Recently several posts have played the "race" card and elicited all manner of responses, but at its root, the fundamental premise had not actually been examined.  Is "race" a valid concept?

I know that many people will immediately experience a knee-jerk reaction to the idea of discussing "race" as pseudo-scientific, because they either have a vested interest in advancing their own "race", or because they wish to use it as a lever against other "races".

However, the premise is quite simple.  If you can't actually define it in scientific terms, then it cannot be science.  Therefore any claims that derive from it are not science.  Similarly, we cannot claim that "race" is valid by simply engaging in arm-waving arguments based on the fact that there are genetic differences between various population groups.  "Race" must be fully quantifiable as specific heritable trait(s) that serves to identify the group in question. 

One argument often made is that "race" has a biological meaning because it is used in the same way "subspecies" is to describe other animals.  It is simply an indicator of genetic differences intrinsic to different population groups.  OK, fair enough.

Human beings are formally classified as Homo sapiens sapiens, with all known subspecies being extinct.  If this is to change, then where's the taxonomic argument for the change?  What should the designations be?  Interestingly enough, advocates are typically silent on this point.  Let's also be clear that a species can be considered monotypic without requiring any further taxonomic designations1.  Even the definition of a species is subject to controversy and subspecies designation is often even more contentious.  However, it is somewhat difficult to argue that humans can be reasonably divided into subspecies, when there is free genetic mixing between groups and no definable taxa that results from such pairings.  So while it may be possible to identify a particular group based on historical, geographic, or cultural lines, this becomes untenable when those groups interbreed with members outside that group.

If the concept of race is to be scientific, then it would need to specifically identify the genetic criteria that is to be used for that differentiation.  Merely claiming some external trait isn't going to do it. 

Such simplistic thinking is insufficient to raise the idea of "race" beyond anything except another convenient [or inconvenient as the case may be] cultural grouping.

Some will complain that this is about political correctness, but that is wrong.  It's about science and it's about having specific scientific evidence on which a classification is based.  Usually whenever you hear someone claim either "inferiority" or "superiority" you can be fairly confident that they are not being scientific.  Basing an explanation of race on a few statistical outliers is not only disingenuous, it is decidedly unscientific.

So what are those genetic differences that are serving as a criteria for defining race? Again, invariably things like the sickle-cell gene, or the propensity for Tay-Sachs disease is hauled out as justification.  However, this isn't justification, since there are a significant number of people within the supposed "racial" classification that do NOT possess those genes.  What are we to make of the classification then?

Sickle-cell is associated with Sub-Saharan Africa, yet this hardly encompasses the entire region populated by blacks, which is the racial designation invariably used. Within the United States, it is estimated that 1 in 500 African-American children will acquire anemia 2.  Should we consider this 0.2% distribution of this disease, a basis for classifying this as a racially specific condition?  Even the most liberal reading of such a statistic would argue that Sub-Saharan Africans would have to be considered a separate race from other blacks.  More importantly, the notion that Sickle-Cell is a "black" gene is simply wrong 3,4.

So, invariably we find that we are drawn to superficial differences to draw the racial lines, without regard for whether there's actually a genetic basis for such a line.

Additionally, some will argue that these "race" studies are important for medical reasons because of population susceptibility to such genetic diseases.  This is most certainly a desirable objective, but it should illustrate how much more important it is that we have clear genetic differentiation and not something so trivial as skin color 5.

For example, if you assumed that the Sickle-cell gene only affected blacks, then scientifically you would have missed the reality of people in India also having it.  How scientific is that?  So as science has actually demonstrated, the issue of the sickle-cell gene is closely tied to the incidences of malaria on a worldwide scale.  Therefore the important point is to target descendents of individuals from those regions, rather than some nebulous claim of their purported race.

So, while there is no question that examining genetic differences in population groups is an endeavor worth pursuing.  Perhaps before we pursue it, we should indicate what those genetic differences actually are, instead of simply assuming that something as trivial as skin color can do that work for us.

Despite all the professed interest in combating "racially distinguished" genetic diseases, this is also a decidedly unscientific tact.  It is clear that as populations move and interbreed, that such "racial" differences will spread throughout the population, since genes don't distinguish "race".  So, while it might be of some historical interest, it is of little medical value.  In fact, diagnosing genetic diseases based on "race" could invariably result in significant misdiagnosis of affected individuals simply because they are members of the, apparently, "wrong race" 6.
"Although opponents of a biological definition of race acknowledge that it is possible to classify geographically defined populations on the basis of clusters of genetic building blocks, they argue that the public health implications of such ancestral clustering of genes is controversial and that race at the ancestral or continental level has not been proven useful in terms of predicting individual diagnoses or individuals’ responses to drugs or causes of disease."
More to the point, there are already numerous diseases that have been identified as having a genetic basis, but they don't neatly divide between "races" because of population mixing.  In effect, the motivation to use "race" as a indicator of genetic diseases is doomed to fail, since these populations cannot be assumed to remain geographically or culturally isolated.

However, let's be completely clear on this point.  Any genetic disease is capable of being passed on to any other human being, without regard for "race".  Despite the sense that Tay-Sachs affects primarily those of Jewish descent or Sickle Cell Anemia affects Sub-Saharan Africans, this only describes those with the highest probabilities of having the trait.  Anyone can have it, so there is no "racial" protection against acquiring the genes.

Let's remember that the issue isn't whether there are genetic differences between individuals; we know there are.  The issue isn't whether there are genetic differences between population groups, especially those that are geographically or culturally isolated; we know there are.  The issue is whether such differences are clear and significant enough to be useful in assessing individual traits.  If these differences are not consistent, nor applicable in all cases, then their utility must be suspect, hence their basis for being considered as indicative of "race" or subspecies.

We should also consider that when talk turns to genetic traits such as those in sports, we quickly begin to invoke nationality just as casually as "race".  So, it's Jamaicans or Kenyans, despite there being little biological basis for such gross distinctions.  All we do know is that a few select individuals that may share a particular population history have demonstrated prowess in areas that others haven't.  That is hardly a basis for claiming genetic relatedness between millions of people.  In other words, if you wish to discuss Usain Bolt, then do so, but to invoke Jamaicans, in general, is unwarranted.

So, if we really want to pursue the topic of "race" or designating subspecies of humans, then lets do so on a scientific basis, and not some arbitrary socio-cultural designation.  If "race" is going to be based on genetics, then it should be intuitively obvious that people will have their "racial" classification changed based solely on their personal family history.  As a result, the designation of any particular "race" could actually change from generation to generation. Therefore any claim at racial knowledge that is based on arbitrary external traits rather than the specific genetic traits, is, by definition, wrong 7

Show me the genes.

Additional note on genetic claims

For an example of how nonsensical such genetic claims can be, consider the following article:

It's fairly obvious that if the slave trade were responsible for selective breeding to produce these "super-athletes" then clearly the early slaver-owners must've been considering the selective traits for running sprints in some slaves, and marathons for others [of course this neglects the number of offspring that the white owner may have fathered].  Since even artificial selection presumes that there is a specific trait being selected for, it is ridiculous to suggest that such selective "breeding" by slave owners would've produced such a pronounced dichotomy between the two different skills.  While it might make for a good story, it is unclear how surviving a horrific ocean voyage at the hand of slave-traders, would translate into superior genetics for running.  
"Only the toughest survived. During one such voyage in 1732, more than 95 per cent of slaves perished — 170 were herded on to the ship and only six got off alive."
Again, this perspective is generally the result of a gross misunderstanding of the phrase; "survival of the fittest", in that it assumes that strength equates to fitness.  As unfortunate as such conditions were, generally it is naive and simplistic to simply presume that the survivors were the strongest.

In truth, our ancestors were likely to be considerably faster than us modern humans, so the more plausible explanation is that there are still some genes [conserved traits] present that allow the occasional outlier to have such a genetic advantage.  It is clearly not intrinsic to everyone, so to infer such a broad genetic interpretation on an entire population is ludicrous.

1  The variation among individuals is noticeable and follows a pattern, but there are no clear dividing lines among separate groups: they fade imperceptibly into one another. Such clinal variation always indicates substantial gene flow among the apparently separate groups that make up the population(s). Populations that have a steady, substantial gene flow among them are likely to represent a monotypic species even when a fair degree of genetic variation is obvious.

2  In the United States, about 1 out of 500 African-American children born will have sickle-cell anaemia.

It is important to note that acquiring the disease is not the same as having the trait.  Since the purpose of using genetics to classify "race", then any individual that possesses the gene would need to be considered part of that racial group.
"Sickle cell is found in people in Greece, the island of Orchomenos, in particular, has very high carrier rates for sickle cell. There are also high carrier rates among people on the Arabian peninsula and people in India. There are parts of India where sickle cell carrier rate is as high as it is anywhere in Africa. On the other hand, people in the southern part of Africa - or people whose ancestry goes back to the southern part of Africa - don't have high sickle cell carrier rates because that allele is selected for only in human beings who have lived where there's a high instance of malaria, which you don't find in southern Africa. People do not carry the sickle cell gene variant because they are of a certain race, but because of some more particularized population history."
3 Regions of Sickle Cell in the world:

    Mediterranean countries (such as Greece, Turkey, and Italy)
    The Arabian peninsula
    Spanish-speaking regions (South America, Central America, and parts of the Caribbean)
"...yet it is common in people of African, Mediterranean and Indian origin."
The sickle-cell gene is found in people from Africa (or African descent), the Mediterranean area (Italy, Greece), Middle East, East India, Caribbean and Central and South America.
4  Consider this survey of the Sickle Cell Trait (SCT) in Uganda, where there isn't even a consistent occurrence within the same country.  Yet, despite such statistical variation, it is still often discussed as if it were a uniform characteristic of a general class of people.

5  Within medicine:
 ...within medicine, knowledge of a patient’s racial and ethnic background is often a significant factor in the appropriate selection of treatment modalities. It is well known for example that the survival rates of transplant patients are influenced by race, as the lack of close ethnic matching between donor and patient is a significant factor influencing tissue rejection
While this may seem like an eminently reasonable position, consider how few people actually know what their racial and ethnic backgrounds are, and how many of them have that information available to a medical professional.  In short, if medicine is based on personal anecdotal information, then the usefulness in medicine is suspect, and becomes muddled when only obvious external racial traits are considered indicative of anything.

6 While anyone can be a carrier of Tay-Sachs, the incidence of the disease is significantly higher among people of eastern European (Ashkenazi) Jewish descent. Approximately one in every 27 Jews in the United States is a carrier of the Tay-Sachs disease gene. Non-Jewish French Canadians living near the St. Lawrence River and in the Cajun community of Louisiana also have a higher incidence of Tay-Sachs. For the general population, about one in 250 people are carriers.

  You would think that that lesson should have been clear to biology when numerous taxonomic changes had to be made when DNA analysis demonstrated that classification by external traits was often wrong.

Let's also remember the huge degree of variation in external traits that we see in animals like dogs, despite there being no biological consideration that the various breeds represent a subspecies.