The Tell-Tale Brain

“Yeah, yeah.”

“You can understand.”

“Right.”

“Very good.”

“Uhhhhhh…but uhhhh…the speech, uhhhhh, ummmmm slowed down.”

“Okay, do you think your speech is slowed down, or your thought is slowed down?”

“Okay. But ummmm [points to head] uhhh…words are beautiful. Ummmmm speech…”

He then made twisting motions with his mouth. Presumably he meant that his flow of thought felt intact, but the words were not coming out fluently.

“Supposing I ask you a question,” I said. “Mary and Joe together have eighteen apples.”

“All right.”

“Joe has twice as many apples as Mary.”

“Okay.”

“So how many does Mary have? How many does Joe have?”

“Ummmmm…lemme think. Oh God.”

“Mary and Joe together have eighteen apples…”

“Six, ahhhh twelve!” he blurted.

“Excellent!”

So Dr. Hamdi had basic conceptual algebra, was able to do simple arithmetic, and had good comprehension of language even for relatively complex sentences. I was told he had been a superb mathematician before his accident. Yet later, when Jason and I tested Dr. Hamdi on more complex algebra using symbols, he kept trying hard but failing. I was intrigued by the possibility that the Broca’s area might be specialized not just for the syntax, or syntactic structure, of natural language, but also for other, more arbitrary languages that have formal rules, such as algebra or computer programming. Even though the area might have evolved for natural language, it may have the latent capacity for other functions that bear a certain resemblance to the rules of syntax.

What do I mean by “syntax”? To understand Dr. Hamdi’s main problem, consider a routine sentence such as “I lent the book you gave me to Mary.” Here an entire noun phrase—“the book you gave me”—is embedded in a larger sentence. That embedding process, called recursion, is facilitated by function words and is made possible by a number of unconscious rules—rules that all languages follow, no matter how different they may seem on the surface. Recursion can be repeated any number of times to make a sentence as complex as it needs to be in order to convey its ideas. With each recursion, the sentence adds a new branch to its phrase structure. Our example sentence can be expanded, for instance, to “I lent the book you gave me while I was in the hospital to Mary,” and from there to “I lent the book you gave me while I was in the hospital to a nice woman I met there named Mary,” and so on. Syntax allows us to create sentences as complex as our short-term memory can handle. Of course, if we go on too long, it can get silly or start to feel like a game, as in the old English nursery rhyme:

This is the man all tattered and torn

That kissed the maiden all forlorn

That milked the cow with the crumpled horn

That tossed the dog that worried the cat

That killed the rat that ate the malt

That lay in the house that Jack built.

Now, before we go on discussing language, we need to ask how we can be sure Dr. Hamdi’s problem was really a disorder of language at this abstract level and not something more mundane. You might think, reasonably, that the stroke had damaged the parts of his cortex that control his lips, tongue, palate, and other small muscles required for the execution of speech. Because talking required such effort, he was economizing on words. The telegraphic nature of his speech may have been to save effort. But I did some simple tests to show Jason that this couldn’t be the reason.

“Dr. Hamdi, can you write down on this pad the reason why you went to the hospital? What happened?”

Dr. Hamdi understood our request and proceeded to write, using his left hand, a long paragraph about the circumstances that brought him to our hospital. Although the handwriting wasn’t good, the paragraph made sense. We could understand what he had written. Yet remarkably, his writing also had poor grammatical structure. Too few “ands,” “ifs,” and “buts.” If his problem were related to speech muscles, why did his writing also have the same abnormal form as his speech? After all, there was nothing wrong with his left hand.

I then asked Dr. Hamdi to sing “Happy Birthday.” He sang it effortlessly. Not only could he carry the tune well, but all the words were there and correctly pronounced. This was in stark contrast to his speech, which, in addition to missing important connecting words and lacking phrase structure, also contained mispronounced words and lacked the intonation, rhythm, and the melodious flow of normal speech. If his problem were poor control of his vocal apparatus, he shouldn’t have been able to sing, either. To this day we don’t know why Broca’s patients can sing. One possibility is that language function is based mainly in the left hemisphere, which is damaged in these patients, whereas singing is done by the right hemisphere.

We had already learned a great deal after just a few minutes of testing. Dr. Hamdi’s problems with expressing himself were not caused by a partial paralysis or weakness of his mouth and tongue. He had a disorder of language, not of speech, and the two are radically different. A parrot can talk—it has speech, you might say—but it doesn’t have language.

HUMAN LANGUAGE SEEMS so complex, multidimensional, and richly evocative that one is tempted to think that almost the entire brain, or large chunks of it at least, must be involved. After all, even the utterance of a single word like “rose” evokes a whole host of associations and emotions: the first rose you ever got, the fragrance, rose