The target article argues that the disciplines drawing on the archaeological record should reformulate their null hypothesis. This should apply to language evolution, where it is still common to assume the absence of language in the absence of hard artefactual evidence – an inferential strategy that Stibbard-Hawkes clearly shows to be invalid. In order to change this, we propose that (a) based on biological continuity, we should start from the assumption of language-readiness in extinct hominins, and (b) we should integrate the existing evidence to assess the likely extent of the realization of this readiness and/or to question the null hypothesis.

The explanans for language evolution remains highly influenced by overall ontological, epistemological, and methodological assumptions, as is captured by Jackendoff's (Reference Jackendoff, Larson, Deprez and Yamakido2010) slogan: “your theory of language evolution depends on your theory of language.” For example, in the field of archaeology, “language” is often tacitly assumed to be akin to modern language (for a discussion, see, e.g., Botha, Reference Botha2010), likely due to the fact that modern humans are our only point of reference, and by that token, an inevitable starting point of comparisons with the extinct species. “Language,” however, can be construed in a variety of widely different ways, which – as our earlier work shows (Wacewicz, Żywiczyński, Hartmann, Pleyer, & Benitez-Burraco, Reference Wacewicz, Żywiczyński, Hartmann, Pleyer and Benitez-Burraco2020) – form a broad family-resemblance mosaic that simply cannot be reduced to a single “correct” definition. In short, speaking of “language” sensu largo is often unhelpful, and we need to use more precise terminology.

A particularly useful distinction is that between the biological or “somatic” readiness for language and the non-biological scaffolding, the former understood as a set of organism-internal traits transmitted mostly through biological inheritance that are necessary but not sufficient to develop language, and the latter as a set of largely organism-external variables – social, motivational, cultural, etc. – that make it possible to develop language based on the former. This is already foreshadowed in Stibbard-Hawkes' distinction between “somatic” and “cultural” models and is in fact present in standard language evolution models. A prime example is Arbib's (Reference Arbib2012) “language-ready brain,” which captures the idea of a minimal cognitive endowment necessary to use a language-like communication system (cf. also, e.g., Burkart, Martins, Miss, & Zürcher, Reference Burkart, Martins, Miss and Zürcher2018, for the importance of biologically grounded adaptations for cooperativity as another sine qua non). But more generally, most scenarios endorsing the hypothetical stage of protolanguage (e.g., Scott-Phillips & Kirby, Reference Scott-Phillips and Kirby2010) assume a relatively greater role of biological evolution in molding a hominin phenotype that becomes capable of using protolanguage, and after that, a relatively greater role of cultural evolution or other external factors (e.g., “language-ready social settings,” Pleyer & Lindner, Reference Pleyer, Lindner, Cartmill, Roberts, Lyn and Cornish2014). The difference between the internal versus external conditions for language has non-trivial consequences. For example, they differ in the rate of change, with the external scaffolding being relatively faster to change, but the evolution of biological language-readiness being relatively slower-paced (Chater, Reali, & Christiansen, Reference Chater, Reali and Christiansen2009). Biological language-readiness can thus be reasonably assumed to have a deep past; hence, we propose that based on biological continuity, it is more parsimonious to assume its presence rather than absence (of course as a default defeasible with evidence).

With this new null hypothesis in mind, the role of the archaeological evidence should also be reassessed. The target article demonstrates how little of the actual material culture would be preserved from modern hunter-gatherer societies, documenting the dangers of “negative” inferences from archaeological material, that is, from the absence of material record to the absence of the underlying cognition. In light of this, we see archaeological evidence as having a primarily confirmatory role, that is, mandating inferences from its presence but not absence: archaeological material should aim to confirm the likelihood of the realization of the language capacity. When this likelihood is small, prehistoric hominins are not denied the capacity per se (or even its realization, as it might simply not be detectable through archaeological remains). This leaves a “gray zone” in which the likelihood of language use in hominins can be probabilistically evaluated with non-absolute but increasing certainty, something that is impossible in dichotomous thinking about the presence versus absence of language.

More generally, we propose that different evidence can contribute to assessing different parts of the new null model and its consequences. The type of evidence most relevant to the assessment of our proposed null – that is, that as a default, extinct hominins should be assumed to be language-ready – is mostly the anatomical, genetic, fossil, and so on, evidence informative about the extent of biological continuity. On the other hand, the totality of available interdisciplinary evidence must be used to estimate the potential use of language by these hominins. Language evolution research is by nature fundamentally interdisciplinary (e.g., Christiansen & Kirby, Reference Christiansen and Kirby2003; Fitch, Reference Fitch2010), meaning that many disciplines play important roles in providing pieces to the puzzle of language evolution (Mithen, Reference Mithen2024). For example, research in comparative cognition as well as animal cognition and communication have an important role in specifying the evolutionary platform on which the evolution of the language-ready brain built on (e.g., Berthet, Coye, Dezecache, & Kuhn, Reference Berthet, Coye, Dezecache and Kuhn2023; Tomasello, Reference Tomasello2008; Zhang & Pleyer, Reference Zhang and Pleyer2024). Further, experimental research in the cultural evolution of language has important contributions to make in specifying the social and cognitive dimensions that support the emergence of communication systems (Delliponti et al., Reference Delliponti, Raia, Sanguedolce, Gutowski, Pleyer, Sibierska and Wacewicz2023; Müller & Raviv, Reference Müller and Raviv2024; Nölle & Galantucci, Reference Nölle, Galantucci and Ibáñez2023; Roberts, Reference Roberts2017; Tamariz, Reference Tamariz2017). However, the target article serves as an important reminder that we have to determine which strands of evidence from different disciplines can constrain hypotheses on language evolution (Johansson, Reference Johansson2005) and how they can be used to advance causal hypotheses that can be empirically investigated (Roberts et al., Reference Roberts, Killin, Deb, Sheard, Greenhill, Sinnemäki and Jordan2020). Most importantly, Stibbard-Hawkes's findings reiterate that we have to critically re-assess which inferences can be drawn from existing evidence not only for archaeology, but for all disciplines involved in investigating the evolutionary emergence of language (e.g., Botha, Reference Botha2020; Botha & Everaert, Reference Botha and Everaert2013).