The author started out with a few misconceptions about paleo dieting that she corrected in the course of researching her article (such as that a paleo diet involves eating a lot of factory-farmed meat).
She is left with two remaining criticisms:
There's no such thing as a uniform diet shared by all of our paleolithic ancestors: common conceptions of the paleo diet are based on an arbitrarily selected period of our ancestral past; and in any case, no matter which period we're talking about, human diets varied by region and by season.
Human dietary choices are not shaped simply by our genes; they are shaped also by culture and tradition.
Her first point is a very good one, and I think she gets less mileage out of it than she could have. Her second point misses the mark.
Taking those points in turn:
1.
This is not a criticism of the nutritional precepts of the paleo diet. It is instead a question about its underlying logic. It identifies a potential miconception of paleo-dieters: that our paleolithic ancestors ate anything resembling a uniform diet (that we can now emulate). They didn't. Two responses come immediately to mind. First, any such misconception is not widespread. For the most part, everybody already knows that human diets have always varied quite a bit. And second, even among people who hold that misconception, it has limited practical significance. While the diets of our paleolithic ancestors varied quite a bit, they all had a number of features in common, and it is these shared features that matter to paleo dieters. For example, none of our paleolithic ancestors ate large amounts of high-fructose corn syrup (and other refined sugars), refined flour from domesticated grains, or industrially processed vegetable oils. While the huge majority of ancestral human diets had plenty of other features in common as well, avoiding just those three neolithic agents of disease gets us a long way toward eating a diet that accords with our evolutionary heritage (and therefore with our metabolic infrastructure).
But there's another aspect to this criticism that I think is generally underappreciated in the paleo community and merits further consideration. The author says: "Our ancestors began to eat meat in large quantities around 2 million years ago, when the first Homo forms began regular use of stone tool technology. Before that, the diet of australopithecines and their relatives was overwhelmingly plant-based, judging from clues in teeth and bones. I could argue that the more genuine 'paleo' diet was vegetarian."
Indeed. If it's always best to eat what our ancestors ate, then how did early humans thrive by incorporating meat into their diets in the first place? Put another way, if switching from a generally vegetarian diet to a thoroughly omnivorous diet was an improvement, then who's to say that switching from a meat-and-vegetable–heavy diet to a grain-heavy diet can't also be an improvement?
This fallacy is somewhat common in the paleo community: "Dairy wasn't eaten in the paleolithic era, so you shouldn't eat butter." "Rice wasn't eaten in the paleolithic era, so it must be bad for you." And so on.
That's not how the universe works. What we can say with some confidence is that foods that have long been common in our ancestral diets are safe for us: we are adapted to them.* It does not follow, however, that foods that have not long been common in our ancestral diets are unsafe for us. Maybe they are, maybe they aren't. To find out, we need to see what happens when we eat them. We need the scientific method — we need experiments — not just stories about our ancestral past.
As it turns out, and as we should expect, most foods that we haven't fully adapted to are problematic in a number of ways. Modern wheat, for example, contains phytates and enzyme inhibitors and potential gut irritants that make it less optimal for most humans than a similar amount of broccoli would be. But that's not something we can divine from our prehistory; it's something we must conclude only from careful observation. Paleo-dietary principles do not, therefore, constitute unyielding conclusions; they merely supply testable hypotheses.
Its hypotheses have generally held up quite well when tested — but with exceptions. I think paleo dieters who avoid butter on prehistorical grounds, for example, are making a mistake.
2.
The author has the reasoning underlying a paleo diet backwards here. The claim isn't that our ancestral dietary choices were shaped by our genes; the claim is that our genes were shaped by our ancestral dietary choices. Modern koala bears are adapted to eating eucalyptus leaves because that's what their ancestors have eaten for many generations; humans are likewise adapted to eating the types of foods that our ancestors have eaten for many generations. Turned around in the proper direction, this reasoning is sound.
*This is one reason why the idea that dietary saturated fat and cholesterol are harmful to us should be highly suspect until it is supported by good empirical evidence — which it isn't.