Descartes' Error: Emotion, Reason, and the Human Brain. By A.R. Damasio. 1994. New York: G.P. Putman and Sons ISBN-0-399-13894-3.
It is the mark of a fashion's zenith that more and more people claim to have suddenly discovered the same original idea. The body, by that test, is definitely in. Though the announcement that the mind and body are not separate entities is rather old news to philosophers, it is still proclaimed as a novel idea in a number of recent books, including the two under review.
"Descartes," in Damasio's title, usually stands for some son of rigid dualism of mind and body: but what exactly might that mean? Descartes himself, as evidenced by his own book on the Passions of the soul, thought emotions had everything to do with the body and served biological functions. Just what it means to say mind and body are nor separate, however, is hardly obvious: the standard form of individualism in recent years has viewed the brain as a computational system. But that view can itself be seen as "dualist": the same software can run on different computers, as souls might inhabit different bodies. If computers are bodiless brains, moreover, they may turn out to be poor models for embodied minds. How do brain and mind, in turn, relate to the body?
Rosenfield's and Damasio's books both address this last question. Like the whimsically titled books of Oliver Sacks (1987, 1995), they illustrate the way that neurological tales have become the Aesop's Fables of our time. Through the barely credible wonders revealed to the high priests of brain research, we hope to learn the deepest truths about our nature and our destiny.
Both books make stimulating reading. Instead of merely undertaking another sisyphean assault on the tired "mind-body problem", they cast the light of neurology onto the question of what makes for the uniqueness of individual persons, what makes you or me the separate and unique persons that we are. Much of the answer, for both, rests on our specific repertoire of emotional responses; the exact nature and extent of that repertoire in turn lies not only in the brain, but in the body, its dispositions and positions, and our own perceptions of it. This view is summed up in almost identical phrases by both authors: Rosenfield describes the body as "the brain's absolute frame of reference" (R45), while Damasio says "the body, represented in the brain, may constitute the indispensable frame of reference for the neural processes that we experience as the mind." (Dxvi). Furthermore, they both go on to suggest that such monitoring of the body's responses constitutes an essential component of our individual subjectivity. For both, then, computers can have nothing that can really count as a mind unless they have a body.
Beyond these similarities, the two books are widely different. Rosenfield is primarily a historian of modern cognitive science; he offers an imaginative reinterpretation of neurological evidence collected over the last century and a half. Damasio is an active researcher whose reinterpretation of classic cases is richly supplemented with accounts of his own specific hypotheses and of ingenious experiments designed to test them. He goes much further than does Rosenfield in presenting a richly integrated hypothesis about the nature and role of emotions in the rational conduct of life.
Rosenfield, like John Locke and David Hume, views our personal identity as essentially residing in individual consciousness and memory. His view is unlike theirs, however, in two important ways. First, he stresses that a memory is never merely retrieved, because it doesn't in any literal sense subsist intact, waiting to be brought to consciousness. Rather it is reconstructed, in a process which fashions both the memory and consciousness itself. Second, our consciousness is somehow rooted in our subjective perception of our own bodies, or what Rosenfield calls our body image.
This last point, which is the central thesis of Rosenfield's book, could be viewed as an extension of the philosophical argument made some thirty years ago by P.F. Strawson in his Individuals. What Strawson demonstrated was that there is no possibility of acquiring the concept of an objective world, or the distinction between self and non-self without a concept of objective space. That in turn, requires that some things - physical objects of perceptible size - be privileged as "basic particulars". And some of those must be identified egocentrically: that is, they must be picked out by "token reflexives" such as I, now, here. Token reflexives identify their referents in terms of the location of the speaker in Strawson's scheme, that location was purely punctual: the self that needs to be identified in this way requires no thickness. Rosenfield quite literally fleshes out the self starting from the obvious fact that the speaker is not punctual but a specific extended body shape. Thus, in order to be conscious, we need to be selfconscious, and the self that selfconsciousness is consciousness of is the body: in the phrase already cited, "the body, not a sense of absolute space, is the brain's absolute frame of reference" (R45). Since memory, in turn, is closely bound up with conscious itself, it then turns out that in order for both memory and ordinary consciousness to function normally we need to identify not only the position of our own body in space, but its shape and posture. Proprioception is the mother of memory.
This chain of reasoning involves some bold leaps over some slipping conceptual crevasses. it is accompanied (if not exactly supported), however, by some interesting observations about some classic but still intriguing neurological celebrities. These include Luria's mnemonist (who turns out on Rosenfield's view, to have "a very poor memory" (R29); la Tourette's coprolalic Madame X, who bequeathed his name to her syndrome; Oliver Sacks himself, who described the eerie unreality of his own leg as a result of a mountain accident that left it temporarily paralyzed; and John Hull, who wrote a moving account of the effects on him of his sudden blindness in his early twenties. My favourite is perhaps Henri Baud, who was unable to remember anything of the 18 previous years, yet amiably claimed to have visited his mistress at the Rue Cambaceres "last Saturday", a tryst encased forever in the amber of his loss of time sense.
Some patients are confused in their illness; others are indifferent. Both attitudes are instructive. Rosenfield points out that confusion is a mark of consciousness, "computers for example, which lack consciousness, do not become confused when they arrive at contradictory conclusions or when part of their 'memory' is lost." (R11). This alone seems to disqualify computer models of consciousness. (True, one might say, but then few computer models have actually been claimed to explain consciousness, as opposed to some aspect or other of mentality.) It also leads to the interesting observation that "the failure of memory in a patient with brain damage is quite different from... our ordinary experience of forgetting." (R21). This is particularly marked in some patients who seem to have no awareness of their deficit. At its most extreme, their indifference amounts to anosognosia, the complete lack of awareness of their own problem. In such patients, Rosenfield claims, what is disrupted is their body image. This is also what happened, he argues, to Oliver Sacks after his accident. (And since Sacks enthusiastically endorses the book on the back cover, one assumes he felt no need to demur.) So, too with the "dual personality" of Mary Reynolds, who was described in 1890 by William James and anticipated the current swarm of MPD sufferers: all these strange phenomena "become less mysterious when one recognizes them as alterations and breakdowns in the mechanisms of subjectivity." (R27) The case that Rosenfield uses to exhibit most directly the mechanism that underlies this sense of subjectivity is the case of Madame I, a patient at the Salpétrière Hospital described in 1905. This patient had lost not only all affect but the sense of familiarity with her own family members. She had also lost the awareness of her own body and kept touching herself to recover some sense of it. she forms the paradigm in terms of which Rosenfield interprets all those neurological symptoms: what unites them in their diversity is that their lesions all cause, in their different ways, disturbances in the patients' sense of their own bodies. This leads, in turn, to disturbances about their own sense of who they are, their own "subjectivity", and even their consciousness.
It's far from clear, however, that these case histories support Rosenfield's claim that consciousness, memory, bodily awareness, time, and anti-modularity really all constitute facets of a single phenomenon dependent on the "body image". Even in the case of Madame I, Rosenfield's description does not bear out his claim that "there are no memories without a sense of self".(R41) For Madame I does not seem to have lost the power to recollect that certain events occurred, but only the power to picture these events to herself in a vivid and emotionally significant way. This, like the case of H.M. who could learn new skills but not remember recent events, seems rather to point to a distinction between semantic and event memory. But Rosenfield minimizes the significance of that distinction (explored since the 1970's by Endel Tulving) without saying anything that establishes that his thesis applies equally to both.
Insofar as one can sustain the distinction between two types of memories is in personal or episodic memories that the connection with consciousness is no doubt most intimate. These, as many writers knew before modern neurology, are best acquired and maintained by a brain pickled in emotion. "Fear, like all the emotions, shapes the self." And while the theme of emotions plays a supporting role in Rosenfield's book, it firmly holds the centre of the stage in Damasio's.
Damasio's opening neurological tale is one especially apt to emit terror and pity. Phineas Gage was a rail road worker whose job involved laying gunpowder charges in rock cliffs to be cleared for the rail road. The fused gunpowder would normally be covered by a layer of sand tamped down with a specially shaped steel bar. One day in 1848, Gage, in a moment of distraction, had tamped the powder before the protective layer of sand had been poured in. The resulting explosion propelled the rod into his left cheek and out of the top of his head. Astonishingly, this accident left him conscious, and able to talk and walk. Indeed, it had apparently had little effect on his mental powers, including his conscious memories and linguistic skills. But it gradually transpired that Gage was allegedly changed in his emotional make-up and his personality: it was as if all the "signs" of his character traits had been electronically switched from plus to minus by a general Search-and-Replace operation. The model employee became unemployable; from dutiful he became irresponsible; from persevering, capricious; from polite, rude and inconsiderate; from temperate, drunken; from gentle, violent; from energetic, to slothful from a good citizen, in short, to a bad one.
Is there, then, a simple brain center for Good Citzenship? The story Damasio has to tell isn't that simple, but it's a fascinating one. In the simplest terms, Damasio's thesis is that emotions have an indispensable part to play in practical rationality. Without emotions, an agent might reason as well as anyone about hypothetical cases, but be wholly unable to act on any of this knowledge in practical life. The reasons are several: one is that reason, by itself, knows neither when to start nor when to stop evaluating costs and benefits: thus it is likely to defeat itself with too much deliberation. Another is that without the "bias" of emotion, the theoretically possible alternatives are too numerous for effective consideration: the framework of our emotional bias selects a small number as what William James called "live options". Most important is the fact that without the ability to care about the (future or hypothetical situations) evoked by the reasoning, the values involved never get to move the agent in accord with this "knowledge". When the capacity for emotions is present, it involves a capacity for the brain to monitor the body's past and hypothetical responses, both at the level of the autonomic and the voluntary systems This capacity rests on "somatic markers" which embody the individual agent's experience as well as their innate disposition to respond to various stimuli. And the capacity to lay down such markers depends especially on certain parts of the prefrontal lobes which, as Hanna Damasio established by minute examination of Phineas Gage's skull over a century later, were just those diced away by his steel rod as it passed through his head.
In some ways, Damasio's Endings tally well with the hypothesis in Rosenfield's book (though Damasio doesn't cite him.) Descartes' error can be thought of as expanding on the hypothesis of the primacy of the body image from a neurological point of view. For Damasio found that besides the prefrontal brain damage responsible for Gage's difficulties with decisions, and the very similar ones studied in his own patient "Elliott", there is another sort of brain lesion that gives rise to very similar symptoms, and that is precisely a lesion in the "somatosensory regions of the brain", located in the right hemisphere (D65)3 . I.e. those "that produce, through their cooperative interactions, the most comprehensive and integrated map of the current body state available to the brains" (D67). But Damasio's findings do not bear out Rosenfield's more radical claim, that any serious disruption of the body image should entail further disruptions of consciousness and understanding. Indeed, this is the most striking feature of cases such as Gage and Elliott: that in all tests of intelligence and rationality that rely on knowledge, they seemed entirely normal. In real life, however - and, as we shall see, in ingeniously devised experiments that reproduce the relevant features of "real life" - they exhibit a profound incapacity to act rationally.
The clue that Damasio followed in his attempt to understand these cases lies in the observation that in a number of patients the impairment of the capacity to make life decisions was generally associated with a failure of certain kinds of emotional response. The patients "know but cannot feel."(D45,D205) Here, as often in this story, philosophers were there first. It was Plato who first spoke of the difference between having a piece of knowledge, and actually using it, thus drawing our attention to the phenomenon of akrisia or "weakness of will". In some ways the troubles experienced by the patients described by Damasio are reminiscent of the akratic's problem. In akrasia, an agent knows the best thing to do, but somehow fails to do it. But sufferers from akrasia differ from both Damasio's patients and from psychopaths, who display similar signs of "myopia for the future." Akratics frequently suffer from remorse: not so psychopaths or victims of prefrontal lesions. In his 1964 book, The Mask of Sanity, Hervey Cleckley had already observed that psychopaths seemed perfectly well able to experience both the punishing effects of pain and - all too well - the attraction of pleasure. What they seemed unable to do was to care about anyone but their own present self ignoring not only the interest of others but their own interests in the future. Damasio may have discovered the underlying neurology of this phenomenon, in terms of what he calls the somatic marker hypothesis.
The somatic marker hypothesis is an elaboration of William James's notorious theory of emotion, cleared of its confusions. James identified an emotion with the subject's awareness of the somatic changes caused by the evoking stimulus, which independently provoked any expressive or functional response. Hence his famous dictum: "We do not weep because we are sad, but we are sad because we weep." That view had two serious defects: first, it made emotion sound purely epiphenomenal - as if the awareness involved could make no difference to my behaviour, since that was independently determined by the bodily dispositions of which my emotion merely monitored the effects. Second, it seemed to place the intention and object of the emotion in the wrong place: as if I could never weep for Hecuba or anyone else, but only for my tear ducts. Damasio avoids both these problems by introducing two distinctions: first, between "primary" and "secondary" emotions, second, between an emotion and the characteristic responses associated with it.
Primary emotions involve innate dispositions to respond to certain classes of stimulant controlled by the limbic system. Secondary emotions involve learned categorizations of representations or "images" of stimuli, associated with past responses, both somatic and behavioral as well as their results, evaluated as good or bad. Secondary emotions involve cortical processes but "utilize the machinery of primary emotions" for their expression, as evidenced, for example, by the fact that spontaneous smiles are quite different from those we muster on purpose. Feelings are the actual "experience of such changes in juxtaposition to the mental images that initiated the cycle" (D145). So we can weep for Hecuba, at least if "juxtaposing her image" has something to do with it and moreover the emotion is far from inert, since it is associated with a whole scenario that is acquired in individual experience, and that involves characteristic responses.
In simple and telling experiments, Damasio has collected evidence for his hypothesis. Patients like Gage and Elliott, it turns out, have normal galvanic skin responses (GSR) to their own immediate emotions. (GSR measures the conductivity of the skin, which reflects minute differences in sweating and signals emotional arousal). But they fail entirely to respond to disturbing images, even while they know - and can remark on quite articulately - that these images should have upset them. More revealing still, in an experiment devised by Hanna Damasio, they proved incapable of adopting a rational strategy in a realtime gambling game, consistently ending up, as Gage and Elliott did in real life, with huge losses.
Thus the tragedy of those who have lost the integrating function normally handled by the frontal lobes and limbic system is that they have become incapable of secondary emotions: incapable, that is, of emotional learning. Rosenfield's suggestive hints and Damasio's more careful neurological hypothesis have brought us back to Aristotle, who taught that the education of a responsible moral adult requires learning the right emotions. For the rational conduct of life, it is not enough to react in accordance with the innately wired "primal emotions." On the other hand, it is counterproductive to seek a form of rationality divorced from emotion: as Damasio wickedly suggests, the only true Kantian is a patient with prefrontal damage (D172). As for Descartes' error, it was not merely dualism: it was putting the mind inside the body. But the body, in truth, is inextricably lodged inside the mind.
Damasio, A.R. (1994) Descartes' error: Emotion reason, and the human brain. New York: G.P. Putman & Sons.
Cleckley, Heney M. (1964) The mask of sanity an affempt to Clarify some issues about the so- called psychopathic personality. 4th ed. Saint Louis: C.V. Mosby co.
Descartes R. (1984 <<1649>>) The philosophical writings of Descartes. (J. Cottingham, R. Stootoff & D. Murdoch, Trans.): Vol. 1. The passions of the soul. Cambridge: Cambridge University Press.
Hull, J.M. (1990). Touching the Rock: An experience of blindness. London: Pantheon.
James, W. (1884). "What is an emotion?" Mind 19: 188-204.
Merleau-Ponty, M. (1962) Phenomenology of perception (C.Smith, Trans) International Library of Philosophy and Scientific Method, London: Routledge and Kegan Paul.
Sacks, O. (1987) The man who mistook his wife for a hat and other clinical tales. New York: Harper and Row.
Sacks, O. (1995) An anthropologist on Mars: Seven paradoxical tales. New York: Knopf.
Strawson, P. F. (1963) Individuals: An essay in descriptive metaphysics. Garden City, NY: Doubled Anchon
Tuling E. (1972). "Episodic and semantic memory." (E. Tulving & W. Donaldson, Eds). In Organization of memory. New York: Academic Press.
Ronald de Sousa is Professor of Philosophy at the University of Toronto. He is the author of The Rationality of Emotion (MIT 1987,1989).