The Articulate Mammal (12 page)

But even apart from problems such as this, brain–body ratio cannot be a decisive factor as far as language is concerned, since it is possible to find young chimpanzees and human children who have similar brain–body ratios – yet the child can talk and the chimp cannot. Even more convincing is a comparison between a 3-year-old chimp and a 12-year-old nanocephalic dwarf – a human who because of a genetic defect grows to a height of around 760 mm (or 2 feet 6 inches).

Source
: Lenneberg 1967: 70

Although the chimp and the dwarf have exactly the same brain and body weights (and so, of course, the same brain–body ratio), the dwarfs speak, in a somewhat limited fashion, but the chimps do not. These figures show conclusively that the difference between human and chimp brains is a
qualitative
, not a
quantitative
one.

Superficially, the brains of a chimp and a human have certain similarities. As in a number of animals, the human brain is divided into a lower section, the
brain stem
, and a higher section, the
cerebrum
. The brain stem keeps the body
alive by controlling breathing, heartbeats and so on. A cat with the upper section of its brain removed but with the brain stem intact could still swallow milk, purr, and pull its paw away from a thorn when pricked. The higher section, the cerebrum, is not essential for life. Its purpose seems to be to integrate an animal with its environment. This is the part of the brain where language is likely to be organized.

The cerebrum is divided into two halves, the
cerebral hemispheres
, which are linked to one another by a series of bridges. The left hemisphere controls the right side of the body, and the right hemisphere the left side.

But the two hemispheres do not function identically. This was first discovered over a hundred years ago. A Frenchman, Marc Dax, read a paper at Montpellier in 1836, pointing out that paralysis of the right side of the body was often associated with loss of speech, while patients whose left side was paralysed could usually talk normally. This suggested that the left hemisphere controlled not only the right side of the body, but
speech
also. Dax’s hypothesis turned out to be correct. Speech in the majority of humans is the concern of the
left
, not the right hemisphere. But it was a long time before this was reliably confirmed. Until relatively recently, statistics could only be drawn up by chance observations, when researchers managed to note cases of people in whom loss of speech was associated with right-side paralysis. But in the twentieth century more sophisticated methods were adopted. One is the sodium amytal test developed by Wada in the 1940s. In this test the patient was asked to count out loud while a barbiturate (sodium amytal) was injected
into an artery carrying blood to one side of the brain. If this was the hemisphere used in speech, the patient lost all track of his counting and experienced severe language difficulties for several minutes. If it was not, the patient could resume normal counting almost immediately after the injection. Although this test was effective, it also carried an element of risk. So it was only used when brain surgery was advisable (as in severe epilepsy) and the surgeon wished to know whether he was likely to disturb vital speech areas. If so, he was unlikely to operate.

Simpler and less invasive methods for discovering which hemisphere controls language are now the norm. The first was the use of dichotic listening tests (Kimura 1967; Obler and Gjerlow 1999). The subject wears headphones, and is played two different words simultaneously, one into each ear. For example, he or she might hear
six
in one ear, and
two
in the other. Most people can report the word played to the right ear (which is directly linked to the left hemisphere) more accurately than the word played to the left ear (linked to the right hemisphere). It is clear that this is not simply due to an overall preference for sounds heard in the right ear, because for non-linguistic sounds the left ear is better. If different tunes are played simultaneously into each ear, subjects will
identify the tune played into the left ear better than the one directed into the right ear. We conclude that the left hemisphere is better at processing linguistic signals – and so is normally the dominant one for speech.

A further technique is tachistoscopic (fast-view) presentation. An image is presented very fast to either the left or right visual field (the area that can be seen to left or right without moving the head or eyes). A linguistic stimulus will normally be processed faster if it is presented to the right visual field, which is then transferred to the left (usually language dominant) hemisphere.

In another twentieth-century technique, electrodes are attached to the skull in order to measure the amount of electrical activity in the area beneath (as will be discussed later). Spoken words produce a greater response in the left hemisphere, whereas noises such as mechanical clicks arouse a greater response in the right (Rosenfield 1978).

The results of the observations and tests described above are surprisingly consistent. The majority of normal human beings – perhaps as many as 90 per cent – have speech located primarily in the left hemisphere. This cannot be due to chance.

A further related discovery is that the location of speech centres in the left hemisphere seems to be linked to right-handedness. That is, most humans are right-handed, and most people’s speech is controlled by the left hemisphere. In the nineteenth century it was commonly assumed that left-handers must have speech located in the right hemisphere, and this seemed to be confirmed by a report in 1868 by the influential neurologist John Hughlings Jackson that he had discovered loss of speech in a left-hander who had sustained injury to the right side of the brain. But this viewpoint turns out to be false.
Surprisingly, most left-handers also have language controlled predominantly by the left hemisphere, though the picture is not completely straightforward. Of the relatively few people who do not have their speech centres located in the right hemisphere, more are left-handed than right-handed.

(Figures averaged from Penfield and Roberts 1959; Zangwill 1973; Milner,
et al.
1964.)

These figures indicate two things: first, it is normal for speech and handedness to be controlled by the same hemisphere, and it has been suggested that speech and writing problems are found more frequently in children where the two are not linked. Second, there is a strong tendency for speech to be located in the left hemisphere even when this appears to disrupt the standard linking of speech and handedness.

Some work has been directed at finding out if
all
speech processing must be located in one hemisphere or whether subsidiary linguistic abilities remain in the non-dominant hemisphere. One group of researchers at Montreal, Canada, found ten patients who had speech abilities in both halves of the brain. The sodium amytal test disturbed speech whichever side of the brain it was injected. Interestingly enough, all these patients were either left-handed or ambidextrous (Milner,
et al.
1964).

Other studies suggest that the right hemisphere contains a limited potential for language which is normally latent, but which can be activated if needed. Patients who have had the whole of the left hemisphere removed are at first without speech. But after a while, they are likely to acquire a limited vocabulary, and be able to comprehend a certain amount, though they always have difficulty in producing speech (Kinsbourne 1975). The right hemisphere is not useless, however. Patients with right hemisphere damage have difficulty with intonation, and in understanding jokes and metaphors (Caplan 1987).

Perhaps the most widely reported experiments on this topic are those involving ‘split brain’ patients (Gazzaniga 1970, 1983). In cases of severe epilepsy it is sometimes necessary to sever the major links between the two hemispheres. This means that a patient has virtually two separate brains, each coping with one half of the body independently. A patient’s language can be tested by dealing with each hemisphere separately. An object shown to the
left
visual field is relayed only to the
right
(non-language hemisphere). Yet sometimes the patient is able to name such an object. This indicates that the right
hemisphere may be able to cope with simple naming problems – but it seems unable to cope with syntax. However, the results of these experiments are disputed. Some people have suggested that the information is being transferred from one hemisphere to the other by a ‘back route’ after the major links have been severed.

This lateralization or localization of language in one half of the brain, then, is a definite, biological characteristic of the human race. At one time, it was thought to develop gradually. But later research indicated that it may be present at birth (Kinsbourne and Hiscock 1987). Even foetuses have been claimed to show traces of it, with some areas of the left hemisphere being bigger than the right (Buffery 1978). The issue is an important one for psycholinguists, since it has sometimes been argued that the period of lateralization coincides with a ‘critical period’ for language acquisition (to be discussed in
Chapter 4
).