1.5. The Double Brain

WHAT HAPPENS in the brain of a bicameral man? Anything as important in the history of our species as a completely different kind of mentality existing only a hundred generations ago demands some statement of what is going on physiologically. How is it possible? Given this profoundly subtle structure of nerve cells and fibers inside our skulls, how could that structure have been organized so that a bicameral mentality was possible? This is the great question of the present chapter. 

Our first approach to an answer is obvious. Since the bicameral mind is mediated by speech, the speech areas of the brain must be concerned in some important way.

Now in discussing these areas, and throughout this chapter, and indeed in the rest of this essay, I shall be using terms suitable only to right-handed people, in order to avoid a certain clumsiness of expression. Thus, it is the left cerebral hemisphere of the brain, controlling the right side of the body, which in righthanded people contains the speech areas. It is therefore commonly called the dominant hemisphere, while the right hemisphere, controlling the left side of the body, is commonly called the nondominant. I shall be speaking as if the left hemisphere were dominant in all of us. Actually, however, lefthanded persons have a variety of degrees of lateral dominance, some being completely switched (the right hemisphere doing what the left usually does), others not, and still others with mixed dominance. But being exceptional, only 5 percent of the population, they can be left out of the present discussion.

Jaynes's Neurological Model

The speech areas then are three, all on the left hemisphere in the great majority of mankind.1 They are: (I) the supplementary motor cortex, on the very top of the left frontal lobe, removal of which by surgery produces a loss of speech which clears up in several weeks; (2) Broca’s area, lower down at the back of the left frontal lobe, the removal of which produces a loss of speech which is sometimes permanent and sometimes not; and (3) Wernicke’s area, chiefly the posterior part of the left temporal lobe with parts of the parietal area, any large destruction of which after a certain age produces a permanent loss of meaningful speech.

It is thus Wernicke’s area that is the most indispensable to normal speech. As we might expect, the cortex in Wernicke’s area is quite thick with large, widely spaced cells, indicating considerable internal and external connections. While there is some disagreement as to its precise boundaries,2 there is none about its importance to meaningful communication.

 Of course it is extremely hazardous thinking to isomorphize between a conceptual analysis of a psychological phenomenon and its concomitant brain structure, yet this is what we cannot avoid doing. And among these three areas on the left hemisphere, or even in their more subtle interrelationships, it is difficult to imagine a duplication of some speech function to the extent and separation which my theory of the bicameral mind would demand.

Let us sit down with this problem a moment. Speech areas all on the left side. Why? One intriguing puzzle which has long fascinated me and anyone else who has considered the evolution of all this is why language function should be represented in only one hemisphere. Most other important functions are bilaterally represented. This redundancy in everything else is a biological advantage to the animal, since, if one side is injured, the other side can compensate. Why then was not language? Language, that most urgent and significant of skills, the pre-emptory and exigent ground of social action, the last communicant thread on which life itself in the post-glacial millennia must often have depended! Why was not this without-which-nothing of human culture represented on both hemispheres?

The problem drifts off into even more mystery when we remember that the neurological structure necessary for language exists in the right hemisphere as well as the left. In a child, a major lesion of Wernicke’s area on the left hemisphere, or of the underlying thalamus which connects it to the brainstem, produces transfer of the whole speech mechanism to the right hemisphere. A very few ambidextrous people actually do have speech on both hemispheres. Thus the usually speechless right hemisphere can under certain conditions become a language hemisphere, just like the left.

And a further range of the problem is what did happen in the right hemisphere as the aptic structures for language were evolving in the left? Just consider those areas on the right hemisphere corresponding to the speech areas of the left: what is their function? Or, more particularly, what is their important function, since it must have been such to preclude its development as an auxiliary speech area? If we stimulate such areas on the right hemisphere today, we do not get the usual “aphasic arrest” (simply the stopping of ongoing speech) which occurs when the normal language areas of the left hemisphere are stimulated. And because of this apparent lack of function, it has often been concluded that large portions of the right hemisphere are simply unnecessary. In fact, large amounts of right hemisphere tissue, including what corresponds to Wernicke’s area, and even in some instances the entire hemisphere, have been cut out in human patients because of illness or injury, with surprisingly little deficit in mental function. 

The situation then is one where the areas on the right hemisphere that correspond to the speech areas have seemingly no easily observable major function. Why this relatively less essential part of the brain? Could it be that these silent ‘speech’ areas on the right hemisphere had some function at an earlier stage in man’s history that now they do not have? 

The answer is clear if tentative. The selective pressures of evolution which could have brought about so mighty a result are those of the bicameral civilizations. The language of men was involved with only one hemisphere in order to leave the other free for the language of gods.

If so, we might expect that there would have to be certain tracts by which the bicameral voices would relate between the right nondominant temporal lobe and the left. The major interconnection between the hemispheres is of course the huge corpus callosum of over two million fibers. But the temporal lobes in men have their own private callosum, so to speak, the much smaller anterior commissures. In rats and dogs, the anterior commissures connect the olfactory parts of the brain. But in men, as seen in my rather imprecise sketch, this transverse band of fibers collects from most of the temporal lobe cortex but particularly the middle gyrus of the temporal lobe included in Wernicke’s area, and then squeezes into a tract only slightly more than one eighth of an inch in diameter as it plunges over the amygdala across the top of the hypothalamus toward the other temporal lobe. Here then, I suggest, is the tiny bridge across which came the directions which built our civilizations and founded the world’s religions, where gods spoke to men and were obeyed because they were human volition.3

Jaynes's Neurological Model
In ancient times, what corresponds to Wernicke’s area on the right
hemisphere may have organized admonitory experience and coded
it into ‘voices’ which were then ‘heard’ over the anterior commissure
by the left or dominant hemisphere.

There are two forms in which this hypothesis can be specified.

The stronger form, and the one I favor because it is simpler and more specific (and thus more easily verified or disconfirmed by empirical investigation), is that the speech of the gods was directly organized in what corresponds to Wernicke’s area on the right hemisphere and ‘spoken’ or ‘heard’ over the anterior commissures to or by the auditory areas of the left temporal lobe. (Note how I can only express this metaphorically, personifying the right temporal lobe as a person speaking or the left temporal lobe as a person listening, both being equivalent and both literally false.) Another reason I am inclined to this stronger form is its very rationality in terms of getting processed information or thought from one side of the brain to the other. Consider the evolutionary problem: billions of nerve cells processing complex experience on one side and needing to send the results over to the other through the much smaller commissures. Some code would have to be used, some way of reducing very complicated processing into a form that could be transmitted through the fewer neurons particularly of the anterior commissures. And what better code has ever appeared in the evolution of animal nervous systems than human language? Thus in the stronger form of our model, auditory hallucinations exist as such in a linguistic manner because that is the most efficient method of getting complicated cortical processing from one side of the brain to the other.

The weaker form of the hypothesis is more vague. It states that the articulatory qualities of the hallucination were of left hemisphere origin like the speech of the person himself, but that its sense and direction and different relation to the person were due to right temporal lobe activity sending excitation over the anterior commissures and probably the splenium (the back part of the corpus callosum) to the speech areas of the left hemisphere, and ‘heard’ from there.

At the present time, it does not really matter which form of the hypothesis we take. The central feature of both is that the amalgamating of admonitory experience was a right hemisphere function and it was excitation in what corresponds to Wernicke’s area on the right hemisphere that occasioned the voices of the gods.

The evidence to support this hypothesis may be brought together as five observations: (I) that both hemispheres are able to understand language, while normally only the left can speak; (2) that there is some vestigial functioning of the right Wernicke’s area in a way similar to the voices of gods; (3) that the two hemispheres under certain conditions are able to act almost as independent persons, their relationship corresponding to that of the man-god relationship of bicameral times; (4) that contemporary differences between the hemispheres in cognitive functions at least echo such differences of function between man and god as seen in the literature of bicameral man; and (5) that the brain is more capable of being organized by the environment than we have hitherto supposed, and therefore could have undergone such a change as from bicameral to conscious man mostly on the basis of learning and culture.

The rest of this chapter will be devoted to these five observations.

1. That Both Hemispheres Understand Language

The gods, I have said with some presumption, were amalgams of admonitory experience, made up of meldings of whatever commands had been given the individual. Thus, while the divine areas would not have to be involved in speech, they would have to be involved in hearing and in understanding language. And this is the case even today. We do in fact understand language with both hemispheres. Stroke patients who have hemorrhages on the left side of the cortex cannot speak, but still can understand.4 If sodium amytal is injected into the left carotid artery leading to the left hemisphere (the Wada test), the entire hemisphere is anesthetized, leaving only the right hemisphere working; but the subject still can follow directions.5 Tests on commissurotomized patients (which I shall describe more fully in a moment) demonstrate considerable understanding by the right hemisphere.6 Named objects can usually be retrieved by the left hand, and verbal commands obeyed by the left hand. Even when the entire left hemisphere, the speech hemisphere, remember, is removed in human patients suffering from glioma, the remaining right hemisphere immediately after the operation seems to understand the surgeon’s questions, though unable to reply.7

2. That There Exists Vestigial Godlike Function in the Right Hemisphere

If the preceding model is correct, there might be some residual indication, no matter how small, of the ancient divine function of the right hemisphere. We can, indeed, be more specific here. Since the voices of the gods did not, of course, entail articulate speech, did not entail the use of the larynx and mouth, we can rule out what corresponds to Broca’s area and the supplementary motor area, to a certain extent, and concentrate on what corresponds to Wernicke’s area or the posterior part of the temporal lobe on the right or so-called nondominant side. If we stimulate it in this location, would we hear then the voices of the gods as of yore? Or some remnant of them? Something that would allow us to think that three thousand years ago its function was that of the divine direction of human affairs?

We may recall that this was indeed the very area which had been stimulated by Wilder Penfield in a famous series of studies a few years ago.8 Let me describe them in some detail. 

These observations were made on some seventy patients with a diagnosis of epilepsy caused by lesions somewhere in the temporal lobe. As a preliminary to the removal of the damaged brain tissue by surgery, various points on the surface of the temporal lobe were stimulated with a gentle electric current. The intensity of the stimulation was approximately the least current needed to excite tingling in the thumb by stimulation of the appropriate motor area. If it be objected that the phenomena resulting from this stimulation are corrupted by the presence of some focal area of gliosis, or sclerosis, or meningo-cerebral cicatrix, all typically found in such patients, I think such objections would be dissipated by reviewing the original report. These abnormalities, when found, were circumscribed in location and were not in any way influencing the responses of the subject as they were being stimulated.9 It can thus be assumed with some confidence that the results of these studies are representative of what would be found in normal individuals. 

In the great majority of these cases, it was the right temporal lobe that was stimulated, particularly the posterior part of the temporal lobe toward its superior convolution, Wernicke’s area on the right side. A remarkable series of responses from the patients was obtained. This is, to repeat myself, the point at which we might expect to hear the gods of antiquity calling to us again, as if from the other part of our bicameral minds. Would these patients hear some vestiges of the ancient divinities?

Here are some representative data.

When stimulated in this region, Case 7, a twenty-year-old college student, cried out, “Again I hear voices, I sort of lost touch with reality. Humming in my ears and a small feeling like a warning.” And when stimulated again, “Voices, the same as before. I was just losing touch with reality again.” When asked, he replied that he could not understand what the voices were saying. They sounded “hazy.”

In the majority of cases, the voices were similarly hazy. Case 8, a twenty-six-year-old housewife, stimulated in approximately the same area, said there seemed to be a voice a way, way off. “It sounded like a voice saying words but it was so faint I couldn’t get it.” Case 12, a twenty-four-year-old woman, stimulated at successive points of the superior gyrus of the posterior temporal lobe, said, “I could hear someone talking, murmuring or something.” And then further on, “There was talking or murmuring, but I cannot understand it.” And then stimulated about three quarters of an inch along the gyrus, she was at first silent, and then gave a loud cry. “I heard the voices and then I screamed. I had a feeling all over.” And then stimulated a little back toward the first stimulations, she began to sob. “That man’s voice again! The only thing I know is that my father frightens me a lot.” She did not recognize the voice as her father’s; it only reminded her of him.

Some patients heard music, unrecognized melodies that could be hummed to the surgeon (Cases 4 and 5). Others heard relatives, particularly their mothers. Case 32, a twenty-two-year-old woman, heard her mother and father talking and singing, and then stimulated on another point, her mother “just yelling.”

Many patients heard the voices as emanating from strange and unknown places. Case 36, a twenty-six-year-old woman, stimulated somewhat anteriorly on the superior gyrus of the right temporal lobe, said, “Yes, I heard voices down along the river somewhere — a man’s voice and a woman’s voice, calling.” When asked how she could tell it was down along the river, she said, “I think I saw the river.” When asked what river, she said, “I do not know, it seems to be one I was visiting when I was a child.” And at other stimulation points, she heard voices of people calling from building to building somewhere. And at an adjacent point, the voice of a woman calling in a lumberyard, though she insisted that she had “never been around any lumberyard.”

When the voices were located as coming from one side or the other, as rarely happened, it was from the contralateral side. Case 29, a twenty-five-year-old man, stimulated in the middle of the right temporal gyrus, said, “Someone telling me in my left ear, ‘Sylvere, Sylvere!’ It could have been my brother.” 

The voices and the music, whether garbled or recognized, were experienced as actually heard, and the visual hallucinations were experienced as-actually seen, just as Achilles experienced Thetis, or Moses heard Yahweh out of the burning bush. Case 29, the same as above, when stimulated again, also saw “someone speaking to another and he mentioned the name, but I could not understand it.” And when asked whether he saw the person he replied, “It was just like a dream.” And when asked further if the person was there, he said, “Yes, sir, about where the nurse with the eyeglasses is sitting over there.” 

In some slightly older patients, only exploratory stimulation produced an hallucination. A thirty-four-year-old French-Canadian, Case 24, after previous stimulations had produced nothing, when stimulated on the posterior part of the middle gyrus of the right temporal lobe, suddenly said, “Wait a minute, I see someone! ” And then about an inch higher, “Oui, la, la, la! It was he, he came, that fool!” And then stimulated somewhat higher though still within what corresponds to Wernicke’s area on the right side, “There, there, j’entend! It is just that somebody wanted to speak to me, and he was going, ‘vite, vite, vite!’

But at younger ages, there is a definite suggestion that hallucinations caused by stimulating the right temporal lobe are more striking, vivid, and admonitory. A fourteen-year-old boy (Case 34) saw two men sitting in armchairs singing at him. A fourteenyear-old girl, Case 15, when stimulated on the superior posterior gyrus of the right temporal lobe, cried out, “Oh, everybody is shouting at me again, make them stop!” The stimulus duration was two seconds; the voices lasted eleven seconds. She explained, “They are yelling at me for doing something wrong, everybody is yelling.” At all stimulation points along the posterior temporal lobe of the right hemisphere, she heard yelling. And even when stimulated an inch and a half posterior to the first point, she cried out, “There they go, yelling at me; stop them!” And the voices coming from just one stimulation lasted twentyone seconds. 

I should not give the impression that it is all this simple. I have selected these cases. In some patients, there was no response at all. Occasionally such experiences involved autoscopic illusions such as we referred to in I.2. A further complication is that stimulation of corresponding points on the left or usually dominant hemisphere may also result in similar hallucinations. In other words, such phenomena are not confined to the right temporal lobe. But the instances of response to stimulation on the left are much less frequent and occur with less intensity.

The important thing about almost all these stimulation-caused experiences is their otherness, their opposition from the self, rather than the self’s own actions or own words. With a few exceptions, the patients never experienced eating, talking, sex, running, or playing. In almost all instances, the subject was passive and being acted upon, exactly as a bicameral man was acted upon by his voices.

Being acted upon by what? Penfield and Perot think it is simply past experience, flashbacks to earlier days. They try to explain the failure of recognition so consistently observed as mere forgetfulness. They assume that these were actual specific memories that with more time during the operation could have been pushed into full recognition. In fact, their questions to the patients during stimulation were guided by this hypothesis. Sometimes, indeed, the patient did become specific in his replies. But far more representative of the data as a whole is the patients’ persistence under questioning that these experiences could not be called memories.

Because of this, and because of the general absence of personal active images, which are the usual kind of memories that we have, I suggest that the conclusions of Penfield and Perot are incorrect. These areas of the temporal lobe are not “the brain’s record of auditory and visual experience,” nor are they its retrieval, but combinations and amalgamations of certain aspects of that experience. The evidence does not, I think, warrant the assertion that these areas “play in adult lives some role in the subconscious recall of past experience, making it available for present interpretation.” Rather the data lead away from this, to hallucinations that distill particularly admonition experiences, and perhaps become embodied or rationalized into actual experiences in those patients who reported them on being questioned.

3. That the Two Hemispheres Can Behave Independently

In our brain model of the bicameral mind, we have assumed that the god part and the man part behaved and thought somewhat independently. And if we now say that the duality of this ancient mentality is represented in the duality of the cerebral hemispheres, is this not personifying parts of the brain without warrant? Is it possible to think of the two hemispheres of the brain almost as two individuals, only one of which can overtly speak, while both can listen and both understand?

The evidence that this is plausible comes from another group of epileptics. These are the dozen or so neurosurgical patients who have undergone complete commissurotomy, the cutting down the midline of all interconnections between the two hemispheres.10 This so-called split-brain operation (which it is not ? the deeper parts of the brain are still connected) usually cures the otherwise untreatable epilepsy by preventing the spread of abnormal neural excitation over the whole cortex. Immediately after operation, some patients lose speech for up to two months, while others have no problem whatever ? no one knows why. Perhaps each of us has a slightly different relationship between our hemispheres. Recovery is gradual, all patients showing short-term memory deficits (perhaps due to the cutting of the small hippocampal commissures), some orientation problems, and mental fatigue.

Now the astonishing thing is that such patients after a year or so of recovery do not feel any different from the way they felt before the operation. They sense nothing wrong. At the present time they are watching television or reading the paper with no complaints about anything peculiar. Nor does an observer notice anything different about them.

But under rigorous control of sensory input, fascinating and important defects are revealed.

As you look at anything, say, the middle word of this line of print, all the words to the left are seen only by the right hemisphere, and all the words to the right only by the left. With the connections between the hemispheres intact, there is no particular problem in co-ordinating the two, although it really is astonishing that we can read at all. But if you had your hemispheric connections cut, the matter would be very different. Starting at the middle of this line, all the print to your right would be seen as before and you would be able to read it off almost as usual. But all the print and all the page to your left would be a blank. Not a blank really, but a nothing, an absolute nothing, far more nothing than any nothing you can imagine. So much nothing that you would not even be conscious that there was nothing there, strange as it seems. Just as in the phenomenon of the blind spot, the ‘nothing’ is somehow ‘filled in’, ‘stitched together’, as if nothing were wrong with nothing. Actually, however, all that nothing would be in your other hemisphere which would be seeing all that ‘you’ were not, all the print to the left, and seeing it perfectly well. But since it does not have articulated speech, it cannot say that it sees anything. It is as if ‘you’ ? whatever that means ? were ‘in’ your left hemisphere and now with the commissures cut could never know or be conscious of what a quite different person, once also ‘you’, in the other hemisphere was seeing or thinking about. Two persons in one head.

This is one of the ways these commissurotomized patients are tested. The patient fixates on the center of a translucent screen; photographic slides of objects projected on the left side of the screen are thus seen only by the right hemisphere and cannot be reported verbally, though the patient can use his left hand (controlled by the right hemisphere) to point to a matching picture or search out the object among others, even while insisting vocally that he did not see it.11 Such stimuli seen by the right nondominant hemisphere alone are there imprisoned, and cannot be ‘told’ to the left hemisphere where the language areas are because the connections have been cut. The only way we know that the right hemisphere has this information at all is to ask the right hemisphere to use its left hand to point it out ? which it can readily do.

If two different figures are flashed simultaneously to the right and left visual fields, as, for example, a “dollar sign” on the left and a “question mark” on the right, and the subject is asked to draw what he saw, using the left hand out of sight under a screen, he draws the dollar sign. But asked what he has just drawn out of sight, he insists it was the question mark. In other words, the one hemisphere does not know what the other hemisphere has been doing.

Again, if the name of some object, like the word ‘eraser’, is flashed to the left visual field, the subject is then able to search out an eraser from among a collection of objects behind a screen using only the left hand. If the subject is then asked what the item is behind the screen after it has been selected correctly, ‘he’ in the left hemisphere cannot say what the dumb ‘he’ of the right hemisphere is holding in his left hand. Similarly, the left hand can do this if the word ‘eraser’ is spoken, but the talking hemisphere does not know when the left hand has found the object. This shows, of course, what I have said earlier, that both hemispheres understand language, but it has never been possible to find out the extent of language understanding in the right hemisphere previously.

Further, we find that the right hemisphere is able to understand complicated definitions. Flashing “shaving instrument” onto the left visual field and so into the right hemisphere, the left hand points to a razor, or with “dirt remover” to soap, and with “inserted in slot machines” to a twenty-five-cent piece.12

Moreover, the right hemisphere in these patients can respond emotionally without the left talking hemisphere knowing what it is all about. If among a series of neutral geometric figures being flashed to the right and left visual fields at random, which means respectively into the left and right hemispheres, and then a picture of a nude girl by surprise is flashed on the left side going into the right hemisphere, the patient (really the patient’s left hemisphere) says that it saw nothing or just a flash of light. But the grinning, blushing, and giggling during the next minute contradicts what the speech hemisphere has just said. Asked what all the grinning is about, the left or speech hemisphere replies that it has no idea.13 These facial expressions and blushings, incidentally, are not confined to one side of the face, being mediated through the deep interconnections of the brainstem. The expression of affect is not a cortical matter.

Similarly with other sensory modalities. Odors presented to the right nostril and so to the right hemisphere (olfactory fibers do not cross) in these patients cannot be named by the talking hemisphere, though the latter can say very well whether the smell is pleasant or unpleasant. The patient may even grunt, make aversive reactions, or exclaim “Phew! ” to a stench, but cannot say verbally whether it is garlic, cheese, or decayed matter.14 The same odors presented to the left nostril can be named and described perfectly well. What this means is that the emotion of disgust gets across to the speaking hemisphere through the intact limbic system and brainstem, while the more specific information processed by the cortex does not.  

Indeed, there is some indication that it is the right hemisphere that commonly triggers the emotional reactions of displeasure from the limbic system and brainstem. In test situations, where the speechless right hemisphere is made to know the correct answer, and then hears the left dominant hemisphere making obvious verbal mistakes, the patient may frown, wince, or shake his head. It is not simply a way of speaking to say that the right hemisphere is annoyed at the erroneous vocal responses of the other. And so perhaps the annoyance of Pallas Athene when she grasped Achilles by his yellow hair and twisted him away from murdering his king (Iliad, 1 .-197). Or the annoyance of Yahweh with the iniquities of his people.

Of course there is a difference. Bicameral man had all his commissures intact. But I shall suggest later that it is possible for the brain to be so reorganized by environmental changes that the inferences of my comparison here are not entirely foolish. At any rate, the studies of these commissurotomy patients demonstrate conclusively that the two hemispheres can function so as to seem like two independent persons, which in the bicameral period were, I suggest, the individual and his god. 

4. That Hemispheric Differences in Cognitive Function Echo the Differences of God and Man

If this brain model of the bicameral mind is correct, it would predict decided differences in cognitive function between the two hemispheres. Specifically, we would expect that these functions necessary for the man-side would be in the left or dominant hemisphere, and those functions necessary to the gods would be more emphasized in the right hemisphere. Moreover, there is no reason not to think that the residuals of these different functions at least are present in the brain organization of contemporary man. 

The function of the gods was chiefly the guiding and planning of action in novel situations. The gods size up problems and organize action according to an ongoing pattern or purpose, resulting in intricate bicameral civilizations, fitting all the disparate parts together, planting times, harvest times, the sorting out of commodities, all the vast putting together of things in a grand design, and the giving of the directions to the neurological man in his verbal analytical sanctuary in the left hemisphere. We might thus predict that one residual function of the right hemisphere today would be an organizational one, that of sorting out the experiences of a civilization and fitting them together into a pattern that could ‘tell’ the individual what to do. Perusal of various speeches of gods in the Iliad, the Old Testament, or other ancient literatures is in agreement with this. Different events, past and future, are sorted out, categorized, synthesized into a new picture, often with that ultimate synthesis of metaphor. And these functions should, therefore, characterize the right hemisphere.

Clinical observations are consistent with this hypothesis. From the commissurotomized patients of a few pages past, we know that the right hemisphere with its left hand is excellent at sorting out and categorizing shapes, sizes, and textures. From braindamaged patients, we know that damage to the right hemisphere interferes with spatial relations and with gestalt, synthetic tasks.15 Mazes are problems in which various elements of a spatial pattern must be organized in learning. Patients in whom the right temporal lobe has been removed find learning the pathways of visual and tactile mazes almost impossible, while patients with lesions of equal extent on the left temporal lobe have little difficulty.16

Another task involving the organization of parts into a spatial pattern is Koh’s Block Test, commonly used in many intelligence tests. The subject is shown a simple geometric pattern, and asked to duplicate it with blocks that have its elements painted on them. Most of us can do it easily. But patients with brain lesions in the right hemisphere find this extremely difficult, so much so that the test is used to diagnose right hemisphere damage. In the commissurotomy patients referred to earlier, the right hand often cannot succeed at all in putting the design together with the blocks. The left hand, in a sense the hand of the gods, has no problem whatever. In some of the commissurotomy patients, the left hand had even to be held back by the observer as it tried to help the right hand in its fumbling attempts at this simple task.17 The inference has thus been drawn from these and other studies that the right hemisphere is more involved in synthetic and spatial-constructive tasks while the left hemisphere is more analytic and verbal. The right hemisphere, perhaps like the gods, sees parts as having a meaning only within a context; it looks at wholes. While the left or dominant hemisphere, like the man side of the bicameral mind, looks at parts themselves.

These clinical results have been confirmed in normal people in what promises to be the first of many future studies.18 EEG electrodes were placed over the temporal and parietal lobes on both sides of normal subjects who were then given various tests. When asked to write various kinds of letters involving verbal and analytic abilities, the EEG records show low-voltage fast waves over the left hemisphere, denoting that the left hemisphere is doing the work, while slow alpha waves (seen on both hemispheres in a resting subject with the eyes closed) are seen over the right hemisphere, indicating that it is not doing the work. When such subjects are given spatial synthetic tests, such as Koh’s Block Test as used in the clinical studies above, the reverse is found. It is now the right hemisphere that is doing the work.

Further deductions can be made about what particular functions might be residual in the right hemisphere by considering what it is that the divine voices of the bicameral mind would have to do in particular situations. To sort out and synthesize experience into directives to action, the gods would have to make certain kinds of recognitions. Throughout the speeches of gods in ancient literature, such recognitions are common. I do not mean recognitions of individuals in particular, but more generally of types of people, of classifications, as well as of individuals. One important judgment for a human being of any century is the recognition of facial expression, particularly in regard to friendly or unfriendly intent. If a bicameral man saw an unrecognized man coming toward him, it would be of considerable survival value for the god-side of his mentality to decide if the person was of friendly or unfriendly intent.

The adjoining figure is an experiment I designed about ten years ago out of such a supposition. The two faces are mirror images of each other. I have so far asked almost a thousand people which face looks happier. Quite consistently, about 80 percent of right-handed people chose the bottom face with the smile going up on their left. They were thus judging the face with their right hemispheres, assuming, of course, that they were glancing at the center of the face. This result can be made stronger by tachistoscopic presentation. With the focal point in the center and flashed at one tenth of a second, the bottom face always looks happier to right-handed persons. 

An alternative hypothesis, of course, is that this tendency to judge facial expression by the left visual field is a carry-over of reading from left to right. And in our cultures it certainly enhances the effect. But that the hemispheric explanation is at the bottom of it is suggested by the results for left-handed people. Fifty-five percent of left-handers chose the upper face as happier, suggesting that it was the left hemisphere making the judgment. And this cannot be understood on the reading-direction hypothesis. Also, in people who are completely left-lateralized, left-handed in every way, the likelihood of seeing the top face as happier seems to be much higher. 

Recently we have made a similar finding, using photographs of an actor expressing sadness, happiness, disgust, and surprise.19 Our subjects, carefully screened for right-handedness, first stared at the fixation point in a tachistoscope, then were presented with one photograph flashed for a few milliseconds in the central position, and then with another either in the right or left visual field for the same duration. The subjects were asked to say whether the photographs were the same or different, and the time taken to make this decision was recorded. Most of the subjects were able to match facial expressions more correctly and in less time when the face was presented on the left and hence to the right hemisphere. In a control condition, scrambled pictures of the same facial expressions (which were really nonsense patterns) also could be matched more quickly and correctly when presented on the left, but not nearly as well as the facial expressions themselves.

Recent clinical evidence is in clear agreement. Failure to recognize faces, not just facial expressions, is much more frequently associated with damage to the right hemisphere than to the left. In clinical testing, the patient is asked to match the frontal view of a face with three-quarter views of the same face under different lighting conditions. Patients with lesions in the right hemisphere find this extremely difficult in comparison with normal subjects or patients who have lesions in the left hemisphere.20 Recognition of both faces and facial expression is therefore primarily a right hemisphere function.

And to tell friend from non-friend in novel situations was one of the functions of a god.

6. A New Look at the Brain

How, it may be argued, can such a system as this, a brain structured into what I have called a bicameral mind, this substrate of human civilization for thousands of years, involving such loci as we have mentioned in the model, how can its function change over so short a period of time, such that the admonitory voices are heard no more and that we have this new organization called consciousness? While the amount of genocide going on in the world during these changes was enough to allow some natural selection and evolution, I in no way wish to rest the case upon that. Such natural selection as did occur during these periods of the development of consciousness certainly assisted in its perpetuation, but could not be said to have evolved consciousness out of the bicameral mind in the sense that a seaPs flipper is evolved out of an ancestral paw.

A true understanding of the situation requires a different view of the brain from that which was usual a few decades ago. Its emphasis is the brain’s plasticity, its redundant representation of psychological capacities within a specialized center or region, the multiple control of psychological capacities by several centers either paired bilaterally or as what Hughlings Jackson recognized as “representations” of a function lying at successively higher and phylogenetically younger levels of the nervous system.21 The organization of the mammalian brain in this fashion allows for those experimental phenomena brought together under the rubric of “recovery of function.” Its emphasis gives a view of the brain much more plastic than usual, with a dramatic surplus of neurons such that, for example, 98 percent of the optic tracts can be cut in the cat, and brightness and pattern discrimination will remain.22 The brain teems with redundant centers, each of which may exert direct influence on a final common pathway, or modulate the operation of others, or both, their arrangements able to assume many forms and degrees of coupling between constituent centers.

All this redundant representation in multiple control gives us the notion of a much more changeable kind of brain than the earlier neurologists described. A particular behavior or group of behaviors engage a host of similar neurons in a given center and may call into play several different centers arranged in various patterns of inhibition and facilitation, depending upon their evolutionary status. And the tightness of the coupling between centers varies tremendously from one function to another.23 In other words, the amount of changeableness that the locus of cortical functions can undergo is different among different functions, but that such changeableness is a pronounced feature of the higher mammalian brain is becoming more and more apparent. The biological purpose or selective advantage of such redundant representation and multiple control and its resulting plasticity is twofold: it protects the organism against the effects of brain damage, and, perhaps more important, it provides an organism of far greater adaptability to the constantly changing environmental challenges. I am thinking here of such challenges as characterize the successive glaciations of primate man’s existence, and, of course, that even greater challenge of the breakdown of the bicameral mind to which man adapted with consciousness.

But this does not mean just that adult man’s behavior is less rigid than his forebear’s, though this is of course true. More important, it provides an organism where the early developmental history of the individual can make a great difference in how the brain is organized. Some years ago, an idea such as this would have seemed very far-fetched indeed. But the increasing tide of research has eroded any rigid concept of the brain, and has emphasized the remarkable degree to which the brain can compensate for any structures missing either by injury or by congenital malformation. Many studies show that brain injury to animals in infancy may make little difference in adult behavior, while similar injury to adults may have profound changes. We have already noted that early injury to the left hemisphere usually results in the switch of the entire speech mechanism to the right hemisphere. 

One of the most astonishing of the cases that demonstrate this resiliency of the brain is that of a thirty-five-year-old man who died of an abdominal malignancy. At autopsy, it was revealed that he had a congenital absence of the hippocampal fimbria, the fornix, septum pellucidum, and the mass intermedia thalami, with an abnormally small hippocampus and abnormally small hippocampal and dentate gyri. In spite of these remarkable abnormalities, the patient had always displayed an “easygoing” personality and had even led his class in school!24

Thus, the growing nervous system compensates for genetic or environmental damage by following other but less preferred developmental paths which utilize intact tissue. In adults, with development completed, this is no longer possible. The normally preferred modes of neural organization have already been achieved. It is only in early development that such reorganization of the systems of multiple control can take place. And this is definitely true of the relationship between the hemispheres so central to this discussion.25

With this as a background, I do not see the difficulty in considering that, in the bicameral epochs, what corresponds to Wernicke’s area on the right nondominant hemisphere had its strict bicameral function, whereas after a thousand years of psychological reorganization in which such bicamerality was discouraged when it appeared in early development, such areas function in a different way. And similarly, it would be wrong to think that whatever the neurology of consciousness now may be, it is set for all time. The cases we have discussed indicate otherwise, that the function of brain tissue is not inevitable, and that perhaps different organizations, given different developmental programs, may be possible.