I think highly of Andreas Wimmer’s project and consider him a friend. In some ways we may be talking past each other, or just have very different ideas about what race, political sociology, and indeed politics, mean. I am not sure whether Emirbayer and Desmond, Bonilla-Silva, or Feagin are responding to Wimmer’s editorial. I can only write here about his discussion of my work. I have argued for a much more contextualized and conscious approach to research on race. This makes measurement more difficult of course, but it at least attempts to understand what is being measured (‘construct validity’). Wimmer confronts the problem that at numerous points he has to ignore complexities in order to get to his data. The flexibility and the instability of the empirical phenomena are not addressed, but worse, the meaning of the data is often lost. Or it just may not be there. For example, boundary making and group classification is shaky and changes over time; Wimmer acknowledges this. What does this do to data on ‘closure’? Another example: discrimination has cumulative consequences; Wimmer recognizes this. Yet he dismisses claims about ‘structural racism’.
