many other differences between us and theapes, and in principleany of themmightbe the factor that makes language
possible. The most prominent candidate is sheer brain size. Eric Lenneberg (1967) discounts this possibility with the
evidenceof“nanocephalic dwarves,”individuals whose brains developonly to about thesize of those of chimpanzees,
with a proportional reductionin number of neurons. Lenneberg claims that, though deeply retarded, these individuals
still learn to speak.^46
Another benchmark for brain size concerns children who have undergone early hemispherectomy, so their brains are
half normal size. They too are deeply impaired in certain ways, but in some cases language does develop—not entirely
perfectly, but prettywell,evenwhenthemissinghemisphere is the left, the one that usuallyspecializes for grammatical
function (Dennis 1980; Vargha-Khade metal. 1991; Curtiss and de Bode 2000).
The other side of the equation comes from the experiments teaching language to chimpanzees, gorillas, and bonobos
(Linden 1974; Premack 1976; Savage-Rumbaugh et al. 1998; Seidenberg and Petitto 1978; Terrace 1979). These
experiments have been subject to ongoing and heated controversy: have the apes achieved language or not? My own
interpretation is in the middle. I am willing to accept that they have achieved the use of symbols for communicative
purposes. But, although they do concatenate symbols into strings, they seem not to have achieved any reliable
combinatoriality in the sense of Chapter 3. As will become clearer in Chapter 8, we are not required to proclaim
whether they“have language”or not. The apparent binarity of the decision is a consequence of seeing Universal
Grammar as a unified“grammar box,”a position to be rejected here. Rather, we can say that the apes are capable of
learning some aspects of language and notothers, and thatthisis partlya reflectionofdifferences incognitive capacity.
4.9.2 Characteristic timing of acquisition
Again this characteristic of language is based on a commonsense observation: adults are not as good at learning
languages as children. Anychild,takentoany linguisticcommunityat an earlyage, will cometospeak thecommunity's
language like a native, while the parents may struggle for years and never achievefluency.
This observation can be nuanced a bit more closely. All normal children acquire the native language(s) of their
community; acquiring a language as a
UNIVERSAL GRAMMAR 95
(^46) Ihaveencounterednomorerecentpublishedresearch onlinguisticabilityinthissyndrome,nowcalled SeckelSyndrome(Shanskeetal. 1997). Anecdotalevidence suggests
it is not as good as Lenneberg thought.