Avoiding an excessively narrow focus on agriculture

One flaw in G&K's analysis is an excessively narrow focus on agriculture in their definition of ultrasociality. The key condition for ultrasociality is a surplus of resources, making division of labor possible. Many species of eusocial insects meet this condition without practicing agriculture (e.g., army ants; Hölldobler & Wilson Reference Hölldobler and Wilson2008). Likewise, eukaryotic cells and multicellular organisms represent major evolutionary transitions, from groups of organisms to groups as organisms, without practicing agriculture. The fact that the cells of a multicellular organism are totipotent at the genetic level, but also capable of extreme specialization through the differential expression of genes, provides a useful frame of reference for thinking about human division of labor. Expanding the comparative view beyond social insects that practice agriculture does not detract from G&K's main thesis.

The importance of multilevel selection and complex systems theory

The majority of authors writing peer-reviewed articles on social evolution agree that the major theoretical frameworks (e.g., multilevel selection theory, inclusive fitness theory, evolutionary game theory, and selfish gene theory) are equivalent ways of accounting for evolutionary change. They offer different perspectives on a single causal process, rather than invoking difference causal processes, and therefore deserve to coexist to the extent that their perspectives offer useful insights. In this sense, the controversy over group selection (e.g., the question of whether traits evolve by between-group selection, despite being neutral or selectively disadvantageous within groups, as these terms are defined within multilevel selection theory) is over (Wilson Reference Wilson2015b).

It is possible for two theoretical frameworks to be equivalent in some but not all respects (Wilson Reference Wilson2015a). As an example, inclusive fitness theory attempts to describe the evolution of a trait in the form rb – c > 0 (Hamilton's rule), where c is the absolute fitness effect on the actor, b is the absolute fitness effect on the recipient, and r is a coefficient of relatedness (which need not be genealogical relatedness). This framework is well suited for the study of traits that can be clearly measured in individuals and have a simple genetic basis (e.g., what Okasha [Reference Okasha2006] calls multilevel selection 1[MLS1] traits) but can become unworkable for traits that can be measured only in groups and/or have a complex genetic basis (e.g., what Okasha [Reference Okasha2006] calls multilevel selection 2[MLS2] traits).

Against this background, it is unsurprising that multilevel selection has become the theoretical framework of choice for the study of human cultural systems as units of selection (Yaworsky et al. Reference Yaworsky, Horowitz and Kickham2014). The group-level properties that are being selected are too systemic to be rendered in the form of Hamilton's rule. This doesn't mean that reductionistic analysis is impossible, only that it reveals so many lower-level traits interacting in such a complex fashion that if the three terms of Hamilton's rule can be calculated for a lower-level trait at all, they take the form of complex multi-term equations that are difficult to measure or interpret and are often frequency-dependent, so their value changes with every generation (e.g., van Veelen et al. Reference van Veelen, García, Sabelis and Egas2012). Multilevel selection theory avoids these problems by measuring phenotypic variation, selection, and heritability at the level of groups, which does not require detailed reductionistic understanding of the traits and their heritable basis. Against this background, G&K's reliance on multilevel selection theory is thoroughly warranted and should be uncontroversial.

Utopic and dystopic scenarios for the future

In my own recent (coauthored) BBS target article titled “Evolving the future: Toward a science of intentional change” (Wilson et al. Reference Wilson, Hayes, Biglan and Embry2014), we stress that left unmanaged, evolution often takes us where we don't want to go. Enlightened public policy requires becoming wise managers of evolutionary processes. One dystopic scenario is that lower-level selection prevails over higher-level selection, resulting in various “tragedies of the commons” such as global warming, extreme inequality, and societal collapses that are already taking place in some parts of the world. G&K outline another dystopic scenario that is equally disturbing: Even if higher-level selection prevails over lower-level selection, the cost of a well-run society might be a loss of individual autonomy and intelligence. The decline in human brain size during the last 10,000 thousand years suggests that this trend is already underway.

Without being naively optimistic, I think that utopic scenarios for the future are still possible. It is encouraging that inclusive nations function better as large-scale societies than nations governed by a small group of elites (Acemoglu & Robinson Reference Acemoglu and Robinson2012; Pickett & Wilkinson Reference Pickett and Wilkinson2009). Evidently, egalitarianism is needed at all scales. All societies must become socially differentiated as they increase in size, but “structure” can embody an egalitarian ethos of “communitas,” as the anthropologist Victor Turner (Reference Turner1969) put it. The moral philosopher John Rawls' (Reference Rawls1971) concept of the “Veil of Ignorance” – which asks us to design a society subject to the constraint that we will be randomly placed within it – makes the same point. The fact that cultural evolution must increasingly be guided by policies informed by science and formulated with the global good in mind means that we must become more mindful, rather than more mindless, if we are to succeed at all.