Music
Still other evidence pointed to a right-hemisphere role for music. Persons made speechless by left-hemisphere injury could still sing or carry a tune, abilities presumably mediated by the intact right hemisphere. Hughlings-Jackson [38] found 3 such persons among 28 with aphasia; Bouillaud [11] found one more, a man who composed and sang without “even a minor mistake [in articulating] the appropriate sounds” (quoted in translation in Della Rocca [43], p 140); and in possibly the earliest report, from 1745, van Dalin [44] described a man who, after a stroke with loss of speech and right-side paralysis, could “sing” and “whistle hymns … learned before he became ill … so purely and explicitly like any other normal person” (quoted in translation in Johnson et al. [45], p 18). All this evidence was only indirect. Direct evidence came from patients with right-hemisphere lesions who showed the symptoms in reverse: they could still speak but not carry a tune, and their speech lacked normal “song,” or prosody. August Knoblauch [46] called the condition “amusia” but, envisioning the motor and auditory centers for tones as being close to those for speech and comprehension, localized music in the left hemisphere, not the right (see Johnson and Graziano [47], p 112).
Localization of Symptoms or Localization of Function?
Along with his reports on imperception, Hughlings-Jackson [38] made another signal contribution. He identified what he saw as a common error in interpreting clinical data: speech loss following F3 damage on the left was widely taken to mean that speech was located there. The error lay in equating localization of symptoms with localization of function:
Whilst I believe that the hinder part of the left third frontal convolution is the part most often damaged (when speech is lost), I do not localize speech in any such small part of the brain. To locate the damage which destroys speech and to locate speech are two different things ([30], p 81).
Hughlings-Jackson instead saw speech, and language more generally, as a dynamic process involving the integrated functioning of the whole brain, such that the more complex the task, the more regions involved, each contributing in its own way. The question was not, where is language located, it was, what is each region’s contribution? His answer was that language is represented in both hemispheres at different functional levels: Emotional utterances (the lowest level) were the least lateralized, or most bilateral, which explained the preservation of “emotional speech,” such as oaths and other “automatic,” “involuntary” phrases in persons with aphasia, persons like Leborgne. Comprehension was more lateralized, and “propositional speech” (the highest level), together with the background of conceptual thought it requires, depended wholly on the left hemisphere.
Explanations of Hemispheric Specialization
How could cerebral specialization be explained? For Broca [20], the challenge was significant because, in declaring that we speak with the left hemisphere, he noted that, except for certain “secondary convolutions,” the hemispheres were “perfectly alike” (p 381). (Malgaigne [10], we recall, saw things differently.) How, then, could asymmetrical control for speech be reconciled with the “general truths [the law of symmetry] it seems to contradict” (pp 381–382)?
Broca saw a clue in something Leurat and Gratiolet [48] found in the brains of human fetuses: frontal convolutions appear earlier on the left. Could that dispose the left hemisphere to take the lead in complex acts like speech and language? Further reports, however, were mixed, some finding earlier development on the left (e.g., [49], others on the right, e.g., [50], and still others, no differences [51]). The discrepancies were never resolved. As the author of one of the reports later remarked, “the question which side is first developed seems still doubtful” ([52], p 293).
Could the difference instead lie in the number of convolutions? In the adult brain, Broca, in an apparent change from his prior [20] position, reported finding “notably more” on the left (quoted in Bateman [53], p 380). He also found the left frontal lobe to be heavier ([54]). Asked whether F3 made the difference, he said he was inclined to this view but was reluctant to say more given the difficulty in surgically isolating this structure (reported in [55]).
Size was considered too. Cunningham [56] measured the upward angle of the Sylvian fissure’s posterior ramus in human fetuses. Finding it smaller on the left, he concluded that the region it bounded ventrally, containing the planum temporale, part of what by then was being called Wernicke’s area, was larger on that side. But seeing the same in great apes – animals with no known language – he questioned whether it is “in any way associated with … localization of the active speech centre …” (p 293). He presumably meant “comprehensioncentre.”
What about the right hemisphere? For the temporal and occipital lobes, Gratiolet found earlier convolutional development on the right, the reverse of that for the frontal lobes. For the occipital lobe, Broca reported the same for number: “the right is richer in convolutions” (in [53] but, with his focus on speech, would have had no reason to see it as significant. Hughlings-Jackson [38] had very good reason: “These anatomical facts, I submit, support the view … that the hinder part of the brain on the right side, is the chief seat … in the recognition of objects, places, persons, &c.” (p 70).
The Role of the Corpus Callosum
By the 1920s, with the discovery and further documentation of lateral differences in