A variety of thinkers have held that human language is the medium of all thought within the human brain and is an essential precondition to us becoming reasoning beings. For example, this claim can be found in Ludwig Wittgenstein's Tractatus Logico-Philosophicus. In a very different school of thought, the literary theorist Kenneth Burke also held that we think in language.
In the present day, some contemporary branches of Critical Theory hold that language is the crucial trait that allows humans to escape from biology and construct their world socially. One formulation of this idea is, "The world must be made to mean." For example, the book Ten Lessons in Theory is an introductory textbook to Critical Theory. In that book's introduction, Calvin Thomas writes, "The phrase 'textual anthropogenesis,' then, involves what's called linguistic determinism, or what I'll call semiotic materialism, the argument, also to my mind permanently revolutionary, that any human reality, and any individual or collective 'subject' thereof, any and all persons, must be made out of language as a specifically 'anti-natural'--unreal or 'anti-real'--form of productive labor." (xxiv).
Steven Pinker, as a neuroscientist and advocate for evolutionary psychology, holds that language is not the stuff of thought. Pinker argues that the mind thinks in its own internal thought-language called mentalese. Pinker started out in the field of neurolinguistics, so he is in an excellent position to make that case. To him, verbal language is a mechanism by which we can package information, originally formulated in mentalese, and share it with other individuals within our language community. It is a communications protocol, not the machine language of the brain's central processor. (To give my own example of non-linguistic thinking, a good chess player can sit down to a game and think hard for hours without using any language at all. A sculptor can do the same while wielding the chisel.)
The cultural evolution of Boyd, Richerson, and Henrich says that cultural learning is the essential precondition for what makes us human, not language per se. Language evolved in a cultural context in which we were already learning from one another.
Henrich argues that the skill for crafting physical tools may have come before verbal language. Sophisticated tool-making requires recursive mental procedures which parallel the recursive mental procedures of language. Therefore, tool-making could have been the cognitive path evolution took to build the genetically programmed neural structures that became the preadaptation for human language (Secret, pp. 250-255).
Language presents a tricky problem for evolution. If a single individual in a proto-human species were to begin to develop language, with whom could she speak? What selective advantage would accrue to this lone proto-speaker? It is difficult to see how language as a stand-alone instinct with specialized brain wetware can evolve. It would be like the sound of one hand clapping.
On the contrary, a social group that has a gifted tool-maker can benefit even if that individual's mental abilities are unique. Everyone else can still use those superior tools, even if they don't have the brains to make them. Brain structures that facilitate a complex Chomskyan grammar might have evolved because of the selective advantage that was conferred on early human communities with singular individuals who could produce superior tools. Only then did the selective advantage for cultural learning take those brain structures and adapt them to create this powerful communication system we call verbal language. Language, once evolved, then turbo-charged cultural learning.
For those theorists who have argued that language is a necessary precondition for humans evolving the capacity to cooperate with one another, Henrich argues that it is the other way around. "For complex communicative repertoires to evolve in the first place, this cooperative dilemma has to have already been at least partially solved. Therefore, language can't be the big solution to human cooperation" (257).
Whether the theory of language-through-toolmaking is confirmed by later research, this scenario at least offers at path by which it could have occurred. In Chomsky's original formulation of his universal grammar, the language instinct popped into existence as a fully formed spandrel, like Athena from the head of Zeus. Pinker's neurolinguistic work shows that the Chomskyan grammar which is universal among humans has some of the "Panda's thumb" imperfections that indicate that it has evolved, but Pinker did not explain how the human brain was able to boot-strap our sophisticated language-related traits from brains that lacked human language. Tool-making-first theories solve the paradox.
Post-script: As for recent research into the evolution of language-enabling structures in the brain and throat, scientists have compared the DNA of Homo sapiens (us) with the DNA of Neanderthals and Denisovans. They have found that the most rapid evolution within the human genome since the split with those other species has targeted the vocal tract and other language-related structures (as well as genes that have flattened the face and eliminated the chimp-like muzzle). See for example: Gokhman, David, et al. "Differential DNA methylation of vocal and facial anatomy genes in modern humans." Nature Communications 11.1 (2020): 1189.
Bibliography
Henrich, Joseph. The Secret of our Success: How Culture is Driving Human Evolution, .... Princeton UP, 2015.
Pinker, Steven. The Language Instinct. William Morrow, 1994.
Thomas, Calvin. Ten Lessons in Theory: A New Introduction to Theoretical Writing. 2nd Edition. Bloomsbury Academic, 2023.
