2
definition encourages the identification of instances where race is used explicitly to
distinguish groups in the population; the social process definition allows for the
highlighting of ostensible race-neutral practices that end up reinforcing racial hierarchies
without ever explicitly invoking them. For instance, if racism is defined as involving
conscious intent to discrim
inate against some people based on an ideology of racial
superiority and inferiority, say between whites and blacks, then evidence that policy is
explicitly designed to be consistent with that ideology would be needed (Sniderman and
Piazza 1993). We can call this “explicit” or “old” racism. Yet, if racism is defined as
involving the reflective participation in social processes that reinforce racial
disadvantages, then even a policy stated in race-neutral terms can be vulnerable to this
charge (Ansell 1997; Bobo, Kluegel and Smith 1997). Under these conditions, we might
want to emphasize processes of “racialization” or “race-making” that recreate racial
hierarchies without resorting to explicit racial categories or theories or race grounded in
biology (Lipsetz 1998; Wacquant 2002). This we will call “implicit “ or “new” racism.
Concerning the data themselves, different findings often lend themselves to some
interpretations more than others. Definitions matter here as well. The categories used to
define the data in the first place predispose data to being interpreted in some ways more
than others (Stone 2000; Hochschild 2000; Schram and Soss 2002). For instance,
research on welfare in recent years has hewed more frequently toward examining data
that have been defined in ways that from the very beginning suggest that differences are
attributable to individual effort more so than societal factors (O’Connor 2001). Such data
are likely to support charges of racism in welfare reform only in those cases where there
is overt evidence of discriminatory treatment against particular individuals (i.e., the “old”
Convention