(1) Egyptian civilization was heavely influenced by the civilizations of the Middle East, long before other civilizations existed in Africa. You purposedly forget that.

I agree that an open mind is necessary, however I prefer to start from zero and add facts as they're confirmed than to start with an assumption and then adjust it as new facts appear.

That’s just it, though. The concept of race was created by attributing undue significance to specific aspects that were deliberately selected to prove the hypothesis of those who created the concept. It was a theory that sought out its evidence, not one that evolved from its evidence. I think it's more reasonable to shelve it pending better evidence than to persist with it.

Those minor variations in size, shape, etc. that do occur in various human populations occur at a much more local level than between the so-called 'racial' divides, and reccur in similar environments accross the globe, irregardless of 'race'. For instance, people from mountainous areas the world over tend to skew towards a shorter, barrel-chested frame. This is not a characteristic of African or Asian or European peoples generally, but a local environmental adaptation that flourishes when it's beneficial regardless of what 'stock' the population comes from. Although humans don't all look like clones of each other, their range of variation isn't terribly different in different areas.

There are a great many skills and behaviours that humans evolved over time that are absolutely inappropriate to the modern context, and classifying people at a glance is one of them. It was very useful (it still is, I suppose) to know who is your close cousin and who is not. It is much less useful – and much less accurate - to try and figure out which group of millions of people are more like you than another.

The mechanism that we use to identify close relations (generally, the shape of the face) becomes less and less accurate the further removed the relation. Not only does it necessitate an increasing focus on smaller and smaller commonalities as the group becomes bigger, it also leaves more and more room for subjectivity in deciding who belongs in which category (among those that share some commonalities, but not others).

While differences between people certainly exist, attempting to force huge groups into artificially delineated 'races' is hugely problematic. If 'population' or 'ethnic group' are vague concepts, 'race' is even moreso, and practically random by comparison.

Incidentally, 'ethnicity' is not a euphemism for race. It refers to a cultural or national identity, not a biological one. I also consider 'population' to be a better word than 'race', because it generally (in my experience) is used to identify localized groups – not giant agglomerations – that can be much more defensibly identified as relatively complete and discrete entities.

I agree, which is why I consider adherence to a rigid classification system inaccurate. As our understanding of human genetics improves, it becomes less and less necessary to classify humanity through broad generalizations.

But there are so many "externally visible elements", you run into the problem of which ones to choose, and how to decide how significant any apparent differences are. The fact, for instance, that a small group of humans crossed Beringia to the Americas, and then re-diversified into almost the full range of human variation in size, shape, colour, etc, shows that there has been very little loss or change in the human genetic code, even in the farthest-flung populations.

I consider language to be as good an ethnic classification scheme as any (though still hardly an exhaustive one) - at the very least, the languages people speak have undeniable significance in their lives.

It’s not a matter of being ‘right’ or ‘wrong’ – it’s a matter of being useful or useless. As far as I’m concerned, the one potential value that can come from classifying humans is for medical purposes. For example, as people from this or that background are more susceptible to this or that disease, they should get this or that test more often. But our ever-increasing understanding of the genes that control susceptibility to disease are bringing us closer to a point where we can individualize these kinds of considerations. This is much better than statistical generalizations, because it’s individuals who get sick, not racial groups.

Even from a standpoint of pure anthropological curiosity, it makes little sense to insist on classifying a person or group of persons according to an artificial standard that almost invariably ignores part of their biological heritage (throughout history, isolated populations have been few and far between).

While craniology is helpful in some cases, it isn't helpful in all of them. It's natural to use craniology to track the evolution of hominids, because the changing shape and size of the head was by far the most noticable and important development in hominid evolution. It's also much more meaningful to track skeletal changes that occur over intervals of hundreds of thousands or millions of years, than to try and pick out the miniscule differences that appear in the space of thousands or tens of thousands of years, and assign them the proper significance.

As an example: sundadonty was identified by observing skeletons. Originally, though, it was presented as evidence that South-East Asians are a hybrid of East Asians and South Asians - the pre-existing bias of those who made the determination. Genetic evidence, however, suggests that this is precisely the wrong conclusion, and that sundadonty is in fact ancestral to sinodonty.

That actually isn't accurate. Diversification is common among animals with small distributions, as competition for the same resources drives different populations to each specify in the exploitation of one resource or another. Darwin's finches on the Galapagos Islands would be the classic example.

The role of space in human development has traditionally been overstated, while the role of time has been understated.

That number is much larger than zero, which is how many breeding pairs existed outside Africa at the time.

My understanding is that modern humans probably developed as coast-dwellers, not savanna-dwellers. Hunting is one of those activities that have traditionally been over-emphasized in human development because of the biases of those (mostly men) who have studied it.

You're being very selective in the facts you choose to paint the picture of a clean, linear expansion out of Africa that supports your views. The population did not "mushroom" after leaving Africa, the non-African population unerwent several bottlenecks as well. The fact of the matter is, there are more extant mtDNA lines from Africa than from any other continent. I agree that this isn't the whole story, but it puts a hole in your theory that Africans have no more diverse a lineage than anyone else (forgive me if I'm wrong, but that seems to be what you're implying here).

Again, I’m not sure that’s entirely accurate. For instance: while the earliest out-of-Africa expansion probably went along the southern coasts of Eurasia and followed a fairly traceable route, once humans moved inland (particularly into Central Asia, which was just as flat and just as conducive to fast migration then as it would remain later) expansion became much more convoluted and back-migration much more extensive. It’s possible to create a straightforward narrative of human expansion by picking one theme out of the whole, but that doesn't give you a complete picture.

Height, weight, body shape and susceptibility to certain diseases, to name a few. These factors either don't vary by region, or vary regionally at such a local level that it's not reasonable to call them 'racial' characteristics.

If people weren't able to interbreed between "regions", then what exactly makes it more likely that they would have been able to interbreed within "regions"? If we take the so-called Caucasian and Negroid races as an example, how is it more likely that interbreeding would happen between people in Spain and India or between people in Sudan and South Africa than between people in Spain and Sudan or India and South Africa?

I would agree that at a local level many populations show small adaptations that were beneficial to them at some point in time, and that point to a peculiar developmental history. But to extrapolate from that and say that huge groups of people, containing almost the full range of human variation and spread across vast and diverse regions, share meaningful traits that differentiate them from other, equally huge and diverse groups of people, is to stretch an idea to its breaking point.

I'm interested to know what ballet has to do with this (and to be clear, I don't mean that sarcastically - I'm genuinely interested)

As you say, we can't realy argue our respective subjective experiences. I will expand, though, on the reasons I believe looks to be a very unreliable way of judging people's ancestries.

I think we have to be vigorous in challenging the way we see things at first glance. Our minds subconsciously strip out a great deal of information when we look at something, so our perception of other people's appearance suffers from a great deal of subconscious weighting and selectivity. We often learn, for whatever reason, to associate particular traits with particular groups. This makes us look for those traits when we expect to find them, and not to look for them when we don't - skewing our perception. This also leads us to take those traits as proof in and of themselves, whether or not they are corroborated by anything else.

For example, I have been informed that you can tell by look who in Skopje is Bulgarian and who is Albanian. I suppose that's possible, but I also doubt this assertion has been verified by asking everyone who 'looks' Bulgarian if they actually are Bulgarian and so on.

This type of perceptual bias is the same that makes people believe that they encounter more red lights when they're having a bad day. They don't actually, but they subconsciously notice and remember things to get angry about when they're feeling angry. When you're expecting to find something, you'll probably find it.

That's a bold analysis you give of my psychology. 'Personality' isn't really what we're talking about here, but rather the professional tightrope people like Dawkins and Hawass have to walk as both scientists and media characters. Both Hawass and Dawkins try to use the media to get their messages out, and in doing so both take the chance that sometimes they will help themselves and sometimes they will hurt themselves - the nature of popular media inherently precludes scientists speaking and explaining in the same way they would to a scientific audience (particularly when, as in Dr Hawass' case, he happens to be representing one of his country's biggest industries in addition to his scientific field). Dawkins has certainly stepped in it on a number of occasions. I believe that this particular debate may be one of those that are more about public perceptions than science - after all, how much relevance does the reconstruction of King Tut's face, or the migrations of peoples thousands of years before Egyptian civilization existed, really have for the study of Egyptian archaeology? I'd argue that it changes not a thing.