>>11821941>race isn't realWhen we talk about race you need to understand that we're talking about a classification with a high degree of arbitrariety. We're taking a big chunk of data, all humans, that can be classified according to genetic distancing(the choice of metric will create different genetic maps, so the metric space choice by itself yelds arbitrariety). After we create a metric space with the chosen genetic distance function we than need to classify clusters of members according to arbitrary definitions, we can use computers for this(Google community detection in graphs to understand better). Of course that the parameters used by the computer will be all arbitrary like "how many clusters do you want", "what is the maximum diatance allowed between two members of a cluster" and so on.
When you do basic cluster analysis in the human genome, using the colors to model clusters you'll get a picture like the one here. The metric chosen is arbitrary, the number of "races" is arbitrary, but once you plug the algorithim you can actually get results that match human intuitive notions of race.
You can chalk all this effort out because of the arbitrary nature of clustering, but by the same token you could throw any grouling out. Why classify red and pink as different colors if they're connected in a continuum? Why classify males and females if we have people with extreme genotipic defects like xxy or xyy?
You need to understand that there is no serious discussion about the reality of genetic diferences, it's just a debate about the arbitrariety of clustering. Saying "that black race isn't real" is just like saying "the color red isn't real", the classification of a cluster of frequencies as "red" in the color spectrum is arbitrary, still it is useful.
