Some keys to understanding Why Only Us
Why only us (Berwick & Chomsky 2016) was published at a time of great development in research on language evolution (see Wacewicz et al. in press for a detailed review of the main contributions of the last decade). The book proposes a theory of the evolution of language that articulates Noam Chomsky’s ideas on the matter already presented before, both in collaboration (Hauser, Chomsky & Fitch 2002) and alone (Chomsky 2007). Despite the scientific pedigree of its authors, I have the feeling that the book has been essentially misunderstood, and that it has been widely ignored by mainstream research traditions on language evolution. And both facts can be explained because of the same reason: the Chomskyan approach to language evolution contradicts two common assumptions in the mainstream literature on the subject, namely (i) that there is a relationship between the change of languages over time and the evolution of the language faculty, and (ii) that language evolved as a communication system.
From now on I aspire to show in what sense these two assumptions are incorrect, and why the model of the evolution of language presented in Why only us, even though it is a speculation (as all its competitors are), should be considered a notable contribution in clarifying the mystery of the origin of our most distinctive attribute as a species.
Language change is not language evolution
We know that languages change over time. Therefore, it is legitimate to ask whether the evolution of languages over time was responsible for the evolution of language in our species. At first, this seems like a common-sense idea. Thus, since we know that Spanish comes from Latin, and we know that Latin comes from Proto-Indo-European, we can imagine that modern human languages evolved through linguistic changes of that type from a kind of primitive pre-human languages.
However, this scenario is implausible. We know what linguistic changes are like, we know how they affect languages, and therefore we know that they cannot explain how the human faculty of language arose, because linguistic changes are cultural changes (see the previous post on this blog for a more detailed argument).
But current human language cannot be explained as a set of historical or cultural changes that have altered the language, for example, of chimpanzees (our closest living ancestors): if that were all, we could not explain why chimpanzees cannot learn human languages (even when intensively trained to do so). Therefore, some changes in the evolution of the human brain must have occurred that explain that all humans (except severe pathology), and only humans, are capable of learning and using languages such as Spanish, or Russian, or ASL, that is, human languages.
The typical reaction according to which these changes (those that explain why people can learn a language while chimpanzees cannot) are changes in general cognition -and not specific to the language faculty- is unsatisfactory if the changes in general cognition responsible for that difference are not specified. Furthermore, any biological change that leads from a brain that cannot learn a language to one that can is, by definition, a part of the biological evolution of the human faculty of language, which cannot, therefore, be reduced to cultural changes.
Language did not evolve for communication, but for thought
This one is undoubtedly the most relevant thesis of Why only us, and the one that most directly confronts both the common-sense vision, and the vision of most prominent modern traditions in language evolution research.
It should be noted that, in reality, what Chomsky stipulates is that one relevant property of human language (discrete infinity) did not evolve in the service of communication, but rather in the service of thought. Why only us adopts a very restrictive notion of the term language in the expression language evolution: only the recursive computational system (the syntax) that provides productivity in any human language. (Undoubtedly, using the term language for an internal computational system is another reason why the theory presented in Why only us is often misinterpreted.)
According to the famous model of Hauser, Chomsky & Fitch (2002), this computational component forms, together with the conceptual-intentional system and the sensory-motor system, the human faculty of language. The central hypothesis is that the evolution of language as a specific trait of our species essentially consists in the development of this (very simple) computational system within a more complex cognitive organ, whose other components are widely shared with other species and have an independent evolutionary trajectory.
More specifically, the Chomskyan model conjectures that the relationship between the computational system and the other two systems is asymmetric: the computational system would be adaptively optimized for its interaction with the conceptual-intentional system and would have formed, together with it, an internal language of thought (ILoT), whereas the relationship between the ILoT and the sensory-motor component (a necessary connection for the “externalization” of the ILoT and its use for communication) would be ancillary. As Chomsky noted, “the earliest stage of language would have been just that: a language of thought, used internally” (Chomsky 2007: 13).
By adopting this perspective, Why only us conflicts with the bulk of research in this area, which assumes that language evolved from ancient communication systems. But the hypothesis that language first evolved as an internal system of thought, and not as a system of communication, has a fundamental advantage: it better explains why the most clearly distinctive property of human language, discrete infinity, emerged.
Indeed, all human languages have a recursive computational component that allows combining a finite set of elements and producing a potentially infinite set of concepts, ideas, and thoughts from them. This ability, which, as far as we know, is unique to our species, is a good candidate to be the facilitator of one of the most characteristic properties of our cognition: the possibility of creating mental representations of entities that do not exist in the world, which are fictitious, but in which we can believe, such as gods, nations, laws or money. It has been aptly argued that humans are basically hunter-gatherers capable of believing in fictitious entities (Harari 2014), or that we are little more than apes capable of cumulative cultural knowledge (Stewart-Williams 2018), but both capabilities are only possible thanks to the core property of language.
Of course, the use of language for communication is essential to explain the development of human cognition and culture. But this is not questioned by Chomsky’s model, which simply states that the communication function does not explain the evolution of the central and distinctive component of human language or the properties it has.
More than Nature needs?
The evolution of a recursive syntax (“more than nature needs”, in the Shakespearean terms used by Bickerton 2014) makes adaptive sense if it evolved for the benefit of thought, but not if it evolved for the benefit of communication. A recursive syntax is not essential for communication, but it is essential for building complex thoughts that can be interesting to communicate.
The theory that human language evolved as a communication system has the difficulty of explaining what circumstances in the life of Homo sapiens about 200,000 years ago (or of any other ancestral species) would explain the need for a communication system more sophisticated and versatile than the one already available. Any adaptive explanation based on communication requires that previously (or at least simultaneously) there has been an equally sophisticated development of internal language or thought. Otherwise, you cannot imagine any scenario in which adding a recursive syntax could actually be adaptive. A sophisticated communication system is only necessary (and therefore adaptive) if there are sophisticated thoughts to communicate. The idea that the complexity of the communication system (the code) precedes the complexity of the message to be communicated (or that the former is responsible for the latter) should be considered surprising, however widespread it may be.
This problem is analogous to what Bickerton (2014) called “Wallace’s problem” (that human beings have much more brain capacity than is necessary to survive). Indeed, the co-discoverer of the theory of evolution by natural selection had to contradict himself when considering human cognition, since he could not imagine how the development of so-called “higher capacities” such as language, music, mathematics, or art could have been adaptive for an archaic primate. But the problem is mitigated if the evolution of human language is conceived as the evolution of a mosaic of relatively independent systems with different functions and evolutionary trajectories.
The uniqueness of human language is not that it is a very sophisticated communication system (although it can certainly be used as such), but that its computational core is a fundamental internal tool for cognition. It is not a part of culture, but the fundamental tool to build it. Thus, it could be concluded, contrary to the prevailing vision in the field of language evolution research, that human beings do not have language because we are more intelligent, but that we are more intelligent because we have language.
Philosophy of Linguistics 