“The problem with racial discrimination, though, is not the inference of a person's race from their genetic characteristics. It is quite the opposite: it is the inference of a person's characteristics from their race. The question is not, can you, given an individual's skin color, hair texture, or language, infer something about their ancestry or origin. That is a question of biological systematics -- of lineage, taxonomy, of racial geography, of biological discrimination. Of course you can -- and genomics as vastly refined that inference. You can scan any individual genome and infer rather deep insights about a person's ancestry, or place of origin. But the vastly more controversial question is the converse: Given a racial identity -- African or Asian, say -- can you infer anything about an individual's characteristics: not just skin or hair color, but more complex features, such as intelligence, habits, personality, and aptitude? /I/ Genes can certainly tell us about race, but can race tell us anything about genes? /i/To answer this question, we need to measure how genetic variation is distributed across various racial categories. Is there more diversity _within_ races or _between_ races? Does knowing that someone is of African versus European descent, say, allow us to refine our understanding of their genetic traits, or their personal, physical, or intellectual attributes in a meaningful manner? Or is there so much variation within Africans and Europeans that _intraracial_ diversity dominates the comparison, thereby making the category "African" or "European" moot?We now know precise and quantitative answers to these questions. A number of studies have tried to quantify the level of genetic diversity of the human genome. The most recent estimates suggest that the vast proportion of genetic diversity (85 to 90 percent) occurs _within_ so-called races (i.e., within Asians or Africans) and only a minor proportion (7 percent) within racial groups (the geneticist Richard Lewontin had estimated a similar distribution as early as 1972). Some genes certainly vary sharply between racial or ethnic groups -- sickle-cell anemia is an Afro-Caribbean and Indian disease, and Tay-Sachs disease has a much higher frequency in Ashkenazi Jews -- but for the most part, the genetic diversity within any racial group dominates the diversity between racial groups -- not marginally, but by an enormous amount. The degree of interracial variability makes "race" a poor surrogate for nearly any feature: in a genetic sense, an African man from Nigria is so "different" from another man from Namibia that it makes little sense to lump them into the same category.”