by Oliver Sacks
Yet he was surprised to find, as a nurse reminded him, that he could still write, even though he could not read; the medical term, she said, was “alexia sine agraphia.” Howard was incredulous—surely reading and writing went together; how could he lose one but not the other?2 The nurse suggested that he sign his name; he hesitated, but once he started, the writing seemed to flow all by itself, and he followed his signature with two or three sentences. The act of writing seemed quite normal to him, effortless and automatic, like walking or talking. The nurse had no difficulty reading what he had written, but he himself could not read a single word. To his eyes, it was the same indecipherable “Serbo-Croatian” he had seen in the newspaper.
We think of reading as a seamless and indivisible act, and as we read we attend to the meaning and perhaps the beauty of written language, unconscious of the many processes that make this possible. One has to encounter a condition such as Howard Engel’s to realize that reading is, in fact, dependent on a whole hierarchy or cascade of processes, which can break down at any point.
In 1890, the German neurologist Heinrich Lissauer used the term “psychic blindness” to describe how some patients, after a stroke, became unable to recognize familiar objects visually.3 People with this condition, visual agnosia, can have perfectly normal visual acuity, color perception, visual fields, and so on—yet be totally unable to recognize or identify what they are seeing.
Alexia is a specific form of visual agnosia, an inability to recognize written language. Since the French neurologist Paul Broca in 1861 had identified a center for the “motor images” of words, as he called it, and his German counterpart Carl Wernicke, a few years later, identified one for the “auditory images” of words, it seemed logical to nineteenth-century neurologists to suppose that there might also be an area in the brain dedicated to the visual images of words—an area that, if damaged, would produce an inability to read, a “word blindness.”4
In 1887, a French neurologist, Joseph-Jules Déjerine, was asked by an ophthalmologist colleague to see a highly intelligent, cultivated man who had suddenly lost the ability to read. Edmund Landolt, the ophthalmologist, wrote a short but vividly evocative portrait of the patient, and Déjerine, in his own paper on the subject, included a long excerpt from this.
They described how in October of that year, Oscar C., a retired businessman, found himself suddenly unable to read. (He had had some brief attacks of numbness in his right leg on previous days, but had paid little attention to them.) Though reading was impossible, Monsieur C. had no difficulty recognizing people and objects around him. Nevertheless, thinking that his eyes must be at fault, he consulted Landolt, who wrote:
Asked to read an eye chart, C is unable to name any letter. However, he claims to see them perfectly. He instinctively sketches the form of the letters with his hand, but he is nevertheless unable to say any of their names. When asked to write on a paper what he sees, he is able, with great difficulty, to recopy the letters, line by line, as if he were making a technical drawing, carefully examining each stroke in order to reassure himself that his drawing is exact. In spite of these efforts, he remains incapable of naming the letters. He compares the A to an easel, the Z to a serpent, and the P to a buckle. His incapacity to express himself frightens him. He thinks that he has “gone mad,” since he is well aware that the signs he cannot name are letters.5
Like Howard Engel, Monsieur C. was unable to read even the headlines of his morning paper, although he nonetheless recognized it, by its format, as his usual newspaper, Le Matin. And, like Howard, he could write perfectly well:
While reading is impossible, the patient … can write fluently and without any mistakes whatever material is dictated to him. But should he be interrupted in the middle of a phrase that he is writing … he becomes muddled and cannot start up again. Also, if he makes a mistake he can’t find it.… He can never reread what he has written. Even isolated letters do not make sense to him. He can only recognize them … by tracing the outlines of the letter with his hand. Therefore it is the sense of the muscular movement that gives rise to the letter name.…
He is able to do simple addition, since he recognizes, with relative ease, numbers. However, he is very slow. He reads the numbers poorly, since he cannot recognize the value of several numbers at once. When shown the number 112, he says, “It is a 1, a 1, and a 2,” and only when he writes the number can he say “one hundred and twelve.”6
There were some additional visual problems—objects appeared dimmer and a little blurred on the right side and completely devoid of color. These problems, along with the specificity of Oscar C.’s alexia, indicated to Landolt that the underlying problem was not in the eyes but in the brain; this led him to refer his patient to Déjerine.
Déjerine was fascinated by Monsieur C.’s condition and arranged to see him twice weekly at his clinic in Paris. In a monumental 1892 paper, Déjerine summarized his neurological findings succinctly and then, in a much more leisurely style, provided a general picture of his patient’s life:
C spends his days taking long walks with his wife. He has no difficulty walking and every day he does his errands on foot from the Boulevard Montmartre to the Arc de Triomphe and back. He is aware of what is happening around him, stops in front of stores, looks at paintings in gallery windows, etc. Only posters and signs in shops remain meaningless collections of letters for him. He often becomes exasperated by this, and though he has been so afflicted for four years, he has never accepted the idea that he cannot read, while remaining able to write.… In spite of patient exercises and much effort, he has never relearned the sense of letters and written words, nor has he ever relearned how to read musical notes.
Despite this, Oscar C., an excellent singer, could still learn new music by ear, and he continued to practice music with his wife every afternoon. And he continued to enjoy and excel at playing cards: “He is a very good card player, calculates very well, prepares his blows well in advance and wins most of the time.” (Déjerine did not comment on how Monsieur C. was able to “read” the cards, but it seems likely that he recognized the iconic images of hearts, diamonds, spades, clubs, jacks, queens, and kings—just as Howard Engel recognized the icon of an ambulance when he arrived at the emergency room. Number cards, of course, can also be recognized by their patterns.)
When Oscar C. died following a second stroke, Déjerine performed an autopsy and found two lesions in the brain: an older one, which had destroyed part of the left occipital lobe and which he presumed was responsible for Monsieur C.’s alexia, and a larger, recent lesion, which had probably caused his death.7
It is always difficult to make inferences from the appearance of the brain at autopsy; one may find damaged areas, but it is not always possible to see their manifold connections with other areas of the brain or to determine what controls what. Déjerine was well aware of this; nonetheless, he felt that by relating a specific neurological symptom—alexia—to damage in a particular area of the brain, he had, in principle, demonstrated what he called a “visual center for letters” in the brain.
Déjerine’s discovery of this area essential for reading would be confirmed over the next hundred years by scores of similar cases and autopsy reports of patients with alexia, irrespective of its cause.
By the 1980s, CT scanning and MRIs made it possible to visualize living brains with an immediacy and precision impossible in autopsy studies (where all sorts of secondary changes may blur the picture). Using this technology, Antonio and Hanna Damasio and, later, other researchers, were again able to confirm Déjerine’s findings, and to correlate their alexic patients’ symptoms with highly specific brain lesions.
With the development of functional brain imaging a few years later, it became possible to visualize the activity of the brain in real time, as subjects performed various tasks. A pioneer PET scan study in 1988 by Steven Petersen, Marcus Raichle, and their colleagues showed the different areas of the brain activated by reading words, listening to words
, uttering words, and associating words. “For the first time in history,” as Stanislas Dehaene writes in his book Reading in the Brain, “the areas responsible for language had been photographed in the living human brain.”
Dehaene, a psychologist and neuroscientist, has specialized in studying the processes involved in visual perception, especially the recognition and representation of words, letters, and numbers. Using fMRI technology, which is much swifter and more sensitive than PET scanning, he and his colleagues have been able to focus even more closely on what he calls the visual word form area or, more informally, “the brain’s letterbox.”
Dehaene’s studies (with Laurent Cohen and others) have shown how the visual word form area can be activated in a fraction of a second by a single written word, and how this initial, purely visual activation then spreads to other areas of the brain—especially the temporal lobes and the frontal lobes.
Reading, of course, does not end with the recognition of visual word forms—it would be more accurate to say that it begins with this. Written language is meant to convey not only the sound of words but their meaning, and the visual word form area has intimate connections to the auditory and speech areas of the brain as well as to the intellectual and executive areas, and to the areas subserving memory and emotion.8 The visual word form area is a crucial node in a complex cerebral network of reciprocal connections—a network peculiar, it seems, to the human brain.
As a prolific writer and an omnivorous reader, accustomed to reading newspapers every morning and many books each week, Howard Engel wondered how he would manage life with his alexia, which showed no signs of clearing. In a world full of traffic signs, printed labels, and directions on everything from a prescription bottle to the television, ordinary life is a continuing, daily struggle for anyone with alexia. But for Howard, this was an even more desperate situation, for his whole life and identity (to say nothing of his livelihood) depended on his ability to read and write.
Being able to write without reading might be all right for a short letter or memorandum, a page or two. But for the most part, he thought, it “was like being told that the right leg had to be amputated but that I could keep the shoe and sock.” How could he hope to go back to his previous work—to write an elaborate narrative of crime and detection, full of plots and counterplots, to do all the corrections and revisions and redrafting a writer must do—without being able to read? He would have to get others to read for him, or perhaps get one of the ingenious new software programs that would allow him to scan what he had written and hear it read back to him by a computer. Both of these would involve a radical shift from the visuality of reading, the look of words on a page, to an essentially auditory mode of perception—going, in effect, from reading to listening and, perhaps, from writing to speech. Would this be desirable—or even possible?
Precisely this question had forced itself on another writer who consulted me ten years earlier. Charles Scribner, Jr., was also a man of letters; he presided over the publishing house established by his great-grandfather in the 1840s. In his sixties, he developed a visual alexia—probably as a result of a degenerative process in the visual parts of the brain. It was a devastating problem for a man who had published the work of Hemingway and others, a man whose life was centered on reading and writing.
As a book publisher, Scribner slightly disapproved of audiobooks, which had recently been introduced to the general public. But he decided nonetheless to reconstruct his entire literary life in an auditory mode. To his surprise, this did not prove as difficult as he expected. He even began to enjoy listening to audiobooks:
It never dawned on me that these spoken books would become a major part of my intellectual life and recreational reading. By now I must have “read” hundreds of books in this way. I was never a rapid reader as a boy, although my retention was high. Paradoxically, now that I was reading books on tape, my reading speed was better than ever and my retention just as good. I can fairly say that for me the discovery of this mode of reading was a kind of “open sesame” to my continued enjoyment of literature.9
Like Howard, Scribner preserved the power to write, but he was so deeply distressed by his inability to read what he had written that he decided to change to dictation, something he had never before tried. Luckily, this too was successful—dictation worked so well that it allowed him to complete more than eighty newspaper columns and two book-length memoirs about his life in publishing. “Perhaps,” he wrote, “it’s another instance of a handicap honing a skill.” Apart from his close friends and family, no one seemed aware that he had accomplished all this by switching to an entirely new mode.
One might have expected Howard, too, to turn to an auditory mode of “reading” and writing, but his course was very different.
After his week at Mount Sinai Hospital, he was moved to a rehabilitation hospital, where he spent almost three months studying himself, what he could and could not do. When he was not trying to read a paper or a get-well card, he found, he could forget about his alexia:
The sky looked blue, the sun shone on the hospital windows, the world hadn’t suddenly become unfamiliar. My alexia existed only when I had my head buried in a book. Print brought it on and reminded me that, yes, there was a problem. Thus was born the temptation to simply avoid reading.
But this, he quickly realized, was unacceptable to him as a reader and a writer. Audiobooks might do for some, but not for him. He still could not even recognize individual letters, but he was determined to read again.
Two months after his stroke, still living at the rehab hospital, Howard had continuing difficulties recognizing places; he would get lost within the hospital three or four times a day and could not find his own room until he finally learned to recognize its floor “by the way the light filled the hall just opposite the elevator.” He continued to have some object agnosia, too—even when he returned home after three months, he noted, “I kept finding cans of tuna in the dishwasher and jars of pencils in the freezer.”
But with reading, Howard noted some signs of improvement: “the words no longer looked like they were written in an unfamiliar alphabet. The letters themselves looked like ordinary English letters, not the Serbo-Croatian I had imagined [after] my stroke.”
There are two forms of alexia: a severe form which prevents even individual letters from being recognized and a milder form, in which letters can be recognized but only one by one, not simultaneously as words. Howard seemed to have moved, at this point, to the milder form—perhaps due to a partial recovery of the tissues affected by his stroke, or the brain’s use (or perhaps even construction) of alternative pathways.10
Given this neurological improvement, he was able, with his therapists, to explore new ways of trying to read. He would slowly and laboriously puzzle out words, letter by letter, forcing himself to decipher the names of streets and shops or the headlines of newspapers. “Familiar words,” he said,
including my own name, are unfamiliar blocks of type and have to be sounded out slowly. Each time a name recurs in an article or review, it hits me as unfamiliar on its last appearance as it does on the first.
Yet he persisted.
Even though the reading was slow and difficult—frustrating as hell at times—I was still a reader. The blast to my brain could not make me otherwise. Reading was hard-wired into me. I could no more stop reading than I could stop my heart.… The idea of being cut off from Shakespeare and company left me weak. My life had been built on reading everything in sight.
Howard’s reading grew somewhat easier with practice, though it might take him several seconds to make out a single word. “Words of different lengths,” he observed, “like cat, table and hippopotamus, are processed in my head at a different rate. Each added letter adds more weight to the load that I am trying to lift.” Scanning a page, reading in the usual sense, was still impossible, and “the whole process,” he wrote, “was exhausting beyond belief.” Sometimes, however, if he looked at a word, a couple of lette
rs would suddenly jump out at him and be recognized—for example, the bi in the middle of his editor’s name, though the letters before and after this remained unintelligible. He wondered whether such “chunking” was the way he had originally learned to read as a child, perhaps the way we all learn to read, before we go on to perceive words, even sentences, as a whole. (Pairs and perhaps clusters of letters are particularly important in the construction and reading of words, and whether reading is being learned for the first time or relearned after a stroke, there seems to be a natural progress from seeing single letters to seeing letter pairs or sequences. Dehaene and his colleagues suggest that there may be special “bigram” neurons in the brain devoted to this.)
“I can make myself see that certain letter groupings are indeed familiar words,” Howard wrote to me, “but that comes only after I have stared at the page.”
Becoming a fluent reader is a difficult and multileveled task; most children need years of practice and instruction to achieve this (though a few precocious ones may learn to read by themselves, and at an early age). In some ways, Howard had been reduced to the level of a child first learning his ABC’s. But with a lifetime of experience as a reader, he could also bypass his disabilities to some extent, for his large vocabulary, his grammatical sense, and his command of literary and idiomatic English helped him to guess or infer words and even sentences from the slightest hint.
Whatever language a person is reading, the same area of inferotemporal cortex, the visual word form area, is activated. It makes relatively little difference whether the language uses an alphabet, like Greek or English, or ideograms, like Chinese.11 This has been confirmed by lesion studies such as Déjerine’s, and by imaging studies. And this idea is supported, too, by “positive” disorders—excesses or distortions of function produced by hyperactivity of the same area. The opposite of alexia, in this sense, is lexical or text hallucination, or phantom letters. People with disorders of the visual pathway (anywhere from the retina to the visual cortex) may be prone to visual hallucinations, and Dominic ffytche and his colleagues estimate that about a quarter of these patients who hallucinate see “text, isolated words, individual letters, numbers, or musical note hallucinations.” Such lexical hallucinations, as ffytche and his colleagues have found, are associated with conspicuous activation of the left occipitotemporal region, especially the visual word form area—the same area that, if damaged, produces alexia.
