7-minute read

Neanderthals have enjoyed quite the renaissance in the last decade or so, with research attributing skills and capacities to them once considered uniquely human. One of the most contested claims in this arena is language. Since (spoken) language does not fossilise, nor leave material traces in the archaeological record, the case for Neanderthal language relies on indirect evidence. In this book, linguist Rudolf Botha takes a hard-nosed look at why this matter is so controversial and offers a framework to properly tackle it.

The introduction to Neanderthal Language might intimidate as it immediately introduces Botha’s conceptual framework and attendant terminology. The book is as much a review of the many claims put forth as it is about the nature of evidence and how to evaluate claims. This is where it all gets a bit meta, so bear with me.

In his previous book, Language Evolution, Botha introduced what he calls the Windows Approach. Basically, since there is no direct evidence available to study the evolution of language, you have to rely on other phenomena for which you do have direct evidence, and then convincingly argue that these offer a window on language evolution. For your inferences and conclusions to be sound, three conditions need to be met. One, your conclusions need to be pertinent: whatever you are describing (here, language) should be properly identified. Two, your inferences should be grounded: grounded, that is, in accurate data that is informative of the phenomenon you are trying to study. And three, your inferences should be warranted: you should be able to justify why you can draw on this indirect evidence to say something about language evolution, i.e. you require a so-called bridge theory. Are you still with me? The first two chapters will require you to wrap your head around this framework, but stick with it, it is quite intuitive.

Botha evaluates two groups of claims that archaeologists and anthropologists use to argue for Neanderthal language: behaviours that are symbolic (personal ornamentation, cave art, body decoration, and burials) and non-symbolic (making stone tools, teaching it, and hunting big game). The last chapter shortly touches on biological attributes such as genetics (e.g. the gene FOXP2) and brain regions, but that is not the focus of this book*.

“[Claims for language rely on] complex inferences: you do not just jump from finding objects at a dig site to the conclusion of language.”

All of these claims rely on what Botha calls complex inferences: you do not just jump from finding objects at a dig site to the conclusion of language. So, for the personal ornamentation argument, the chain of inference runs something like this: we find objects associated with Neanderthal skeletons, we conclude these were worn as ornaments, from which we conclude in turn that they were treated as symbols, from which we conclude in turn that Neanderthals had language. The problem is that upon close inspection most claims already fall at the first hurdle, writes Botha.

The examples of cave art attributed to Neanderthals might have been made by other hominids; we just cannot date them accurately enough. Plus, their meaning remains a mystery—maybe they were symbols, maybe something else. The blocks of manganese dioxide yielding black pigment might have had other explanations than Neanderthal make-up, fire-starters being one likely candidate. And those burials were probably intentional, but the lack of clear grave goods undermines their symbolic status as funerary practices. Out of this lot, only personal ornamentation fares reasonably well. This is the largest chapter as there is much empirical work to make a pretty convincing case that objects found associated with Neanderthal skeletons were, indeed, used as jewellery.

Where all these claims fail is the inference that they are examples of symbolic behaviour. Botha asks probing questions, such as “what are the distinctive properties of symbols according to [these authors]?“, and “To what theory of symbolism do they subscribe?” (p. 54). Many authors are either not explicit about this, do not specify what would count as evidence for or against, define symbolism but present data that has no bearing on it, or, in the case of ornamentation, simply state that ornaments are symbolic by definition, excluding them from empirical verification.

“[…] there is much empirical work to make a pretty convincing case that objects found associated with Neanderthal skeletons were, indeed, used as jewellery.”

Even if, writes Botha, we assume for argument’s sake that these behaviours were symbolic: how do you go from there to language? The often unspoken assumption is “because language is also symbolic”. This lands you into a hornet’s nest of linguistics. I quote Botha: “By what bridge theory is [the inference] underpinned—a theory specifying the ways in which semantically non-combinable cultural symbols and semantically combinable linguistic signs are interlinked?” (p. 105). Without this, it is merely “an arbitrary inferential leap” to go from symbolic behaviour to language. Further problems are the vague and frequently different definitions of what exactly would characterise Neanderthal language, and the failure to distinguish between Neanderthal groups. “The Neanderthal” does not exist: they were not a monolithic entity, but a hominid species that were around for roughly 400,000 years, distributed over a large area. So who, exactly, are we talking about?

It is much the same for the non-symbolic behaviours, according to Botha. The stone-tool claims start with empirical data obtained on modern human volunteers. In the former, there is the assumption that tool behaviour and speech are similar enough that, because we knap stone tools using a series of actions involving hierarchies and recursions, so did Neanderthals, and from there that this also applies to their language. The latter argues that, since teaching stone-tool making by modern humans benefits from verbal instructions, it is likely that Neanderthals used language too. The only claim that Botha deems likely is that of big-game hunting. We have unequivocal evidence of cooperative ambush hunting of large prey, which would have required cooperation, which would have required communication, which would have required language. Even here, though, by mentioning a grammatically simple language such as Riau Indonesian, he points out it need not have been a complex language.

“[Botha] is not unfriendly to the idea of Neanderthal language […] but he calls for better science and offers a framework to do so.”

I mentioned this book gets rather meta, talking of conceptual frameworks and bridge theories, the nature of evidence and the evaluation of claims, and a healthy dollop of linguistics. Fortunately, it is also clearly structured, both within and between chapters, making good use of lists, headers, and figures to make its points. Though aimed at professionals, this book is perfectly accessible to a more general audience.

Books such as The Smart Neanderthal and Kindred have popularised the idea that Neanderthals are not all that different from us, so you might think Botha a bit of a party pooper. Note, however, that he is not unfriendly to the idea of Neanderthal language and explicitly says his criticism does not disprove it, but he calls for better science and offers a framework to do so. As such, Neanderthal Language provides a healthy dose of caution and scepticism for readers of general works such as the above and should be a valuable guide for practising archaeologists and anthropologists.

---

* After this review was published, a user on Reddit by the name of Denisova commented that they found it “a bit awkward that genetic and anatomical arguments are left out in his assessments.” This is a good point of critique that, in hindsight, I did not raise sufficiently. Botha mentions in his introduction they will not be the focus of this book, but it would have made for a more complete treatment if he had, possibly seeking the help of a co-author better versed in those fields.

Disclosure: The publisher provided a review copy of this book. The opinion expressed here is my own, however.

Neanderthal Language

Other recommended books mentioned in this review: