Osiurak and Reynaud's (O&R's) theory of technical reasoning does not advance our understanding of cumulative cultural evolution. There are four important deficiencies. First, it does not make a novel contribution. We agree that the study of cumulative cultural evolution would benefit from a better understanding of technological reasoning. However, this idea is not novel. A number of authors have suggested that theories of cultural evolution need to better integrate reasoning processes (Claidière et al. Reference Claidière, Scott-Phillips and Sperber2014a; Heyes Reference Heyes2012). It is not enough to present another a verbal argument for the importance of technological reasoning. Real progress requires an account of how such reasoning works. What existing cultural evolutionary work does well is provide a theory that explains how individual-level decisions scale up to population-level dynamics (e.g., Mesoudi Reference Mesoudi2016; Morgan Reference Morgan2016), including experimental studies that investigate individual cognition, phylogenetic approaches that characterize macro-evolution, and theoretical models that link the two. A theory of technological reasoning should link how individuals reason about a task to the innovations they produce, and finally to cultural dynamics. The manuscript is lacking in this regard.

Second, the authors make numerous empirically testable claims without sufficient evidence to move relevant debates forward. For instance, they criticize the cultural niche theory for downplaying the importance of technological reasoning. However, this is a disagreement that can be solved empirically: to what extent does data suggest that cumulative cultural change is constrained and/or directed by technological reasoning? The authors offer frustratingly little hard data for the relevance of technical reasoning. A compelling theory would go beyond stating that technical reasoning is important and, instead, offer evidence that quantifies this importance, shows how technical reasoning shapes cultural change, and makes clear the limitations of cumulative cultural change in its absence. At present, the authors cite only a single (self-authored) study, which does little to convince the reader of the bold statements made in the manuscript.

Third, we question the assertion that individuals can always produce any single solution produced by cultural evolution. To quote the authors, “all of us are smart enough to acquire each piece of information – as well as to produce any kind of innovation – necessary to survive in any single habitat.” This claim is empirically dubious. For example, the Polar Inuit lost a number of important skills during an epidemic in the early nineteenth century, including kayaks, bows, and important aspects of snow house construction. For 40 years or so these people lived without these important tools, and none of the several hundred people in the group were able to reinvent them. When a group of Baffin Island Inuit visited them around 1860, they immediately reincorporated them into their technological repertoire (Boyd et al. Reference Boyd, Richerson and Henrich2011). There are many other examples (Earl & McCleary Reference Earl and McCleary1994). This claim also neglects the fact that cumulative cultural evolution changes the nature of the problems that must be solved. Cumulative cultural evolution not only involves building upon and improving technology from previous generations, it also opens up new niches with new problems. The invention of the wheel, writing, the internet, all introduce new possibilities and new problems to solve. Perhaps more broadly, life in larger groups generated both a need for norms and institutions that maintain cooperation, as well as more potential for innovation. Would a single human, of infinite lifespan, with infinite time, be able to produce the entirety of the current human repertoire? The authors seem to suggest that this would indeed be the case, but competing accounts suggest that it is not individual cognition, but social complexity and inter-connectedness that are responsible for our technological success (Muthukrishna & Henrich Reference Muthukrishna and Henrich2016). Whether the authors' claim is true or not remains a question unanswered here.

Fourth, the authors emphasize technological reasoning as key to technological change in cumulative cultural evolution, but cumulative cultural evolution involves a large range of domains other than technology: cooking, language, norms, art, knowledge, abstract or not, all change as a result of cultural transmission and become increasingly complicated. A great deal of knowledge essential for survival, like knowledge about hunting techniques and plant detoxification, does not involve technical reasoning of the kind described by O&R. The domain of language, as well, is very different. Although technological products are shaped by functional pressures (a good bow shoots far and accurately), languages serve communication with symbols defined by convention. The exact form of words is irrelevant, as long as everybody agrees on the same form, and languages are shaped by expressivity and learnability instead (Kirby et al. Reference Kirby, Cornish and Smith2008). Art is, perhaps, at the opposite end of the spectrum from technology. It is free to change according to our esthetic preferences, without any objective benchmark to measure against, and intuitive physics is no help in identifying successful designs. Nonetheless, complexity increases as a result of cultural transmission processes, without any need for technical reasoning abilities. If technological reasoning is the key to human uniqueness, it must explain this plethora of culturally governed processes that fall outside of the technological domain.

Finally, the authors' focus on technological reasoning as “necessary and sufficient” conditions for cumulative cultural change is misguided. Existing theory clearly shows that cumulative cultural change can occur through a number of mechanisms, ranging from a guided-variation/technical-reasoning model to a selective-transmission-of-random-variation model (Mesoudi et al. Reference Mesoudi, Chang, Dall and Thornton2016). There is no need for “magic bullet” theories, especially when the current state-of-the-art theories suggest that complex cycles of co-evolution of cognition, life-history, social structure and culture are responsible for our species' success (Boyd Reference Boyd2017; Henrich Reference Henrich2016; Laland Reference Laland2017a).