Many of the popular diets that Americans follow are based on sweeping theories that neatly separate food types into categories of “good” and “bad.” The Paleo Diet, for example, classifies foods humans are presumed to have eaten before the agricultural revolution as “good” and foods that are presumed to have entered the human diet within the past 12 millennia as “bad.”
All theory-based diets share the common assumption that it is possible to predict the effects of different food types on human health on the basis of a single principle derived from general knowledge. None of these theories stands up to rigorous testing, however. In each case, there are discrepancies between the lists of foods that, according to the theory, “ought” to be healthy or unhealthy and those that actually are.
What if we were to abandon a comprehensive theory of healthy eating and instead define it heuristically through the identification of eating patterns common to the healthiest humans? I will employ the science of self-organizing systems to explain why this approach is more sensible than the theory-based approach and will reference the dietary patterns of elite endurance athletes to identify universal dietary best practices for humans.