Violent aggression is overwhelmingly a problem of adolescence. Around the world, adolescents commit proportionally more violent aggression than any other age group. Violent crime rates peak around age 20, and youths between the ages of 15 and 25 (with age 25 marking the approximate end of adolescent brain development [Blakemore Reference Blakemore2008; Giedd Reference Giedd2004]) engage in anywhere from 2 to 10 times more violence than adults or younger children (Moffitt Reference Moffitt1993; Snyder Reference Snyder2012).

This difference can be attributed to a range of cultural, social, and neurodevelopmental factors, including – of particular relevance to the CLASH model – poor self-control (Cohen et al. Reference Cohen, Breiner, Steinberg, Bonnie, Scott, Taylor-Thompson, Rudolph, Chein, Richeson, Heller, Silverman, Dellarco, Fair, Galván and Casey2016; Moffitt Reference Moffitt1993). Poor self-control in adolescence reflects in part the rapid development and heightened activity of reward-sensitive regions of the brain in this age group, even while executive regions responsible for balancing risk and reward, and weighing short- and long-term rewards, remain underdeveloped. These dynamics increase the risk not only of aggression, but also of related outcomes like substance use, accidental injury, and unintentional drowning (Steinberg Reference Steinberg2013). All of these phenomena are particularly pronounced in adolescent males (Scheidt et al. Reference Scheidt, Harel, Trumble, Jones, Overpeck and Bijur1995), in whom patterns of brain development are somewhat delayed relative to those of females (Lenroot & Giedd Reference Lenroot and Giedd2006), and who are also much more likely to engage in violent aggression (Snyder Reference Snyder2012).

The authors have made a strong case that rates of violent aggression rise with geographic proximity to the equator, which is used as a proxy for climate. But the model may require amending to incorporate the mediating or moderating role that the proportion of adolescents across regions may play in geographic variation in violent aggression.

As it happens, more equatorial countries also contain large – sometimes much larger – proportions of adolescents. Almost without exception, those nations with median ages less than the global median age of 29 are equatorial nations of Africa, Asia, and Central and South America. The youngest countries in the world include equatorially proximate African nations like Uganda, Niger, and Mali. By contrast, the oldest countries include Japan, Germany, Monaco, and other European and Asian countries closer to the poles (Central Intelligence Agency 2016b).

So-called “youth bulges” in countries proximal to the equator could explain increased violence in these countries without reference to life history theories or climate. It has been observed that the proportion of individuals within a society who are between the ages of 15 and 24 is predictive of the prevalence of various forms of violence in that society, including homicide, domestic armed conflict, terrorism, and rioting (Bricker & Foley Reference Bricker and Foley2013; Mesquida & Wiener Reference Mesquida and Wiener1999; Urdal Reference Urdal2006). In the United States, a significant proportion of changes in violent crime over time can be explained by fluctuations in the proportion of adolescents (Phillips Reference Phillips2006). Globally, the relationship between violence and the proportion of a country's population composed of adolescents has also been found across several investigations (Cincotta & Leahy Reference Cincotta and Leahy2011; Pampel & Gartner Reference Pampel and Gartner1995; Urdal Reference Urdal2006).

Van Lange and colleagues might argue that countries like Guatemala, Belize, and Honduras are among the most violent countries in the world because of their equatorial climates, which result in the population of these regions adopting relatively fast life history strategies, characterized by “short-term planning, greater risk taking, a focus on immediate gratification for short-term benefits, and more aggression” (sect. 3.1, para 3). But one could just as easily argue that the reason these quintessentially adolescent traits are pervasive in these countries is that their populations are disproportionately composed of adolescents. These are among the world's youngest countries, and are the three youngest nations in the Americas. Nearly a quarter of the population of Guatemala, for example, is between the ages of 15 and 24 (Central Intelligence Agency 2016b).

It is possible, then, that the patterns the authors have observed do not reflect climate-induced variation in life history strategies, but are instead an artifact of geographic fluctuations in the proportion of youths, owing to reasons that are unrelated to climate. Alternately, it is possible that fluctuations in the proportion of youths could be incorporated into the CLASH model. Youth bulges are thought to emerge during the stage of a nation's development when infant mortality has been successfully reduced, yet fertility rates remain relatively high (Bricker & Foley Reference Bricker and Foley2013), but no generally accepted explanation exists for why youth bulges currently cluster around the equator.

Any attempt to incorporate the relationship between youth and violence into the CLASH model should reflect the fact that the relationship between youth bulges and violent aggression may not be a simple one. Variables like access to education and jobs are critical influences on the behavior of youths within a society (Bricker & Foley Reference Bricker and Foley2013). Likewise, the effect of youth bulges on violent aggression may be mitigated by protective cultural factors like collectivism, which can transform large youth populations into civic opportunities (Pampel & Gartner Reference Pampel and Gartner1995). Understanding the role of adolescence in societal variation in aggression therefore requires considering not only how many adolescents a society contains, but also how those adolescents are faring: Are they civically engaged? Educated? Impoverished? Optimistic about their future prospects (Bricker & Foley Reference Bricker and Foley2013; Hart et al. Reference Hart, Atkins, Markey and Youniss2004; Pampel & Gartner Reference Pampel and Gartner1995)?

The essential fact remains that a model of cultural variation in violent aggression that does not consider the role of adolescence remains an incomplete model.