Chapter 6 asks what language evolved for. The first half starts with how to pose that question.
What adaptationism is (6-1)
Darwin’s theory of natural selection is persuasive because it explains, in naturalistic terms, why organisms look so well fitted to their environments (p. 255). Applying it to the human mind gives evolutionary psychology; the two together are called adaptationism. The author says the term is easily misunderstood, so he clarifies it first (p. 256).
The point is that adaptationism is not the claim that every trait is an adaptation. It is a research tool. Suppose a trait has some function. Then one can ask: what design would serve that function? How does the actual design differ from what other functions, or no function, would predict? Testable predictions follow (p. 257).
Example: motion-induced blindness
The author’s example is motion-induced blindness. When a pattern moves behind a clearly visible object, the object drops out of awareness. The usual reading is that this is a flaw of vision.
An alternative hypothesis: the eye can have blind spots (parts of the visual field that are missing), and the visual system fills them in from the surroundings. On this view motion-induced blindness is the result of the filling-in process treating a moving pattern as background. This is testable. If filling-in is the cause, it should not occur in other situations. When the experiments were run, the results did not fit the flaw account. The visual system seems to be designed to ignore small objects in a changing field (pp. 257–259).
His conclusion: adaptationism yields testable predictions about function, and experiments then get at the actual mechanism. Language-evolution research should use the same method (p. 259).
Pinker and Bloom (6-2)
A classic of language evolution is Pinker and Bloom’s paper “Natural Language and Natural Selection” (1990). Against the mainstream linguistics of the time, especially Chomsky, they argued (pp. 260–261):
- The language faculty is a complex biological adaptation.
- Therefore natural selection is the only thing that can explain its design.
Chomsky’s camp held that nothing definite can be said about natural selection here (p. 260).
The author sees two premises in the argument (pp. 261–262).
- If nature shows evidence of design, the only naturalistic way to explain it is natural selection.
- The language faculty really is an innate cognitive mechanism of humans (Universal Grammar).
He accepts the first. The problem is the second.
The premise of Universal Grammar
Whether Universal Grammar exists is still disputed. The Chomskyan argument is that children could not acquire language from the data they meet alone. Against this stands the reply that data alone suffice (pp. 262–263).
The author stresses that the issue is not whether humans can acquire language; they clearly have that disposition. The issue is the content of the disposition, in particular whether it contains something specific to grammar and syntax (p. 263).
His logic then runs (p. 264):
- If a mechanism like Universal Grammar exists, it is meaningful to ask whether it is a product of natural selection.
- But even so, the important question is when and why it evolved.
- If ostensive communication (chapter 2) came first, natural selection could act on mechanisms that help language acquisition. Whether that is Universal Grammar stays an empirical question.
Is cultural attraction an alternative to Universal Grammar?
Cultural attraction from chapter 5 (article 281) can explain why languages resemble one another without assuming Universal Grammar, since it works at a different level from biological evolution (pp. 264–265).
The author calls the “biological versus cultural evolution” dichotomy a mistake (pp. 265–266). The two answer at different levels.
| Level | Question |
|---|---|
| Ultimate | Why did a trait (bias) spread in the population? (natural selection) |
| Proximate | By what mechanism does the trait work? (cultural attraction, etc.) |
In one computer simulation, for example, agents with a weak bias toward structured languages are subject to natural selection. The widespread structure of languages can be explained by combining the two as answers at different levels.
He likens this to Mendelian genetics and Darwin’s natural selection. They once seemed opposed. Later Mendel’s work was seen to explain the concrete mechanism that produces the variation selection needs, and the synthesis followed (p. 267). For language too, separating the two levels dissolves the quarrel, in the author’s view.
What is linguistic communication for? (6-3)
Next is language’s evolutionary function. Earlier research concentrated on internal cognitive mechanisms (grammar and so on). The author asks instead about use for communication (pp. 268–269).
Ostensive communication consists of two kinds of behaviour, showing and inferring. So whether it is an adaptation must be asked first about producing signals, then about understanding them (p. 269).
Hearers must sort out what grabs their attention; speakers must save the hearer’s effort. These are the design requirements built into the two principles of relevance (article 279). If so, the mechanisms of ostensive communication should be shown to work as designed. The author says the evidence shows they do (pp. 270–271).
Direct and derived functions
The author frames the evolutionary function of linguistic communication as follows.
- Direct function: the function that explains why a trait was reproduced. The direct function of the heart is pumping blood.
- Derived function: the various concrete functions a trait serves on that basis.
He says the direct function of ostensive communication is reading and manipulating other minds (pp. 272–274). “Manipulation” here is not necessarily bad; it covers any attempt to influence a hearer’s behaviour. The many uses of language, such as sex, politics, gossip, courtship, and hunting plans, are all derived functions. So theories that trace the origin of language to one of them (say, “for planning hunts”) fail to make this distinction.
Note
Criticism of adaptationism is plentiful outside this book; I present only the author’s framing. The details of the motion-induced blindness experiments and the Universal Grammar debate are the book’s summary and have not been checked against the original papers.