Osiurak and Renaud defend the bold position that the invention of cumulative technological culture (CTC) does not depend so much on social processes than on intrinsic cognitive abilities, or technical-reasoning skills (TRS). This is clearly bold, because their position goes against the most dominant schools of thoughts, represented by Boyd and Richerson (Richerson & Boyd Reference Richerson and Boyd2005), Tomasello (e.g., Tomasello et al. Reference Tomasello, Carpenter, Call, Behne and Moll2005), their associates, and many others in the field, who explain the emergence of CTC through particular social-learning processes such as imitation and teaching (Tennie et al. Reference Tennie, Call and Tomasello2009). I very much agree with the authors that too much weight has been given to social-learning processes at the expense of other cognitive mechanisms, leading to circular arguments that only grant CTC to species displaying both imitation and teaching, and therefore excluding all nonhumans from CTC by definition (Gruber Reference Gruber2016). Yet, although I am supportive of decoupling CTC from its social components and generally agree that TRS must occupy a central place in the debate, this discussion should not be limited only to modern humans. Although the authors discuss intriguing findings in some nonhuman species at the beginning of section 2, they soon assert that (all) nonhumans have limited cognitive abilities connected to tool use, following Vaesen (Reference Vaesen2012), and evacuate rather than solve the problem of CTC in nonhumans. More problematically, the wording changes throughout the article, with CTC “at best, minimal in nonhumans” (sect. 2.2.1., para. 1) becoming completely “absent” (sect. 2.2.4., para. 1). There is, yet, a crucial difference between absence and acknowledging that some nonhuman behavioral variation is close to CTC, triggering discussion of limitations. In particular, such limitations probably applied to our early hominin ancestors, which are themselves seldomly discussed by the authors. This is problematic because these hominins likely built the foundations of our amazing and unparalleled CTC on the existing biological building blocks. In fact, the refocus on TRS could be a good way to approach behavioral continuity for CTC. My comment, thus, aims to extend the model of the authors to nonhumans, allowing sustaining an evolutionary continuity with our own ancestors.

Although impressive findings in animal tool cognition are discussed, they are only used to promote the idea that nonhumans are cognitively limited to understand their own tool use (sect. 2.1, para. 2). This position fails to acknowledge the large inter- and intra-specific variation between animal species, which is in fact suggestive of different cognitive skills, even within the same population, not so different than what the authors propose for modern humans. Failing to acknowledge differences in animal tool use from the most constrained and hardwired to the most “creative,” as proposed by Call (Reference Call, Sanz, Call and Boesch2013), appears to me a biological mistake. In addition, starting from a null hypothesis where tool use might be cognitive to some extent in some species appears theoretically more sound in terms of evolutionary continuity, without hindering the fact that human tool use plays in a whole different category. I will exemplify this with one particular property of CTC: opacity, which the authors discuss in the context of theory of mind. Opacity, following Csibra and Gergely (Reference Csibra and Gergely2009; Reference Csibra and Gergely2011), has been hailed as a marker of CTC in humans, as opposed to apparent “transparent” animal cultures. Yet, some animal cultural traits, such as termite-fishing in the Goualougou chimpanzees (Pan troglodytes, Sanz et al. Reference Sanz, Call and Morgan2009), combine complexity with limited information for the new learner to acquire visually (or “transparently”). Intriguingly, a recent report shows that this community displays more scaffolding of knowledge compared to a community with less complex tool use (Musgrave et al. Reference Musgrave, Lonsdorf, Morgan, Prestipino, Bernstein-Kurtycz, Mundry and Sanz2020). In field experiments with wild chimpanzees in Uganda, stick-savvy Kanyawara chimpanzees did not evaluate sticks as opaquely as the stick-less Sonso chimpanzees (Gruber et al. Reference Gruber, Muller, Strimling, Wrangham and Zuberbühler2009; Reference Gruber, Muller, Reynolds, Wrangham and Zuberbühler2011). Even more surprisingly, in the same experiment, male bearded capuchin monkeys (Sapajus libidinosus) do not evaluate sticks with the same opacity as females (Cardoso & Ottoni Reference Cardoso and Ottoni2016). Hence, there is evidence of variation in cognitive opacity between individuals across and within nonhuman species.

Given these results, it seems difficult to argue that modern humans are completely unique in having cognitive tool use, and subsequently, CTC. If this were true, then, an important question would be to determine when it occurred during our evolution. The classic view, heavily biased towards social learning, sees CTC mostly arising around the Acheulean period: It is argued that the complexity of bifaces makes it necessary for imitation to have been present to allow conserving the regularity in manufacturing these tools (Morgan et al. Reference Morgan, Uomini, Rendell, Chouinard-Thuly, Street, Lewis, Cross, Evans, Kearney, de la Torre, Whiten and Laland2015). Yet, imitation is also used to justify the subsequent stasis that is observed for more than a million year (Tennie et al. Reference Tennie, Premo, Braun and McPherron2017), being in fine responsible for both fostering and stalling technological progress. This illustrates that imitation or advanced social-learning skills remain mostly convenient one-fits-all explanations in the circular reasoning mentioned above. If, as proposed by the authors, one discards the idea that social mechanisms such as theory of mind or metacognition are essential to CTC and should rather be seen as “boosters”; and one agrees, as I argue here, that there is some kind of cognitive technological skills in nonhumans, there is much less of a theoretical conundrum regarding the appearance of sudden cognitive skills uniquely in the human lineage that allowed CTC to emerge. In addition, the problem of stasis mentioned above is no more about social learning, but rather about cognitive limitations. As illustrated by the authors, TRS also allows testing CTC with neurological data: If the endocasts of early possible human ancestors suggest that the parietal areas responsible for a cognitive appraisal of tools were already developed (Bruner Reference Bruner2018), we may conclude that CTC was already present in these species. The same reasoning could be applied to nonhumans that have controversially been granted CTC to assess whether homologous (Hopkins et al. Reference Hopkins, Meguerditchian, Coulon, Misiura, Pope, Mareno and Schapiro2017) or analogous brain structures (e.g., for corvid species) may indeed sustain CTC in nonhumans, offering an alternative but complementary evolutionary route to explain the unique rise in behavioral complexity in our lineage.