The target article necessitates reconsideration of the “ancestral null” hypothesis. This hypothesis states that “without positive [archaeological] evidence to the contrary, past humans should not be considered cognitively or behaviorally sophisticated” (target article, Sect. 4, para. 6). The ancestral null is tacitly accepted by the interdependence hypothesis (Tomasello, Melis, Tennie, Wyman, & Herrmann, Reference Tomasello, Melis, Tennie, Wyman and Herrmann2012). Consequently, that hypothesis suggests an unnecessarily vague and temporally inaccurate phylogeny of shared intentionality. An alternative links psychological with ethnographic and fossil data to suggest that shared intentionality may have been favored by persistence hunting in H. erectus.

The interdependence hypothesis describes the phylogeny of shared intentionality. Shared intentionality is the motivation and skills to align mental states. Arguably, shared intentionality underlies humans’ derived communicative and cooperative behavior (Tomasello, Reference Tomasello2019). The evolutionary model is the stag hunt (Skyrms, Reference Skyrms2004). In stag hunts, cooperation maximizes individual payoff. Thus, partners have a stake in ensuring that both cooperate adequately (Roberts, Reference Roberts2005). However, partners are fallible, and their intentions are uncertain. Thus, cooperation is risky. Sharing intentions by communicating reduces uncertainty and, therefore, risk. Consequently, cooperators can outcompete noncooperators by sharing intentions with likeminded partners.

As an evolutionary narrative of shared intentionality, the above has two shortcomings. First, there is unnecessary vagueness about the form of stag hunts. It is “unnecessary” because some evidentiary sources, like ethnographies and fossils, are neglected. Tomasello, Hare, Lehmann, and Call (Reference Tomasello, Hare, Lehmann and Call2007) demonstrate the cooperative function of white sclera, but sclera do not fossilize. Instead, Tomasello et al. (Reference Tomasello, Melis, Tennie, Wyman and Herrmann2012) only consider archaeological evidence for interdependent hunting. Moreover, all hypothetical examples include only material culture (e.g., spears). This suggests acceptance of the ancestral null, which invites inaccurate phylogenetic timelines. Indeed, this is the second problem. Tomasello et al. (Reference Tomasello, Melis, Tennie, Wyman and Herrmann2012) suggest that interdependent foraging first occurred in H. heidelbergensis (also, Tomasello, Reference Tomasello2014, Reference Tomasello2019, Reference Tomasello2022). This understates the antiquity of shared intentionality.

Rather, shared intentionality may have been favored by persistence hunting in H. erectus. Persistence hunters use endurance running to chase prey to exhaustion (Carrier, Reference Carrier1984). Exhausted prey can be safely killed with simple weapons, for example, rocks. Persistence hunting by H. erectus explains its cursorial adaptations (Bramble & Lieberman, Reference Bramble and Lieberman2004), patterns of meat acquisition, and encephalization (reviewed in Pontzer, Reference Pontzer2017). In short, if (i) contemporary persistence hunting is an interdependent, energetically profitable subsistence strategy and (ii) fossil evidence does not preclude persistence hunting by H. erectus (i.e., insofar as cursorial traits are attested), then shared intentionality may have been favored by persistence hunting in H. erectus.

Is there evidence for (i)? Ethnographies portray persistence hunting as interdependent and profitable. Lieberman et al. (Reference Lieberman, Mahaffey, Cubesare Quimare, Holowka, Wallace and Baggish2020) argue that effective persistence hunting requires cooperation. Those authors discuss Rarámuri individuals’ recollections of persistence hunting. All persistence hunts were performed by groups. Sometimes, groups drove prey into traps while others ran alongside to prevent its escape, or else groups chased prey to exhaustion and killed it with rocks. Aboriginal peoples have reportedly jointly pursued kangaroos (Tindale, Reference Tindale1974). One individual chases while the other intercepts the kangaroo's path, reversing roles until the kangaroo becomes exhausted. Liebenberg (Reference Liebenberg2006) discusses !Xo and /Gwi hunters (central Kalahari, Botswana) alternating tracker or chaser roles in persistence hunts. All three citations suggest the importance of (joint commitment to) shared goals and individual roles for persistence hunting. Moreover, tracking is often collaborative. Liebenberg (Reference Liebenberg1990) discusses how “tracks are commented on by… gesture [and] soft whispers” so as not to spook prey (p. 55). He argues that “success… depends on how quickly the animal can be tracked down” (Liebenberg, Reference Liebenberg1990, p. 61). This implies ecological pressure for derived communicative skills and motivations, like shared knowledge (“What do we both know and what do only I know?”), the motivation to increase shared knowledge (e.g., via linguistic reference; Vasil, Reference Vasil2023), and false belief and joint reasoning (“Why should we follow which tracks?”). The metabolic expense of using the body as a foraging tool – compared to, say, shopping carts or guns – probably contributes to the rarity of persistence hunting, today (Lieberman, Bramble, Raichlen, & Shea, Reference Lieberman, Bramble, Raichlen and Shea2007). However, persistence hunting can kill prey large enough to require cooperation to transport and consume (for prey sizes, see Morin & Winterhalder, Reference Morin and Winterhalder2024). Consequently, persistence hunting is likely to be profitable, on average (Morin & Winterhalder, Reference Morin and Winterhalder2024). Altogether, this supports (i) that contemporary persistence hunting is interdependent and energetically profitable.

Is there evidence for (ii)? H. erectus possessed derived cursorial traits, like a plantar medial longitudinal arch (reviewed in Holowka & Lieberman, Reference Holowka and Lieberman2018). The arch is maintained by bony and soft tissue structures (Huang, Kitaoka, An, & Chao, Reference Huang, Kitaoka, An and Chao1993). The vertical loads of running compress bony structures, and compression stretches soft structures, like the plantar aponeurosis (Ker, Bennett, Bibby, Kester, & Alexander, Reference Ker, Bennett, Bibby, Kester and Alexander1987). The stretched plantar aponeurosis stores elastic strain energy. Releasing this energy at push-off propels runners forward, like hopping on springs (Holowka, Richards, Sibson, & Lieberman, Reference Holowka, Richards, Sibson and Lieberman2021). This increases locomotor efficiency because elastic strain is generated passively (Alexander, Reference Alexander1991). Importantly, the arch's spring is likely a cursorial adaptation, and not a spandrel associated with bipedal walking. Walking compresses the arch (Caravaggi, Pataky, Günther, Savage, & Crompton, Reference Caravaggi, Pataky, Günther, Savage and Crompton2010). However, only running causes compression sufficient to engage the spring mechanism (Stearne et al., Reference Stearne, McDonald, Alderson, North, Oxnard and Rubenson2016). Moreover, similarities in the walking kinematics of humans and nonhuman apes (who lack the arch) suggest that the arch is not required for hominin stiff-lever walking (Holowka & Lieberman, Reference Holowka and Lieberman2018); and the transverse arch greatly stiffens the foot during walking and predated Homo (Venkadesan et al., Reference Venkadesan, Yawar, Eng, Dias, Singh, Tommasini and Mandre2020). Thus, evidence of a longitudinal arch in H. erectus suggests cursorial adaptedness. Perhaps the arch supported persistence hunting by cooperative H. erectus groups (Hatala et al., Reference Hatala, Roach, Ostrofsky, Wunderlich, Dingwall, Villmoare and Richmond2016). Altogether, this supports (ii) that fossil evidence does not preclude persistence hunting by H. erectus.

In conclusion, shared intentionality may have been favored by persistence hunting in H. erectus. This “first step” in the evolution of shared intentionality enabled its “second step” (Tomasello et al., Reference Tomasello, Melis, Tennie, Wyman and Herrmann2012), partially preserved in the archeological record as evidence of “mosaic cumulative culture” and “technological ratchets” (target article, Sect. 13, para. 4). This discussion excluded necessary questions of life history. Perhaps shared intentionality was also favored by alloparenting in H. erectus (O'Connell, Hawkes, & Blurton Jones, Reference O'Connell, Hawkes and Blurton Jones1999; see Hrdy, Reference Hrdy2009; relatedly, Lieberman, Kistner, Richard, Lee, & Baggish, Reference Lieberman, Kistner, Richard, Lee and Baggish2021).