Today I noticed that both Tim Ferriss and Gary Taubes, both of whose work I largely respect, announced the debut of a new organization called the Nutrition Science Initiative (NuSI).  As Tim Ferriss put it in this blog post, “No hidden agenda, no corporate interests, nothing to do with food subsidies or ulterior motives. Just good science. It’s about time, right?”
In general, I like the approach of this organization, but I still remain skeptical of its universal dietary optimism. To make claims about certain foods being universally healthy or unhealthy seems dubious — what is healthy for one person may actually be unhealthy for another (e.g., milk).
Over the last 10,000 years some people have dietarily evolved to tolerate certain products of agriculture that others have not (e.g., lactose).  Mention this on the Web in public discourse, however, and you will soon understand where Godwin’s Law came from (“As an online discussion grows longer, the probability of a comparison involving Nazis or Hitler approaches 1.” As Godwin’s Law informs us, many people will fear where this line of evolutionary reasoning heads because of the potentially politically incorrect logical implications. In one important sense, though, it is absolutely true that those who have the allele that allows them to tolerate lactose are indeed genetically different than those who do not have it. When it comes to the ability to tolerate lactose, not all people are created equal — as one might suspect, this is most likely true in other areas of human biology as well.
I, politically incorrect as it may be, maintain the belief that evolution is accelerating its work on the human species and will continue to do so until we humans eventually speciate. Again, I think the last 10,000 years of human development provide plenty of evidence to support this claim.
Anyway, when it comes to diet and health, many people seem to want to ignore the elephant in the room, which is that science cannot provide us with a one-size fits all diet recommendation that works for everyone. However, I’m not suggesting that certain modern techno-foods are not bad for all humans across the board either. I’m also not trying to say that we should do away with an empirical approach to dietary issues all together either — in fact, I’m suggesting quite the opposite.
If we increase the sample of people being studied to include a massive amount of others who have evolved radically different food tolerations than certain others (like ourselves), then we will end up diluting the most important data point (ourselves) into obscurity. What I’m suggesting is that when it comes to what dietary choices will make you healthy, the most important data point is you. In other words, anecdotal evidence is not always inferior to empirical evidence, which is difficult for the scientific minded to remember when arguing about what foods are good or bad.
 See The 10,o00 Year Explosion