This paper improves arguments for recent and sophisticated views concerning biological realism about race, but also helps move many of us beyond this issue altogether.

I first argue that if we are going to inquire about whether races are biologically real, we should reconfigure the way this issue has been framed and pursued till now. My new way uses recent work on property clusters to make common but unclear appeals to non-essentialist natural kinds more precise. This allows us to uncover and articulate several distinct and often confounded sorts of biological realism about race, some more interesting than others. I then show how this helps improve on existing reasons to think that clinal biological races are unreal in some respects, and in others are real but uninterestingly so. Likewise for cluster biological races, and clade biological races. More generally the needed innovative reconfiguring more convincingly points evidence in support of the view that each biological realism about race is either false or vacuous.

However I next argue that many of us have less reason than often supposed for inquiring in the first place about whether races are biologically real. For instance, even when a particular biological realism about race is false, any favored anti-realist counterpart is often of less interest than supposed. When race matters practically and theoretically, it is often because of associated claims about causation, intervention, and prediction, the truth-values and justifications of which are not influenced by verdicts about realisms.