The Making of Mind. A R Luria

Disturbance of Brain Functions

WE HAD two strategies for discovering and describing the nature of higher psychological functions. The first was to trace their development; the second was to follow the course of their dissolution under conditions of localized brain damage. In the mid 1920s Vygotsky first suggested that an investigation of localized brain damage could provide a way to analyze the cerebral structure and development of the higher psychological processes. At that time neither the structure of the higher psychological processes themselves nor the functional organization of the brain was clear.

Two diametrically opposed principles then prevailed for explaining how the brain worked. On the one hand, there were the localization theorists who attempted to relate each mental function to a specific cortical area; on the other, there were the holistic theorists who assumed that the brain functions as a whole to produce the psychological functions expressed in behavior. According to this view, the amount of brain tissue damaged, rather than the location of a lesion, determines the nature of the resulting defects.

The scientific investigation of disturbances of the complex mental processes began in 1861 when the French anatomist Paul Broca described the brain of a patient who for many years had been kept in the Salpetriere Hospital because he was unable to speak, although he could understand speech. When the patient died, Broca was able to obtain precise information about the area of his brain that had been damaged. Broca was the first to demonstrate that speech production, that is, the motor coordinations which produce speech, are associated with a localized region of the brain, namely the posterior third of the left inferior frontal gyrus. Broca postulated that this location is the “center for the motor images of words” and that a lesion in this region leads to a distinctive type of loss of expressive speech, which he originally called “aphemia” and which later came to be called “aphasia,” the term still used today. Broca's discovery represented the first time that a complex mental function like speech was localized on the basis of clinical observation. It also led Broca to the first description of a marked difference between the functions of the left and right cerebral hemispheres.

Broca's discoveries were followed by those of Carl Wernicke, a German psychiatrist. In 1873 Wernicke published descriptions of cases in which lesions of the posterior third of the left superior temporal gyrus resulted in a loss of the ability to understand audible speech. He claimed to have found “the center for the sensory images of words,” or the center for understanding speech.

The discovery that a complex form of mental activity can be regarded as the function of a local brain area aroused unprecedented enthusiasm in neurological science. Within a short time many other brain centers for intellectual functions were found, including a “center for concepts” in the left inferior parietal region and a “center for writing” in the posterior part of the left middle frontal gyrus. By the 1880s neurologists and psychiatrists were able to draw “functional maps” of the cerebral cortex. They thought that they had settled the problem of the relation between brain structure and mental activity. Such research persisted well into the 1930s.

From the beginning some scientists disapproved of this kind of theorizing. Prominent among them was the English neurologist Hughlings Jackson. He maintained that the cerebral organization of mental processes differs according to the complexity of the process in question and the representation in the brain for processing that complexity.

Jackson's ideas grew out of observations that seemed to defy the kind of localization theory advocated by Broca. In studies of motor and speech disturbances Jackson noted that circumscribed lesions of a particular area never caused complete loss of function. A paradox occurred: sometimes the patient moved or spoke in ways that, according to a strict localization position, ought to be impossible. For example, the patient might be instructed, “Say the word no,” and would be unable to do so. Yet a little later in the same interview the patient might, in response to some quite different request, say, “No, doctor, I can't do that.”

Jackson resolved paradoxes of this sort, where “no” is both possible and impossible, by suggesting that all psychological functions have a complex “vertical” organization. Each function has a “low” level representation in the spinal chord or brain stem, is also represented in the “middle” or sensory and motor levels of the cortex, and finally is represented at a “high” level, presumably in the frontal lobes.

He advocated careful study of the level at which a particular function is carried out, not its localization in particular areas of the brain.

Jackson's hypothesis, which greatly influenced our work, was not taken up and developed until fifty years later when it reemerged in the writings of such neurologists as Anton Pick (1905), von Monakow (1914), Henry Head (1926), and Kurt Goldstein (1927, 1944, 1948). Without denying that elementary psychological “functions” such as vision, hearing, cutaneous sensation, and movement are represented in clearly defined areas of the cortex, these neurologists expressed doubts about the applicability of the principle of narrow localization to the brain mechanisms of complex forms of human mental activity. However, forgetting Jackson's teaching, they approached complex mental activity from the opposite extreme of the strict localizationists. Pointing to the complex character of human mental activity, Monakow attempted to describe its specific features with as vague a term as the “semantic character of behavior;” Goldstein talked about “abstract sets” and “categorical behavior” to emphasize the same idea. They either postulated that complex mental processes-which they termed “semantics” or “categorical behavior” are the result of activity of the brain as a whole, or divorced complex processes from brain structure altogether and emphasized their special “spiritual nature.”

From our point of view neither of these two positions seemed to provide the necessary basis for further scientific research. We rejected the holistic theories because we felt it was absurd to maintain an obsolete separation between “spiritual life” and the brain and to deny the possibility of discovering the mind's material basis. Uncritical “mass potential” ideas revived what we considered to be equally unacceptable notions of the brain as a primitive, undifferentiated nervous mass. Our reasons for rejecting the strict localization theories were arrived at over the course of many years of work and are somewhat more complicated.

Most investigators who have examined the problem of cortical localization have understood the term function to mean the “function of a particular tissue.” For example, it is perfectly natural to consider that the secretion of bile is a function of the liver and the secretion of insulin is a function of the pancreas. It is equally logical to regard the perception of light as a function of the photosensitive elements of the retina and the highly specialized neurons of the visual cortex connected with them. However, this definition does not meet every use of the term function.

When we speak of the “function of respiration,” this clearly cannot be understood as a function of a particular tissue. The ultimate object of respiration is to supply oxygen to the alveoli of the lungs to diffuse it through the walls of the alveoli into the blood. The whole of this process is carried out, not as a simple function of a particular tissue, but rather as a complete functional system, embodying many components belonging to different levels of the secretory, motor, and nervous apparatus. Such a “functional system,” the term introduced and developed by P. K. Anokhin in 1935, differs not only in the complexity of its structure but also in the mobility of its component parts. The original task of respiration-restoration of the disturbed homeostasis – and its final result – transportation of oxygen to the alveoli of the lung, followed by its absorption into the blood stream – obviously remain invariant. However, the way in which this task is performed may vary considerably. For instance, if the diaphragm, the principal group of muscles working during respiration, ceases to act, the intercostal muscles are brought into play, but if for some reason those muscles are impaired, the muscles of the larynx are mobilized and the animal or person begins to swallow air, which then reaches the alveoli of the lung by a completely different route. The presence of an invariant task, performed by variable mechanisms, which bring the process to a constant invariant conclusion, is one of the basic features distinguishing the work of every “functional system.”

The second distinguishing feature is the complex composition of the functional system, which always includes a series of afferent (adjusting) and efferent (effector) impulses. This combination can be illustrated with reference to the function of movement, which has been analyzed in detail by the Soviet physiologist-mathematician N. A. Bernshtein. The movements of a person intending to change his position in space, to strike at a certain point, or to perform a certain action can never take place simply by means of efferent, motor impulses. Since the locomotor apparatus, with its movable joints, has many degrees of freedom because different groups of articulations participate in the movement, and since every stage of the movement changes the initial tonus of the muscles, movement is in principle uncontrollable simply by efferent impulses. For a movement to take place, there must be constant correction of the initiated movement by afferent impulses, which give information about the position of the moving limb in space and the change in muscle tone. This complex structure of locomotion is required to satisfy the fundamental conditions preserving the invariance of the task and its performance by variable means. The fact that every movement has the character of a complex functional system and that the elements performing it may be interchangeable in character is clear because the same result can be achieved by totally different methods.

In Walter Hunter's experiments a rat in a maze achieved its goal by running a certain path, but when one element of the maze was replaced by water, the rat achieved its goal by swimming movements. In some of Karl Lashley's observations a rat, trained to follow a certain movement pattern, radically changed the structure of its movements after removal of the cerebellum. The rat was unable to reproduce the movements learned through training, but still it was able to reach its goal by going head over heels. The same interchangeable character of movements necessary to achieve a required goal can be clearly seen if any human locomotor act is carefully analyzed, such as hitting a target, which is done with a different set of movements depending on the initial position of the body, manipulating objects, which may be performed by different sets of motor impulses, or writing, which can be performed with either pencil or pen, by the right or left hand, or even by the foot, without effecting the meaning of what is written.

This “systemic” structure is characteristic of complex forms of mental activity as well as of simple behavioral acts. Although elementary functions like the registration of sensations from the retina could legitimately be said to have precise localization in particular groups of cells, it seemed absurd to us to think that complex functions could be regarded as the direct function of limited cell groups or could be localized in particular areas of the brain. Our approach to the structure of functional systems in general, and of the higher psychological functions in particular, led us to believe that the ideas on localization provided by the theorists of the early part of the century had to be radically revised.

Applying what we knew and surmised about the structure of higher psychological functions based on our work with children, Vygotsky reasoned that higher psychological functions represent complex functional systems which are mediated in their structure. They incorporate historically accumulated symbols and tools. Consequently, the organization of higher functions must differ from anything seen in animals. Furthermore, since the human brain took millions of years to evolve but human history is limited to thousands of years, a theory of the cerebral organization of higher functions must account for processes, such as those involved in writing, that depend in part on external, historically conditioned mediators. In other words, Vygotsky assumed that his historical approach to the development of such psychological processes as active memory, abstract thought, and voluntary actions would also hold for the principles of their organization on a cerebral level.

His theory of the development of higher psychological functions in children also led to the conclusion that the role played by a cerebral region in the organization of a higher psychological process would change during the course of an individual's development. Our research had shown that all complex, conscious activities are initially carried out in an expanded way. In its early stages, complex thinking requires a number of external aids for its performance. Not until later in the life of the child or in the course of mastering a particular form of activity does thinking become condensed and converted into an automatic skill. It seemed logical to suppose that in the course of development not only the functional structure of thinking changes but its cerebral organization as well. The participation of auditory and visual areas of the cortex, essential in the early stages of the formation of many cognitive activities, no longer plays the same role in their later stages when thinking begins to depend on the coordinated activity of different systems of cortical zones. For example, in the child the sensory areas of the cortex provide the basis for the development of cognitive processes, including speech. But in adults, in whom speech and complex cognitive processes are already developed, the sensory areas lose this function, and cognition becomes less dependent on sensory input. Reasoning in this way, Vygotsky explained why circumscribed lesions in cortical areas can have opposite effects depending on whether the lesion takes place in early childhood or in adulthood. For example, a lesion of the visual sensory areas of the cortex in early childhood results in an underdevelopment of cognition and thought, while in an adult an identical lesion can be compensated for by the influence of the already developed higher functional systems.

Our initial observations were strongly influenced by the English neurologist Head, who summarized a great deal of nineteenth and early twentieth century research on aphasia and provided us with a tempting interpretation of the relation between disturbance of speech and disturbance of thinking. In his classic monograph on aphasia, Head concluded that disturbances in language function produced disturbances in thinking. Aphasia caused a reduction of intellectual power, Head would have us believe, because thinking was no longer mediated by language but must depend on primitive, direct relations between objects and actions, on the one hand, and language on the other.

For example, Head pointed out that an aphasic patient who could easily match an object shown to him with a similar one lying on a table might fail if the task were complicated by adding another object to the one held up for inspection and asking the patient to select the two objects on the table that matched them both. Head attributed this difficulty to the possibility that, when presented with two objects, the patient tried to register them in words and to make his choice on the basis of their remembered names. In this case, Head noted, “A symbolic formula has been interjected and the act is no longer one of direct matching” (p. 518). Elsewhere Head made the point, in a manner completely consistent with our own theorizing, that “an animal, or even man under certain conditions, tends to react directly to the perceptual or emotional aspects of a situation; but symbolic formulation enables us to subject it to analysis and to regulate our behavior accordingly” (p. 525).

This testimony of a leading expert in the study of the brain fitted so closely with our own distinction between mediated and natural processes that at first we thought it possible that aphasia, by disturbing language – man's primary means of mediating his experience – acted to force the injured individual to operate on a natural, unmediated basis. We were reinforced in this presupposition by evidence presented by Guillaume and Meyerson, who claimed that their aphasic patients solved problems in a manner characteristic of young children. This position turned out to be incorrect, as many subsequent investigations have shown. We were greatly oversimplifing both the nature of aphasia and the psychological processing in brain-injured patients. At the beginning, however, these ideas were a strong motivation for assuming that the study of brain injury would lead us to an understanding of the nature of man's higher psychological functions and would provide us a means for understanding their material basis in the brain as well.

We were more successful when we began to observe patients suffering from Parkinson's disease. Parkinson's disease affects the subcortical motor ganglia so that the flow of involuntary movements is disturbed. We observed that tremors occurred shortly after patients suffering from this disease started to carry out an action. When we asked them to walk across a room, they could take only one or two steps before a tremor set in and they could walk no further.

We noted the paradoxical fact that patients who could not take two successive steps when walking on a level floor were able to climb stairs without difficulty. We hypothesized that in climbing stairs, each step represented a signal to which the patient's motor impulses responded. When climbing stairs, the successive, automatic flow of movement in walking on a level surface is replaced by chains of separate motor reactions. In other words, the structure of the motor activity is reorganized, and a conscious response to each link in a chain of isolated signals replaces the subcortically organized, involuntary system that guides ordinary walking.

Vygotsky used a simple device to construct a laboratory model of this kind of reorganization of movement. He placed a series of small paper cards on the floor and asked a patient to step over each one of them. A marvelous thing happened. A patient who had been able to take no more than two or three steps by himself walked through the room, easily stepping over each piece of paper as if he were climbing a staircase. We had helped the patient to overcome the symptoms of his disease by getting him to reorganize the mental processes he used in walking. He had compensated for his defect by transferring the activity from the subcortical level where his nerves were damaged to the cortical level which was not affected by the disease.

We then tried to use the same principle to construct an experimental model of self-regulating behavior, but our experiments were very naive and the results obtained were somewhat inconclusive. We asked a patient suffering from Parkinson's disease to tap sequentially for half a minute. This was impossible for him to do. In less than half a minute a muscle tremor set in, and his movements were blocked. But we found that if we asked the same person to tap in response to the experimenter's cues “one,” “two” - standing for “tap once,” “tap twice,” and so on - he could tap for a short time.

We wondered what would happen if a patient were to produce his own cues that would act as stimuli for his actions. We chose blinking as a signal because it was a physical system that seemed to be less affected by the disease than either walking or hand movements. We asked each patient to blink and after every blink to press a rubber bulb which recorded his movements. We discovered that the blinks served as a reliable self-regulating device. Patients who could not keep up a steady stream of squeezing movements under ordinary circumstances, could blink on command and then squeeze a rubber bulb in response to their blink.

Our final series of experiments with Parkinsoman patients used the patient's own speech to regulate his behavior. Our first attempts failed. The patients listened to the verbal instructions and started to press, but muscle hypertension and the concomitant tremors set in almost immediately, preventing them from completing their actions.

We found that we needed to reorganize the Parkinsoman patient's motor act so that the decisive stimulation for it would come from his higher cortical processes. We accomplished this by arranging matters so that a motor reaction was produced as an answer to an intellectual problem that the patient had solved mentally. We asked patients to give us their answers to problems by tapping their solutions. The questions were of the following sort: “How many angles occur in a square?” “How many wheels does a car have?” “How many points are there on the red stars of the Kremlin?”

We found that although the limitations of movement associated with muscular hypertonicity remained, the structure of the patient's motor act changed under these conditions. When we had instructed the patient simply to “press five times,” his first movements were strong but his subsequent movements diminished in intensity and muscular hypertension and tremors took over. Now when the patient signaled his mental solutions through movements, he showed no signs of exhaustion.

These early pilot studies were encouraging, but they also showed us how much we needed to learn if we were to make the study of the dissolution of higher psychological functions an integral part of our effort. We realized that we needed to undertake a study of the brain and its functional organization and to conduct clinical investigations in place of the experimental approach on which we had been relying. We also knew that the success of our work depended on a far better understanding of the structure of higher psychological functions, a line of investigation that was still in its infancy.

Undaunted, we decided to enter medical school. I resumed medical training in the late 1920s, beginning where I had left off many years before in Kazan. Vygotsky also began medical training. Professors in one school and students in another, we simultaneously taught, studied, and conducted our research.

In the early 1930s, a promising base for our work appeared when we were asked to set up a psychology department in the Ukrainian Psychoneurological Academy of Kharkhov. I began commuting between Kharkhov and Moscow, while Vygotsky commuted on a triangular route between Moscow, Leningrad, and Kharkov. It was in Kharkov that I first began to create new methods for the psychological analysis of the consequences of local brain lesions. But my time was still heavily occupied with other work. I lived this double existence until 1936 when I entered medical school on a full-time basis.

Immediately after passing my examinations at the First Moscow Medical School in 1937, I approached N. N. Bourdenko, a neurosurgeon who was head of the Neurosurgical Institute (now named in his honor), to ask him if I could intern at the institute. My plan was to train as a practical neurologist and at the same time to develop psychological methods for the diagnosis of local brain lesions. I do not know if Professor Bourdenko understood or approved of my plans. But he must have thought it worthwhile to have a psychology professor as an intern on his staff, because he accepted me.

My two years as an intern at the Neurosurgical Institute were the most fruitful of my life. I had no staff and no scientific responsibilities except routine medical work. In my free time I turned to my own research. It was during this period that I began to devise my own approach to the neuropsychology of local brain injuries.

In 1939 I moved to the Neurological Clinic of the Institute of Experimental Medicine, which later became the Neurological Institute of the Academy of Medical Sciences, to become head of the Laboratory of Experimental Psychology. With the hindsight of many years I can see that the move was a mistake. It would have been much more productive to remain as a staff member at the 300-bed Bourdenko Neurological Institute with patients whose local brain lesions had been verified by operations or post mortem. As events would have it, this error was corrected in the fullness of time, for as I write these lines I once again have a laboratory at the Bourdenko.

The period from 1937 to 1941 was taken up by my first serious work in neuropsychology. I soon found that in order to accumulate adequate clinical data I had to revise the basic style of my research. In experimental work a scholar usually begins by choosing a specific problem. Then he constructs a hypothesis and selects methods for testing his hypothesis. He arranges matters so that he can more easily focus his attention on those facts that will prove or disprove it. He is able to ignore all data that do not contribute to his analysis of the problem and to the proof of his hypothesis. By contrast, in clinical work, the starting point is not a clearly defined problem but an unknown bundle of problems and resources: the patient. The clinical investigator begins by making careful observations of the patient in an effort to discover the crucial facts. In the beginning, he can ignore nothing. Even data that on first glance seem insignificant may turn out to be essential. At some point the vague contours of factors that seem important begin to emerge, and the clinician forms a tentative hypothesis about the problem. But it is still too early for him to say definitely whether the facts he has picked out are important to the problem or extraneous. Only when he has found a sufficient number of compatible symptoms that together form a “syndrome” does he have a right to believe that his hypothesis about the patient might be proved or rejected.

At first I found it difficult to change from the logic of ordinary experimental investigation, which was imprinted on me, to the logic of clinical work. It took a while to learn to pay attention to those small events that can become a turning point in such investigations. The procedures and reasoning of such investigations seemed more like those used by detectives solving a crime than like the problem-solving behavior that prevails among psychologists and physiologists. In addition to giving up my reliance on experimental methods, I also felt it necessary to reject any use of the psychological tests of the era that had been created for the evaluation of an individual's intellectual level and which some investigators employed in the clinic. I found such test instruments as the Simon-Binet and other “measures of intellect” inadequate either to the task for which they had been designed or to the new applications that we had in mind.

The first problem to which I turned my attention was the tangled knot of disorders that were and still are referred to under the general rubric “aphasia.” At the time I began this work, three general classes of aphasia were recognized – sensory, motor, and semantic or amnestic – although there was great disagreement about the specific localization of each class and the character of the different capacities associated with each location. The first syndrome that we singled out for careful study, called “sensory aphasia,” was a form of speech disturbance associated with damage to the left temporal lobe, predominantly in secondary zones. This disability was termed sensory aphasia because the patient's ability to comprehend speech was disturbed, an observation that had led Wernicke to say that the “sensory images of speech” were decoded in this area. Our observations soon showed that the basic difficulty which underlay the other symptoms associated with sensory aphasia was an inability to discriminate the distinctive features of phonemes, the basic units of word sounds. Difficulties in understanding words, in naming objects, in retrieving words while speaking spontaneously, and in writing were secondary, or system-related, consequences of the primary defect in phonemic hearing.

The second form of aphasia that we tackled, called “motor aphasia,” was the disorder initially studied by Broca. Again we found that we were dealing not with a single syndrome, namely a center housing the motor images of speech, but with a variable set of symptoms within which we were able to distinguish two fundamentally different classes. Since our research on the motor aphasias illustrates neatly the basic logic of virtually all of my work, I shall digress somewhat to explain the distinctions we were forced to make and the broader approach to understanding the brain and its relation to psychological processes that resulted.

Speaking is only one of many voluntary acts that an individual performs. We therefore supposed that speaking would have a great deal in common with all kinds of complex, voluntary movements but would, like any other kind of movement, also have components that are specific unto itself. Thus, in order to understand motor aphasia, we needed to know more about voluntary motor responding in general and about the specific aspects which apply to speaking in particular.

Here we had the important advantage of being able to draw on the work of Bernshtein, who pointed out that movements require innervation not only by efferent nerve impulses that get muscle neurons to fire, but also by afferent nerve impulses that yield information about the state of the limbs. These afferent signals indicating the position of the limbs and the tension of the muscles are essential to restrict the infinite number of innervations that can occur and to reduce the degrees of freedom of movement. In cases where this system of afferent impulses is defective, no organized movement can take place. On the assumption that what was true of movement in general would be true of speech-related movement in particular, we hypothesized that motor aphasia would consist of two distinct varieties – one associated with disturbance of the efferent motor system, as Broca had surmised, and one associated with afferent defects.

True to our speculations, we found that there exists a kind of afferent motor aphasia, which I term “kinesthetic aphasia,” in which the principal symptom is the mispronunciation of individual speech sounds called “articulemes.” If the disturbance is severe, the patient may say k in place of b and t, which are different with regard to articulation. Less severe lesions produce more localized substitutions, such as b for p. The basic cause of this difficulty is that the brain does not register feedback from movements producing the articulemes, articulatory actions lose their selectivity, and the patient cannot assume correct positions of the tongue and lips.

Another form of motor aphasia concerns the serial organization of movements that are needed in pronunciation. In order to speak normally, it is necessary that the links between individual articulemes be organized so that smooth transitions are possible in our terminology, the entire kinetic melody that links words must be intact. However, when the lower parts of the premotor cortex of the speech areas are damaged, although the articulemes themselves remain intact, the patient cannot provide the required transition from one articuleme to the next. This “kinetic motor aphasia” is what Broca was referring to in his early observations.

The third classical form of aphasia that I began to study at this time was termed “semantic” or “amnestic aphasia,” after the word amnesia, “a state of forgetfulness.” Amnestic aphasia was supposed to be a special form of speech disorder, in which neither the sensory nor the motor defects are present, but patients find it difficult to remember the names of objects. Some neurologists of the period had speculated that this defect was caused by sensory disorders which destroyed presumed “sensory word traces.” Others hypothesized that the disorder reflected impairment of a special center where traces of language were stored. Still others, who generally favored the mass action approach to brain function, assumed that semantic aphasia was a result of deterioration in categorical thinking and the abstract attitude.

As in other classes of aphasic disorders, we were skeptical about hypotheses which assumed that all of the symptoms included under the rubric of semantic aphasia were really a single disturbance that could be precisely localized in a single area. We first satisfied ourselves that we were not dealing with some variety of motor or sensory aphasia, because patients with one or another of the semantic aphasia symptoms rarely showed signs of disturbed articulation or phonemic hearing. Their lesions also tended to occur in the parietotemporal area, above and behind the lesions characteristic of the aphaslas studied so far.

Next we searched the literature to see what kinds of symptoms arose in connections with semantic aphasia. In previous research and in our own observations we found that these patients had no difficulty grasping the meaning of complex ideas such as “causation,” “development,” or “cooperation.” They were also able to hold abstract conversations. But difficulties developed when they were presented with complex grammatical constructions which coded logical relations.

As Head pointed out in his work, these grammatical constructions require the coordination of details into a coherent whole. Such patients find it almost impossible to understand phrases and words which denote relative position and cannot carry out a simple instruction like “draw a triangle above a circle.” This difficulty goes beyond parts of speech that code spatial relations. Phrases like “Sonya is lighter than Natasha” also prove troublesome for these patients, as do temporal relations like “spring is before summer.”

Analysis shows that all of these logical-grammatical relations share a common feature: they are verbal expressions of spatial relations, although in some cases the spatial factor is more obvious than in others. The examples involving “above” and “to the right of” are clear, but on closer examination we found that in addition to linear relations expressed in such words as “before,” there are also spatial factors in such expressions as “the master's dog” or “brother's father.” As one patient put it in a particularly diagnostic manner, “Of course I know what brother and father are, but I can't imagine what the two mean together.”

All these examples demonstrate the error in assuming that semantic aphasia is a single, unitary syndrome. We found no evidence for uniform intellectual dissolution. What we did find was that a variety of mental operations requiring a component of spatial comparison and synthesis are disturbed.

My initial work on the three basic types of aphasia posited by neurologists in the 1920s and 1930s brought me to the end of my schuliabren. At that time I tried to summarize my ideas in what was intended to be a three-volume work, with one volume devoted to each kind of aphasia. I completed the first volume on sensory aphasia and defended it for my second higher degree, Doctor of Medicine. Although I began the second volume on semantic aphasia, analyses were too fragmentary, and this volume, like the first, remained unpublished. I also began to write about the forms of motor aphasia, but here too I found that I had no more than begun my work. All of these manuscripts remain in my desk. I can remember feeling that if only Vygotsky had lived, he would have penetrated far more deeply into the complex problems I had encountered. Not until the appearance of my Traumatic Aphasia in 1947 was a full treatment of these ideas brought together in print.

In June of 1941 the course of my work was permanently altered. World War II began.