Ontogenetic niche construction, object play and play objects

Niche construction theory posits that not only genetic and cultural information is passed on from generation to generation but that also environmental modifications are inherited. This process is not only true in Homo, albeit peculiarly and extensively so in this lineage (Odling-Smee et al., Reference Odling-Smee, Laland and Feldman2003). Anthropologists have long argued that material culture constitutes an ‘extra-somatic means of adaptation’ (Binford, Reference Binford1962: 218), a notion that goes back to at least Leslie White (Reference White1959) and, loosely, V. Gordon Childe (Reference Childe1936), and which presaged Dawkins’ (Reference Dawkins1982) extended phenotype. The idea of the extended phenotype re-entered the anthropological discussion in the context of the nascent evolutionary archaeological paradigm of the 1990s (O'Brien & Holland, Reference O'Brien, Holland, Skibo, Walker and Nielsen1995). While the tight linkage between genotype and extended phenotype turned out to be less useful for investigations of human culture – with perhaps the exception of its earliest variants (Corbey et al., Reference Corbey, Jagich, Vaesen and Collard2016) – it did become clear that the actions of organisms on the environment, via their extended phenotypes, critically modify those organisms’ physiological niche parameters. Critically, the longevity of some of these modifications, often across multiple generations, entails selection-modifying legacies. As new members are born into a niche-constructing population, the modified niche components and its resources become ‘ecologically inherited’ (Figure 1).

Figure 1. The three domains of inheritance of niche construction theory: genetic, cultural, and ecological with the respective resources (R_p, R_i) that are transferred. Redrawn and adapted from Odling-Smee (Reference Odling-Smee2007).

Niche modifications can target the ontogenetic environment or act on environmental components that mollify and direct selection; they can buffer organisms against environmental changes and so create adaptive lags (Laland & Brown, Reference Laland and Brown2006) or create entirely novel interactions that can result in further behavioural or genetic change. A final important feature of niche construction is that such niche modifications can have unintended positive or negative selective effects in the long term (see Riede, Reference Riede and Prentiss2019 for a recent review).

The majority of studies concerned specifically with human niche construction have focused on model systems such as animal and plant domestication (e.g. Altman & Mesoudi, Reference Altman and Mesoudi2019; Bentley & O'Brien, Reference Bentley, O'Brien, Saqalli and Vander Linden2019; Boggs, Reference Boggs, Emmett and Lekan2016; Boivin et al., Reference Boivin, Zeder, Fuller, Crowther, Larson, Erlandson, Denham and Petraglia2016; Fox et al., Reference Fox, Pope and Ellis2017; Zeder, Reference Zeder2016). However, niche construction can also operate at a more intimate scale and with effects not so much on other species but on individuals belonging to the niche-constructing species itself. During niche construction purely informational resources (R_i) as well physical resources (R_p) can be transmitted (Figure 1). Physical resources, such as artefact structures, themselves will often also contain semantic cues (Jeffares, Reference Jeffares2010).

That the ontogenetic niche is especially important in humans is well established. Homo sapiens newborns are helpless and take a long time to mature; infant and childhood survival demand considerable care, which in obligate tool-users results in a rich array of trappings such as child-carrying devices, cots, slings and toys. The criticality of this ontogenetic niche is also appreciated in the context of cumulative cultural evolution. Tomasello (Reference Tomasello1999: 512) asserts: ‘The major part of the ratchet in the cumulative cultural evolution of human societies takes place during childhood. That is, each new generation of children develops in the “ontogenetic niche” characteristic of its culture … mastering the artifacts and social practices that exist at that time’ (see also Tomasello, Reference Tomasello2020). However, children are not simply empty vessels for adult culture. Increasingly children's agency is seen as a key force in human niche construction (see Nowell, Reference Nowell2021). As Flynn et al. (Reference Flynn, Laland, Kendal and Kendal2013: 303) argue ‘two important points that are particularly evident in human populations are that children are not passive recipients of an adult's instruction and that instructors are not always adults … Thus, to a degree, and consistent with NCT, children direct their own learning by shaping their own learning environment’.

This means that, while cumulative human culture is often addressed as a species-wide phenomenon, there is significant lineage-specific variation because different human populations furnish their ontogenetic niches differently. The agency of children as they move through these niches will result in cultural evolutionary trajectories of technological, social and epistemic change that differ in space and time.

Life-history, cognitive plasticity and object play

The unusual life-history of humans has long been acknowledged (e.g. Key, Reference Key2000; Mace, Reference Mace2000; Thompson & Nelson, Reference Thompson and Nelson2011). In this context, the significant protraction of the pre-reproductive life stages of infancy, childhood, juvenility and adolescence is particularly noteworthy as they are argued to facilitate the extensive and flexible learning strategies that underwrite human culture (Bogin, Reference Bogin1997, Reference Bogin, Hawkes and Paine2006; Högberg & Gärdenfors, Reference Högberg and Gärdenfors2015; Nowell, Reference Nowell, Haidle, Conard and Bolus2016). Articulated with this period of extended childhood/adolescence is the notion of ‘extended parenting’ that provides the appropriate niche environment for youngsters to develop cognitively (Uomini et al., Reference Uomini, Fairlie, Gray and Griesser2020) and to be exploratory prior to the onset of reproductive demands (Gopnik, Reference Gopnik2020), although these studies do not address the specific role of objects in these ontogenetic niche spaces. Studies of non-human primates have found that social learning is positively correlated with longevity, suggesting that there was selection for increased time for social learning, opportunities to make the most of that learning, and for passing on that knowledge to offspring (Street et al., Reference Street, Navarrete, Reader and Laland2017).

The experimental studies of Iriki and colleagues offer interesting additional insights (Iriki & Sakura, Reference Iriki and Sakura2008; Iriki & Taoka, Reference Iriki and Taoka2012). By provisioning captive Japanese macaques – who also use a range of tools in the wild (Leca et al., Reference Leca, Gunst and Huffman2008) – with rakes with which to obtain food rewards, they were able to demonstrate how using such tools actually modifies neural connections within the lifetime of a single individual. Similar neural plasticity has been demonstrated for humans by Mithen and Parsons (Reference Mithen and Parsons2008), whose experiment focused on musical ability, and by Stout, Hecht, and colleagues, focusing on flint tool production (Hecht et al., Reference Hecht, Gutman, Khreisheh, Taylor, Kilner, Faisal and Stout2015; Stout & Chaminade, Reference Stout and Chaminade2007). These studies, working from evolutionary questions, align with numerous advances in the cognitive and medical sciences that have demonstrated how different forms of neural plasticity are fundamental to acquisition, mastery and specialisation in complex motorial, cognitive and social activities (e.g. Blakemore & Frith, Reference Blakemore and Frith2005; Kolb & Whishaw, Reference Kolb and Whishaw1998; Magee & Grienberger, Reference Magee and Grienberger2020).

In humans, such activities commonly involve material culture. Malafouris encapsulates the feedback relationship between material culture and cognition in his example of the ‘Blind Man's Stick’, in a striking parallel between the Japanese macaques’ rakes and the most ubiquitous and probably most ancient of human tools, the stick (cf. Oswalt, Reference Oswalt1976; Rios-Garaizar et al., Reference Rios-Garaizar, López-Bultó, Iriarte, Pérez-Garrido, Piqué, Aranburu and Libano2018). Specifically, Malafouris (Reference Malafouris2013) asks where a blind man's mind ends and his world begins, arguing that it is at the tip of his cane where the tactile is transformed into the visual. Through this example, he advocates for an extended cognition wherein the mind is not relegated to the skull but instead stretches into the material world.

While experimental studies offer valuable insights into the mechanisms of neural plasticity and pruning, they rarely if ever capture such dynamics across multiple generations and in ecologically realistic settings. Processes of neural adaptation respond to social and material cues that channel neural connections into particular, historically and culturally specific forms, which over time may form lineages. The ecologically inherited, constructed niche furnishings serve as combined physical and informational resources that condition the formation of these lineages, where the baseline of what constitutes the inherited ecology shifts in each generation (Figure 2).

Figure 2. A conceptual model of how niche furnishings change over time within (a) beavers and (b) humans. Prior to any niche construction, the organism interacts with an unmodified environment at t, for instance when moving into a new territory. Incipient niche construction begins at t + 1, where many of the niche furnishings can also be seen as the extended phenotype of the organism in questions. At t + 2, the original organism has offspring that are born into a niche that already is modified, including the ontogenetic environment. These furnishings are no longer extensions of the new generations phenotype but rather part of their modified environment. This feedback-rich relationship continues into t + 3 (and t + n), where the original organism is dead but the niche provisioning continuous, now along specific historical trajectories.

Domain-specific innovation through play object priming in Arctic prehistory

Greenlandic prehistory is characterised by a succession of colonisation episodes – the Paleoeskimo (Saqqaq, Independence I/II, Dorset) and subsequent Thule cultures – beginning around 2500 BCE. Although long-lived, the various Paleoeskimo occupations eventually ended around c. 0 CE, albeit with some regional holdouts such as the Late Dorset persisting in some areas until c. 700 CE (Appelt et al., Reference Appelt, Damkjar, Friesen, Friesen and Mason2016). After a lengthy hiatus from about 1200 CE a new cultural complex appeared in Greenland: the Thule. These later peoples are the direct cultural and biological ancestors of living Inuit peoples today (Raghavan et al., Reference Raghavan, DeGiorgio, Albrechtsen, Moltke, Skoglund, Korneliussen and Willerslev2014). Over time, Arctic technologies were refined to include sophisticated weaponry (Grønnow, Reference Grønnow, Morrison and Pilon1994, Reference Grønnow2012), instruments, facilities, sledges and different kinds of watercraft. The two cultures practised broadly similar subsistence economies and experienced similar environmental conditions. Yet Paleoeskimo (especially Saqqaq) material culture is ‘remarkably uniform’ throughout their tenure in Greenland (Gulløv et al., Reference Gulløv, Andreasen, Grønnow, Jensen, Appelt, Arenborg and Berglund2004: 105) and may even have included the loss of, for instance, bow technology in certain regions and periods. This contrasts with Thule material culture, which included much larger, seaworthy umiak boats, but which was also highly dynamic in the development of many diverse harpoon forms, kayak designs and clothing styles – all of which are also pervasively present as play objects (see also Figure 5a–d).

In comparison with the Thule, play objects are scarce in Paleoeskimo contexts. Knuth (Reference Knuth1968), for instance, reports the presence of a ‘toy’ ivory harpoon fore-shaft from Independence II contexts, but this identification could be a misinterpretation of the artefact function based singularly on its small size. Appelt and colleagues (Reference Appelt, Damkjar, Friesen, Friesen and Mason2016: 787 and 792) do note the possibility of some miniature tools from Late Dorset contexts, such as unusably tiny soapstone lamps and harpoon heads, as being toys, but they also acknowledge that their exceptional craftsmanship may preclude a purely play-oriented function (but see Langley, Reference Langley2018). Objects such as miniature weapons and tools, boats and human and animal dolls could have played an important part in the establishment of gender roles and identities (e.g. Fienrup-Riordan, Reference Fienrup-Riordan1983), as well as in the transmission of specific cosmological (i.e. animistic) notions (Fienrup-Riordan, Reference Fienrup-Riordan1994). By the same token, these objects could also have aided in active ecological (e.g. Mithen, Reference Mithen1991; Sugiyama & Sugiyama, Reference Sugiyama and Sugiyama2009) and technological learning and hence guided variation through their inherent mnemonic qualities and as examples of the specific affordances of these complex materials and tools.

This point is all the more evident in the astoundingly rich record of children's material culture from Thule sites (Park, Reference Park1998, Reference Park2005; Park & Mousseau, Reference Park and Mousseau2003) and is further reflected in ethnographic reports of children's behaviour and equipment in Arctic societies (e.g. Hawkes, Reference Hawkes1916). Thule children can quite readily be identified in the archaeological record through, for instance, their spatial signature (off-site miniature tent rings with pebble meat and fat pieces – see Hardenberg, Reference Hardenberg2010; and Cory, Reference Cory2021; Langley, Reference Langley2020), as well as small but fully functional tools and weapons as well as human/non-human dolls (Figure 3; see also Park, Reference Park1998). However, throughout the North American Arctic, it should be noted that ‘dolls’ and figurines – while they certainly served as objects of amusement and learning – also represent other potential functions beyond ‘play’ (see, for instance, Boas, Reference Boas1964: 157–166), having served in numerous ritual contexts as well (Boas, Reference Boas1964: 152).

Figure 3. A diminutive harpoon fashioned from wood, from Ainu Creek site, Urup Island, Kuril Islands, Russian Far East. Photo: Matthew J. Walsh.

Ethnographically, the presence of such diminutive implements and objects is ubiquitous across the North American Arctic, from Alaska (Birket-Smith, Reference Birket-Smith1953; Gubser, Reference Gubser1965; Nelsen, Reference Nelsen1900) across the Canadian Arctic (Balikci, Reference Balikci1970; Boas, Reference Boas1964; Mathiassen, Reference Mathiassen1927) and throughout Greenland (Kroeber, Reference Kroeber1899; Mathiassen, Reference Mathiassen1933). Langley and Litster (Reference Langley and Litster2018) and Kamp and Whittaker (Reference Kamp, Whittaker, Rebay-Salisbury and Pany-Kucera2020) provide extensive reviews of this literature. As Park (Reference Park2005) notes, this difference between the miniature record across Paleoeskimo and Thule cultures cannot be readily reduced to differential preservation. These aspects of material culture are usually discussed in terms of socialisation, gender roles and the sexual division of labour. No doubt, play objects are important in this regard. Given the evident correlation between what kinds of miniatures can be identified archaeologically and the technological domains in which Thule cultures are particularly innovative, their role as salient niche furnishings also stands out. For example, miniature hunting tools such as harpoons and bows-and-arrows mirror their adult prototypes and their use by children is ethnographically directly linked to the learning of specific skills (e.g. Laughlin et al., Reference Laughlin, Heath and Arima1991; Losey & Hull, Reference Losey and Hull2019); miniature bows and tiny ‘toy’ harpoon heads are frequently recovered as part of archaeological assemblages (e.g. Larsen & Rainey, Reference Larsen and Rainey1948). Park and Mousseau (Reference Park and Mousseau2003) show that miniatures smoothly scale up into adult sizes, reminding us that childhood and adolescence are themselves graded life-stages. The abundant presence of differently sized, adult-manufactured and fully functional miniatures of these complex technologies attests to the consistent niche provisioning in these cultural contexts.

Growing up in these two cultures would have afforded different degrees of innovation potential owing to a greater degree of familiarity with the affordances of specific technologies amongst children and adolescents. Ethnographically, youngsters’ learning in Arctic societies is highly experimental and not subject to significant pedagogic interventions by adults. This absence of formal modes of teaching underlines, we suggest, the critical importance of material culture as pedagogical scaffolds, an observation further underlined by the loss of certain specialised technologies among some Paleoeskimo groups (see Maxwell, Reference Maxwell, Gilbert and Gulløv1997; McGhee, Reference McGhee1996; and discussions of implication in Prentiss et al., Reference Prentiss, Walsh, Foor and Barnett2015). From the ethnographic and archaeological evidence, there can be little doubt that object play was vital in the maintenance and refinement of technologies within Thule societies.

Wheels, vehicles and journeys of life and death in the Eurasian Neolithic

Although enshrined in popular culture as perhaps the most iconic invention, the wheel and its application in wheeled vehicles in reality only emerged late in prehistory. The earliest data points for fully fledged wheeled vehicles (vehicle parts, wheel tracks, iconography) are scattered across western Eurasia and all date to the middle of the fourth millennium BCE or slightly later (Burmeister, Reference Burmeister, Hansen and Müller2011; Fansa & Burmeister, Reference Fansa and Burmeister2004; Mischka, Reference Mischka2011). Potentially reflecting a rapid spread of this technology, this pattern has left ample room for debate between mono- and polycentric models of its origin (e.g. Sherratt, Reference Sherratt1996; contra Vosteen, Reference Vosteen1996a, Reference Vosteenb). Pertinent here is the observation that a miniature application of the wheel and axle combination predates the appearance of cattle-drawn carts and wagons by at least a century: in the fourth-millennium BCE Tripolye contexts of the north-western Pontic region (mainly present-day Ukraine), a range of small, zoomorphic ceramic vessels with holes for two axles have been found (Figure 4). While these presumed axle holes may in principle have served other purposes, their character and location on the figurines are most likely explained by the former presence of axles and wheels made from more perishable materials. These objects are generally regarded as precursors of wheeled vehicles (Maran, Reference Maran, Fansa and Burmeister2004; Matuschik, Reference Matuschik, Pétrequin, Arbogast, Pétrequin, van Willigen and Bailly2006). Conflicting or perhaps complementary interpretations of these as ritual paraphernalia, quotidian objects or playthings abound (cf. Langley & Litster, Reference Langley and Litster2018). Yet, regardless of whether they were designed specifically as play objects or with some other intent, they are likely to have been handled and played with by children.

Figure 4. Clay figurine from Late Tripolye context at Karolina, Ukraine. The holes in the legs suggest that this figurine was once wheeled. After Gusev (Reference Gusev1998).

Although not very precisely dated, it is clear that the apparently wheeled figurines from Tripolye contexts predate the first full-scale wooden wheels and date to a period when these societies saw significant cultural and socioeconomic changes and innovations. During the centuries 4100–3600 BCE, Tripolye settlements developed into proto-urban communities (Menotti & Korvin-Piotrovskiy, Reference Menotti and Korvin-Piotrovskiy2012; Müller et al., Reference Müller, Rassmann and Videiko2016) and a number of significant technological changes occurred: new forms of in-house weaving production, novel techniques of large-scale pottery production employing three-channelled pottery kilns and cattle-drawn sledges for the transportation of materials and goods within, around and to the urban area appear (Kirleis & Dal Corso, Reference Kirleis, Dal Corso, Müller, Rassmann and Videiko2016; Korvin-Piotrovskiy et al., Reference Korvin-Piotrovskiy, Hofmann, Rassmann, Videiko, Brandtstätter, Müller, Rassmann and Videiko2016; Müller & Rassmann, Reference Müller, Rassmann, Müller, Rassmann and Videiko2016).

The latter technology is particularly significant because the co-existence of an animal-drawn (non-wheeled) form of vehicle and wheeled miniature items presented preconditions for developing full-scale wheeled vehicles. However, the two needed to be combined creatively, and in this process inquisitive and entrepreneurial children and adolescents could plausibly have played a role. Youngsters in this cultural niche had probably observed, handled and played with wheeled objects and thus acquired some familiarity with the mechanical affordances of wheel-and-axle technology, and they would have been equally familiar with the affordances of an animal-drawn transport. Fusing these two sets of experience and translating the combination into a new, operational technology would have required both the ability and openness to associate the two cognitively, and the time, freedom and curiosity to follow up with trial-and-error exploration at full scale. In contrast to Tripolye adults, some youngsters in this context are likely to have simultaneously fulfilled all of these requirements. At the same time, however, youngsters’ interaction with adults was probably key in allowing implementation of innovations in society, requiring both resources and authority held by adults mainly – their tinkering may have laid the foundations for innovations and improvements implemented by adults.

While the origin of full-scale wheeled vehicles in all likelihood lies in the period immediately preceding the middle of the fourth millennium BCE, and while Tripolye is the strongest candidate for the cultural niche in which the development took place, the downstream consequences of this innovation unfolded gradually. It is not until the very end of the fourth millennium that some cultural niches across this vast region are significantly impacted by this technology. From around 3100 BCE, the archaeological finds of wheels and vehicle parts increase dramatically in the Pontic Steppe and in Central and Northern Europe (Burmeister, Reference Burmeister, Hansen and Müller2011; Piggott, Reference Piggott1983) and their importance grows significantly, not least in groups that adopt mobile pastoralism around this time (Anthony, Reference Anthony2007; Johannsen et al., Reference Johannsen, Nielsen, Jensen, Furholt, Grossmann and Smyt2016; Schroeder et al., Reference Schroeder, Margaryan, Szmyt, Theulot, Włodarczak, Rasmussen and Allentoft2019).

Interestingly, there follows a secondary process of innovation relating to the wheel, resulting directly from the presence of wheeled transport in the cultural niche. In some groups, the dead are now buried in or with wheeled vehicles, sometimes complete with a team of draught oxen – ‘animal machines’ that were themselves under direct niche-constructing pressures that are observable in their skeletons (Johannsen, Reference Johannsen2006, Reference Johannsen and Both2007) – indicating that the transition from the world of the living to that of the dead is undertaken by wheeled vehicle. A particularly clear manifestation of this belief is found on the Jutland Peninsula (Denmark), where individual cart burials accumulate into linear cemeteries alongside the roads used by the living (Johannsen et al., Reference Johannsen, Nielsen, Jensen, Furholt, Grossmann and Smyt2016; Johannsen & Laursen, Reference Johannsen and Laursen2010). This conceptual innovation relates directly to the affordances of wheeled transport and to the experience of such transport in life, i.e. to the niche effects of this technology. Yet the new journey of death represented in the funerary rituals and structures of these communities is an entirely unintended consequence of the development of wheeled vehicles, underscoring the complexity of downstream (material and cognitive) niche effects of technological innovation on subsequent behaviour and innovation.