Biology

Adolescence as a human life history stage is defined as the period from the onset of the adolescent growth spurt to its cessation (Bogin, Reference Bogin, Thompson, Krovitz and Nelson2003, Reference Bogin2009). The adolescent growth spurt is a uniquely human phenomenon. During this time individuals gain 50% of their weight and 20% of their final adult stature and it encompasses sexual maturation, i.e. puberty. While a boy is fertile before he develops any secondary sexual characteristics (e.g. enlargement of larynx and deepening of voice, growth of facial and body hair including pubic hair, increased stature), a girl ‘develops nearly all of her secondary sexual characteristics [e.g. breasts, underarm and pubic hair, increased stature] to their fullest extent before acquiring her fertility’ (Lancaster, Reference Lancaster, Lancaster and Hamberg1986: 21). In addition, girls only achieve adult pelvic dimensions at the very end of puberty. What this difference in patterning suggests is that, evolutionarily, there has been selection for delaying fertility in human females until they have had time to function socially as adults (Lancaster, Reference Lancaster, Lancaster and Hamberg1986). With the onset of puberty, another noticeable characteristic of adolescence emerges – a significant shift in sleep patterns. Alterations to circadian and homeostatic systems lead to increasingly later releases of melatonin throughout puberty. This means that teens feel the need to sleep later and wake later than others do (Galván, Reference Galván2020).

The order of the biological changes that mark adolescence is universal, but the timing and speed of change varies between the sexes and is influenced by an individual's social environment (see discussion below). Definitions of adolescence based on these biological changes suggest that in twentieth- and twenty-first-century contexts this life stage begins at around age 10 (Sawyer et al., Reference Sawyer, Azzopardi, Wickremarathne and Patton2018). The end of adolescence is less clearly defined biologically. The cessation of the adolescent growth spurt (~16–17 years) provides a clear biological and developmental marker for the end of puberty, but whether this corresponds with the end of adolescence is debated, and other developmental features that characterise adolescence suggest a later end. For example, Roenneberg et al. (Reference Roenneberg, Kuehnle, Pramstaller, Ricken, Havel, Guth and Merrow2004) propose that another abrupt change in sleeping patterns at age 20 marks the biological end of adolescence.

Social

Adolescence is universally recognised as a distinct social, as well as biological, stage of development (Schlegel & Barry, Reference Schlegel and Barry1991). Biological age, which is estimated from biological changes in the body, is not necessarily the same as social age, which is defined as ‘the culturally constructed norms of appropriate behaviour and status of individuals within an age category’ (Halcrow & Tayles, Reference Halcrow and Tayles2008: 192). However, these two types of age are interrelated as social and environmental factors such as degree of physical activity, access to nutritious food, familial stress, life expectancy and socioeconomic stress can impact the onset and duration of puberty (Halcrow & Tayles, Reference Halcrow and Tayles2008; Nabi et al., Reference Nabi, Amin, Sultan and Kmil2014; Lewis et al., Reference Lewis, Shapland and Watts2016). Furthermore, the development of secondary sexual characteristics may signal readiness to take on behaviours associated with specific social age categories (see, for example, Lewis et al., Reference Lewis, Shapland and Watts2016).

The social aspect of adolescence is no less important than the biological in terms of defining this life stage. As part of the recognition of having reached a new life stage, adolescents are treated differently from both younger children and adults. These differences usually relate to preparing adolescents for their future economic and family roles and responsibilities, and accordingly, vary between the sexes (Schlegel & Barry, Reference Schlegel and Barry1991, pp. 33–35). Social definitions of adolescence are more variable than those based on biology but they both share a difficulty in defining the end of the adolescence phase. In recent industrialised contexts, Sawyer et al. (Reference Sawyer, Azzopardi, Wickremarathne and Patton2018) propose extending the end of adolescence to 24 years (from earlier definitions of 19 years) to reflect social trends including a later age at marriage, longer periods of education and delayed economic independence. We need to remember, however, that while almost all cultures recognised adolescence as a distinct life-stage, ‘the modern adolescent is not the modal adolescent’ (Schlegel & Hewlett, Reference Schlegel and Hewlett2011: 287); who is considered an ‘adolescent’ and the roles and responsibilities of these individuals are culturally variable. In this regard, the social investment theory suggests that adolescents transition to adulthood and mature more quickly in cultures where they assume adult responsibilities at an earlier age (Bleidorn et al., Reference Bleidorn, Klimstra, Denissen, Rentfrow, Potter and Gosling2013 and references therein).

Cognitive

As Arain et al. (Reference Arain, Haque, Johal, Mathur, Nel, Rais and Sharma2013: 451) observe, ‘adolescence is one of the most dynamic events of human growth and development, second only to infancy in terms of the rate of developmental changes that occur in the brain’. While there is individual variation in adolescent brain maturation, in broad strokes, throughout puberty, cortical grey matter decreases as a result of synaptic pruning while white matter gradually increases possibly as a result of increased myelination (the process by which white matter tracts are covered with a fatty coating to conduct signals more efficiently; Khundrakpam et al., Reference Khundrakpam, Lewis, Zhao, Chouinard-Decorte and Evans2016; O'Rourke et al., Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020). Thus during adolescence, in what has been described as a ‘use it or lose it’ process (O'Rourke et al. Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020), pruning removes excess neurons and synaptic connections that are no longer needed by the brain in order to improve the efficiency of neuronal transmission (Khundrakpam et al., Reference Khundrakpam, Lewis, Zhao, Chouinard-Decorte and Evans2016). Pruning represents ‘the behavioural, and ultimately, the physiological suppression of competing, irrelevant behaviours’ (Casey et al., Reference Casey, Giedd and Thomas2000: 246). This process reflects the importance of both biology and environmental experience on an adolescent's brain maturation (O'Rourke et al., Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020).

Different regions of the brain mature at different rates. For example, during adolescence, there are significant changes in the subcortical limbic system governing emotion and mood. However, the prefrontal cortex (PFC), governing executive functions such as attention, inhibition, cognitive flexibility and the ability to plan and think through the consequences of an action, continues to develop throughout adolescence. In fact, the PFC is actually the last region of the brain to fully mature at around age 25 (Burnett & Blakemore, Reference Burnett and Blakemore2009; Khundrakpam et al., Reference Khundrakpam, Lewis, Zhao, Chouinard-Decorte and Evans2016; O'Rourke et al., Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020). This diversity in maturation rates can lead to greater risk taking, impulsivity and emotionally driven decision making in teens without the necessary checks and balances to constrain that behaviour being fully available.

Social cognition in teens is equally complex. Activation of the amygdala during pubertal development leads to an increased ability to read emotions in the faces of others (O'Rourke et al., Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020) and in taking on other emotional perspectives – what Burnett and Blakemore (Reference Burnett and Blakemore2009: 52) describe as the ability to ‘step into someone else's shoes’. During puberty, gonadal hormones involved in reorganising neural circuitry in both males and females lead to greater motivation for sexual and romantic partners (O'Rourke et al., Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020). At the same time, fMRI studies demonstrate that teens activate different regions of the brain than adults when processing situations that require social understanding. This suggests that, over time, understanding social situations becomes more automatic and less laborious, as teens garner more life experience and as neural circuitry is fine-tuned (Burnett & Blakemore, Reference Burnett and Blakemore2009; Blakemore, Reference Blakemore2018). As Burnett and Blakemore (Reference Burnett and Blakemore2009: 54–55) note, ‘an unexplored implication of this could be that the period of life when arMPFC [anterior rostral medial prefrontal cortex] and other social brain regions are still developing – the teens and early 20s – might be a period of particular open mindedness to new ideas and different types of people’. Similarly, O'Rourke et al. (Reference O'Rourke, Whalley, Janes, MacSweeney, Skrenes, Crowson and Schwannauer2020) argue that the unique plasticity of the adolescent brain renders them more open to acquiring new skills. This plasticity also creates an opportunity for learning and motivation relevant to romantic and sexual behaviour (ibid.).

Social adolescence in recent hunter–gatherer societies

Much as is the case within archaeology, the study of adolescents in recent foraging societies is less common than the study of childhood (e.g. Montgomery, Reference Montgomery2009; Konner, Reference Konner2010; Meehan and Crittenden, Reference Meehan and Crittenden2016). Furthermore, adolescence is mostly either studied as part of childhood sensu lato, or otherwise undifferentiated. Many of the key features of hunter–gatherer adolescence do appear to have important precursors in childhood, and from an evolutionary perspective, developmental stages cannot be understood in isolation (Stambler, Reference Stambler2017: 25). Nonetheless, the adolescence of hunter–gatherers, like that of all human societies, is a distinct, marked stage of social life and development.

Hewlett and Hewlett (Reference Hewlett, Hewlett and Hewlett2012) have conducted the most comprehensive cross-cultural research on hunter–gatherer adolescence. In addition to characteristics of adolescence shared with other higher primates generally (e.g. an increase in sexual activity, an increase in time spent learning complex skills) and humans specifically (e.g. sexual division of labour, marriage), they identify multiple features of adolescence they consider unique to foragers (Table 2). We highlight a few of these here.

Table 2. Key characteristics of adolescence unique to hunter–gatherers (after Hewlett & Hewlett, Reference Hewlett, Hewlett and Hewlett2012)

Many of the key features of forager adolescence are the result of demography (Hewlett & Hewlett, Reference Hewlett, Hewlett and Hewlett2012: 96). While group composition is fluid (Dyble et al., Reference Dyble, Salali, Chaudhary, Page, Smith, Thompson and Migliano2015), the small size of forager camps (cross-cultural mean = ~25–30 people; Marlowe, Reference Marlowe2005; Hill et al., Reference Hill, Walker, Božičević, Eder, Headland, Hewlett and Wood2011) means that hunter–gatherer children and adolescents are almost always physically close to a small number of people, within a network of widely dispersed groups living at low population densities. Furthermore, infant and child mortality rates are high in hunter–gatherer communities. Cross-culturally, ~43–49% of children die before the age of 15 (Hewlett, Reference Hewlett1991; Marlowe, Reference Marlowe2001, Reference Marlowe2005; Volk & Atkinson, Reference Volk and Atkinson2013). This combination of small group sizes and high childhood mortality means that, in contrast to many other societies, adolescents are rarely spending their time only with other adolescents; there are simply too few of them for this to occur. Adolescent socialisation and maturation thus occurs in an intergenerational context, alongside both younger children and adults, a factor that contributes notably to the ‘physical and emotional intimacy with parents and other adults’ that characterises hunter–gatherer adolescence.

This physical and emotional intimacy that characterises forager adolescence has its roots in childhood. Combined with two other important features of hunter–gatherer childhood – the development of individual autonomy and a strong sense of self and belonging caused by homogeneity of world view (Hewlett, Reference Hewlett and Hewlett2017) – this intimacy manifests in one of the most notable features of forager adolescence: the lack of the teenage ‘rebellion’ or identity crisis that is considered the hallmark of social adolescence in many other societies (see, for example, Blakemore, Reference Blakemore2018).

An example of social and economic changes caused by contact with twentieth-century Western cultures demonstrates the importance of hunter–gatherer demography and worldview in shaping the experience of social adolescence. Condon (Reference Condon1990) describes how the behaviour of teenagers of the Canadian Inuit community of Holman Island altered drastically following interventions by the federal government. Some of the key changes were demographic. Centralisation of settlements and accompanying population growth meant that, for the first time, there was a sizable peer group of adolescents. In combination with the beginnings of formal schooling, teenagers began to spend more of their time with their same-aged peers, rather than in mixed-age groups with their parents and other Inuit adults. Schooling also exposed the teenagers to other value systems and non-Inuit economic and social roles to aspire to, roles and employment that were nonetheless very difficult for Inuit youths to obtain. The concomitant increased influence of peers and decreased parental and elder influence, combined with their exposure to new values and social expectations that contrasted with those of Inuit society, led to a stark increase in troubling adolescent behaviour including the drinking of alcohol and violence. Notable changes were also witnessed in other domains, including the selection of marriage partners (now less influenced by parental choice) and a delay in marriage age, which slowed down the social transition from childhood (sensu lato) to adulthood in Inuit society.

Adolescence among hunter–gatherers is a time during which individuals seek sexual and potential marriage partners, a feature they share with teenagers in other societies and well as higher primates who have just attained reproductive maturity. The search for marriage and sexual partners, combined with the low population densities of foragers – the cross-cultural mean of which is 0.25 persons/km² (Marlowe, Reference Marlowe2005) – is responsible for another of the characteristic features of hunter–gatherer adolescence: long distance exploration. Among foragers, there is a significant inverse correlation between mating distance and population density; the lower the regional population density is, the greater the distance travelled to find a suitable partner (MacDonald & Hewlett, Reference MacDonald and Hewlett1999). Furthermore, it is during adolescence or young adulthood that the greatest distances are travelled, and men usually travel further than women. This latter finding accords with sexual selection theory that proposes that the sex with the higher potential reproductive rate will compete more strongly for access to potential partners (ibid.). While it is hard to state with certainty the reason for specific instances of long-distance travel (see for example, Hagino & Yamaucho, Reference Hagino, Yamauchi, Akazawa, Ogihara, Tanabe and Terashima2014), there is no doubt that this is a frequent occurrence among teenage hunter–gatherers. Ethnographers commonly note the high levels of absence of adolescent boys in hunter–gatherer camps (e.g. Hagino & Yamaucho, Reference Hagino, Yamauchi, Akazawa, Ogihara, Tanabe and Terashima2014; Milner et al., Reference Milner, Gurven, Kaplan and Gaulin2014). Evolutionarily, there may have been selection for navigation skills in adolescent males. In a study of male and female West Point cadets, Munion et al. (Reference Munion, Stefanucci, Rovira, Squire and Hendricks2019: 1933) note that ‘traveling longer distances without changing course, pausing less, and fewer returns to previously visited locations were significantly related to the ability to locate the correct target … [and] the significant relationship between gender and navigational success is fully accounted for by men and women producing different wayfinding behaviors, which in turn predict differences in navigational success’.

Finally, the shifts in sleeping patterns that occur during adolescence are of particularly importance within forager communities. Samson et al. (Reference Samson, Crittenden, Mabulla, Mabulla and Nunn2017) studied the relationship between chronotype variation (the times of day that a person feels the most alert or the greatest need for sleep) and group sleeping among the Hadza, a people who practice a foraging lifestyle in Northern Tanzania. Over a 20-day period, they found that there were only 18 one-minute episodes in which everyone was simultaneously asleep, while 99.8% of the time one person or more was awake. Chronotype varied by age only and not by sex, co-sleeping, nursing or study day. They argue that variation in age helps to generate variation in chronotype which facilitates sentinel-like behaviour which evolutionarily would have been key to group survival (e.g. for detection of danger from predators, other humans and the environment). This variation in sleeping patterns is particularly important for the small groups in which hunter–gatherers usually live and camp (12 or fewer). In models where they controlled for chronotype variation (i.e. they assumed that all members of a group fell asleep and woke at roughly the same time), groups had to be significantly larger for individual variation (e.g. someone wakes because they are too hot, have a nightmare, are hungry, need to urinate) to produce the same ‘sentinel effect’ (Samson et al., Reference Samson, Crittenden, Mabulla, Mabulla and Nunn2017). Thus, as teens enter puberty, they begin to play an increasingly important role in overall group safety.

Adolescence in the European Upper Palaeolithic

The features listed in Table 2 are particularly characteristic of mobile, immediate-return hunter–gatherers – those who lack storage, are generally egalitarian, have a culture of extensive giving and sharing, and live in small (<30 individuals) groups at low population densities (Hewlett & Hewlett, Reference Hewlett, Hewlett and Hewlett2012). Along the simplified spectrum from ‘simple’, egalitarian, immediate-return hunter–gatherers to ‘complex’, hierarchical, delayed-return hunter–gatherers (Woodburn, Reference Woodburn1982; Keeley, Reference Keeley1988), European Upper Palaeolithic hunter–gatherers – with their evidence of substantial storage, some semi-permanent dwelling structures and the presence of social hierarchies (e.g. Soffer, Reference Soffer1985; Fišaková, Reference Fišáková2013; Iakovleva, Reference Iakovleva2015; cf. Wengrow and Graeber, Reference Wengrow and Graeber2015) – fall closer to the latter. Nonetheless, we still consider the features listed in Table 2 to be highly relevant to our understanding of the experience of social adolescence in the Upper Palaeolithic, with variability in adolescence resulting from different foraging adaptations being of a degree rather than of a kind.

The relationship between demography and the key characteristics of forager adolescence provides the main support for this assertion. The demography of European Upper Palaeolithic foragers is hard to infer from archaeological and paleoanthropological data (French, Reference French2016), but in general terms resembles that of recent foragers. The European Upper Palaeolithic metapopulation fluctuated across its ~28,000-year span, with absolute estimates ranging from ~1,500 to 11,000 people in Western and Central Europe (Maier, Reference Maier2017). Outside of some key areas (e.g. southwestern France, the Danube basin) regional populations were small and lived at low densities in small, mobile groups (Bocquet-Appel et al., Reference Bocquet-Appel, Demars, Noiret and Dobrowsky2005; French, Reference French2015; French & Collins, Reference French and Collins2015; Kretschmer, Reference Kretschmer2015; Maier & Zimmerman, Reference Maier and Zimmerman2017; Schmidt & Zimmerman, Reference Schmidt and Zimmermann2019). Estimates of Upper Palaeolithic fertility and mortality rates come with many caveats, but similarly indicate high rates of both adult and child mortality (Trinkaus, Reference Trinkaus2011). Accordingly, we propose that the behaviours described above, and listed in Table 2 are an appropriate ‘baseline’ for social adolescence in the Late Pleistocene.

While the uniformitarian principle allows us to draw broad parallels between past and present in the developmental and biological aspects of adolescence, the aforementioned plasticity in the onset, pace and duration of puberty – plasticity that is strongly influenced by social and environmental factors – should prompt us to ask if and how hunter–gatherer adaptations influence biological adolescence. The age of onset of female adolescence has received particular attention, with an increasingly early age of puberty (marked by the onset of menarche) recorded in industrialised countries over the last 100 years, a trend that has been linked to, among other factors, energy balance and nutrition (Garn, Reference Garn1987). The link between onset of menarche and energy probably explains the contrasts seen between hunter–gatherer and modern industrial societies in this variable. Jones and Lopez (Reference Jones and Lopez2006: 47) report an average age at menarche for hunter–gatherer girls of 16.1 years (compared with 12.1 years for girls in the USA), 19.5 years for age at first birth (26 years in the USA) and an interval of 3.4 years between menarche and first pregnancy (13.5 years in the USA). While the latter is heavily influenced in modern industrial populations by age at marriage, among hunter–gatherers it is more likely to be the result of adolescent subfecundity (the period of up to several years of irregular/anovulatory cycles that follows the onset of menstruation, which may be compounded by low calorie intake), as marriage may precede menarche (Hochberg & Konner, Reference Hochberg and Konner2020: 7). If, as is the case among the !Kung, the adolescent life stage (for women) is defined as the period between the onset of menarche and the birth of the first child, ‘adolescents’ are those between approximately 16 and 21 years of age (Howell, Reference Howell2010). Among !Kung men the period of adolescence is longer, lasting on average from 16 to 27, and similarly ending with the birth of the first child (ibid.).

Direct data on the onset and duration of puberty and adolescence among Upper Palaeolithic children is lacking. To the best of our knowledge, the aforementioned skeletal maturation indicators of puberty of Lewis et al. (Reference Lewis, Shapland and Watts2016; Shapland and Lewis, Reference Shapland and Lewis2013, Reference Shapland and Lewis2014) have not been applied to any Upper Palaeolithic individuals. Some European Upper Palaeolithic populations appear to have been in reasonably good health and nutritional condition (e.g. those of the Mid-Upper Palaeolithic or Gravettian; Holt & Formicola, Reference Holt and Formicola2008; Formicola & Holt, Reference Formicola and Holt2015), which may have contributed to lowering the age of onset of puberty. The Ostuni 1 woman, who was heavily pregnant at the time of her death at age 20 years (Nava et al., Reference Nava, Coppa, Coppola, Mancini, Dreossi, Zanini and Bondioli2017), provides the only firm evidence for age at pregnancy in the European Upper Palaeolithic, giving a terminus ante quem for the end of the period of adolescent subfecundity for this individual (although it is unknown whether this was her first pregnancy). If the end of adolescence was defined in similar ways in past and present hunter–gatherer societies (i.e. by the age at first birth), then it is likely that female Upper Palaeolithic hunter–gatherer adolescence similarly lasted until the late teens/early twenties, given the role of biology in determining this variable; adolescence may have lasted even longer for men. Further study of adolescents in the European Upper Palaeolithic should focus on those older teens identified in Table 2, rather than the young teens (~10 years) (cf. Gluckman & Hanson, Reference Gluckman and Hanson2006).