Ethnicity as a construct in leisure research: A rejoinder to Gobster

Our primary purpose in our paper was to question the efficacy of ethnicity as a variable in examining differences in leisure preferences, choices, and behaviors among groups of people. While we agree with Gobster on many of his points, we do not agree that the political or folk use of ethnic or racial classifications automatically qualify them as "natural kinds" that are useful as scientific categories. Indeed, as Gobster points out, research in genetics shows that within-race variation greatly exceeds between-race variation. In what sense, then, are races, and by extrapolation, ethnic groups, natural kinds that carve nature at its joints? That such groups have been given names does not guarantee that they are homogenous or functional as constructs. We need to look at within-group and between-group variance in our research and evaluate the implications of the results, not merely assume that the groups are invariably different because of their labels. We are concerned with an additional issue, as well, that neither we discussed in our paper nor Gobster mentioned in his commentary. This is the distinction between methodological individualism and methodological holism. The former stance refers to the idea that groups are simply sums of their parts; that is, that group properties can be completely determined by studying characteristics of the individuals who comprise the groups. Methodological holism, on the other hand, is the position that groups may have emergent properties that cannot be determined by examining individual characteristics alone. Are we concerned, when discussing ethnic and/or racial groups and leisure, with the actions of individuals who we group together under some label or, alternatively, with the actions of groups that themselves constitute our unit of analysis? This is a major issue in the philosophy of social science (see, e.g., Kincaid, 1996) but one that seems to fly completely under the radar of leisure researchers. Is it possible that the variation sometimes observed among ethnic or racial groups with respect to leisure preferences, choices, or behaviors is an emergent property of the groups to which members self-ascribe and are ascribed by others rather than individual characteristics of members? We do not know the answer but believe the question to be worthy of investigation. Finally, Rosenberg (1988) claims, "... in the social sciences, there has been almost universal agreement that the descriptive categories that common sense has used since the dawn of history are the right ones" (p. 11). We maintain that ethnic group labels are the kind of common sense descriptive categories to which Rosenberg refers. He goes on to say "... the social sciences are rather like chemistry before Lavoisier: trying to describe combustion in terms of 'phlogiston,' instead of 'oxygen,' and failing because there is no such thing" (p. 13). We are not claiming that ethnicity is unquestionably the phlogiston of leisure research or social science, more generally. We claim only that, first, whether ethnic groups are homogeneous, second, what they are homogenous in terms of, and, third, how this homogeneity, or lack of it, affects leisure preferences, choices, and behaviors, are important empirical questions that leisure researchers have failed to ask.