Diet, natural selection, the zerg rush and the zombie apocalypse

Unnecessary personal introduction

I have been reading a lot recently on diets related to reconstructions of what prehistoric humans supposedly ate. Under names like The Primal Blueprint or the Paleo Diet, they make the basic argument that the evolution we have undergone since the domestication of agriculture is minimal compared to the evolution we underwent prior to that. Meaning, pragmatically, that we are not evolutionarily selected to do things like eat grains or wear shoes. This strain of thought seems to be based in a long stream of Romantic views of nature, but also in some hard science. Indicating, for example, that hunter-gatherers who don't eat much sugar or grains have better tooth and jaw structure (based in the research of Weston Price); that there is little connection between high dietary cholesterol and heart disease and so on.

The fact is, I've tried reducing my sugar consumption to near zero, and my grain-based carbohydrate consumption (esp. at lunch) to low levels, and I have found myself rewarded with higher energy levels and getting sick very rarely. My wife suggests that I have some deep discontent with society that leads me to seek outsider views, especially relating to my diet. And this is doubtless true - I was vegetarian on and off for about four or five years, played around with raw food briefly, etc. Nonetheless, I have been more satisfied with the short term effects of reducing sugar and grain consumption and upping animal fat consumption; short, intense, whole-body workout; minimal shoes and so on than I ever was eating lots of beans and jogging. The question of the long-term effects is still a somewhat open question, as far as I am concerned, although I am coming around to the opinion (belief) that limiting inflammation is much more important than minimizing blood cholesterol.

Personal experience aside, there is something deeply unsatisfying about the argumentation of a lot of the paleo party. On the one hand, they delight in ridiculing the laboratory reductionism of big science, pointing out population studies (like Weston Price's surveys of native populations, the Framingham Study, etc.) that support their positions, and selectively choosing examples of the most successful remaining hunter gatherers. On the other hand, they delight in citing instances of laboratory science that support their conclusions, or arguing study against study. Most existing criticism of this line of thinking has been even less self-critical, largely relying on what paleos dismiss as a "conventional wisdom" supported by an interested government-agribusiness-pharmaceutical alliance of interests. The lab study vs. lab study type of argument reminds me of nothing more than that line about statistics; the specifics of study design and similar factors mean that there is a huge tendency toward confirmation bias, even when researchers don't intend it. Population studies are notoriously difficult to unpack, and I can't really tell whether either side is selectively citing examples of hunter-gatherers. I will arbitrarily call these positions a wash. To me, the argument rests largely on the application of theory, in particular evolution.

The evolutionary argument for paleo

The best argument that the paleos trot out, for which I would direct you to Mark Sisson, is that agriculture is:

1. Too recent a development to have effects on evolution.
2. A system that makes it too easy for the weak to survive to reproductive age - by producing an easy surplus - thereby allowing all genes to be perpetuated.

It's a bit more complex than this, but basically the idea is that we still have the genes of prehistorical hunter-gatherers, and by having a very different diet and activity pattern, we are substantially less healthy. The reason we don't simply diminish and die is that we are surprisingly effective omnivores, and our gene expression changes so that we exist at a less effective, but still minimally effective, level of homeostasis. This minimal level of effectiveness allows us to pass on our genes without evolution playing a strong role - which means that we also pass on the genes that make grain-based diets unhealthy.

Implicit in this argument is the assumption that we have moved to a fundamentally different mortality regime. The prehistoric mortality regime was essentially random and selective - death was due primarily to accidents, seasonal starvation and only rarely infectious disease (essentially similar to most wild animal mortality patterns). This produced a population curve that tapered off rather steeply, but evenly across all ages (after especially steep infant mortality). This meant that barring accident or poor hunting performance, fifty or sixty-year-olds were nearly as healthy and able as twenty or thirty-year-olds, otherwise they would be unable to provide for themselves and die. It also provided for natural selection to take place on the individual level - poor hunters or gatherers were more likely to die of starvation or accident - and selected for people who do well eating meat and fish, fruits, vegetables and tubers.

The historic, premodern mortality regime was random but not selective, infectious disease based. Its shape was steep among infants and children, shallower during late childhood and early adulthood, and steeper again in middle age and beyond. The poor diet in this period meant a faster aging process, and an illness- and malnutrition-related die-off starting in the late 30s due in large part to things like tooth decay. The randomness was more likely to wipe out entire populations, but beyond infant mortality, it meant that people were more or less equally likely to live to childbearing age. Thus a continuation of poor health among the majority of the population, still selected to eat lots of meat and veg but unable to do so.

The modern mortality regime is both genetic and environmental, but not selective, chronic disease-based. The population curve has a much smaller infant die-off, some limited accident-based mortality, and very few deaths to infectious disease - basically a shallow slope into middle age. At middle-age, the curve steepens due to the beginnings of chronic disease mortality and again in 70s and 80s as chronic illness and age-related illness combine. Those with "bad genes" die of heart attacks in their forties and fifties, and those with "good genes" of stroke, organ failure, and decrepitude-related disease/accident combination in their 80s and 90s. The conventional wisdom is that we now die of cancer and heart disease because we live long enough to. The paleo critique is that we die of cancer and heart disease because we live long enough, and our behavior promotes it; and that both middle-age heart attacks and old age decrepitude are results of our behavior as much as our genes. In either case, we live long enough to pass on our paleo genes - its just that now most people live that long, rather than a large, random subset of them dying due to infectious disease.

My critique

As I see it, the problem with this argument is that it overlooks the effect of group selection on evolution. At the individual level, the changing mortality regime would seem to indicate that we are still hunter-gatherers at heart (and stomach, and DNA). But group-level and population-level effects are enormous! The transition to agriculture made major changes by building big enough population bases for infectious disease. This changed the pattern of our mortality, but it also changed the pattern of our responses. The last post was about some of these - the various things we have called "hygiene." Most of these responses are now operating on the level of "culture" rather than "nature" - memes rather than genes - but they still have genetic effects!

Consider this: John McNeal asserts that 20% of all human years lived in the past 40,000 years were lived in the past 100. This is not the same as saying that 20% of all people who lived in the past 40k lived in the past century, because our lifetimes are longer, so let's conservatively say that this is equivalent to 5-10% of people. This means that the vast majority of population expansion happened in the modern era. If we assume that people in the past had significantly more than two children per woman on average (i.e. above the replacement rate), this means that mortality was very high to prevent wild population expansion. This is not a controversial argument. This means that mortality was the dominant control on population. Non-random mortality is the hand of evolution. We have established mortality, if we establish non-randomness, then we have evolution.

Randomness may be a much more difficult question. We tend to think of disease as relatively random, but it is not the case. Recall the massive demographic effect of the plague on Europe or smallpox on the Americas. These had the effect of greatly reducing certain genetic pools while providing population vacuums for the immune to fill - i.e. for the demographic expansion of the Mongols, or of the Spanish, British, French genetic material. So it seems likely that we are selected on the group level for the types of people who make good conquerors. Who makes good conquerors? Nomads? Farmers? Definitely not hunter-gatherers.

In the early stage of Chinese history, it appears that farmers and hunters were rather balanced - see the long wars between the Shang and Zhou states and the four barbarians. But over time, the farmers won out. This was probably not because the farmers were individually better warriors, evidence seems to indicate quite the contrary, that the hunters were much better fighters, larger in stature and probably healthier. But over time, the weight of population, of energy, of technology came out on the side of the farmers - there were more of them, with better weapons. Not only this, but disease was on the side of the farmers. The farmers definitively won, there are very few hunter-gatherers left in the world, and only in areas where extreme climate or disease regimes keep agriculture at bay.

Later, the main opponent of the farmers was the nomads. It is not clear that this period of competition lasted longer than the farmer-hunter wars, but it is much better documented. Again, this appears to be a case where the nomads were probably more healthy. The nomads also had another energy reserve - livestock - on their side. They also seem to have been less susceptible to epidemic disease genocide. But the lost in the end to a gradual rise in demographic pressure and shifting of the technological balance. The role of non-organic energy may have also been critical - they were generally tamed by gunpowder empires and destroyed by fossil-fueled industrial farming and mining concerns.

There were also the wars between farmers and other farmers, or farmers and hybrid farmer-herders. This is where I would be interested to look for group selection. In the case of farmer vs. hunter, the energy balance was too uneven to promote any clear evolutionary pressure - even unhealthy farmers could defeat hunters. On the other hand when two different pools of farmers competed, the ones better able to use grain-based energy would have been less susceptible to disease, made healthier fighters, and seen fewer children die before they could become soldiers. This makes me wonder whether farming groups that adapted genetically, not just culturally, to farming would have survived better.

Finally, there are several pieces of evidence for evolution in historical time. Lactase persistence (i.e. not developing lactose intolerance as an adult), which would give those with the mutation an advantage over others, has clearly developed since the domestication of livestock among farming communities. Some native Mexican and South Chinese groups exhibit a "thrifty gene" that allows them to maximize nutritional content from grain consumption, but leads to diabetes when high levels of grain are consumed. This would have enabled easier population expansion of this group, at the expense of those who did not extract nutrients as readily.

There is probably also a great deal of genetic selection based on memetic selection. In other words, groups that produced good social ideas were likely to out-compete those that didn't. As these groups were generally based, at least to some degree, in kinship groups. Even within relatively static, peaceful populations, the expanded technological ability, capital resources etc. of successful families would likely promote their genetic material at the expense of the less sucessful. These forms of memetic evolution would tend to promote associated genetic material. While this selection is not directly on nutrition-related attributes, the connection between the political economy and food production is too close to assume that social behavior and genetic adaptations are completely unrelated.

The zerg rush and the zombie apocalypse

At base, I take the argument that individual evolution has not proceeded at the same pace, but I argue that group evolution has still occurred. The paradox of group evolution is that it seems to promote something like the zerg rush, the idea that hordes of lesser individuals can overcome a smaller group of more powerful ones. I feel that this has more potential to explain the entire system that both paleo and conventionals are part of - one in which the power of the large group depends on a system that results in the relative poverty and poor health of its constituent individuals. At the same time, the demographic power seems more like a slow creep than a sprint, something more like the zombie apocalypse. This has the metaphorical advantage of suggesting the role that the very illness of that group plays in its success.

This post has grown too long and incoherent, but I have a final, more hopeful suggestion. It seems that the power of the farm - the poor and dense imperative - was rather well balanced against that of the nomad - more individual power and sparser settlement - until the advent of fossil fuels. The biggest problem with the paleo solution is that it remains largely individual and dependent on volition, in the face of the power of the state and market. I have done little to address the social and moral aspects of paleo in the modern world (a topic for another post), but I would suggest that it is a small-scale solution at best, until paired with a viable political economy. Hunter has little to offer in those terms, but herder might.