Thomas, J., concurring
have more diverse outlooks on this metric than two students from Manhattan’s Upper East Side attending its most elite schools, one of whom is white and other of whom is black. If Harvard cannot even explain the link between racial diversity and education, then surely its interest in racial diversity cannot be compelling enough to overcome the constitutional limits on race consciousness.
UNC fares no better. It asserts, for example, an interest in training students to “live together in a diverse society.” Brief for University Respondents in No. 21–707, p. 39. This may well be important to a university experience, but it is a social goal, not an educational one. See Grutter, 539 U. S., at 347–348 (Scalia, J., concurring in part and dissenting in part) (criticizing similar rationales as divorced from educational goals). And, again, UNC offers no reason why seeking a diverse society would not be equally supported by admitting individuals with diverse perspectives and backgrounds, rather than varying skin pigmentation.
Nor have amici pointed to any concrete and quantifiable educational benefits of racial diversity. The United States focuses on alleged civic benefits, including “increasing tolerance and decreasing racial prejudice.” Brief for United States as Amicus Curiae 21–22. Yet, when it comes to educational benefits, the Government offers only one study purportedly showing that “college diversity experiences are significantly and positively related to cognitive development” and that “interpersonal interactions with racial diversity are the most strongly related to cognitive development.” N. Bowman, College Diversity Experiences and Cognitive Development: A Meta-Analysis, 80 Rev. Educ. Research 4, 20 (2010). Here again, the link is, at best, tenuous, unspecific, and stereotypical. Other amici assert that diversity (generally) fosters the even-more nebulous values of “creativity” and “innovation,” particularly in graduates’ future workplaces. See, e.g., Brief for Major American Busi-
