discipline has benefitted so much from studying AAVE. But there are more effective ways to improve African-American lives that draw directly on their expertise; they can directly impact policy through data. Researchers can also help by following the guidelines in their research: openly share their goals with their research subjects, ask questions generated by them, and present the research in a way so that their subjects can understand it.

The Lupyan reading explored the question of whether humans have one single canonical representation for abstract concepts like the triangle. He found through different experiments that different verbal prompts like “triangle” or “three-sided polygon” elicit different levels of recognition of triangles. More typical descriptions like “triangle” elicited stronger typicality effects than “three-sided polygon.” Also, humans seemed to view certain triangles as more correct than others: they were most likely to draw equilateral or isosceles triangles with horizontal bases when asked to draw a triangle. One of Lupyan's most provocative conclusions is that the “representation” of a concept for a human is the representation that best distinguishes between the category and other similar, commonly confused categories.

This reading reminded me of the brief study of philosophy of language in Phil 80 last quarter. George Berkeley took this problem to extremes, concluding that since everyone sees different aspects of a triangle and triangles are slightly different every time someone views them, then there is no objective external definition of a triangle. There are no external things; everything is a subjective mental perception. The Lupyan experiments don't necessarily dispute this argument because they found that people had very different, sometimes subconscious, internal definitions of a triangle. Berkeley is very hard to refute!