>For example, if for a job we have 100 candidates in group A and 100 in group B, and we only hire from group A, then requiring some hiring from group B likely i

>For example, if for a job we have 100 candidates in group A and 100 in group B, and we only hire from group A, then requiring some hiring from group B likely increases the hiring based on merit and decreases discrimination. This is borderline disingenuous. In the vast majority of institutions accused of discrimination because of disproportionate representation, the underrepresented groups are already underrepresented among applicants. Especially in the military - what proportion of military applicants have historically (or contemporaneously) been female? More importantly, your repeated justification for discrimination against white males is predicated upon the dangerous and ubiquitous conflation of equality of opportunity with equality of outcome. There is absolutely no reason to expect one to imply the other in a diverse, even perfectly meritocratic civilization. Culture heavily influences career/employment goals. That's one of the major issues with the entire diversity movement. The bulk of evidentiary support for reverse racism comes from statistical underrepresentation, which is effectively impossible to conclusively blame on discrimination. Equal representation is not a viable goal at least because, ignoring other reasons, not all demographics are equally interested in all jobs.