What you will get from this article
- What the view “language is part of general intelligence” predicts
- What Broca’s aphasia, Specific Language Impairment (SLI), and Williams syndrome each show
- How strong the argument is, and where caveats apply
The answer first. If language were part of general smarts, damage to the brain should make people less capable across the board, language included. The cases Pinker reviews do not fit that prediction. Some people (Broca’s aphasia, SLI) have intact intelligence but broken grammar. Others (Williams syndrome, Denyse) have severely impaired intelligence but fluent language. This “double dissociation,” Pinker argues, suggests that the brain has machinery specific to language.
The logic: two dissociations
Pinker sets out the argument (pp. 45–46).
If, on the other hand, language is just the exercise of human smarts, we might expect that injuries and impairments would make people stupider across the board, including their language.
So what must be checked is both language broken while the rest is fine and, conversely, language fine while the rest is broken (pp. 45–46, 50). Pinker plainly admits that, as of 1994, no one had yet located a language organ or a grammar gene, but says the search is on (p. 46).
1. Broca’s aphasia: intelligence intact, grammar broken (pp. 46–48)
Damage to the lower left frontal lobe (from stroke or a bullet wound, for instance) can produce Broca’s aphasia. Pinker gives two cases.
- The memoir of a man who later recovered. One morning his right arm was numb, he fell to the floor, and he could not make a sound to call his wife. His mind was completely clear and he knew what he wanted to say, but the words seemed wrong and not what he meant (p. 46).
- Mr. Ford, a Coast Guard radio operator who had a stroke at thirty-nine. The neuropsychologist Howard Gardner interviewed him three months later. His speech was slow and effortful, and his sentences were fragments like “Speech … can’t say … talk, you see” (pp. 46–47).
Mr. Ford’s problem was not in controlling his vocal muscles, says Pinker. He could blow out a candle and clear his throat, and writing was as impaired as speech. Most of his handicap centered on grammar itself (p. 47).
- He omitted endings such as “-ed” and “-s” and function words such as or, be, the.
- Reading aloud, he skipped function words but read content words.
- He understood questions whose gist could be deduced from content words (“Does a stone float on water?”) but not ones that required grammatical analysis (“The lion was killed by the tiger; which one is dead?”).
Yet Gardner wrote that he was alert, attentive, and fully aware of where he was and why. Intellectual functions not closely tied to language (knowing right from left, maps, clocks, calculation, carrying out commands) were preserved. His nonverbal IQ was in the high average range (pp. 47–48).
2. Specific Language Impairment (SLI): a grammar deficit that may run in families (pp. 48–50)
Not only adult brain injury: otherwise healthy children may be slow to develop language alone. When obvious causes (retardation, deafness, autism) are ruled out, the label is “Specific Language Impairment” (p. 48).
Pinker describes in detail the British “K” family, studied by the linguist Gopnik and others.
- The grandmother was language-impaired. Of her five adult children, only one was linguistically normal. The other four also had the impairment, and of their twenty-three children, eleven were impaired and twelve normal (p. 48).
- Impaired children were randomly distributed among families, sexes, and birth orders (p. 48).
- If the cause were environmental (poor nutrition, hearing an impaired relative), why would it strike some siblings and spare an age-mate (in one case a fraternal twin)? The geneticists working with Gopnik noted that the pedigree suggests a trait controlled by a single dominant gene (p. 49).
Overall intelligence is not impaired. Many score in the normal range on nonverbal IQ tests, and an unrelated child with the syndrome routinely received the best grade in a mainstream math class (p. 49).
Their problem is language. Conversation is like a tourist struggling in a foreign city: slow and deliberate, asking others to complete sentences. Speech contains frequent grammatical errors, such as misuse of pronouns and of suffixes like the plural and past tense (p. 49). Examples include The boys eat four cookie (missing plural) and Carol is cry in the church (wrong verb form).
In experiments the wug test is characteristic. A child is shown a birdlike creature and told it is a wug, then shown two and asked to complete “There are two ___.” A typical four-year-old says “wugs” at once. One SLI adult became confused (“Wug … wugness, isn’t it? No … pair it up”). For the next animal, zat, she said “Za … ka … za … zackle”; for sas she deduced “sasses,” and, flushed with success, overgeneralized, turning zoop into “zoop-es” and tob into “tob-ye-es” (pp. 49–50). Pinker reads this as the gene somehow impairing the development of rules normal children acquire unconsciously, with adults compensating by conscious reasoning, with clumsy results (p. 50).
The wug test is also a classic demonstration that children do not learn language by imitating parents, since no one has ever heard the plural of “wug” (p. 49).
3. The opposite dissociation: low intelligence, fluent language (pp. 50–54)
Broca’s aphasia and SLI are cases where language alone breaks. But that leaves the reading that “language simply puts greater demands on the brain than other problems” (p. 50). The decider is the opposite dissociation: good language with poor cognition.
Denyse (age fourteen). Pinker presents a transcript of a conversation with the psycholinguist Cromer. Denyse tells, fluently and in refined British speech, how she lost her bank book. But in fact she has no bank account and no boyfriend, has never been to Scotland, is severely retarded, and cannot read, write, or handle money (pp. 50–51). She was born with spina bifida and had hydrocephalus. Children like this, Pinker says, can be significantly retarded yet have unimpaired, even overdeveloped, language (p. 51). The condition has been called “cocktail party conversation” or “chatterbox syndrome” (p. 52).
Williams syndrome (Crystal). Studied by Bellugi and colleagues, Williams syndrome is a rare form of retardation associated, Pinker writes, with a defective gene on chromosome 11 involved in calcium regulation. The children are small, with narrow faces, broad foreheads, and star-shaped patterns in the irises, sometimes called “elfin-faced.” Their IQ is about 50 and they are incompetent at tying shoes, finding their way, retrieving things from a cupboard, adding, or drawing a bicycle. Yet they are fluent conversationalists (p. 52). Crystal, at eighteen, told a story about elephants that contained complex sentences (pp. 52–53).
In tests the children understand complex sentences, fix ungrammatical ones, and show a fondness for unusual words. Ask a normal child to name animals and you get dog, cat, horse, cow, pig; a Williams child gives unicorn, pteranodon, yak, ibex, water buffalo, sea lion, saber-tooth tiger (p. 53).
Pinker’s conclusion
Pinker likens the people met so far to “a field guide to language users”: the highlanders of New Guinea, the Hawaiian laborers, Nicaraguan children, Simon, Sarah, Mr. Ford, the K family, Denyse, Crystal. They show that complex grammar appears across the full range of human habitats.
- You do not need to have left the Stone Age.
- You do not need to be middle class.
- You do not need to do well in school.
- You do not need to be old enough for school.
- Your parents need not bathe you in language or even command a language.
- You do not need the intellectual wherewithal to function in society.
- But you cannot be a competent language user if you lack the right genes or the right bits of brain (pp. 53–54).
Cautions: how far can this go?
- The argument is strong in structure (double dissociation), but whether the two are “fully independent” is a separate question. Later research has pointed out that real cases are blurrier at the edges (a general ToL note, not covered in this series).
- At the time of the 1994 book the causal gene in the K family was unknown. Later work showed that mutations in a particular gene (FOXP2) are involved in the disorder of the K family (known as the KE family), though the usual careful view is that this is not “a grammar gene.” (Also a ToL note outside the book’s scope; consult specialist literature.)
- Case details follow the book; I have not checked the original papers.
Summary
- If language were part of general intelligence, injury should cause across-the-board decline.
- Yet there are people with intact intelligence and broken grammar (Broca’s aphasia, SLI) and people with low intelligence and fluent language (Denyse, Williams syndrome).
- From this Pinker argues that language has a dedicated basis in brain and genes.