The Sleep of Others and the Transformation of Sleep Research 9781442627789

Citation preview

THE SLEEP OF OTHERS AND THE TRANSFORMATIONS OF SLEEP RESEARCH

This page intentionally left blank

KENTON KROKER

The Sleep of Others and the Transformations of Sleep Research

UNIVERSITY OF TORONTO PRESS Toronto Buffalo London

© Kenton Kroker 2007 University of Toronto Press Toronto Buffalo London Printed in Canada ISBN 978-0-8020-3769-5

Printed on acid-free paper

Library and Archives Canada Cataloguing in Publication Kroker, Kenton, 1969– The sleep of others and the transformations of sleep research / Kenton Kroker. Includes bibliographical references and index. ISBN 978-0-8020-3769-5 (bound) 1. Sleep – Physiological aspects – Research – History. QP425.K76 2007

612.8c21

I. Title.

C2006-906422-9

University of Toronto Press acknowledges the financial assistance to its publishing program of the Canada Council for the Arts and the Ontario Arts Council. University of Toronto Press acknowledges the financial support for its publishing activities of the Government of Canada through the Book Publishing Industry Development Program (BPIDP). This book has been published with the help of a grant from the Canadian Federation for the Humanities and Social Sciences, through the Aid to Scholarly Publications Programme, using funds provided by the Social Sciences and Humanities Research Council of Canada.

Contents

Acknowledgments Introduction

vii

3

1 The Persistence of Privacy

18

2 Analogize and Experiment 3 The Ends of Darkness

121

4 Inhibition and Disease 5 Performing Sleep

205

6 Sleep Finds a Groove 7 Begin the Begin 8 Insomnia Returns 9 Breathe Epilogue

395 429

Notes 433 Bibliography Index 525

503

178

255

325 349

71

This page intentionally left blank

Acknowledgments

Sole authorship is but a pleasant fiction. The engaged support of both faculty and students at the Institute for the History and Philosophy of Science and Technology at the University of Toronto helped me, as a graduate student in the 1990s, to imagine what a history of sleep might look like. Pauline Mazumdar and Ian Hacking, in particular, provided the necessary blend of critical insight and good humour to help me complete my doctoral dissertation on the discovery of REM, which became the framework for several chapters of this book. Many others soon helped transform some of my ideas. When they were not offering useful comments on my own work, I tried to follow the lead of Rhodri Hayward, Simon J. Williams, Peter Keating, Tara Abraham, and my colleagues in the Science and Technology Studies program at York University. Outside the academic world, Rob MacDonald and Duncan MacDonald influenced my thinking in subtle yet important ways. The enthusiasm of my editor, Len Husband, never seemed to waver. Ilya and Emilia Kabakov’s image from their 2005 installation ‘The House of Dreams’ perfectly reflected the core idea of this book. I was very pleased when they generously granted permission to use it on the cover. Institutions, of course, also have tremendous creative force. The greater part of the book took shape in libraries and archives, with the assistance of the staff of the Robarts and Gerstein Libraries at the University of Toronto, the Countway Library at Harvard Medical Schools, the Rockefeller Archive Center, Special Collections at the University of Chicago, and the Neuroscience History Archives at the University of California, Los Angeles. Clara Lecadet uncovered some valuable correspondence in the French National Archives, for which I am also grateful.

viii

Acknowledgments

Doctoral and post-doctoral fellowships from the Social Science and Humanities Research Council of Canada gave me the wherewithal to start this project, and York University provided research monies and assistantships to embellish it. The Celebrations Committee of the Associated Professional Sleep Societies also provided substantial financial support. Their grant came with no restrictions regarding the content, form, or argument of the manuscript, which meant that I could work with complete freedom at a crucial stage in my research and writing. Such an unambiguous statement of support for historical scholarship is rare among scientific societies, but there are encouraging signs of similar developments elsewhere. The nature of their support testifies, I believe, to the maturation and dynamism of the field of sleep research that I have attempted to outline in this book. I hope similar groups can extend this process through creating a centralized archive that might somehow contain the bewildering trajectories of sleep research for future scholars. In the bad old days, historians thanked their wives for typing, proofreading, indexing, and assorted other tasks. Leanne Davies performed none of these. She might never read the thing. But no matter – as its inspiration, she must have intuited its contents long ago. It is to her that this work is dedicated.

THE SLEEP OF OTHERS AND THE TRANSFORMATIONS OF SLEEP RESEARCH

This page intentionally left blank

Introduction

This investigation is dominated, from start to finish, by a single question. Tonight, thousands of patients around the world will have their sleep observed in a laboratory. Why? The simple answer, of course, is that these people are ill. They sleep in a special lab, generally found within a sleep clinic. The laboratory contains special equipment and medical staff that together observe the patient’s sleep in order to help determine the nature of his or her illness, and perhaps also to prescribe and test a course of treatment. But it takes little effort to consider a world in which such an event could not or did not occur. Many people have never heard of ‘sleep disorders,’ and without that concept, the need for a sleep laboratory to diagnose disease disappears, just as the need for a bacteriological laboratory cannot exist without the concept of ‘germs’ or ‘infectious disease.’ Many who read this book have lived in just such a world, for the sleep laboratory is a relatively recent invention. It simply did not exist before the 1960s and was fairly uncommon even around 1980. Yet, today, sleep laboratories are a mundane part of medical practice, as are the patients who sleep in them. How did such a curious practice become routine? I attempt to answer this question by proposing that, to become part of an interdisciplinary, laboratory-based biomedical field, sleep needed first to emerge as a scientific object. We all know what an object is, since objects are integral parts of the realm of experience. They are the things we control and manipulate, more or less successfully. They can be tangible, like wooden blocks, toys, or power tools. They can be images, like the vision that greeted us this morning in the bathroom mirror. They can also be the more ephemeral things mediated by language, such as ideas. In any

4

The Sleep of Others

event, objects reveal the nature and limitations of our subjectivity, of our ability to control or transform or understand the world. Scientific objects, however, are somewhat different. They are typically (although not always) exotic and unusual. They are not accessible by everyone but are usually handled by experts who have gone through an extensive and more or less homogenized form of training. Scientific objects are often found in extremely specific locations, like laboratories. If they are impossible to contain physically, they can be modelled, like climate. If they lack material existence, they can be represented by a highly refined and specialized language, like mathematics. Recent descriptions of this way of framing developments in science, technology and medicine invoke the terms ‘applied metaphysics’ or ‘historical ontology.’1 Instead of describing the things that science handles as eternal and unchanging, this approach examines how objects in the sciences come into and fade out of being. This is an inherently historical enterprise that takes seriously the idea that history is about the attempt, however unlikely, to use reliable evidence to recreate the past as it was. It does not necessarily ignore or set aside present scientific knowledge in its quest to recreate the past. But it does refuse to admit that current scientific understanding provides the best tools for examining the dynamics of the scientific enterprise. This is not, after all, what contemporary scientific objects were designed to do. Yet, though these objects are not historical tools, they can direct and influence historical inquiry by prompting questions about how such a state of affairs came to be. Sleep, always part of human experience, has been rendered into an object in myriad ways. We can sleep in groups, or on our own. We can ingest substances to encourage sleep, or defer it to a later time. We can sleep on mattresses of foam rubber, spring coils, cotton batting, or feathers. We can sleep on the floor. We can sleep during the day, at night, in one single block of time, or at several times during the day. In short, human societies try to control when and how people sleep, just as they do with eating. Because it is a physiological necessity, controlling sleep is often portrayed as occurring at the very limits of normal experience. To take but a single example, the protagonist of Jerzy Kosinski’s 1975 novel Cockpit shocks readers in its opening pages by revealing two ways in which he controls his environment by breaking with social convention: he has sex with his mother well into adulthood; and he sleeps in two four-hour blocks, one during the day, and the other at night. What makes sleep an unusual form of experience, however, is that it is, by definition, an annihilation of experience. Sleep is typically experi-

Introduction

5

enced and described as the absence of phenomena (of consciousness, of movement, of sensation), rather than the presence of anything at all. It was just this property, observed the historian of science Georges Canguilhem, that had made sleep so resistant to biomedical investigation. Despite the vast amount of time that people spent in sleep, they experienced it not as part of life that could fail, generate pain and suffering, and demand the attention of biomedical researchers. Sleep was, on the contrary, defined as a temporary and passive diminution of life itself, so it was effectively established outside the realm of the vital forces that physiology investigated and medicine attempted to protect and restore.2 As a passive state that temporarily suspended subjectivity, sleep could never be part of Canguilhem’s analysis of how the very concept of health inevitably implicated subjectivity. But, even as he made these remarks during the early 1940s, this conception of sleep was changing. Sleep had begun to enter the biomedical laboratory. To this point, the origins of knowledge about sleep came from personal experience. Although sleep had been, at different times and in different ways, a public performance, it was an object of knowledge only by virtue of the evidence from personal testimony, either of one’s own sleep or of an analysis of such reports usually compiled by physicians.3 Knowing sleep was thus a matter of ‘I’ and ‘thou’ and was refracted through the prism of individual experience that depicted sleep as a negative state, as the quiescence of the body and the absence of consciousness. People typically paid attention to sleep only when it was disrupted or somehow compromised by dreams, insomnia, or fatigue, all of which originated with an individual’s sensation. The practices and technologies that eventually came to constitute the sleep laboratory changed all this. Relying on the testimony of instruments rather than individuals, investigators interested in sleep began to create a new series of sleep phenomena that refashioned sleep as a scientific object. Their use of instruments and animal subjects introduced a ‘third-person’ perspective to the problem. Phenomena generated by experimental manipulations or instrumental observations of sleep were initially correlated and calibrated to subjective sensations. But soon enough, this practice became unnecessary, and such practices were deemed capable of representing sleep as it really was, apart from any link to subjective experience. Knowing something about sleep implied that the knowledge originated with the study of the sleep of others. Whether or not the personal experience of sleep was transformed in the process is an empirical question that is just beginning to be

6

The Sleep of Others

answered by ethnographic, sociological, and historical analysis.4 It seems likely that sleep practices have changed quite dramatically since the early nineteenth century. One recent historical study argues that, in pre-industrial societies, the division of rest into ‘first’ and ‘second’ sleeps was in no way unusual, and possibly even the norm.5 The bearing of medical knowledge on this transformation, however, seems to have been slight. Unlike chronic and supposedly constitutional illnesses like tuberculosis or acute epidemic outbursts like cholera, there was little sense that sleep was an appropriate subject for medical research for most of the nineteenth century. It was instead a vehicle through which physicians could dispense indeterminate, conflicting, and paternalistic advice about ‘sleep hygiene’ supposedly derived from clinical experience and inevitably framed in familiar terms of personal experience. Sleep was not a problem of populations, but of individuals. The sleep of others began to coalesce only around 1900, once investigators began to describe sleep as a positive, rhythmic process that served a biological function. It seems counter-intuitive to suggest that anyone could ever have imagined that sleep did not serve a function. We become tired, we fall asleep, and we awake, refreshed. But sleep was generally framed in terms of individual psychology before 1900. And here we find the paradox of sleep as an experience that annihilates experience having a substantial impact on the evolution of sleep as a scientific object. Sleep, so the reasoning went, is the (temporary) elimination of consciousness. It is little more than nothingness, and how can nothingness have a function? Of course, most of us dream, and on this basis we suspect that sleep is not the simple elimination of consciousness. Dreaming has, in fact, played a pivotal and diverse role in sleep’s history, particularly regarding the issue of sleep’s function. For it was through the problem of dreaming, not of sleep, that psychologists and psychiatrists first took an interest in understanding night-life in terms of function. The dynamic interaction of dreaming and sleep as scientific objects has not, however, been at all straightforward, and part of the intent of this study is to show how the intimacy of their relationship has oscillated over the past twenty-five hundred years. When sleep emerged as an object of modern biomedical analysis over the past century, dreams were effectively set outside the investigative domain. They were reunited, briefly, in mid-century by virtue of the discovery and dissemination of rapid eye movements (REM) around 1953. But, by the mid1970s, they had begun to separate again, as the sleep clinic began to

Introduction

7

take shape and biomedical interest in establishing sleep as a publichealth problem trumped engagement with the idea of using laboratories to comprehend dreaming. Of course, science is a dynamic and creative enterprise, and there is always the possibility that this relationship will change again in the future. Sociological studies of sleep generally forgo attention to this dynamic interaction between sleep and dreaming and instead focus on sleep research as an example of the flexible interactions between biomedical categories, technologies, and patient populations. Some historical studies, generally those written by physicians, typically treat sleep and its associated phenomena as semantic containers, waiting to be filled up with scientific knowledge over time.6 Readers are presented with a vast array of facts from diverse historical periods whose relationship to each other is presupposed rather than explained; or they focus on the history of particular aspects of sleep (such as narcolepsy or encephalitis lethargica), without invoking the intellectual, social, or technological environment in which such facts took shape.7 Philosophers and neuroscientists have, unsurprisingly, focused less on sleep and more on the question of dreaming, in both its historical and its epistemological dimensions.8 The present study tries to incorporate aspects of all these approaches by concentrating on the diversity and dynamism of the ways in which sleep has been made into an object of scientific knowledge. Extreme specialization in historical research tends to shun the utilization of sweeping historical periods often favoured by physician-historians, but I will nevertheless make the attempt here. Detailed analyses of the technologies of sleep research demand careful social analysis, particularly insofar as they visualize aspects of sleep not so much in space as in time. With the introduction of electroencephalography, sleep (and later, dreaming) entered the field of physiognomy through biopsychiatry.9 It achieved a kind of morphology that it had never had before. This shape was dynamic, and, as such, the electroencephalography of the twentiethcentury sleep laboratory becomes part of the history of time, and timekeeping.10 The rhythmic analysis of mind instigated by the electroencephalogram (EEG) has even been proposed as a means of framing a natural history of consciousness.11 And, although dreaming is no longer high on the agenda of sleep research, it played a pivotal role in establishing the field, and any biography of sleep as a scientific object cannot ignore dreams. Both dreaming and sleeping were found, in different configurations, within the spheres of the personal and the public, of the visible and the invisible, of the animal and the human, of the clinical

8

The Sleep of Others

and the experimental, and even of the sacred and the profane. In this book, I endeavour to offer an overview of how those different configurations ultimately generated a new view of sleep that transformed it from an intimate experience to an impersonal and public object. Temporal Structure Sleep had, of course, been part of medical practice since antiquity. As the first chapter indicates, ancient medical interests in sleep were encapsulated by the role dreams played in the Asclepian cults. Galen acknowledged the value of such dreams for medical diagnosis, but it was his emphasis on sleep as a ‘non-natural’ that was ultimately formalized by the medieval medical schools, as dreaming was separated from medical practice and became part of philosophical psychology. In both instances, sleep and dreams were intrinsically personal things. Physicians understood sleep only as part of the patient’s idiosyncratic constitution upon which diagnosis and therapy were based. For their part, philosophers took dreams to be the workings of an individual’s consciousness distorted by sleep. If the solipsistic nature of dreaming went unrecognized, argued some Enlightenment scholars, erroneous belief, manifest in the delusions of the mad and the devout, was the inevitable result. Chapter 2 introduces some nineteenth-century aspects of the scientific study of sleep. There were two dominant approaches to the problem of sleep in this period: one from the clinical point of view, and one from the perspective of physiological psychology. The most prominent clinical problem concerning sleep was insomnia. In the hands of neurologists like William Alexander Hammond, insomnia became part of a continuum of sleep ‘derangements’ that ranged from an occasional inability to fall asleep all the way to a full-blown neurosis. In popular books and in the margins of physiology texts, clinicians described the different ways that one could fall asleep, but they gave hardly a thought to what natural sleep actually was. This question was eventually formulated during the last two decades of the century, in the midst of the largely French debates over hypnosis. At first, many investigators were enthusiastic about using hypnosis as a tool that could probe the unconscious mind. The hypnotic state was thought to be equivalent to sleep, and hallucinations experienced while under hypnosis were seen as analogous to dreams. By the late 1890s, this analogy had begun to unravel, as hypnosis became tantamount to

Introduction

9

‘suggestion’ – the ability of the hypnotizer to influence the behaviour of his subject. By the time war broke out in Europe in 1914, hypnotism had fallen from grace, and its status as a form of ‘artificial sleep’ was thoroughly rejected. The study of fatigue, on the other hand, was just getting started. Its origins lay in the application of Étienne-Jules Marey’s ‘graphical method’ to the scientific study of labour. But one of Marey’s students, Angelo Mosso, soon turned fatigue into an independent object of study. Sleep, as the product of intellectual and physical fatigue, was granted a minor but significant status in this program. Marey’s graphical method offered a means of describing physiological phenomena without disturbing the organism’s natural state, regardless of whether it was an animal, a patient, or a labourer. This was tremendously important for sleep research, since it provided a way of obtaining sleep’s ‘natural signatures’ (blood pressure, pulse, respiration, and the like) without awakening the sleeper. The graphical method allowed the construction of an image of sleep from the perspective of the body, rather than the mind. But, if sleep was becoming more and more grounded in the body in the early twentieth century, psychoanalysis, which took dreaming as a model of mental activity, seemed headed in the opposite direction. My third chapter deals with two distinctive approaches to dreaming in the early twentieth century: those of Sigmund Freud and Henri Bergson. Both produced important contributions to the study of dreaming around 1900, although only Freud has attracted much attention from historians. Freud developed a clinical technique based on the interpretation of dream content as an elaboration of repressed desire. For my purposes, his original significance lies in the fact that he turned dreaming into a vital function, linking it to health rather than to a pathological alteration of consciousness. Bergson, on the other hand, took dreams to be a primitive experience of time as duration, which he contrasted to the mechanized, scientific vision of time as a succession of instants that could be represented by space. Bergson had a tremendous impact on French philosophy, which in turn affected the growth of psychological knowledge there, because most workers in that field were first trained as philosophers. His metaphysics encouraged psychologists to rigorously separate objective experiment and subjective experience. The fate of sleep as a function is charted in chapter 3 through the early career of Henri Piéron, a psychologist who had studied under Pierre Janet and Théodule Ribot. He turned to physiology and in 1912 produced a dissertation on sleep, which he published the following year.

10

The Sleep of Others

Piéron’s study of sleep revolved around his method of ‘experimental insomnia,’ a technique that involved depriving his animal subjects of sleep until they slipped into a coma and died. Such a method was admirably suited for demonstrating a physiological need for sleep that operated outside of the parameters of fatigue laid down by Mosso in the 1890s. It also enabled Piéron to examine sleep from a physiological, rather than psychological, perspective. In this respect, Piéron’s work rode on the earlier success of the ‘biological’ theory of sleep proposed in 1904 by the Swiss psychologist Edouard Claparède, who suggested that sleep was an ‘active defence’ against fatigue. Such claims reinforced the belief that sleep should be studied through the observation of comportement, or ‘behaviour,’ rather than through the methods of introspective psychology. At this point, our story shifts, in part, to the New World. In the United States, sleep was little more than an appendage of clinically oriented physiology until well after the First World War. Chapter 4 discusses the epidemic of ‘sleeping sickness’ that spread through North America in the wake of the influenza pandemic of 1918–19. Although influenza initially received the bulk of medical attention, by the early 1920s, American neurologists (who had little to say about influenza) began to focus on this mysterious disease that often left its victims in varying states of dementia. Their interest in sleep brought the topic out of the doldrums of hygiene and into the world of organic brain disease. To add to the high degree of visibility of sleep during this period, Constantin von Economo, the Viennese clinician who first described ‘encephalitis lethargica’ in 1917, visited the United States in 1929. Drawing on Pavlov’s description of sleep as inhibition, Economo argued that natural sleep was the product of a regulatory ‘sleep centre’ in the brain, which was somehow damaged over the course of the disease. His ideas resonated with neurologists, who relied on the concept of brain localization for their own diagnostic authority. Physiological interest in sleep had already been piqued by Ivan Pavlov’s visit to the United States in the summer of 1923, also discussed in chapter 4. Pavlov, depicted in the American press as ‘the last free man in Russia,’ was probably the most famous living physiologist at the time. The application of his method of conditioned reflexes to the study of sleep was at the top of Pavlov’s agenda, and, at every opportunity, he spoke about his theory of sleep as ‘generalized inhibition.’ Pavlovian conditioning, which dovetailed with James B. Watson’s behaviourism, was already well known to American physiologists and psychologists by

Introduction

11

this time. But, for my purposes, Pavlov’s significance lay in the currency his interest gave to the physiological study of sleep. Chapter 5 examines the very local conditions of early sleep research at the University of Chicago during the interwar period. When Pavlov visited the A.J. Carlson’s Physiology Department in 1923, a young student named Nathaniel Kleitman was just publishing the first of a series of papers on the physiology of sleep. Kleitman, who would eventually establish himself as the first physiologist to dedicate his entire career to sleep, had adapted Piéron’s method of ‘experimental insomnia’ to the unique conditions of physiological research at Chicago. He used human subjects for many of his experiments, preferring to rely on innovative physiological recording techniques rather than post-mortems to frame his questions about sleep. His non-interventionist methods were more faithful to Marey and Mosso than they were to Piéron or Pavlov. They also reflected the growing interest, sponsored by the Rockefeller Foundation, in establishing a department of neuropsychiatry at Chicago. Alan Gregg, the head of the Medical Sciences Division of the Rockefeller, hoped that Kleitman’s research would lend neuropsychiatry some scientific credibility. To this end, he funded Kleitman’s study of sleep for several years in the mid-1930s. But the neuropsychiatry project collapsed, along with Kleitman’s funding, in 1939, just as Kleitman was publishing his famous textbook on sleep. The professional conflicts that destroyed the Rockefeller project played themselves out on a smaller scale in Carlson’s physiology department. Edmund Jacobson, a clinician with an active interest in sleep, left the department in the wake of this disaster. He took with him a remarkable skill in electrophysiological recording, a talent that would be resurrected by one of Kleitman’s graduate students more than a decade later. While no amount of Rockefeller money seemed to be able to make the neuropsychiatry project work, the electroencephalograph was bringing together neurology, psychiatry, physiology and psychiatry as never before. The EEG, the topic of chapter 6, was first discovered by a German psychiatrist named Hans Berger in 1925. But it did not receive much scientific attention in the United States until 1934, when the Nobel laureate Edgar Adrian began experimenting with ‘Berger waves’ in his laboratory at Cambridge. Adrian’s interest in the EEG was short-lived. Yet there was also a distinctive American tradition of research, led by the millionaire financier Alfred Lee Loomis from his private laboratory in Tuxedo Park, New York. Unlike Adrian, the Loomis group used the EEG as a research tool, rather than attempting to explain its origins. Before their work

12

The Sleep of Others

ended in 1939, sleep had become one of the major focal points at Tuxedo Park. Loomis, who had a long-standing passion for precision instruments and timekeeping, used his extensive financial resources to construct an enormous kymograph that was able to record eight hours’ worth of sleep at a time. By late 1935, his work had established a standardized set of five ‘sleep stages’ based on distinctive EEG tracings, through which the brain ‘cycled’ throughout the night. This image of sleep as a phenomenon of brain-regulated timekeeping was to dominate sleep research for the remainder of the century. In 1936 Loomis was joined by Hallowell Davis, a neurophysiologist from Harvard who had just begun to dabble in psychoanalysis. His hope was to apply the EEG as the basis for a study of individual differences, modelled on ego psychology. His neo-phrenological aspirations illustrate the vast potential this instrument was thought to have in the years before the Second World War. Chapter 6 also examines the conditions under which REM finally appeared to scientific perception in the year that led up to Eugene Aserinsky and Kleitman’s publication of their discovery in 1953. Aserinsky, on the advice of Kleitman, had originally set out on a project to study eye movements as an index of wakefulness in infants. After observing regular periods of eye quiescence in these infants, he turned to observing adults. Here he found periods of eye movement instead, and he quickly associated these with dreaming. After soliciting the advice of Edmund Jacobson, Aserinsky reconfigured his experimental assemblage to include EEG and Jacobson’s technique of measuring eye movements. Within months, he had managed to convince most people around him, including Kleitman, that dreaming was closely related to these periods of eye movement. Although he himself was no Freudian, Aserinsky’s discovery came in the midst of a revival of scientific interest in the dream in the United States. Once the province of an elite group of neurologists and ambitious psychiatrists, psychoanalysis had assumed a dominant role in the sciences of mind and brain by the early 1950s, and the systematic study of dreams outside the clinic was high on the agenda. Aserinsky, anxious to leave Kleitman’s laboratory, abandoned his research to William Dement, a young medical student with a keen interest in psychiatry. Dement recognized the potential of REM to bridge the gaps between the physiological laboratory and the psychiatric clinic. By the mid-1960s, his experiments with psychiatric patients and his theories about ‘dream deprivation’ had brought sleep research out of the backwaters of physiology and onto the centre stage of the neurosciences.

Introduction

13

This revival of the dream as an investigative object dominated the early years of the first scientific organization to form around the phenomenon of REM. When it emerged in the early 1960s, members of the Association for the Psychophysiological Study of Sleep (APSS) took seriously the idea that laboratories would reveal the truth about dreaming. Many of them also felt that the ‘dream laboratory’ would finally bring psychoanalytic concepts and practices into the domain of the biomedical sciences, precisely because they took REM to be an unproblematic and objective index of dreaming that could be effectively detached from its instrumental and cognitive context. Indeed, one psychoanalyst recalled a meeting with Kleitman in 1957 in the following terms: ‘I told him [Kleitman] that I was interested in setting up a dream research laboratory and was, of course, particularly interested in the phenomenon of rapid eye movements (REM), which had been described as accompanying dreaming. He immediately replied, “Then let me show you them.” He leaned forward in his chair, closed his eyes, and I could clearly see his eyeballs moving back and forth under the closed lids. He opened his eyes again and said, “You saw that, didn’t you? It is unmistakable. Yet no observer had made that observation for many, many centuries!” I realized that this phenomenon, despite all the electronic complications of its measurement, was basically simple and observable without complex instrumentation.’12 An analysis of the history of the APSS, however, indicates the fallacy of such an assumption. Despite initial enthusiasm from practising psychoanalysts and psychiatrists with an active interest in Freudian theory, such individuals were in short supply within the APSS by the 1970s. They either abandoned the practices of the ‘dream laboratory’ entirely or radically recalibrated their expectations of it. For his part, Dement became instrumental in shifting the trajectory of the laboratory towards clinical problems. Narcolepsy, in particular, became a pivotal disease that was remade by laboratory-based sleep research. Since its initial description during the 1880s, narcolepsy had enjoyed a dubious existence as a minor battlefield for those who considered it a manifestation of mental illness and those who counted it as a brain disorder. The sleep laboratory, following the tradition of psychophysiological research, attempted to mitigate this conflict by offering a new level of diagnostic specificity for the disease by instrumental surveillance. The ‘laboratory test’ for narcolepsy ultimately redefined it as a state-specific pathology that could be accurately diagnosed only through identification of its victims’ aberrant REM periods. With this development, sleep research

14

The Sleep of Others

began to shift towards the biomedical field and away from neuroscientific investigation. In the process, the APSS was itself transformed, as an ever-greater proportion of its members identified themselves as clinicians, not psychophysiologists. Narcolepsy, however, had always been understood to be a relatively rare disorder. It was not until insomnia was brought within the purview of the sleep laboratory that sleep medicine began to emerge as a medical specialty and sleep itself came to be depicted as a public-health issue. This process, described in chapter 8, was by no means a linear progression from laboratory-based knowledge to its application in clinical practice. On the contrary, seemingly ‘external’ issues, such as the redefinition of addiction as a medical rather than a criminal issue and the expansion of the concept of ‘community psychiatry,’ had an enormous impact on the transformation of insomnia from a common and thoroughly ignored aspect of medical practice to a widespread concern recognized by the highest levels of government. As consumers of a vast amount of pharmaceuticals to treat their condition with a minimum of medical supervision, insomniacs were perceived to be at risk for addiction and even death by accidental or intentional overdose. While narcolepsy had justified the sleep laboratory’s entry into the clinic, the evolution of insomnia as a public-health issue presented the prospect of an entirely different scale of expansion. Hypnotics, new and old, were evaluated for their differential impact on sleep stages, while insomniacs themselves were ushered into the sleep lab for an objective diagnosis of their disorder, which, to that point, had been almost entirely selfascribed. The final chapter evaluates the rise of sleep apnea following sleep medicine’s first forays into public health through the problem of insomnia. The latter disorder was cursed, as it had always been, by indeterminacy. Repeated attempts to standardize its diagnosis by either physiological or psychological means failed. Insomnia was both mental and physical, and such a position was difficult to negotiate during the 1980s, when the American psychiatric establishment began a full-scale purge of psychoanalytic ideas from its diagnostic canon. Insomnia also fell victim to a recriminalization of drug addiction in the United States, which meant that it became increasingly difficult to dedicate resources to the study of those who had become addicts while working within the accepted boundaries of medical practice. Instead, biomedical interest turned towards sleep apnea. Practically unknown before the 1980s, the problem of sleep apnea had blossomed into a major public-health

Introduction

15

debate ten years later. Unlike narcolepsy or insomnia, the parameters of sleep apnea (and its subsequent significance as a public-health issue) depended much less on patients’ symptoms than upon laboratorygenerated signs. Sleep apnea thus became a symbol of a new form of sleep research, detached from the vagaries of dreaming and capable, like any other biomedical field, of generating its own set of distinctive experimental practices that transcended and even defied individual, subjective experience. Sources This is an essay in cultural history, written from the perspective of a history of knowledge. As such, it relies on an array of sources. Historians typically work from textual evidence that has been organized in large part by others. Currently, no archive for sleep research exists, though I hope that this book may prove instrumental in establishing such a collection. In any event, sleep research and sleep medicine draw upon the concepts and practices of numerous scientific fields and medical specialties. The potential scope of archival sites that might hold fruitful data on sleep’s epistemological history is thus vast, even within a national context. The University of Chicago, however, was an obvious place to start, since it houses both Nathaniel Kleitman’s papers (recently donated by his daughter, Esther) and the institutional records of the Associated Professional Sleep Societies, the successor of the original APSS. But much of the recent and highly transformative period of sleep research is within living memory. I was grateful to have the opportunity to interview a number of investigators who began their work in the field within a decade of REM’s discovery. Although I have relied far more on published and archival documents in creating this history, my conversations with sleep researchers did help to shape my ideas on how a history of knowing sleep might be written. In particular, I found my interviewees’ appeal to visual, instrumentally generated evidence so frequent as to be almost routine. But it was not necessarily consistent among different interviewees. In some cases, this appeal took place outside of any ‘laboratory tour’ that had been offered to me. When our interview turned to a discussion of REM sleep and its discovery, for example, one Canadian researcher immediately reached for a paper and pencil and began sketching out the various waveforms that made up sleep stages.13 In other cases, researchers would spontaneously discuss social factors in their research apparatus. After showing me

16

The Sleep of Others

her experimental system to study how neuronal mechanisms in the rostral ventromedial medulla of a rat’s brain modulate pain in sleep and wakefulness, my interviewee, a neurobiologist at the University of Chicago, was moved to account for the differences in her recordings and those of some of her colleagues in terms of sociological factors.14 The roughly ninety-minute microelectrode recordings from her lab sometimes suggested that the animal was sleeping when video recordings indicated that it was active and awake. Her undergraduates, paid to make just such comparisons of the two sorts of visual evidence, were instructed to classify such periods as ‘wakefulness,’ despite the evidence of the microelectrode recordings. Her colleagues studying the function of sleep deprivation, she claimed, could never dispense with their graphical recordings in this way, since such studies involved hundreds of hours of recordings taken over several months, making it prohibitively expensive to compare every second of this data against videotaped evidence. Another investigator, recalling the early days of the APSS, made a point of noting how the dissemination of EEG recording among those interested in sleep and dreaming was greatly facilitated by the fact that it generally took no more than two weeks to learn how to ‘score’ polygraphic recordings to detect REM periods.15 In these and in other cases, it became clear to me that the ability to visualize sleep in terms of recorded graphical data was absolutely pivotal to the status of sleep as an object of modern scientific research and biomedical practice. Admittedly, recordings could play different epistemological roles in different investigative situations, be they a quest to understand sleep’s physiological function, a study of highly specific brain mechanisms, a clinical diagnosis of a sleep disorder, or a laboratory-based analysis of dreaming (to name but a few). But in each instance, graphically recording sleep was an indispensable part of the life-world of sleep researchers. It was part of their habitus – a practice that was so routine as to exist below the threshold of cognition, while at the same time thriving as a cognitive object that explicitly distinguished their activities from those of other investigators.16 Recording, in short, had become integral to who sleep researchers thought they were, as sleep researchers. That scientists of any stripe should place such a priority upon instrumentally generated visual data is no less surprising to the historian or sociologist than it is to anyone who has worked in a laboratory. For nearly thirty years, sociologists, historians, and philosophers have paid close attention to the diverse roles such evidence has served in the scien-

Introduction

17

tific enterprise. This study of sleep, and the questions I asked my interviewees, was consequently informed by such attempts to fit ‘seeing’ into scientific work. The discovery of REM obviously looms large in such a schema, but in a rather curious way. Charting the course of REM periods through the night was, at least in its initial formulation as the physiological counterpart to dreaming, as much about the process of seeing as it was (and is) part of the practice of seeing. The analogy between dreams and seeing – as vision – is so pedestrian as to hardly warrant comment. Dream ‘images,’ which are by convention visual and by exception otherwise, thus take advantage of the most powerful tool humans seem to have to turn the world around us into a collection, more or less orderly, of objects. Yet we appear to do this to our selves in dreams, and it is this fact that has made dreams such a potent and perennial source of self-knowledge over the centuries. Seeing in dreams is perhaps no less a habitus (though one so broad as to be almost universal) than seeing REM in a sleep laboratory or clinic. In this sense, REM’s discovery might be read, not as a transformative moment that turned sleep into a scientific object, but as a second-order objectification, merely another chapter in the long history of turning ourselves, as individuals, as organisms, or as populations, into objects. I am not, at this point, convinced that such a metaphysics is either appropriate, accurate, or even useful. But I must confess that my attempt to create a coherent history of sleep has occasionally fallen under its spell.

1 The Persistence of Privacy

Before concerns about physiological efficiency began to take hold of the biomedical imagination in the nineteenth century, sleep was little more than the matrix in which dreams could form. The observation of sleeping bodies, be they human or animal, played no practical role in understanding the origin or the meaning of dreams. The creation and communication of knowledge about dreams was grounded in evidence generated by private consciousness. Regardless of what position one took on dreams, all writing on the subject invoked or assumed varying degrees of self-observation on the part of both author and reader. Third parties – that is to say, others subjected to observation – were not involved in the creation of knowledge about sleep or dreaming. But, if these others did not exist as potential sources of information, they did emerge as objects of considerable anxiety. In antiquity, physicians and philosophers alike worried that their own authority over the knowledge of nature, health, and disease was threatened by the popularity of dream interpreters and the Asclepian cults. Early church fathers and Christian humanists, on the other hand, emphasized the spiritual dimension of dreams even as they acknowledged the natural origins of many dreams. Threats of heterodoxy obliged them to accept the revelatory potential of dreams, but, in their hands, dreams offered a diagnostic window into the state of the dreamer’s soul, rather than his or her body. Philosophers ultimately jettisoned this approach in the wake of the religious turmoil of the sixteenth and seventeenth centuries. In their analyses, dreams became exemplars of the irrational that threatened to undermine the new social order grounded in the individual’s exercise of reason in daily life. Eighteenth-century physicians extended this approach by focusing on sleep as a regimen that could, if properly practised, eliminate dreaming altogether.

The Persistence of Privacy 19

Empirical Trajectories How is it that people feel justified in claiming that they know anything at all about dreams? Such claims fly in the face of experience. After all, what could be more absurd than listening patiently to someone recount their dream, only to respond with ‘You’re lying – you couldn’t have dreamed that.’ When this very scene unfolded in the dark comedy The Young Poisoner’s Handbook (1995), the joke was on the psychiatrist who claimed to be able to distinguish genuine dream accounts from fabricated ones. The dreamer in question here is a character named Graham Young, who poisons most of his family and then is caught, tried, and sentenced to life in custody as an incurable psychopath. He aspires to escape and kill again, so he enters into a rehabilitation program. The film is set in the 1960s, so, not surprisingly, Young’s psychiatrist (Dr Ziegler) soon begins to inquire about his dreams. But, because Young doesn’t dream, he has no stories to tell. In a panic, he makes up one on the spot. When Ziegler is unconvinced, Young begins to extort dreams from his cellmate, who eventually hangs himself after suffering Young’s nocturnal interrogations for several months. Ziegler recognizes these dreams as genuine but fails to detect his patient’s plagiarism. Young is soon released and proceeds to poison the staff at a photo shop where he has recently been employed. Since getting caught was part of Young’s plan from the outset, Ziegler’s initial detection of the fabricated dream is the first instance of a public authority in the film successfully thwarting Young’s schemes. But Ziegler’s success is short-lived. Young is ultimately able to reassert his inalienable authority over his own dreams even though (if the audience is to believe Young’s narration of the events) there were no such dreams to be had. The psychiatrist and parole board are ultimately duped, and this seems a logical conclusion, given the faulty premise upon which they built their knowledge of the criminal: namely, that they could somehow know what he was really dreaming. Yet there have been multiple systems of knowledge built out of the private experience of dreaming. Such systems appear and reappear throughout recorded history, so much so that it would be surprising indeed to hear of a culture or epoch that lacked authorities claiming to know the truth about dreaming. How have these systems of knowledge been created and defended? Current knowledge of dreaming is dominated by experts – profession-

20

The Sleep of Others

ally trained individuals who attempt to analyse dreaming according to the precepts of their discipline. Psychologists, psychiatrists, neuroscientists, philosophers, literary scholars, and historians will almost certainly disagree as to what dreams really are or how they are produced. But each group attempts to craft its own knowledge of dreaming by turning to the tools, techniques, and concepts common to its field, be they by recording the activity of a single neuron in the brain of a sleeping rat or by a hermeneutic analysis of a text from the second century A.D. In either event, the investigator will draw primarily, if not exclusively, from sources that are accessible to other investigators. The experiment can be replicated in another laboratory; the very same text can be read and interpreted by another scholar halfway around the globe. Whether or not this sort of replication actually occurs is beside the point. The fact of the matter is that the experts tacitly agree that it should be possible for another investigator to recreate empirically what was originally observed. If this wasn’t the case, we would all dispense with things like references and footnotes. This sort of expertise simply did not exist in antiquity. Of course, there were philosophers, historians, physicians, and the like, many of whom we know by name and creed. And, as is true today, the methods and rhetorical strategies used by these individuals varied widely. But their social situation was completely different from that of modern-day investigators. They had no state-sanctioned universities with students following a formalized curriculum. They had no mandate to educate and thereby improve a wide spectrum of the citizenry. There were no international congresses and conferences to standardize knowledge according to a general consensus. There was no peer review of journals and books. There were no codified methods of determining how knowledge could or could not be generated, and there was no research imperative driving them to push these methods into ever more obscure parts of the world in pursuit of originality. The transmission of knowledge was a personal affair. The student was expected to master the knowledge held by his teacher through a lengthy and intimate process similar to an apprenticeship. In such a world, knowledge of dreaming amounted to a public accumulation of private experience; there were no attempts to render the experience itself public. There were no empirical codes, save one: the dreamer was the final authority. Dreams were treated primarily as lived experience, and only secondarily (if at all) as impersonal objects of investigation. But this is not to say that studies of dreams in antiquity were derived purely from the individual caprice of the investigator.

The Persistence of Privacy 21

Explicit or implicit appeals to empiricism were commonplace in such studies, just as they were in the magic and natural philosophy of the period. But observing phenomena in order to obtain more data or to challenge hypotheses is a complex affair that can produce surprising and often counter-intuitive results. Observation is not only enmeshed in theoretical ideas about how the world works, it also reflects how we classify. The notion that garlic inhibits the attractive force of magnets was, for example, a commonplace among ancient and early modern natural philosophers.1 Its success as an observed phenomenon depended in part upon how both garlic and magnets were perceived as the kinds of things capable of displaying the virtues of ‘sympathy’ or ‘antipathy’ that ruled the cosmic order of the day. When magnets were reclassified as the sorts of things that displayed ‘magnetism,’ the link between garlic and magnets was severed and new associations appeared (one of which, ‘animal magnetism,’ will be examined in chapter 2). Thus, dreams, like magic, were subjected to procedures of empirical observation that eventually opened up into experimentalism (or something like it). How such procedure evolved has been a long-standing concern for historians of science.2 But, in any event, the trajectory of empirical knowledge was rather different in the case of dreams than it was for, say, magnets or garlic. The latter were quite obviously things that existed as objects in the world. Accounts of their powers involve manipulation, usually to some great public effect. Dreams, on the other hand, are more akin to experience. The thing itself – the dream – is inevitably hidden from public view, even if the results of the manipulation (such as the cures occasioned in the Asclepian temples, discussed below) were very public. Knowledge of dreaming was therefore inevitably projected towards the domain of the personal. Readers (or patients, or clients) expected that knowledge about dreaming could and should be directly correlated to their own experiences. Sleep, the externally observable counterpart to dreaming, played no direct empirical role in studies of dreaming, precisely because sleep involved an annihilation of the perception and personal experience that was so remarkably modified in dreams. This personal and experiential trajectory of dreaming was linked to three domains of inquiry and practice: those of prophecy, therapy, and epistemology. These domains were by no means clearly delineated. Prophetic and revelatory dreams co-existed on the same phenomenological plane as did those dreams caused by an upset stomach or by the faded sense impressions of the day’s events. Many dream theories exploited

22

The Sleep of Others

the relationship between prophecy and medical prognosis, and also drew conclusions about the nature of reason from observations about the physiological origins of some dreams. But, in each domain, the relationship between dreams and sleep was minimized, and the orientation of the theories was inevitably personal. Dreams were treated less as objects to be manipulated than as events to be experienced. The Reign of Dreams The greatest testament to the authority of the dreaming subject and the pervasive interest in dreaming in antiquity is more architectural than it is textual. The cult of Asclepius enjoyed considerable popularity from about the fifth century B.C., and its ritualistic emphasis on dream incubation indicates the degree to which the private world of dreaming governed the more public world of sleep. Asclepius’ origins are controversial, as befits a deity whose fame spread so quickly from the local to the universal.3 Both Hesiod and Homer, writing near the end of the fifth century B.C., concurred that Asclepius was born in Thessaly. But they disagreed as to whether he was born naturally of his mother, Coronis, or whether Apollo, furious at Coronis’ marriage to a mortal, killed her and surgically removed the infant from her womb, leaving him to be educated by the centaur, Chiron. Later accounts shifted the scene to Epidaurus, on the Peloponnesian peninsula. In any event, it was the cult at Epidaurus that transformed Asclepius from a local deity responsible for protecting the health and well-being of the town’s citizens into a figure whose very image came to represent healing and its attendant arts. Epidaurus boasted a temple to Asclepius dating back to the end of the sixth century B.C., and it was probably Dorian colonists from this region who brought the cult with them when they resettled on the island of Cos, where another temple was soon raised. A similar site appeared farther north, in Pergamum, and by 420 B.C. an Asclepian temple was established at Athens. A century later, the Macedonian conquests spread the cult throughout the ancient world as an integral part of Hellenic culture. In 291 B.C. an Asclepian temple was established on Tiber Island at Rome. The complexes at Epidaurus, Cos, and Pergamum were particularly lavish, but the spectacular beauty of their temples, baths, artwork, altars, and theatres masked the relative simplicity of the healing rituals themselves. Patients, typically those suffering from a variety of chronic ailments, visited the temple during the day to offer a sacrifice, sing songs

The Persistence of Privacy 23

extolling the god’s virtues, purify themselves by bathing in sacred waters, and don white robes.4 At nightfall, they entered the abaton, where they lay down on a wooden pallet and fell asleep. Their cure could result from a number of interventions. Patients might awaken, miraculously healed by the touch of the god himself or of one of his symbolic agents (usually a snake or a dog). More frequently, Asclepius would appear to them in a dream, offering instructions as to how they could be healed. Occasionally, temple priests would need to decipher the dream, but more often they acted as facilitators, carrying out the needs of the patient. The patients awoke the next morning, cured or optimistic of such an end, and proceeded to show their thanks by having their experiences memorialized in tablets and votives placed at the shrine. A recent analysis of the Epidaurian tablets depicts these narrative inscriptions as derived from earlier oral traditions. Personal details are frequently minimal or completely lacking, and even the most miraculous cures are communicated in an almost pedestrian fashion.5 Even when the patients were well known, Asclepian dreams often followed a rigid, impersonal, and straightforward chronicle of events that ended in cure. Such was the case with Aelius Aristides, an orator who recorded his dream cures at the Pergamum and later published them as The Sacred Tales. Aristides reported having the following dream involving the cure of his chronic pain in his back and neck in August 148 A.D.: ‘He [Ascleipus] said that there was a royal ointment. It was necessary to get it from his wife. And somehow after this, a servant of the palace, clad in white and girdled, appeared at Telesphoros’ Temple and statue, and escorted by a herald, went out by the doors where the statue of Artemis is, and bore the remainder of the ointment to the Emperor.’6 The fulfilment of the dream was not a great deal more intriguing than going to the pharmacist to fill a prescription. After he awoke, Aristides described the dream to a temple priest, who then retrieved the ointment from the base of a statue of Hygeia (Asclepius’ wife), where it lay after being deposited there by a noblewoman. The almost pedestrian tone of some of Aristides’ narratives indicates just how closely his sense of self-identity, both as a patient and as an orator, was linked to the rituals of dream incubation.7 Aristides was surely unusual among Asclepian supplicants in terms of the frequency of his visits and in the eloquence and publicity of his narratives, but his close identification with his illness and with the god who relieved his myriad symptoms was likely shared by his fellow devotees. Dream incubation was more than simply a means to an end. It was a way of life.

24

The Sleep of Others

All divinities could be involved with healing, and Asclepius’ escalating fame after the fifth century was probably as much an appropriation of the traditions of other gods as it was a novel invention. But the Asclepian cult was distinct in two important ways: it was dedicated exclusively to healing; and it entailed the use of the singular method of dream incubation.8 It was not just the image of the god, but the methods of his supplicants, that spread with such rapidity throughout the ancient world. If, as the historian of medicine Vivian Nutton suggests, neither Epidaurian propaganda nor Athenian plagues are sufficient to account for the rapid spread of the cult, perhaps the technique itself might warrant further consideration as an explanation for its success.9 Dreams are extraordinarily intimate: like pain, they are fundamentally private experiences. They can be interpreted by others in search of a meaning, but, unlike a suppurating wound or broken limb, they can signify nothing until they are spoken of by the subject who has first experienced them. Personal authority reigns supreme in such matters, and a healing cult based on such experience leaves a considerable degree of power in the hands of the individual seeking a cure. This focus on individual authority was sometimes reflected in the level of intimacy occasioned between the supplicant and the god at the Asclepian temples: one account from Pergamum described how an epileptic ‘struck up a conversation’ with Asclepius in a dream, with the subsequent result that he fell ill with a quartan fever and then recovered, cured of his epilepsy.10 Putting aside the question of the possible psychosomatic benefits of such a system of healing, it seems reasonable to suggest that the popularity of the Asclepian cult was at least in part due to its ability to reflect the self through routine intimacy with a god.11 This egalitarianism was reinforced on a social level by making sleep the focal point of the ritual. Sacrifices obviously differed according to one’s means, but all supplicants, be they peasants or governors, shared the common practices of bathing, dressing, chanting, and sleeping in the Asclepian shrines. The role of sleep in such rituals, however, was limited to that of a mediating factor. There is no indication that either the temple priests or the supplicants were interested in sleep for its own sake. No doubt, a restful night’s sleep was desired by many a weary pilgrim, but sleep’s primary role was to function as the vehicle through which the god could intervene. All rituals were designed with this end in mind, and there is no indication that such sites were ever used for the systematic observation of sleeping patterns or habits. In fact, the very architecture of the Asclepian shrines and the nature of the rituals made such a task imprac-

The Persistence of Privacy 25

ticable, if not impossible. The typical abaton (or adyton, ‘the inaccessible’) was a long, covered structure of individual rooms.12 In less lavish shrines, where patients were obliged to sleep in the temple itself, or in a nearby grove of trees, their privacy was protected by none other than the god himself. One cautionary testimonial recalled the punishment meted out when a curious man climbed a tree in the hope of seeing what went on in the abaton at night: he fell and landed on a fence, nearly losing his sight.13 The observation of sleeping patients (examined in chapter 4) lay in the distant future, emerging well after the medical clinic became the focal point of biomedical knowledge in the early nineteenth century. The Medical Use of Dreams in the Hippocratic Tradition Sleeping and dreaming played somewhat less integral roles in the Hippocratic tradition of medical practice.14 Nonetheless, these phenomena are worth investigating, since their position within the Hippocratic system of medicine is indicative of the enduring dominance of the categories of the private and the idiosyncratic for healing in antiquity. For its part, sleep appears primarily as a kind of matrix in which health can be said to exist. Like summer winds that blew unusually hot and dry, or persistent spring rains that saturated the ground below, an alteration in sleeping patterns could have an adverse effect on health. Just as the Hippocratic physician had to know something about the physical environment in which he practised, so too did he need to pay attention to his patients’ accounts of their individual sleeping habits. Such details offered clues concerning the humoral balance that defined the very nature of health for each individual. Galen’s reinterpretation and codification of Hippocratic practice in the second century A.D. emphasized the role of sleep as a ‘non-natural,’ or a bodily practice that could be altered to maintain and improve health. Dreams were also significant for diagnosis, but, unlike sleep, they provided no corresponding route of therapeutic intervention. Physicians frequently acknowledged that dreams had potentially divine origins with sometimes miraculous results, but their resolute focus on the body led them to define dreams as signs requiring interpretation, not as portents or as a set of instructions waiting to be carried out. According to Hippocratic tradition, the soul was cut off from external sensation during sleep and was thus capable of perceiving signs of disease that otherwise remained hidden to both patient and practitioner. Thus, the

26

The Sleep of Others

‘place’ of the dream15 was quite different in Hippocratic practice than in the Asclepian cults. In the latter case, supplicants were compelled to incubate their dreams in well-defined physical sites. The narration of their dreams was secondary, often taking place only after they had been cured. The dreams of a physician’s patient, on the other hand, were not specific to any particular site; rather, they emerged as part of the patient’s presentation of his illness in a dialogue. This initial verbal expression made dreams in their Hippocratic context more like objects: they were, like a rash or a cough, primarily experienced as shared things in the world – signs subject to discussion, dispute, negotiation, and interpretation. Yet the differences between the medicine practised by the physicians and the healing rituals of the Asclepian temples were more a matter of emphasis than of opposition. Almost all physicians, from the Dogmatists (who derived their medical theories from rational philosophy) to the Empiricists (who denied the validity of all medical theory), agreed that miracle cures were possible, in principle if not in practice. 16 The Hippocratic writers acknowledged that talismans and other magical devices could heal, but they emphasized the manipulation of humoral balance through bodily regimens and herbal remedies as a superior means of restoring and maintaining health. Such a conciliatory position made sense, given the fact that physicians were in the minority among healers in antiquity and did not enjoy a social status much above that of a craftsman.17 It would have been difficult indeed for a physician to rail against all forms of religious healing when many of his patients were as likely to visit a shrine or call in a dream interpreter as they were to submit to the physician’s intimate interrogations and humoral analyses. The bedside, particular that of the wealthy patients who could actually afford the services of a physician, was a busy place, filled with diverse practices and opinions: ‘To put it bluntly,’ suggests Nutton,’ the bedside is not always a scene of isolation, where doctor, patient, and disease fight out a lonely battle. Rather, we must imagine a crowded sickroom, filled with friends, relatives, visitors, and healers, competing but also co-operating.’18 Nutton also reminds us that physicians could be found among another medical crowd – the one that gathered around the shrines dedicated to Asclepius, where physicians like those at Athens made their twice-yearly sacrifices to the god.19 Despite this pervasive influence of the divine, physicians nonetheless attempted to distinguish their modes of healing from those found in the temples. In the Hippocratic corpus, the most significant writing in this

The Persistence of Privacy 27

regard has long been On the Sacred Disease, a text that has traditionally been upheld by medical historians as offering a definitive case for setting Hippocratic and other proto-scientific physicians apart from all religious healers.20 The naturalistic account of epilepsy offered there is typical of the Hippocratic corpus, in that it identifies a constitutional predisposition to the disease as well as the environmental circumstances that tend to exacerbate it. In this case, seizures are said to be brought on in phlegmatic people by exposure to moist south winds that prompt the descent of cool, wet phlegm from the brain into the body. The disease, despite its spectacular symptoms, is thus no more divine than any other, and the charms and purification methods recommended by wandering priests and conjurors are ill-advised, since they are founded on false premises. Ill-advised, but not necessarily ineffective. Some of the cures recommended by the itinerant religious healer in On the Sacred Disease were not particularly different in kind from those offered by physicians. Abstaining from bathing or avoiding certain foods was common enough medical advice. What seems to have offended the Hippocratic author was the impiety of these other healers, who claimed to act as the agents of the gods in either dispensing advice or in procuring their magical charms and devices.21 Their interventions implied that they could manipulate the gods to cure their clients – for a price. Such cynicism differed considerably from the intimate requests for Asclepius’ personal intervention by the ill on the temple floor. The very ritual of sleeping in a temple was, in part, a public demonstration of the piety and devotion of the individual making the request, something that could not be recaptured by a magician hired to dispense his wares at the patient’s bedside. To the extent that The Sacred Disease can serve as an example of the Hippocratic attitude to supernatural healing, then, it seems that it was one of delineation of the sacred and the profane, rather than a denigration of the role of the former in the healing arts. The discussion of dreams in the Hippocratic treatise Regimen IV (also known as On Dreams) offered a similarly melioristic attitude towards the divine. The author of Regimen IV readily acknowledged that prophetic and revelatory dreams existed and that they must be interpreted by dream interpreters (discussed below). The diagnostic dream, on the other hand, could be properly analysed only by a Hippocratic physician, for only such experts understood the humoral imbalances signified by such dreams.22 This distinction between the divine and diagnostic dream, however, was hardly straightforward, and most of Regimen IV was taken up with generic examples of dream images that were amenable to

28

The Sleep of Others

medical interpretation. Dreaming of a star’s course being altered by rain or hail, for example, suggested that phlegm-like secretions were adversely affecting the ‘outer circuit’ of the body. The solution was to eliminate excess moisture by increasing perspiration through vapour baths, eating dry foods, and exercising while wearing a cloak. But, along with such changes of regimen, precautionary measures were also to be taken in the form of prayers to the gods, thus suggesting that the same dream could be amenable to both supernatural and medical interpretation.23 The fluidity of the boundary between the immanent and the divine is hardly surprising, given physicians’ and diviners’ common reliance upon analogy to make sense of the dream.24 Both practitioners needed to draw similarities between the dream images and the dreamer’s own life in perform their work of interpretation. The most tangible difference between the two was precisely the one advertised by ancient physicians: namely, their claim to superior understanding of natural philosophy as reflected in their assessment of the humoral imbalances that disrupted health. But what was the role of sleep in the physician’s understanding of these humoral imbalances? Although the timing and duration of sleep were sometimes important to Hippocratic cures, the causes of sleep were generally discussed only insofar as alterations in sleep influenced dreaming. The author of Regimen 3, for example, suggested that an excess of exercise and food initially produced long and pleasant sleeps owing to an increase of moisture in the flesh. But, as soon as the body became unable to assimilate this excess, sleep was disturbed and dreams of struggling ensued.25 Excess moisture was also described as a cause of sleeplessness in On the Sacred Disease; sudden shouts and cries in sleep, on the other hand, were attributed to an overabundance of dryness and heat. The author of Regimen 4 offered a more sophisticated account of the soul’s activities in sleep. In this treatise, the soul continued to perceive in sleep, but its object changed. The soul ceased being a ‘servant’ of the body and was no longer obliged to carry out the various offices of sensation and motion. It was free to perceive things and events of the body, rather than through the body, with the results expressed as dreams. The author of Regimen 4 repeatedly used active sensory verbs to describe these perceptions, indicating that he considered dreaming to be a cognitive process: one did not have a dream, one saw a dream.26 The contrast between such a view and its modern counterpart (discussed in chapter 2) could hardly be more explicit. In the latter, the mind in sleep became ‘trapped’ in the body because it was ‘cut off’ from the world of objects.

The Persistence of Privacy 29

This imprisonment produced dreams that, in the modern approach, became identified as bizarre and distorted versions of normal perception, rather than traces of the soul’s tentative liberation. Movement, Sensation, Teleology The genealogy of this transformation can be traced back to a handful of Aristotelian treatises on sleep and dreaming. But the influence of Aristotle’s work in this regard should not be overestimated: unlike either of the healing traditions already discussed, Aristotle’s theoretical analysis of sleep and dreaming lacked a corresponding positive practice. In keeping with his work in natural history and his investigations into the nature of the soul, Aristotle’s arguments were an amalgam of observation and reason, with an emphasis on teleological forms of explanation. Such an approach was utterly incapable of inspiring further investigation when it was turned upon sleep, because it suffered from both an internal contradiction and an external obstacle. The very condition of sleep – the abeyance of sensation – meant that this condition of the soul could not be directly observed by the individual experiencing it. This contradiction could be overcome in two ways: by inferring that the cessation of movement in animals was a sign of sleep, and by accepting dreams as evidence that could somehow illuminate the nature of the soul in sleep. Progress in the first direction was hindered by Aristotle’s analysis of sleep as a negation. Progress in the second was restricted by the inherent corruption of its source, since the only practice remotely resembling a ‘natural history’ of dreams in antiquity came from the two most prominent cultural dream practices that Aristotle explicitly criticized as irrational: dream incubation and divination. In contrast to the medical writers, Aristotle paid considerable attention to sleep apart from its relationship to dreaming. But he nonetheless continued to frame the problem of sleep in negative terms as the privation of wakefulness.27 And, since waking consisted of ‘nothing else than the exercise of sense-perception,’ the essence of its opposite, sleep, could be defined as the periodic powerlessness of the organs of sensation.28 Sleep, in this analysis, was not common to all life. Plants, for example, underwent growth and decay, but, because they had no capacity for sense perception, they did not sleep. Conversely, Aristotle reasoned, all animals must have periods of wakefulness and sleep, since their organs of sense perception must become exhausted at some point and cease to perform.29 Further, since sleep involved the whole organ-

30

The Sleep of Others

ism simultaneously, it must be the powerlessness of the central organ of common sense that led to sleep, and not merely the cessation of function in the individual special senses.30 It was not reasonable to suppose, argued Aristotle, that sleep was an alteration of any one particular sense, because even those animals that lacked smell, sight, taste, or hearing must also sleep. Sleep was thus an alteration of the sensitive soul; in particular, it was an alteration of qualities perceptible by touch.31 Aristotle argued in several different treatises that touch was fundamental to all the other senses, and that it was the one sense that belonged to all animals, because it was the only one that was essential for the procurement of nutrition.32 Touch and nutrition were united in Aristotle’s treatise under the opposing qualities of ‘hot’ and ‘cool.’ When paired with the opposing (and equally tactile) qualities of ‘moist’ and ‘dry,’ these four qualities were capable of accounting for the composition of the material world; but, in his account of sleep, Aristotle relied only on an analysis of how internal heat regularly changed in the body. This alternation between hot and cold in the body was, Aristotle reasoned, due to the processes of nutrition. Changing ingested food into blood, he argued, produced an ‘evaporation’ that was hot and so naturally rose upwards, along with the blood, towards the head.33 This movement initially produced a heavy sensation in the head, causing the eyelids to close and the head to nod. But true sleep would come only when this ‘spirit’ condensed upon reaching the brain, which was, by its nature, cold. The spirit and blood was thus forced downwards into the veins, towards the heart, which Aristotle considered to be the primary sense organ. The resulting concentration or ‘natural recoil’ of spirits towards the centre of the body caused a ‘seizure’ of the heart, inhibiting its sensory functions and causing sleep. In addition, the peripheral parts of the body were simultaneously deprived of the blood’s natural heat, thus making movement impossible.34 Hot and cold, like moist and dry, were qualities readily attributed to both food and internal states of the body, so Aristotle had little difficulty making his case. The feeling of doziness, for example, typically came on after digesting substances with known properties of heating, which included all foods (particularly when taken in large quantities) as well as wine.35 Likewise, sleep could result from feverish illnesses and fatigue caused by exercise, both of which tended to produce hot evaporations that acted exactly as those produced in digestion.36 Soporific substances, such as preparations of poppy or mandrake, worked according

The Persistence of Privacy 31

to similar principles.37 The idea that ingesting naturally hot things could generate sleep, which was itself a cooling process, was not contradictory for Aristotle, since there were several possible explanations. Sleep could, for example, result from a rapid change between hot and cold, just as ‘those who have hot water poured on them feel a sudden shiver of cold.’38 Or the process of sleep could be analogous to how the heat of the sun produced vapour that ascended to the sky, only to later fall as cool rain.39 But these were merely analogies. At no point in his treatise did Aristotle suggest that the rhythms of sleep were in any way affected by heating and cooling caused by the alternation of day and night. Sleep was not a habit or an instinct for Aristotle. Rather, it was a property of the sensitive soul, and was caused by its movements insofar as they were affected by the nutrition required by the body. Sensation also served as sleep’s teleology for Aristotle. The final cause of sleep – its purpose – was to conserve the animal by ‘remission of movement.’40 But, while this remission was ‘necessary and beneficial,’ it was not an end unto itself. Animals did not sleep simply to rest their musculature. The end of sleep was to make the waking state possible, ‘since the exercise of sense-perception or of thought [which was only possible during wakefulness] is the goal for all beings to which either of these appertains.’41 Thus, sleep was not merely the inability to exercise sense perception, as was the case in fainting, asphyxiating, or in an epileptic seizure.42 These latter conditions were caused by chance or accident, whereas sleep was caused by necessity, governed on the one hand by the daily nutritive requirements of the body, and on the other by the perceptual apparatus’s need for rest. Aristotle united the two by appealing to evidence drawn primarily from the exercise of self-perception. For Aristotle, the experience of feeling alterations of heat and cold during ingestion of foods and drugs indicated how the material and efficient causes of sleep relied upon eating; the very ability to perceive in wakefulness, on the other hand, was itself a gesture towards sleep’s final cause as the preservation of sense perception and of reason. This is not to imply that Aristotle completely ignored the observation of others in his analysis of sleep. Although such observations played little role in On Sleep, observations of sleeping animals figured more prominently in his History of Animals. Closed eyelids and motionlessness in animals offered proof that they were sleeping, and in animals without eyelids, such as fish, molluscs, crustaceans, or insects, lack of movement was enough to signify sleep.43 Aristotle’s most extensive set of observations of animal sleep described the sleep of different species of fish. The

32

The Sleep of Others

fact that the sleep of fishes was often a condition of their being caught for food – it was well known that they could be easily caught at night by encircling them with nets or by scooping them up under torchlight – must certainly have contributed to Aristotle’s focus on this one group of animals.44 Domesticated animals, on the other hand, furnished Aristotle with evidence of what he seems to have considered a commonplace: that animals dream. As examples, he cited only domesticated animals, such as horses, oxen, sheep, and goats, but he extended this claim to cover all viviparous animals. Yet the only evidence he put forward was his observation that ‘dogs show their dreaming by barking in their sleep.’45 This relative absence of observational evidence and its unsystematic organization testifies to the minimal role that observations of animals played in Aristotle’s causal accounts of sleep and dreaming. As we have already seen, Aristotle took movement as merely a token of the existence of sense perception and of thought, the exercise of which was the highest end for all sentient beings. Explaining the relationship between sense perception, thought, and sleep was the goal of Aristotle’s analysis. It was how these things were tied together in human, not animal, experience, that counted. To this end, accounts derived from personal experience played the most powerful role in Aristotle’s analysis of the causes of sleep. Such accounts also appealed to the personal experiences (in particular, those involving eating, which was of course universal) of those who read On Sleep. The appeal to a private consciousness should come as little surprise, given the overall orientation of Aristotle’s treatise towards problems of dreaming, rather than sleep itself. ‘Of all animals, man is most given to dreaming’ ran the culmination of Aristotle’s brief chapter on animal sleep.46 Understanding the nature of human dreams was at the very core of his causal observations of animal sleep. Against his Platonic critics, Aristotle justified his observation of the behaviour of lower forms of life by appealing to the ends such behaviour served.47 Reproduction and growth, for example, could just as well be studied in bees as in man, despite the enormous differences between their relative proximity to the divine. In either case, the beauty of the purpose of such processes could be revealed. Thus, the purpose of sleep – the temporary cessation of movement – could be known by observing sleeping animals. But dreaming was another matter entirely. Dreams themselves did not serve a purpose: for Aristotle, they were accidental products of imagination that occurred during sleep. But, because dreams were alterations of the sensitive soul, they could be examined in order to draw conclusions about the nature of perception. And only

The Persistence of Privacy 33

man was a fit subject for such observations. Observations of dreaming animals, despite their seeming ubiquity, therrefore figured nowhere in Aristotle’s treatise On Dreams.48 Aristotle defined dreams according to their relationship to both thoughts and perceptions. Dream images were not perceptions, since the special senses were unable to function in sleep as they did in wakefulness.49 The soul was capable, however, of acting upon dream images in exactly the same way as it acted upon waking perceptions. Personal experience testified to this fact, for it was ‘manifest to any one who should attend and try, immediately upon arising from sleep, to remember.’50 Thought, in the form of judgment or opinion, could also be had in sleep: some people, reported Aristotle (perhaps thinking of his own classificatory labours), even felt themselves to be organizing lists of subjects in their sleep according to some preconceived plan or ‘mnemonic rule,’ just as though they were awake.51 But instances of thought in sleep and similarities between dream images and waking perceptions served only to reinforce the accidental nature of dreams themselves, which were best considered ‘in the light of the circumstances attending sleep.’52 Sleep, as discussed above, altered the functioning of the sense organs. The nature of this alteration upon sense perceptions could be compared to the nature of the movement of objects in general: just as projectiles continued to move long after the cause of that movement had ceased, so too did the effects of perceptions remain in the sense organs long after the perception itself had ended.53 In wakefulness, these residual perceptions were easily ignored by the more powerful perceptions continually presented by the sense organs, just as a small fire was obscured by a larger one, or as trifling sensations were concealed in the presence of great pains or pleasures.54 But in sleep, the reflux of blood towards the interior of the body not only enfeebled the sense organs but also carried their residual presentations towards the heart. The nature of this organ of common sense meant that the slightest movement, like that remaining in a past sense impression, could be magnified, just as very bright mirrors were turned cloudy by the reflection of a menstruating woman, or as highly polished bronze showed even the slightest smear of the hand.55 When confronted with such degrees of perfection, these ‘stains,’ slight as they might be, made an indelible impression that was analogous to the reception of dream images in the soul. That Aristotle considered these ‘stains’ in a pejorative sense was confirmed by other, equally negative, analogies: oils and wines, for example, were like-

34

The Sleep of Others

wise spoiled or ‘infected’ by being placed nearby small amounts of odour.56 Dreams also appeared more like waking perceptions according to the pathological nature of the dreamer’s soul. Just as cowards consumed by fear saw everywhere their enemies approaching, so too did people unduly influenced by their emotions or in the grips of fever experience the most vivid dreams.57 Aristotle was not, however, pursuing some sort of proto-psychoanalytic scheme of dream interpretation. Nor was he attempting to confirm the utility of dreams in the diagnosis of disease, although he acknowledged that this was, if unreliable, at least possible and reasonable.58 Finding a fit between dream and dreamer was not part of Aristotle’s project. If it had been, he certainly would have included at least one dream narrative in any of the his treatises on sleep or dreaming – yet he did not. Such an absence would have been unthinkable in a medical text on a similar topic, where problems of the idiosyncratic and its relationship to practice dominated.59 But Aristotle’s primary interest was in integrating dreaming into a causal natural philosophy; and here, the question of divination by dreams reigned supreme. The rise of natural philosophy in the fifth and fourth centuries B.C. involved a new conceptualization of what, precisely, nature was. PreSocratic philosophers began to formulate an explicit idea of nature ‘as implying a universal nexus of cause and effect,’ in contrast to the older assumption that what was natural corresponded to what was regularly or normally encountered in practice.60 The latter conception appealed to divine or magical intervention to account for events outside ordinary experience. Some dreams, for example, could be significant for prophecy and for healing precisely because such anomalies existed outside nature. Natural philosophers, on the other hand, were more all-encompassing. They attempted to bring anomalies such as monstrous births within the domain of nature by claiming that they were intelligible in principle, because they were nothing more than accidental interruptions of the universal set of causes that ruled over nature.61 They were the exceptions that proved the rule. Even by the fourth century, philosophers and physicians were ambivalent about the extent to which a uniform set of causes could account for all natural phenomena. Dreams, as we have seen, were perhaps the most potent example of how normal experience could fall outside the domain of nature. The Epidaurian cult of Asclepius had become an integral feature of urban Greek culture by the fourth century B.C., and even rival Hippocratic physicians refused to draw definitive boundaries

The Persistence of Privacy 35

between the natural and the divine when it came to dreams. Even more than the procurement of magical charms and cures, or divination by oracle, dreaming had become a definitive part of private life. Not only had it received public sanction both in the temple and at the patient’s bedside, it had emerged as a normal encounter with what was contrary to nature. In dreaming, the exception was the rule. Aristotle made this paradoxical perspective on dreaming clear at the outset of his On Divination in Sleep. The ubiquity of dream divination meant that it could be neither uncritically accepted nor contemptuously dismissed: ‘The fact that all persons, or many,’ he observed, ‘suppose dreams to possess a special significance’ stemmed from the fact that it was ‘founded on the testimony of experience.’62 Personal experience was, for Aristotle, a reasonable foundation of knowledge in the case of dreaming, as it was in other aspects of his natural philosophy. Indeed, he appealed to just such experience when he argued that dream images seemed to be reflections of things truly present because, in sleep, the soul was unable to perceive directly the imagination at work: ‘That what we here urge is true, i.e. that there are such imaginative movements in the sensory organs, any one may convince himself, if he attends to and tries to remember the affections we experience when sinking into slumber or when being awakened. He will sometimes, in the moment of awakening, surprise the images which present themselves to him in sleep, and find that they are really but movements lurking in the organs of sense.’63 But this emphasis on the priority of personal experience cut two ways, since it also served as an equally powerful court of appeal for dream interpreters and Asclepiad cultists.64 Pay careful attention to your dreams, interpret them properly, or incubate them sincerely, they promised, and their divine nature would be apparent. Aristotle attempted to circumvent this problem not by arguing that dreaming was a fundamental and normal part of sleeping. Instead, he insisted that the experience of those who claimed not to have ever dreamed was, like that of infants or those who slept immediately after meals, explicable in natural terms. In all these instances, excessive heat (due either to the sleeper’s constitution or to the nature of the food) produced violent motions that obscured the dream images in the soul, just as great disturbances in liquid annihilated any reflected image.65 Changes in age or emotional experience (or, one assumes, in eating habits, discussed below) would ultimately reverse the situation and create a sleep replete with dreams. A lack of interest on the part of the gods had nothing whatsoever to do with the absence of dreaming in the sub-

36

The Sleep of Others

ject. The ubiquity of dreaming made it appear reasonable to extend particular accounts of authentic dream divination to cover all instances of dreaming. But this, thought Aristotle, was an account based on mere appearance without the exercise of reason. Such was the logic of a world in which a god, like Asclepius, seemed to intervene routinely in human affairs. In a fully naturalized world, such explanations fell short: ‘The fact of our seeing no reasonable cause to account for such divination,’ Aristotle admonished, ‘tends to inspire us with distrust.’66 Aristotle’s critique of dream divination did not turn upon empirical evidence drawn from dreams. It could not, because to have done so would have been tantamount to engaging dream interpreters and cultists on their own ground. They were the ones who held the richest collections of empirical accounts of dreaming; their texts were the natural histories of dreaming. Nowhere in his critique did Aristotle suggest that stories of dreams of naval battles or of ‘things destined to take place at the Pillars of Hercules, or on the banks of the Borysthenes’ were in any way false.67 He implicitly accepted such dreams took place but rejected outright the implied causal connection. Dreams were not ‘causes’ of events in the world in the same way that the moon caused the eclipse of the sun; nor were they ‘signs’ of events, as fatigue was a sign of fever.68 They were simply coincidences that followed no general rule. Aristotle offered a number of explanations to account for the apparent frequency of such coincidences. People frequently pondered activities before actually performing them, so, argued Aristotle, it was no surprise that dreams representing an individual’s future activities often came true.69 Conversely, dreams could suggest a course of future action to the dreamer, and thereby appear to be a ‘sign’ or ‘cause’ of that action in this weak sense. Given the origin of dream images in the residual impressions met in the senses, it was also reasonable, thought Aristotle, that ‘familiar friends should thus have foresight in a special degree respecting one another.’70 Such people saw and thought about each other frequently, and knew each other’s habits; their dreams merely reflected the intimacy of this knowledge. Aristotle’s primary concern, however, lay beyond the domain of such individualized dreams. Indeed, he was perfectly willing to acknowledge diagnostic dreams as a manifestation of the personal and idiosyncratic nature of true dreams as a whole. It was rather those dreams interpreted as portents of momentous events that did not directly concern the dreamer that were the central focus of Aristotle’s critique. In three sepa-

The Persistence of Privacy 37

rate sections of this short treatise, Aristotle returned to his central theme that dreams of momentous prophecy could not possibly have divine origins because they were most often sent to highly imperfect recipients. The most obvious instance of this was the fact that animals dreamed. This could be interpreted only as an abomination, if it were granted that the divine purpose of dreaming was to communicate important messages from the highest being.71 More significantly, such supposedly prophetic dreams did not regularly appear to ‘the best and wisest,’ as would befit such wighty messages about the future, but to ‘people at random.’72 In fact, the only generalization Aristotle could draw about the nature of the recipients of such divine dreams was that this latter group contained a disproportionate number of ‘persons of inferior type.’ Those whose physical temperament was ‘garrulous and melancholic’ seemed particularly susceptible to such dreams because they regularly experienced emotional upsets and thus tended to ‘see sights of all descriptions.’73 Like gamblers, they tried their luck more frequently than others; their dreams appeared prophetic only because they paid more careful attention to them and reported them far more often than other people. The only concession Aristotle was willing to make regarding dreams of future events, or ‘foresight,’ was one that discounted entirely the notion that such dreams were divinely sent. Dullards and the deranged, he noted, were susceptible to such dreams because their minds were relatively vacant and were thus highly receptive to the extremely slight movements and emanations that might conceivably be caused by future events.74 In any case, Aristotle’s dedication to understanding purpose as cause made it impossible for him to acknowledge the divine origins of dreams. A divine being could not possibly create a phenomena that acted so randomly upon subjects so ill-suited for such a purpose. Nor, for that matter, would it act at night, when the soul’s highest office of thinking was largely incapacitated.75 The social and empirical implications of such a thesis are as unmistakeable as they are paradoxical: because those people who pay greatest attention to their dreams are likely dullards, the deranged, or melancholics, their testimony cannot serve as the foundation of any systematic study of dreaming. Yet such a claim also implied that any selfobservation of dreams was doomed to failure. A natural history of dreaming was impossible in principle, because such a pursuit could be conducted only by making the observer more like the cowards who experienced fearful dreams or the fools who claimed to see the future.

38

The Sleep of Others

Aristotle’s approach to the problem of dreaming, like his rejection of the divine origin of dreaming, was considerably out of step with the ubiquity of dream divination and diagnosis within his culture.76 Certainly, his argument that dreams were merely the traces of waking perception attended to by the soul as the body slept provided ammunition for critics of divination. In the first century A.D., Pliny the Elder’s Natural History briefly reiterated Aristotle’s ideas, while Cicero expanded upon these same arguments in On Divination.77 Some Neo-Platonist philosophers, in their attempts to reconcile Christian orthodoxy with pagan Greek thought, used Aristotle as a foil against those who claimed that all dreams were somehow prophetic.78 The hierarchy of dreams described by Calcidius and Macrobius in the fourth and fifth centuries reflected Neo-Platonic tenets about ontological order more generally: dreams, like being itself, ranged from those originating in the base, immanent processes of the body to those emanating from the highest spiritual essence. Only the latter could be considered truly prophetic or divine. Such a compromise fulfilled an important ideological goal for patristic writers struggling to assimilate or dispense with diverse pagan practices in a Christian empire. On the one hand, the reality of prophetic dreams had to be acknowledged in Christian culture, since there were more than forty recorded in the Old Testament alone. Yet such a possibility had to be severely constrained, lest popular rituals such as dream divination and incubation fracture the authority of a priesthood to mediate between a monotheistic God and humanity through interpretation of revealed text. When taken as an accurate, yet incomplete, account of dreaming, Aristotle’s theory allowed patristic writers to construct a hierarchy of dreams according to their origins, thus replacing an older typology that ordered dreams according to their veracity as prophecy.79 A Medical Habitus If Aristotle’s attempts to incorporate dreaming into a natural history of the soul foundered on the idiosyncrasy of experience, this very same terrain proved to be a fruitful ‘habitus’ for medical practitioners and their patients in Hellenistic culture. That is to say, practices of sleep and dreaming became routines that, because of their regularity, appeared completely natural, yet they also served to structure how individuals perceived both themselves and others within a social system.80 This had, of course, already occurred to some extent in the Asclepian cults, which

The Persistence of Privacy 39

continued to expand in imperial Rome. But the Galenic transformation and codification of the Hippocratic corpus turned this practice from one of supplication to one of self-creation. While dreams maintained their status as one of many potentially useful diagnostic signs, there are also indications that dreams came to be incorporated into certain physicians’ own autobiographies in a way that reinforced their social ambitions. Alterations in sleeping patterns, on the other hand, emerged as both signs of disease and a means of restoring and promoting health. The Greek physician Galen of Pergamum (129–c. 216), is the unavoidable centrepiece of this discussion. The sheer volume, comprehensiveness, and forcefulness of his writing ‘imposed upon later learned physicians an idea of what medicine was (and what it was not).’81 Throughout much of his writings, Galen made it clear that he saw the preservation of health as a problem of moral responsibility.82 This, coupled with the massive amounts of autobiographical detail in his works, suggests that his writings also functioned as a prescriptive account of what it meant to be a physician. Dreams were of considerable significance to Galen. They not only were diagnostic tools but, because they could be sent from God, were also of tremendous personal importance. Galen’s own career as a physician was set in motion after his father, Nikon, had ‘vivid dreams’ that his sixteen-year-old son should study medicine. Pergamum, Galen’s birthplace, was a logical choice for such an education. An Asclepian temple had existed there since about 350 B.C., and, when Galen began to study medicine around 145 A.D., a new temple was being built by Costumius Rufus, a close friend of Galen’s new teacher, Satyrus.83 Until an earthquake destroyed the city a little over a century later, the temple at Pergamum was the most important of its kind in the Roman empire. Given such a context, it is hardly surprising that Galen frequently referred to the role of dreams in his personal and professional life. At age twenty, he suffered from an abscess, which he cured by opening an artery between his thumb and forefinger after seeing Asclepius describe such a cure in a dream.84 The god offered career advice as well, which was a useful resource for someone as well connected with the elite of Roman society as Galen was. In his thirty-eighth year, Galen dreamed that Asclepius forbade him from accompanying Marcus Aurelius to war, so he asked if he could abandon plague-ridden Rome and return to Pergamum. Marcus, then embroiled in conflicts with the Parthians in the east and the Teutonic tribes in the north, frequently acted upon his own dreams and so could hardly deny Galen’s request (although he

40

The Sleep of Others

renewed his imperial contract again a short time later, making Galen physician to his son, Commodus). Dreams even influenced Galen’s writing: in On the Usefulness of the Parts of the Body, he described how he quickly completed his tract on the optic nerve because the demiurge admonished him in a dream not to abandon it.85 Dreams were equally an important part of Galen’s clinical practice. Like Hippocrates and Aristotle, Galen thought that the soul descended into the centre of the body in sleep and was there able to perceive the humoral imbalances that presaged disease. Pliny the Elder had reiterated such views in the first century A.D. in his Natural History, in which he described sleep as ‘nothing but the retreating of the soul into its own midst.’86 In one of his numerous commentaries on the Hippocratic corpus, Galen mentions having written a tract on dreams, which is now lost, save for a brief fragment.87 The fragment reiterates many of the points outlined by the author of Regimen IV, but it also hints at a more sophisticated taxonomy of dreams through which the skilled physician had to navigate. To the standard grouping of prophetic and diagnostic dreams Galen added dreams that were merely the reappearance of daytime thoughts and those that repeated the habits of the soul during wakefulness. Understanding which dreams were diagnostic and which were not was a tricky affair requiring considerable skill, particularly when it was a matter of separating out those dreams that merely recapitulated waking life from those that were truly indicative of illness. This problem had not been addressed by the author of Regimen IV, but, by the first century A.D., it had firmly established itself in the minds of many physicians, many of whom were prominent devotees of Asclepius.88 Rufus of Ephesus, for example, implored his colleagues to interview their patients thoroughly before making a diagnosis, and this rule equally applied to the problem of interpreting a patient’s dreams. In his Medical Questions, Rufus insisted that a physician must first find out about a patient’s sleeping habits before attempting to use their dreams for diagnosis.89 There were no hard-and-fast rules to such a method, since each case differed according to the individual. The physician nonetheless was expected to be attentive towards such matters, and Rufus offered the cautionary tale of Myron the Ephesian, a wrestler whose dream of sleeping in a black marsh of fresh water was ignored by his trainer. After relating his dream, Myron continued to exercise, became short of breath, lost control of his limbs and voice, and soon died. Had the trainer instead ordered a massive evacuation of blood, Rufus suggested, Myron might have lived.90 Galen similarly enjoined physicians to assay their patients’ dreams

The Persistence of Privacy 41

carefully and incorporate them into an evaluation of their habits as a whole: ‘One must pay particular attention both to the time the dream occurred and to any nourishment that the dreamer has consumed,’ he argued.91 Dreaming of being covered in snow, for example, could have various meanings, depending upon the patient’s health and temperament. Such a dream could easily be attributed to shivering in a chill during the night. But, if such feverish symptoms had passed, it more likely indicated the true state of the dreamer’s body as one dominated by cold humors, particularly if the dream had not followed the ingestion of phlegm-producing foods. All the elements of Aristotle’s natural history of dreaming were here, but differently arranged. The interplay of eating, sleeping, and dreaming bespoke less of the universal nature of the soul than of the unique balance of the individual patient attended by the learned physician. The use of dreams as a window through which experts could perceive the nature of the private self was not unique to Hippocratic physicians of the second century. This period witnessed a revival of Hellenic culture, particularly in the Greek cities of Asia Minor, which ultimately took root in the very heart of the Roman empire following the rule of Hadrian (117–38 A.D.). Perceptions of the value of dreams and their relationship to the individual had another Greek source in the form of Galen’s contemporary Artemidorus of Daldis. The five books of Artemidorus’ Oneirocritica soon emerged as a definitive source for learned dream interpreters who were, like the Hippocratic physicians, struggling to distinguish themselves from their illiterate counterparts who plied their trades in the marketplace. In the Hippocratic tradition of Airs, Water, Places, the Oneirocritica contained numerous references to the deeply contextual nature of dreams and charged itinerant dream interpreters to pay careful attention to the particular environmental, social, and physiological circumstances of their clients. Differences in custom, Artemidorus argued, rendered general claims about any particular dream images problematic. Tattoos, for example, were indicative of high social status among the Thracians, while, farther north, the Getae of the lower Danube tattooed only their slaves.92 The Mossynes in the territory of Pontus were said to have ‘sexual intercourse in public and mingle with their wives just as dogs do, whereas in the eyes of other men, this behavior is considered to be shameful.’93 Such cultural variability had dramatic repercussions for how a dream could be interpreted, as did the disposition of the individual dreamer. Persons of high moral quality regularly experienced

42

The Sleep of Others

oneiros, or meaningful dreams, while those of base character would dream only enhypnion, which arose from ‘irrational desire, an extraordinary fear, or from a surfeit or lack of food.’94 Social status was equally important. Only kings, magistrates, or the nobility could have dreams of ‘cosmic’ or ‘public’ significance. It defied reason to believe that such dreams could issue from insignificant people.95 Where Aristotle had used similar observations to conclude that dream interpretation was in principle impossible, Artemidorus used them to delimit further the dream interpreter’s expertise. The adept interpreter not only had to possess a deep knowledge of dream symbols – he also had to know his client well enough to apply this knowledge properly. Dreams were thus a natural crossroads where dream interpreters and physicians would meet, both in cooperation and in competition. The distance between medical prognosis and predicting the future was not, after all, particularly great in second-century Rome. They shared a common source: the need (perceived by those who could afford to purchase it) of having learned interpretation of one’s private nature by ex-perts. As we have already seen, the very practice of finding out about a patient’s sleep was implicit in the medical use of dreams as diagnostic signs. But sleep was a regulatory force in its own right, and Galen repeatedly emphasized the need for a proper balance between the time spent in sleep and wakefulness. His commentary on Epidemics VI, for example, encouraged physicians to make particularly careful inquiries into their patient’s habits in this regard.96 The distinction between sleep as a regulatory force in health and sleep as the context in which diagnostic signs were produced was thus blurred. As Galenism developed over the following fifteen centuries, however, these two aspects of sleep began to separate. As the Christian tradition developed a concept of an abstract, immortal, and unified soul as the locus of human morality, dreams emerged as potential signifiers of the soul’s status. Sleep, on the other hand, remained well within the physician’s purview but, as a strictly hygienic principle, became estranged from dreaming. Although the designation of sleep as one of the ‘six non-naturals’ has traditionally been attributed to Galen, it would be more accurate to associate the term ‘non-naturals’ with the later Alexandrian, Arabic, and Christian translations and commentaries that formed the basis of Galenism for the next fifteen centuries.97 Galen used the term ‘non-natural’ to describe a group of things that could affect the pulse, but the redaction and codification of his work ultimately turned this minor aside into

The Persistence of Privacy 43

a crucible from which all Galenic diagnosis and therapeutics would emerge. Texts such as the Medical Questions and Answers of Hunain ibn Ishaq (also known as Joannitius, 809–73) and the Pantegni of Haly Abbas (d. 944), formalized Galenic medical theory around the ‘naturals,’ ‘the contra-naturals,’ and the ‘non-naturals.’ Naturals and contra-naturals were innate factors essential to health and disease, respectively. The six non-naturals – the air and environment, food and drink, sleep and wakefulness, motion and rest, evacuation and repletion, and the passions of the soul – on the other hand, were factors that altered the health of the body through their quantitative and qualitative management. As aspects of health that could be altered through regimen, the ‘six things’ were an integral feature of Galenic medical practice, while an understanding of the naturals and the contra-naturals were more important for medical theory. Thus, while Galenism began to break down at a theoretical level with the rise of normal anatomy in the sixteenth and seventeenth centuries, the regulation of health through manipulation of the six non-naturals continued as part of medical therapeutics into the early nineteenth century. Morality and Scepticism Christian tradition, on the other hand, was beginning to reorient the significance of dreams away from the future and towards the present and even the past of the dreamer. This reorientation also entailed a shift away from the depiction of dreams as physiological or medical objects. Instead, their problematic status as moral objects became a mainstay of patristic scholars. Some of this same territory had already been opened up by Plato. He had argued that moral responsibility could be delimited by examining the correlation between dream images and a balance of internal forces that affected everyone. Was the dreamer responsible for his dreams of gluttony, incest, and murder? In the Republic, Plato had suggested that such images arose in sleep because, in this state, the rational soul slept while the desiderative soul, which governed the appetite, continued unrestrained.98 Generally speaking, dreams were ‘lawless’ and thus the lowest form of perception. It is probably for this reason that dreams received so little comment in Plato’s cosmology, as laid out in the Timaeus. When approached from an ethical perspective, however, dreams could be indicative of moral temperament. Immoderate, wicked, and foolish people saw gross distortions and immoral acts in

44

The Sleep of Others

their dreams, while temperate and wise individuals could have true dreams that accurately represented the past, present, and future.99 But Plato did not suggest that the dreamer was culpable for acts depicted in a dream as though they had actually been committed; rather, individuals were responsible for what they saw in their dreams only insofar as they were capable of assuming the more temperate course of bringing their soul under the control of reason. Galen transformed Plato’s theory of the tripartite division of the soul and its attendant localizations (reason in the brain, energy in the heart, and appetite in the belly) into physiological functions dominated by a specific organ system.100 But Galen did not, in the process, elucidate any clear notion of what a disease of the soul might look like, or what role dreams might play in diagnosing such an illness. These things were, as far as Galen was concerned, largely beyond the grasp of the physician’s expertise.101 Galen did not practise some variant of modern-day psychotherapy. As we have already seen, his position on dreams was ambivalent, a stance that suited the physician’s somewhat tenuous social situation and hesitant cooperation with the healers of the temple. For the Galenic physician, dreams could be signs of disease in a patient but they could equally be taken as signs from the gods regarding the physician’s own career. The problem of whether or not the patient was in some sense responsible for the content of his or her dreams was thus rendered irrelevant. Dreams could indicate the nature of the patient’s temperament, but this knowledge merely enhanced the reliability of the physician’s diagnosis: the temperament was not, in itself, the disease, nor was the discussion of the dreams with the patient any part of a cure. Patristic scholars had different concerns from physicians, and they thus treated dreams in a quite different manner. Their reconstruction of dreams as primarily moral, rather than medical, objects reflected the Christian transformation of religious healing in the later Roman empire more generally. Early Christianity was not necessarily inimical to medical practice grounded in natural philosophy. There are several examples of physicians who were also deacons or priests, and, despite the fracturing of the empire, the cultivation of learned medicine continued apace during the fourth and fifth centuries along with theology. But, while the early church fathers were introducing new traditions and debates about the nature of the soul, Christian medical writers were innovative only insofar as they successfully reduced Galen’s work into a learned code to be systematically digested by the educated physician.102 This focused medical attention on the non-naturals, and sleep hygiene eventually

The Persistence of Privacy 45

flourished as a result. The same process, however, transferred any practical interest in dreaming from the physicians into the hands of theologians, who were frequently hostile to pagan healing practices. By the end of the fifth century, the major urban Asclepiads in Aegae, Epidarus, Pergamum, and Athens had been razed, and a cult of Christian saints flourished in their place.103 This was not a simple process of substitution, since the function of learned medicine differed substantially in the two traditions. As already noted, temple medicine entailed an intimate, if sometimes antagonistic, relationship with the works of physicians. Christian healing, on the other hand, was typically depicted as a matter of simple faith, exemplified in the lives of the saints.104 The theological isolation of dreams from healing, be it by a physician or by a priest, was merely an extension of these larger processes. The life and work of Augustine of Hippo (354–430) indicated the degree to which dreams had emerged as an index to moral state of the soul. Augustine’s account of dreaming, especially in his Confessions, proffered dreams as diagnostic tools that could indicate the state of grace in which the dreamer found himself. Understanding one’s dreams could thus suggest a proper course of moral action for a Christian.105 Initially, however, Augustine had treated dreaming as a metaphysical problem, drawing upon Neo-Platonist metaphysics to refute the philosophical scepticism to which he was attracted in his youth.106 In an argument first presented in Against the Academicians (386) and revisited in On the Trinity (399–419), he confronted the suggestion that no knowledge was certain because the possibility always existed that one was sleeping and being continually deceived by a dream. But the very act of sleeping, countered Augustine, was essential to life. Thus, even if one was dreaming, that would imply that one was both alive and sleeping. Thus, there was certain knowledge – the knowledge that I live – and with this knowledge, universal scepticism could be rejected.107 In Augustine’s later work, the problem of dreaming took on an intensely personal significance. As we have already seen in Galen, Hellenistic culture cultivated a close linkage between one’s autobiography and one’s dreams. Conversion dreams were one aspect of this linkage, and early Christians readily adopted this tradition.108 In the early third century, Tertullian spoke of an ‘invocatory dream’ (somnia uocatoria) to describe this experience, and dreams figured prominently in the conversions of several saints from the same period. Augustine’s own conversion, as described in the opening books of Confessions (397–400), was foretold by a dream his mother had shortly after hearing him utter

46

The Sleep of Others

his heretical views.109 Acknowledging this as a divinely sent dream, Augustine shifted to more mundane oneiric experiences and their relationship to God’s activities in the world. Dreams provided Augustine with an opportunity to raise the problem of moral responsibility in a radical fashion. God, he argued, granted the faithful the strength to resist the temptations of the flesh while awake. For Augustine, these temptations included images of his previous licentiousness drawn from his memory.110 When these images ‘attacked’ him while he was awake, Augustine successfully resisted their charms. While asleep, however, they were able to arouse pleasure in him and even ‘elicit consent,’ thus making dreams ‘very like the actual act.’ Augustine’s extreme intentionalist ethics made such a situation intolerable. Even if the immoral deed remained uncommitted, he still entertained the idea, and took pleasure in it. He was thus morally responsible for the carnal act, just as though he had committed it while awake. Significantly, this was the only example of such culpability discussed in the Confessions – he does not mention, as Plato did, other immoral or unethical acts entertained in dreams. Perhaps his focus on the sexual nature of dreaming was due to the fact that erections, ejaculations, and orgasm are all possible in sleep as in wakefulness, thus making erotic dreams ‘very like the actual act’ in a way unlike dreams of murder. In any event, Augustine used these dreams as a dramatic example of his argument that God’s grace was a necessary condition for human ethical conduct. Helpless in his dreams, Augustine still sinned when he consented to indecent sexual acts in his sleep. Against the ethical perfectionism advocated by the British monk Pelagius, Augustine insisted that ‘ought’ did not always imply ‘can.’ Human will was not enough: only God’s continual grace could ultimately keep Christians from sin.111 Augustine’s work appears at the early stages of a period during which orthodox thought on dreaming was marked by a double movement. On the one hand, dreams were generalized. In contrast to the pagan traditions of dream interpretation, Christian theologians accepted the notion that God could send dreams to anyone, regardless of their social status.112 Hagiographies came to be filled with such examples. On the other hand, divination by any means (including that by dreams) was officially condemned by the Council of Ancyra in A.D. 314.113 The faithful were left to their own devices to interpret their dreams without the benefit of the expertise that had developed in this domain across centuries and across civilizations. This double movement, which situated the dreamer not only between

The Persistence of Privacy 47

heresy and orthodoxy but between the immanent body and the transcendental divinity, persisted well into the twelfth century.114 Its strategies paralleled that of learned medicine, which had so effectively constructed its authority by celebrating the individuality of its patrons. This similarity is more than a passing coincidence. It was written directly into the Confessions, in which Augustine invoked dreams as diagnostic resources even as he repeatedly referred to God as a ‘physician’ who would ‘cure all the sicknesses of my soul and, by a more abundant outflow of your grace ... extinguish the lascivious impulses of my sleep.’115 This depiction of God and Christ as physicians was becoming a commonplace in the fourth and fifth centuries.116 Beneath the centrality of this metaphor lay the church’s deep-seated ambivalence towards medical practice. Christian theologians frequently drew an intimate association between sin and illness, and some even admonished followers for accepting aspects of worldly medical practice against the will of God.117 Yet many of the early church fathers, while acknowledging that illness could be part of God’s judgment or his divine plan, described medicine as a gift from God that Christians could pursue in earnest, provided they did not subordinate their faith to it.118 In any event, the miraculous cures recorded in the Bible and in the lives of the saints may have well have created a culture of supplicants who turned to religion, not physicians, for healing. Yet the Christian concept of a unified, personal, and eternal soul implied that dreams no longer spoke the language of the body – they were effectively divorced from sleep. This early form of dualism reformulated the autorepresentational capacity of dreams. Dreams could still represent the hidden states of the self to the self – they were still cryptic and transparent in turns. The ninthcentury revival of dream interpretation epitomized by the appearance of Achmet’s Oneirocriticon reflected the new individualism of dreaming that was grounded in (or confounded by) private faith, not the idiosyncrasies of health.119 The self reflected by the Christian tradition was now immortal, immaterial, and in perpetual peril; it was no longer an amalgam of the venal and the venerable. The torment visited upon Augustine through his licentious dreams did not, after all, so much represent the arousal of his body as they did the revitalization of his memory. The Public Functions of Private Experience Sleep and dreams continued to inhabit separate worlds in the early modern era. Philosophers and theologians valorized or debased dream-

48

The Sleep of Others

ing in turns, while physicians and patients continued to develop sleep hygiene as a cornerstone of health. These worlds rarely intersected, but they held in common the privilege of privacy. Sleeping and dreaming retained their intimate natures. They were biographical paths of selfobservation, self-discovery, and self-improvement, and could not be divorced from individual experience. They were not part of systematic, objective investigation. The significance of such experience first began to shift in the case of dreaming. Augustine’s approach had, in many ways, cloaked Asclepian dreams in a Christian phenomenology. Dreams were a personal link to a God who enabled the maintenance of health, be it physical or spiritual. By the seventeenth century, this domain had expanded to encompass questions about social order. If the stability of society was based upon the premise that individuals could and did exercise their reason in their public intercourse, what was to be made of these nightly transgressions of cognitive order? Augustine had not needed to raise this question. For him, human society existed not by virtue of reason but through the continual outpouring of God’s grace at the level of the individual sinner. The notion of God’s continual intervention in human and natural affairs ran like a trope throughout the medieval and Renaissance periods, just as his withdrawal from the world was thought to generate disorder in the form of warfare, disease, or madness. Monsters, too, signified the abeyance of grace. An anencephalous birth, a child with the face of a frog, a sea creature with the appearance of a bishop, flying fish, and whales could all be explained by reference to diverse causes. Too much seed could cause a child to be born with two heads, while ‘sodomists and atheists’ who mated with animals produced hideous hybrid beings.120 But, in every instance, the study of such monsters ultimately testified to the caprice (and therefore the power) of the Divine Craftsman, whose work of creation was never-ending. Cosmology, teratology, anatomy, and natural history all followed this same pattern of desperately combing through the natural world in search of meaningful signs of a harmonious, symmetrical, and ultimately poetic cosmic order.121 Dreams, of course, had long existed as texts that could be read for meanings that gestured towards a divine order. The inscribed tablets at the Asclepian temple at Epidaurus are almost over-determined examples of this. Dreams were equally powerful literary devices that invoked a universal experience for communicating meaningful information about the world, as was the case with Cicero’s divine cosmology, which he described through the medium of a dream in his Republic. Sceptics

The Persistence of Privacy 49

and the faithful alike could partake in the dream as a textual resource. This pluralism of use continued into the seventeenth century, even as the dream’s capacity to signify began to change. The defence of the concept of a sun-centred universe by Copernicans and the subsequent reinvention of cosmological order around the concept of ‘force’ marked the beginning of the end of divine intervention as the metaphysical foundation of natural order. Cartesian epistemology, which emphasized both the subjective foundations of knowledge and the mechanical structure of the universe, offered a similar trajectory for natural philosophy. The dream appeared as a key player in both scenes of inquiry. The Dream of Johannes Kepler is not as well known as the dreams of René Descartes, but it should be. This work, published posthumously in 1634 by Kepler’s son, Ludovic, exemplifies the imaginative appeal of Copernicanism in the first decades of the seventeenth century. Whereas Galileo’s telescopic observations prompted a dramatic change in astronomical practice that suggested, but did not require, fidelity to a heliocentric hypothesis, Kepler’s Dream could capture its reader’s imagination whole because it was expressly a work of fiction.122 Moreover, Kepler set the narrative at a double remove from that of his reader: the tale is opened by a narrator describing a dream in which he read a book, which is itself narrated by yet another character, Duracotus. Duracotus’s story describes, among other things, an encounter with a demon who offers substantial information about the moon, the life it sustains, and the means whereby one can travel between it and the Earth. The demon’s account effectively transposed representations of an Earthcentred universe to the moon. The moon’s inhabitants were incapable of sensing their planet’s motion, yet they had constructed a cosmology that, though successful and accurate, was nonetheless false from the perspective of anyone on Earth who had seen the moon rise and set. One might suspect that Kepler chose this genre for its easy ambiguity in an age of religious intolerance, but it seems likely that the dream allegory or fable had more positive aspects to recommend it.123 Kepler had begun drafts of the story as early as 1610, and he died four years before Galileo’s conviction for suspicion of heresy in 1634, so prudence is unlikely as a motivating factor. Besides, Kepler fully recognized that fiction, like fact, could provoke harsh responses from authorities. Indeed, those same authorities felt that Kepler’s own mother so closely resembled Fiolxhilde, Duracotus’s mother who invoked the demon, that the last five years of her life were spent on trial for sorcery. What the dream genre did offer Kepler was the perspective of a mov-

50

The Sleep of Others

ing observational platform that utterly disrupted knowledge in a way that was simply not possible otherwise. After Galileo published The Starry Messenger in 1610, visual accounts of the moon began more and more to appropriate the ‘rhetoric of reality’ that had been an integral part of anatomical illustration since the middle of the sixteenth century.124 In anatomy, illustrations had come to act as surrogates for eyewitness accounts of dissections, and the same was true in telescopic astronomy, but only in a narrow sense. For the most part, astronomy continued to be a largely mathematical pursuit based upon the ‘rhetoric of precision,’ where the work of instruments figured more than any artistic rendering of celestial objects. Kepler’s use of the dream genre overcame these limitations by directly appealing to his reader’s experience of the unreal that was private (like looking through a telescope) yet universally accessible (like watching a public anatomy). Just as a dreamer in the Renaissance could witness events that might be superficially false but still hide certain truths about the world, so could a reader of dream fiction weave fact around illusion. The ability of dream narratives to tell a coherent story without a fixed point of view reflects precisely what Kepler wanted to communicate about the Copernican hypothesis. The central message of The Dream – that the true structure of the world could be perceived only through interpretation, not experience – was located in its formal aspects at least as much as it was in its content. The very appearance of the text, for example, placed enormous significance upon interpretation: the length of Kepler’s commentary on his own story, expressed in his footnotes, frequently exceeded that of the text itself. The notes guide the reader through Kepler’s own interpretation of his fable, even as they showcase the novelty of his astronomy. The most prominent feature of the latter was Kepler’s comparison between magnetic effects observed on Earth and the attractive force that keeps bodies on the surface of the Earth, causes tides, and renders the planets’ orbits elliptical.125 In the text of the story, the demon informs his interlocutors that it is this magnetic force that characterizes almost every aspect of the journey between Earth and the moon. The first part of the journey is the most difficult, but the force weakens the farther one goes from the Earth’s surface.126 Indeed, humans who undertake the voyage (with help from the demons) must be put to sleep with opiates and narcotics, so that their limbs can be arranged in such a way that the initial shock required to force them away from the Earth is transmitted evenly over their bodies.127 Once a sufficient distance from the Earth has been achieved, the trip becomes much

The Persistence of Privacy 51

easier, and the demons find they need little more than their own willpower to transport their slumbering human cargo.128 Notes to this passage repeatedly refer the reader to the analogy between magnetic and gravitational force, with the example of the tides as the strongest argument in its favour.129 The journey is an obvious allegory for the paradoxical amount of effort required to conceive of a heliocentric universe. The journey is extremely difficult, as evidenced by the efforts required of the demons to transport their human cargo, yet it is also such a simple matter as to afford its participants the possibility to make the journey in their sleep. By utilizing the dream genre to make his case for a new cosmology, Kepler thus emphasized the dream’s epistemological virtues. The liberation of the dreamer’s imagination was no longer a ready source of caustic scepticism, idiosyncratic diagnosis, or self-flagellation. The private experience of dreaming now had a positive public function: harnessing the very irrationality of dreaming in the name of reason suggested that God did not need to intervene continually in human affairs to maintain order in the world. The implication was that, if sufficient attention were paid to the formal qualities of dreams and less to their actual content, the disorderly idiosyncrasies of dreaming would resolve into universal harmonies. Like Kepler, René Descartes (1596–1650) carefully integrated dreaming at several key junctures in his philosophy. On the night of 10 November 1619, Descartes, full of enthusiasm for the new method of philosophy he had just discovered, had three dreams, which he immediately wrote down in a notebook upon awakening. This notebook later provided Adrien Baillet with the requisite material to canonize Descartes as a philosophical saint in his 1692 biography. The notebook has since been lost, but no matter: Descartes would surely have agreed with how his dreams were handled by Baillet, since both Augustine and the neo-Platonist philosopher Proclus offered suitable precedents for grounding one’s philosophy in dream experiences.130 When we examine Descartes’s own dreams and the place of dreaming in his epistemology, we need to remember that we are dealing with problems of narrative and genre no less than when we look at a fictional work by Kepler.131 Naturally, the first interpretation of these dreams was provided by Descartes himself, and his analysis fell within the well-established genre of ‘conversion’ dreams leading to dramatic changes in one’s life. His dreams of being blown about by a strong wind as he walked down the

52

The Sleep of Others

street, of thunder, and of reading a dictionary and a collection of poems all led him to believe that he must pray for enlightenment from God, abandon his training as a lawyer, and dedicate his life to the search for truth. Recent debate has focused on the origins of the ‘evil genius’ identified as the wind that slammed Descartes against the wall of a chapel in his dream. But did Descartes actually use the term ‘malo Spiritu’ to describe this wind, or was this a latter addition by Baillet? The ‘evil genius’ traditionally associated with Cartesian rationalism is the one that infects his mind with a radical scepticism in the Meditations, making him doubt the existence of the world, of others, and even of having a body. Descartes’s famous remedy for such scepticism came in the form of his recognition that, even as he was nearly overwhelmed by doubt, he was unable to doubt that there was an ‘I’ that doubted. This certain knowledge then became the bedrock of his method of philosophy. The cogito was a rational proof accessible to all, implanted in the human mind by God. But if its origins were in Descartes’s dream, would this not make a mockery of rationalism by locating the font of reason in the irrational?132 Ian Hacking has suggested that such paradoxical qualities are part and parcel of how the genres in which we dream are reformulated differently according to ‘place,’ in culture, time, and space. In the case of Descartes, the place of dreams was at one and the same time a material and temporal site (the overheated room in which he dreamt on that November night) and a transcendental, universal path that could lead to reason even as it obliterated its own origins.133 Certainly, the paradoxical qualities of dreaming were exploited by Kepler’s fiction, which took a view from nowhere as the starting point for a true description of the universe. But Descartes’s use of dreaming was somewhat different. For Descartes, as for Kepler, dreams served as an interface between author and reader. But for Descartes, dreams were equally a means of self-discovery. Dreams were a guide to personal experience. They generated enlightenment. Consider, as an example, the dream described in the First Meditation: ‘As though I were not a man who habitually sleeps at night and has the same impressions (or even wilder ones) in sleep as these men do when awake! How often, in the still of the night, I have the familiar conviction that I am here, wearing a cloak, sitting by the fire – when really I am undressed and lying in bed!’134 As Descartes acknowledged in the text, this was a generic dream, one he claimed to have had many times before. Whether Descartes ever actually had this dream on any particular occasion was irrelevant: the point was that he could have dreamed

The Persistence of Privacy 53

such a dream. The mundane nature of the dream further emphasized its universal qualities. Anyone could experience it, or something like it. From there, Descartes proceeded to introduce his method of radical doubt, and the rest of the Meditations continued apace. By the end of the Sixth Meditation, the reader found Descartes happily concluding that he could now dismiss all doubts about being able to distinguish his waking from his sleeping life, because events in the former were connected by memory in a way that events in his dreams never were. Led by Descartes’s arguments that certain knowledge was to be had when the evidence of the senses was joined to reason, his readers could now join him in this cheery dismissal of scepticism. Descartes’s use of dreams in the Meditations was too personal to be allegorical, yet it was too generic to be confessional. By paradoxically implanting dreaming at the very heart of an autobiographical route to reason, Descartes invented a new way of thinking about dreams that capitalized on their idiosyncrasies by converting them into a universal sign of an absent cause. Dreams were idiosyncratic precisely because they could thereby lead everyone to the same conclusions about the nature of reason. The modern alternative to this position, one that suggested a rather opposite public role for dreams, was staked out by Thomas Hobbes (1588–1679) in his objections to Descartes, first published with the Meditations in 1641. Ignoring the formal novelty of Descartes’s work, Hobbes criticized Descartes for raising the hoary old problem of dream scepticism without contributing anything new to it. Descartes admitted his lack of originality on this point but pleaded that he ‘could no more leave them out than a medical writer could leave out the description of a disease for which he wanted to explain the method of treatment.’135 Hobbes further attacked Descartes’s optimistic conclusions about how easy it was to distinguish dreams from wakeful perception on the basis of memory and reason. After all, argued Hobbes, could someone not dream that their dream events followed a smooth continuum, just as events did in waking life? Descartes’s response – that such a delusion could happen only in sleep and would be recognized as a mistake upon awakening – merely presupposed, insisted Hobbes, the very difference between the mind’s activity in sleep and wakefulness that he was attempting to demonstrate. Hobbes made his own claims about dreaming explicit in Leviathan (1651). Medical analysis provided Hobbes with more than a rhetorical model for framing the problems of dreaming; they gestured towards what a materialist account of dream images might actually look like, even as they offered Hobbes a means of diagnosing the source of the

54

The Sleep of Others

religious turmoil that was then undermining the social order of his native England. In sleep, argued Hobbes, the sense organs were ‘benumbed’ and unable to receive the external impressions from objects that caused sensation in waking life. When the inward parts of the body were ‘distempered,’ they kept the brain and the nerves in motion, producing the images of dreams.136 These images were specific to the distemper: ‘lying cold,’ for example, engendered fearful dreams, while heat in certain body parts conjured up images of an enemy, and in other parts brought up images of desire or kindness. Dreams were, in effect, ‘the reverse of our waking imaginations.’ In wakefulness, an angry reaction to a perception spread heat from the brain to certain parts of the body; in sleep, these same parts, if heated, communicated this motion to the brain, creating images of anger. Hobbes took for granted that his readers could, like himself, distinguish the images that appeared during wakefulness from those that appeared in dreams. Perhaps he despaired of ever being able to explain this difference, because he simply appropriated the same ‘coherence’ argument that Descartes had taken such pains to demonstrate: ‘For my part, when I consider that in dreams I do not often nor constantly think of the same persons, places, objects, and actions that I do waking, nor remember so long a train of coherent thoughts dreaming as at other times; and because waking I often observe the absurdity of dreams, but never dream of the absurdities of my waking thoughts – I am well satisfied that, being awake, I know I dream not, though when I dream I think myself awake.’137 In any event, Hobbes’s real concerns were elsewhere. While Hobbes himself claimed that he could, like his readers, easily distinguish between sleep and wakefulness, this distinction, he thought, might not reliably be made by everyone in every situation. Fear and a troubled conscience, for example, might produce insomnia, thus preventing a person from going through the typical ritual of removing one’s clothes in preparation for sleep. If this person then dozed off, perhaps while sitting up in a chair, he might later awaken and, not knowing he had slept, mistake his dreams for apparitions of spirits. This scenario, unlikely as it might seem, was, for Hobbes, the true source of the false and superstitious elements of religion. Those who aspired to power had taken advantage of a common confusion between visions and dreams to encourage belief in (among other things) witches, fairies, the magical power of crosses, holy water, exorcism, dream prognostications, and false prophecy. Hobbes was clear that

The Persistence of Privacy 55

there was no question as to whether or not God had the power to manipulate the world in such occult ways. But he was, argued Hobbes, no more likely to raise spirits or send divine dreams than he was to change the very course of nature. The widespread belief that worldly events were guided by malevolent or benign inventions from a spirit world was false, and such beliefs undermined civil society: ‘If this superstitious fear of spirits were taken away, and with it prognostics from dreams, false prophecies, and many other things depending thereon by which crafty, ambitious persons abuse the simple people, men would be much more fitted than they are for civil obedience.’138 In contrast to Descartes’s use of dreams as part of a universal foundation of all knowledge, Hobbes depicted dreams as potentially corrupting obedience to a sovereign power, which was a key component of his recipe for social stability, be it religious, monarchical, or parliamentarian. Confounding dreams for true visions tended to unravel the very foundations of social order, because it encouraged diverse, spontaneous, and idiosyncratic beliefs that engendered religious strife. Intellectual, political, and religious authorities neglected educating the public as to the true nature of dreams at their own peril. Hobbes was certainly not the first to suggest that dreams and other seemingly spontaneous acts of the imagination were responsible for heresy and superstition. His contemporary, Robert Burton (1577–1640), had made similar claims in his Anatomy of Melancholy, first published in 1621.139 But Burton’s popular treatise spoke of the elements that disrupted one’s personal and emotional life, not the transformation of political and social order. Reflection, not revolution, was his central concern. Hobbes, of course, had good reason to be worried about the latter. Leviathan was published just as ten years’ of civil unrest and open warfare were concluding, with the victory of Cromwell’s parliamentarians over the royalists supporting Charles I. Both sides had appealed to providence and personal revelation to defend their positions, so Hobbes’s emphasis on how dreams could have a malign influence on the course of events was particularly timely.140 Hobbes seemed to find some sympathy on both sides, as he attempted to steer a pragmatic course between religious belief and the governance of the state. God was ultimately unknowable, he argued, so religion was best administered in uniform doses by the governing power. The enthusiasm of radical Puritanism under Cromwell’s Protectorate, he argued, was no better than the mysticism and irrationalism presupposed by the sacraments and prayers of the Roman Church or the Church of England. All led down the path of

56

The Sleep of Others

religious disagreement and dispute that ended in the dissolution of order, as did any account that suggested dreams were anything more than images raised by ‘distempers’ of the body. Those who propagated the one set of beliefs would naturally manipulate the other to their own ends: ‘For when Christian men, take not their Christian sovereign, for God’s prophet; they must either take their own dreams, for the prophecy they mean to be governed by, and the tumour of their own hearts for the Spirit of God; or they must suffer themselves to be led by some strange prince; or by some of their fellow-subjects, that can bewitch them, by slander of the government, into rebellion, without other miracle to confirm their calling, than sometimes an extraordinary success and impunity; and by this means destroying all laws, both divine and human, reducing all order, government, and society, to the first chaos of violence and civil war.’141 For mid-seventeenth-century Puritans, God’s providence was a reality that permeated every aspect of life. Illnesses, weather patterns, and business dealings were as likely to be attributed to divine intervention as were the more dramatic fare of political intrigue, rebellions, or wars. It is hardly surprising, then, that dreams became caught up in this net of events guided by God’s unwavering hand. Hobbes, in contrast, characterized dreaming as a mechanical, law-like, and therefore predictable process. He felt this was a reasonable premise that led to respectable conclusions about the nature of mind, political order, and the divine. It was not something that called for empirical investigation, and Hobbes never proposed such a program of research into dreaming. Of course, Hobbes was no experimentalist, and he later took issue with those who insisted that experimentalism provided the proper foundations of natural philosophy.142 In any event, dreaming and sleeping were never proposed as problems that could be resolved by experiment. It was as though the essence of sleeping and dreaming remained stubbornly personal, and beyond the pale of measurement, intervention, and instrumental demonstration that quickly began to characterize the public worlds of blood, air, light, and gravitational force by the end of the seventeenth century. Taking Measure of the Soul Yet Hobbes’s pragmatic analysis of religious belief as a tool of the state did have a counterpart, and an empirical one at that. Dreams, according to some Puritan writers, could be used to better comprehend one’s ever-

The Persistence of Privacy 57

changing relationship with God. This was a diagnosis of the soul, patterned after the ancient use of dreams for diagnosing the body. Where Hobbes had seen in the medical use of dreams an argument that the senses were the origin of all dream-images, Philip Goodwin (d. 1699), a Hertfordshire preacher, saw an analogy: ‘Those famous Physicians, as Galen, Hypocrates, they to their high praise, proved and improved their excellent knowledge in Dreames, thereby discerning the Symptomes of severall diseases, and so perceived what proper Meanes to propose to their Patients, for their more quick recovery. Aquinas and others, give out evident Instances of such admirable advantages, by which was raised the fame of Physicians in former times; And if it were laudable in them, to look into Dreames to learn out the state of mens bodies, may it not be commendable in others, thereby to discover the case of mens souls?’143 Goodwin, who had sided with parliament in the Civil War, had been appointed a ‘trier’ in Hertfordshire, meaning that he had the power to regulate who was able to preach in public in that region. Following the Restoration of 1660, he was ejected for non-conformity, but he eventually conformed and continued to preach until his death. His treatise on dreams was published in 1658, just as the Protectorate was coming to a close with Cromwell’s death and the interregnum was fast dissolving into a millenarian anarchy populated by radical Levellers, Anabaptists, and Fifth Monarchists. Goodwin’s effort to discipline dreams by encouraging individuals to attend to their own dream life is testament to the exceptional circumstances surrounding the English soul at the close of the tumultuous 1650s. In one sense, Goodwin’s book was a veritable inversion of all of Hobbes’s misgivings about dreams. For Goodwin, the material causes of dreams were incidental to their content, which came directly from either God or the devil. Attempting to understand one’s dreams was not a point of idle mischief, but a solemn Christian duty, duly recognized in Old Testament prophecy and Hellenic philosophy alike. ‘Man as an Animal creature in Dreaming sleepes: And man as a Rational creature in sleeping Dreames,’ aphorized Goodwin, implying that there was no contradiction between the exercise of reason and the acceptance of dreams as divine messages.144 While this was clearly at odds with Hobbes’s summary dismissal of dreams as ill-founded prophecy, Goodwin nonetheless acknowledged that false prophecy, inspired by the devil, could emerge: Manichees and Familists ‘from such Dreaming delusions first received their rotten opinions[.] [And, as for] those prodigious preachings and practices of the boisterous Anabaptists in Germany, their first original

58

The Sleep of Others

was drawn from deluding-Dreams.’145 He equally recognized the potential dreams had for fracturing social order: ‘Dreames declaring how God did advise them, counsell them, command them what wayes to forbear, and which wayes to bend themselves. So some have of late yeares deserted our publike Assemblies, into separated companies, as being warned of God in Dreames so to do.’146 Like Hobbes, Goodwin saw dreams as a potentially dangerous source of false belief. But Goodwin also felt that dreams were a useful venue of empirical activity. Like other natural phenomena, dreams could be observed; more important, they could also be manipulated. What the pious Christian needed to learn, according to Goodwin, was not the sensory origin of dream images but how to control the content of dreams in order to ensure that they were always good and pure. This involved a tempering of the soul, mostly through prayer but also by keeping in memory those good dreams that came from the grace of God.147 Dreams were not only an index to God’s grace. They were also part of the soul’s regimen – a form of divine self-reflection the improvement of which was its own end. By the close of the century, the use of dreams for self-diagnosis had been integrated into a new asceticism in the writings of Thomas Tryon (1634–1703). Tryon is best known as an outspoken vegetarian who rejected the use of animals for food on religious grounds in his popular The Way to Health, first published in 1682.148 A hatter by trade, Tryon underwent a radical conversion near the end of the 1650s. Records of similar transformations exist: Tryon’s fellow haberdasher, Roger Crab (c. 1621–80), for example, had equally promoted vegetarianism as both a statement against social inequality and as a means of self-purification in his The English Hermite, or, Wonder of This Age (1655). The eating of flesh, according to Crab, not only escalated one’s penchant for violence but also caused the price of corn to rise, thus leading to higher rents and increasing the misery of the poor. Crab, who gave away all his property, walked around in a suit of sackcloth and enjoyed a reputation as a gifted herbalist. He claimed to have hundreds of patients routinely crowding his cottage at Ickenham, near Uxbridge. How many of them eventually subsisted only on dock leaves and grass, as Crab himself claimed to have done, is unclear. But what is certain is that Crab’s disciples found the political possibilities of such ascetic self-mastery less palatable after the Restoration. Over the next half-century, vegetarianism, like dietary regimens in general, reoriented itself away from the social body and towards individual health as its end. Dreams were readily incorporated into such a schema. Tryon’s A Trea-

The Persistence of Privacy 59

tise of Dreams and Visions, first published in 1689, did for dreaming what the efforts of Crab and others had done for diet. Deeply influenced by the writings of the German mystic, Jacob Boehme (1575–1624), Tryon argued that dreams had once been the most accessible source of individual illumination but, with the advent of modern culture, had become corrupted through intemperance and immorality. Goodwin’s millenarianism mingled with Hellenic naturalism on almost every page of Tryon’s Treatise. On the one hand, Tryon agreed that sleep was unquestionably physiological, since it was caused by vapours arising from the concoction of food. But this ‘rude and blind age’ had failed to recognize the supernatural origins of dreams, something that had been readily acknowledged by the ancients.149 The soul communicated with both good and evil angels in sleep, as well as with foolish and conniving ‘elemental spirits’ that inhabited woods, streams, and mountains, and the clarity or deceitfulness of dreams varied accordingly. The ability to perceive this difference and thereby to incorporate the self-knowledge gleaned into one’s own life depended, thought Tryon, upon the temperance of the dreamer. Error, false belief, and, ultimately, civil strife did not originate in the tendency to confound dreams with divine visions, as Hobbes seemed to have suggested. Rather, it was the luxury and gluttony that pervaded his countrymen’s lifestyles that was the true source of the social unrest that had pervaded England for much of the seventeenth century. Tryon thus insisted that personal reformation must precede the dissemination of any objective knowledge about dreaming. His subjectivist position was in line with that of the hermetic alchemists and Paracelsians, which had sometimes been interpreted as inimical to experimentalist philosophy. By Tryon’s puritanical yardstick, true natural philosophy (that is to say, proper method) could arise only as a result of self-inquiry: ‘It is highly convenient for everyone that applyes his Thoughts to this Science of Dreams, or indeed to any sort of Knowledge that is truly Philosophical and Divine, to be serious and sober, and to learn first the mysteries of his own World, before he lets his Eyes and Imagination ramble into, or gaze after, the Wonders or Vanities of the great External World, being certain of this, that if he do not in some competant measure know himself, and continually endeavor the Advancement and Encrease of that Knowledge, he will never come to understand any thing without himself.’150 As we have seen, personal regimens such as vegetarianism had already been moulded into a form of political protest by the 1650s. Tryon extended the range of political actions by incorporating the issue of sla-

60

The Sleep of Others

very into his doctrines, while at the same time transforming such regimens into epistemological devices.151 He repeatedly called upon his readers to give up the eating of flesh as a way of expressing a ‘purity of intention’ that would lead to sure knowledge, first of the self through dreams and, later, of the external world.152 As evidence that such progress was possible, Tryon cited the example of Daniel, the greatest dreamer of the Old Testament, who had dined without violence or oppression by refusing to eat unclean meat. Restoring a similar form of asceticism through vegetarianism could, Tryon hoped, likewise bring enlightenment to the modern world. Medical Mechanism and the Revival of Sleep A chasm soon emerged between medical orthodoxy and the mysticism and millenarianism encapsulated by Tryon’s austere dietary regimens. Personal enlightenment and dreaming did not appear to be a reliable foundation for building a secure knowledge of man, despite radical claims to the contrary. Undoubtedly, the growing influence of mechanism among British physicians during the first decades of the eighteenth century was in part responsible for this change.153 The extent to which medical treatises of the period continued to be influenced by hermeticism continues to be a point of debate among historians.154 But one point is abundantly clear: the rise of mechanism as medical orthodoxy brought with it a decline in interest in the diagnostic dream over the course of the eighteenth century. Like Tryon, medical men of this period emphasized diet and regimen as a vital and even primary way of restoring and maintaining health. But eighteenth-century physicians tended to abandon arguments about the wider social significance of individual regimen and social health, at least insofar as dreaming was concerned. Instead, they turned to a non-natural – sleep – as one of several means of intervening to promote and improve health. If physicians raised the question of dreams at all, they tended to do so only in terms of how dreams might disrupt sleep. The absence of dreams in sleep, rather than the extent or nature of a dream’s verisimilitude, became the signifier of health. Following the publication of the Principia mathematica in 1687 and his subsequent dominance of the Royal Society, Isaac Newton’s stature grew to almost mythological proportions across Britain and Europe. Many medical thinkers attempted to follow suit and occupied themselves with a program to reduce all physiological questions to an analysis of the

The Persistence of Privacy 61

quantitative relationships between the body’s ‘solids’ and ‘fluids.’ Archibald Pitcairne (1652–1713), a prominent Edinburgh physician who had a brief stint as a professor of medicine at Leiden in 1692, was one of the most outspoken proponents of this program of ‘iatromathematics.’ His analysis of fever as a function of the circulation of blood, rather than the product of an internal ferment, won him a number of admirers, not the least of whom was George Cheyne (1671–1743), whose own 1702 treatise perpetuated the ‘fever disputes’ between Pitcairne and his detractors. Unlike Pitcairne, however, Cheyne was a talented medical popularizer, able to weave his own experiences of ill health into his medical philosophy.155 After arriving in London from Edinburgh in 1702, Cheyne began frequenting the lively circles of the local gentry. Celebrated for his good temper and quick wit, Cheyne soon picked up a large clientele and established a successful practice. He also developed an enormous bulk. He blamed the latter on his recently acquired intemperance, generated in a seemingly endless stream of coffee houses and taverns in search of new patrons. Cheyne recanted his vice but lost most of his patients and only some of his 32 stone (448 pounds) in the process. He plunged into a deep melancholy and did not recover until he abandoned London for the curative waters of Bath. At the height of his crisis, Cheyne penned his Essay of Health and Long Life, in which he laid out the regimens according to which disease could be cured and health maintained. The Essay, which first appeared in 1724, was a rather conventional ‘long-life’ manual that revisited the well-worn territory of the nonnaturals, but it nonetheless sold extremely well and made its author famous. What Cheyne offered his readers was the passion of a convert combined with the vogue for mechanistic explanations of the workings of the ‘animal oeconomy’ in the human body. Although Cheyne emphasized temperate drinking and a diet of vegetables and milk in the Essay, sleep had its place within this scheme. Overeating, or taking a meal just before going to bed, filled the guts with ‘Wind, Choler, or Superfluous Chyle,’ making sleep practically impossible because ‘sharp and crude Humours, twitching and twinging the nervous Fibres, and Coats of the Bowels, become like so many Needles and Pins, constantly running through them.’ The flow of this ‘unconcocted Chyle’ through the bowels and into the smaller vessels manifested itself in ‘Convulsions, Flatus, Night-Mares, and Oppressions of Spirits.’156 Worst off were those ‘hypochondriacal and hysterical People,’ who, ignoring the rhythms of night and day, ‘begin to live and breath, become cheerful and hungry,

62

The Sleep of Others

about ten, eleven, or twelve a clock at Night; eat a hearty, various, and luxurious Supper; drink a cheeruping Cup of the best, become as merry as Crickets, and long to sit up later.’ Eating habits could explain the disruptions of sleep, but the regular alteration between sleep and wakefulness was another matter. Cheyne treated sleep as a problem of matter in motion and, to this end, attempted to bring sleep into a larger system of a natural and social economy: ‘All animal Bodies, from an active and self-moving Principle within them, as well as from the Rubs of Bodies without them, are constantly throwing off some of their superfluous and decayed Parts; so that animal Bodies, are in a perpetual Flux. To restore this Decay and Wasting of animal Bodies, Nature had wisely made alternate Periods of Labour and Rest, Sleeping, and Watching, necessary to our Being.’157 Sleep was also linked to the rhythms of night and day by the dictates of health and labour as interpreted through the circulation of fluids in the form of perspiration. Respiration through the mouth and skin was facilitated during the day by two factors: the sun’s heat, which rarefied and thinned the earth’s vapours, lowering the pressure upon the body; and movement and labour, which encouraged breathing and a strong flow of sweat. When, at night, vapours condensed and sunk back to the earth, they became noxious and obstructive, preventing proper perspiration. Watching at night (that is to say, staying up late) was to be avoided at all costs, even if it meant taking a drink: ‘Your true Topers [that is, drunkards] are so sensible of this, that by Observation they have gathered it to be more safe for their Health and better for prolonging their Lives, to get drunk betimes and go to Bed, than to sit up and be sober.’158 Although Cheyne never failed to draw conclusions about what such lifestyles could do to one’s health, he avoided entirely the issue of what greater social or political consequences might issue from such behaviour. His natural philosophy was informed by mechanical doctrine, not, as had been the case with Goodwin and Tryon, self-knowledge. Save his singular comment about nightmares that might appear upon overeating before sleeping, Cheyne said nothing at all about dreams in his Essay. This lacuna is even more pronounced in his explicitly autobiographical treatise on nervous disease, The English Malady (1733). Here, where one might expect at least some marginal reference to dreams as signs of physical or moral decay among the anecdotes of patients gripped by convulsions, terror, dread, grief, melancholy, lethargy, hysteria, and assorted other distempers, there is an astounding silence. Even when recounting his own crisis, which featured ‘Restlessness, Watchings,

The Persistence of Privacy 63

and want of Tranquility of sound Sleep’ as symptoms, Cheyne said nothing of his own dreams.159 Eighteenth-century medical writers perpetuated Cheyne’s mechanistic analysis of sleep through their writings on hygiene.160 Sleep remained a ‘balmy restorative’ to fatigue in which the efficient concoction of food into blood could take place, provided sleepers followed certain rules about moderating their diet, sleeping at night, not sleeping too much or too little, and minimizing their exposure to night-time vapours. This last point was of particular concern to Francis de Vallangin (c. 1724–1805), a member of the Royal College of Physicians of London, who worried about the pernicious habit of city folk to stay awake at all hours of the night and then to proceed to sleep until noon.161 Vallangin related a cautionary tale about two French cavalry officers who, during a campaign, disagreed over whether to march during the cool of the night or the heat of the day. Needless to say, the former course of action led to a great loss of life, while those who chose the latter option, after marching six hundred miles in the summer heat, lost not a single man or horse. Such stories mingled with clinical anecdotes of fabulous or unusual sleepers to drive home the point that the duration and rhythms of sleep could indeed be described as idiosyncratic, constitutional, or habitual, but only up to a point. Alterations of the environment were ultimately more important in delineating the nature of sleep. Bizarre exceptions to this rule formed part of the popular sleep literature, a physiological counterpart to the scientific teratology then replacing medieval and early modern ‘fantastic anatomy.’162 Such were the exceptions that proved the rule that, in the end, natural forces regulated the animal economy and prevented disruption, just as they did in the political economy of nations. If sleep was a period during which the routines of natural recuperation could take place, dreaming was depicted as little more than an accidental intrusion upon this process. Whether bad dreams were the cause or the consequence of poor sleep was uncertain. Cheyne had suggested the latter, while John Burton (1710–71), a York physician, argued the former case: ‘For Dreaming is a State between Waking and Sleeping, wherein, although the Mind does not exercise such a Power over the Body, as to direct its Motions in the same Degree as when awake; yet by it’s Attention to those confus’d Ideas which pass thro’ it, the Solids are kept in some Degree of Contraction, greater than is agreeable with sound Sleep.’163

64

The Sleep of Others

In any event, a shift in the medical analysis of dreaming was taking place, as physicians began to frame the problem of dreaming in terms of its influence on health, rather than sensation. To mechanistic hygenists, health was an equilibrium maintained by the internal fluids of the body that circulated through ducts and vessels, nourishing the ‘solid’ organs and muscles.164 Burton, a former student of the Newtonian physician Hermann Boerhaave (1668–1727), meticulously applied the mechanistic approach that he had learned in Leiden. Boerhaave, inspired by the new experimentalist philosophy, had replaced Galen’s nutritive, vital, and animal spirits with chyle, blood, and bodily secretions. Whereas the former had maintained a healthy balance by virtue of their innate powers, the latter could be measured and, in theory at least, subjected to a purely mathematical analysis. The work of Santorio Santorio (1561– 1636) exemplified this sort of reductionism, and mechanists like Burton cited it approvingly.165 In De statica medicina (1614), Santorio described experiments in which he had spent long periods of time in a chair scale, weighing everything he ingested and excreted. The daily variations in his weight, he argued, were due to a quantifiable but insensible ‘perspiration’ that fluctuated according to various internal and external factors, including cold, the onset of fever, and sleep. It was generally held that sensation depended upon the ethereal animal spirits that circulated in the nerves, but these were imponderable and could thus play no role in a mechanistic physiology dominated by the rhetoric and practices of measurement. As partial or corrupt sensations, dreams now suffered a double removal from sleep: as sensations, they were immeasurable; as a notorious source of religious error or enthusiasm, they were unreliable. The anarchy of dreaming could have no place in the orderly animal economy of eighteenth-century hygiene literature. But, if they had lost their place in medical practice, dreams retained their importance as a means of framing claims about the nature of knowledge. John Locke’s An Essay concerning Human Understanding (1690), for example, definitively set dreaming apart from reason by arguing that dreaming was not merely a diminution or alteration of thought but rather the complete absence of thinking.166 Immediate selfreflection or self-awareness was a requisite part of reason, and this was entirely lacking in sleep, be it filled with dreams or not. Its equivalent, argued Locke, was the mental activity of a newborn child, which could not possibly qualify as thinking. In the latter case, the mind had not yet been exposed to sensations, which were the origin of all ideas; in the former, the mind was incapable of reflecting upon those sensations out

The Persistence of Privacy 65

of which all dreams were built. But even as he rejected Descartes’s notion that innate ideas exist in the mind and could serve as foundations of reason, Locke upheld the traditional function of dreams as reason’s counterpart – the readily accessible means by which reason could be known, if only by its absence.167 Dreams would not be forgotten, even in the Age of Reason. Locke’s psychological analysis was given a physiological treatment in David Hartley’s Observations on Man, His Frame, His Duty, and His Expectations, first published in 1749. Hartley (1705–57) had studied mathematics and divinity at Cambridge but ended up practising medicine in Bath instead, despite the fact that he never earned an MD.168 Here, he developed a close relationship with Cheyne, whose quietist views on religion Hartley may have shared. Certainly, they agreed on the need to apply mechanistic notions to physiology. But, whereas the corpuscular Cheyne treated physical and moral reform separately in his writings, Hartley was intent on using physiology to unite ethics and epistemology. To this end, Hartley developed his doctrine of vibrations, which melded together suggestions Newton had made regarding the origins of the sensation of light in the ‘Queries’ appended to his Opticks (1704) and the natural history of ideas that Locke had presented in his Essay. Newton had speculated that sensations arose by vibrations communicated through the ether, a ‘subtle spirit’ that permeated all matter.169 The sense organs received these vibrations from external objects and communicated them through the nerves to the brain, where sensation occurred. On the strength of this account, Hartley explained sleep as an absence of sensation in physiological terms. Vibrations, he argued, were transmitted through the solid nerves to the medullary substance of the spinal marrow and the ventricles of the brain, where the mind recognized the vibrations of the ether there as sensations. In sleep, however, the muscular activity that squeezed the blood out of the veins subsided, allowing blood to accumulate, which in turn put pressure on the brain.170 This pressure was compounded by the fact that, in sleep, the blood was heated, rarefied, and expanded. The subsequent compression of the medullary substance prevented sensation from occurring, thus creating the state of sleep. By a similar token, children slept a great deal, because their brains were constantly under a high degree of pressure as they attempted to grow within the skull. This pressure was so great that, according to Hartley, the two hemispheres of the brain would actually touch at the moment sleep arrived, engendering a sudden diminution of sensation.

66

The Sleep of Others

Dreams, on the other hand, fell under the rubric of the association of ideas. Unlike most hygienists, Hartley paid careful attention to the content of dreams. Dream images were taken by the dreamer to depict real scenes because, argued Hartley, ‘we become quite inattentive to external Objects’ in sleep, leaving no possibility for comparison.171 The ‘Wildness and Inconsistency’ of dreams were the result of unnatural associations, made possible because the brain was in a radically different state in sleep than in wakefulness. The dreamer was never offended by these inconsistencies, because the associations were governed exclusively by ‘bodily causes’ which had no recourse to memory: ‘It is to be observed likewise, that we forget the several Parts of our Dreams very fast in passing from one to another; and that this lessens the aparent Inconsistencies, and their Influences.’172 Individual images (rather than their connection through the process of association), however, were frequently composed ‘of Fragments of visible Appearances lately impressed,’ since such impressions maintained more residual vibrations than older ones.173 Hartley explained the repetition of specific dream scenes throughout one’s life, on the other hand, by arguing that such scenes were probably first established in youth and then successively ‘fixed’ each time the brain returned to them. Hartley’s ambitious attempt to reunite sleep and dreaming under the rubric of vibrations was stillborn. While his project of extending Locke’s doctrine of ideas was welcomed by many, his doctrine of vibrations was almost universally rejected.174 Mechanists like Boerhaave had depended upon the circulation of the animal spirits within the hollow tubes of the nerves in their accounts of the animal economy. Hartley’s notion that the nerves were solid and transmitted sensation through vibrations was anathema to the mechanists’ tidy hydraulic analogies, and the influential Boerhaave (who had died by the time the Observations on Man was published) condemned the doctrine. When the English chemist Joseph Priestley (1733–1804) re-edited the Observations on Man in 1775, he opted to omit the vibration theory altogether. Thus, dreams and sleep continued to be explained in different terms: the former remained marginalized within associationalist psychology and rationalist epistemology; while the latter languished as an incidental part of hygiene, subservient to the rigours of a good diet and a temperate lifestyle. The opportunity for sleep, dreams, and morality to coalesce within a physiological matrix seemed to have passed. Tryon’s efforts to rewrite the ancient art of dream diagnosis as a regimented form of spiritual introspection fell on barren soil during the Enlightenment. Even those

The Persistence of Privacy 67

medical writers who explicitly connected personal hygiene with morality eschewed all but the most incidental discussion of dreaming. The Lausanne professor of medicine Simon-André Tissot (1728–97), for example, discussed hygiene at length in his enormously popular Avis au peuple sur la santé (1761). Tissot’s concern about European depopulation turned on his understanding of the enormous importance of the six non-naturals. If country folk, the group hit hardest by disease, could only grasp the basic principles of hygiene, their health might be preserved, and depopulation prevented. As befitted a treatise that examined the health of the labouring classes rather than the idle rich, Avis au peuple made special mention of the role of perspiration. The animal economy, Tissot argued, depended upon the unrestricted circulation of this fluid to the exterior of the body. Suddenly stopping it by exposure to cold after heavy labour was, he thought, one of the most frequent causes of diseases. Entering a cold room or drinking cold liquids while perspiring fixed and thickened the blood in the capillaries, causing a variety of diseases, including ‘quinsey [tonsillitis], rheumatism, inflammations of the breast, pleurisies and inflammatory colicks.’175 Equally pernicious was the tendency of physicians to prescribe the sweats for acute diseases, which only increased the inflammatory condition by forcing out the ‘watery part’ of the blood and thereby thickening it.176 Given the role of sleep in regulating perspiration already established by Santorio, Cheyne, and Burton, one might expect Tissot to have made some mention of it. But sleep hardly figured in Avis au peuple. Its only mention came in Tissot’s comment that the ill should avoid hot chamber rooms, which was coupled with advice about eating little before retiring to bed and sleeping no more than eight hours a night.177 The latter issue was completely formulaic, although we should note that it was taken for granted by many commentators on sleep before around 1800 that a normal night’s rest was divided into ‘first’ and ‘second’ sleeps, with a not insubstantial period of wakefulness between the two.178 Locke, for example, challenged the possibility that the soul could think during sleep by citing the instance of an intelligent man who never remembered his dreams: it was highly unlikely, argued Locke, that such a man could be entirely unaware that his mind was busy for four (rather than seven or eight) hours while he slept.179 One of Tissot’s more infamous publications had mentioned in passing that some hygienists were so enthusiastic about early awakening that they recommend starting the day after the end of the ‘first sleep.’180 Nothing new, it seems, could be said of sleep’s hygiene. This same eva-

68

The Sleep of Others

sion also applied to dreaming. In his treatise on masturbation (one of the many activities in which people would sometimes engage between their two sleeps), Tissot’s concerns about the humoral economy were explicit. He argued that the activity of the ‘seminal humors’ had an enormous effect on the body, since the loss of just one ounce of this fluid weakened a person as much as losing forty ounces of blood. Madness, memory loss, tabes dorsalis (a type of syphilus), and a host of other diseases were common outcomes.181 Insomnia or sleepiness could also result, but these garnered little attention from Tissot, who trotted out the same old platitudes about the duration, timing, and conditions of sleep necessary to good health.182 He was much more interested in diet, which was the linchpin of his curative regimen. Frightful dreams and nocturnal emissions entered into Tissot’s discussion only by a secondhand route. Citing Onania, an anonymous English anti-masturbation treatise first published in 1710, Tissot recalled the case of an epileptic patient who experienced a seizure every time he masturbated or had a nocturnal emission.183 Tissot was uncompromising in his condemnation, and equated the epileptic’s self-pleasuring practices to suicide. But Tissot’s central concern was not, as it had been with Augustine, the problem of the dreamer’s responsibility for the moral content of his dreams. The issue at stake here was the waste of that most precious of all bodily resources, the seminal fluid. None of this is to say that dreams were ignored in the Enlightenment, for this was far from the case. Rather, dreaming was synonymous with the absence of reason and all the ills that naturally accompanied such a state. The Anglo-American radical and revolutionary pamphleteer Thomas Paine (1737–1809) proposed biblical prophecy as a case in point. In his ‘Essay on Dream’ that prefaced his Examination of the Prophecies (1807), Paine took Hobbes’s arguments about dreams to their naturalistic conclusions. It was not the case that dreams and visions could be confounded: there was not any difference between the two, and all prophecy had been derived from dreams, and, as such, were illusory foundations for religious belief.184 Without committing himself to any speculative claims about the exact nature of the brain’s physiology (Hartley’s doctrine of vibrations included), Paine compared the operation of the mind to that of a watch: the imagination provided the impetus as the mainspring, judgment maintained balance as a pendulum, while memory, like a hand and dial, recorded the motions. Dreams were the result of an imagination run riot while judgment and memory slept. The same experience in wakefulness was tantamount to lunacy. But for

The Persistence of Privacy 69

the fact that dreams appeared with such regularity every night, Paine argued, men would likely have confounded dreaming with madness. It was only the machinations of ‘priestcraft and superstition’ that prevented common sense from recognizing biblical accounts of prophecy and the miracles of Christ for what they truly were: the delusions of dreaming. This analogy between mind and mechanism, while primitive in comparison with the associationist psychology that dominated the mental philosophy of the same period, served Paine’s purposes well nonetheless. For deists like Paine, God was simply a watchmaker, one who had so perfectly ordered nature that he hardly, if ever, interfered with its operation. Man’s divine duty was to use his reason to comprehend this order, thereby eliminating irrational beliefs in witches, spirits, miracles, or mankind’s need for salvation in the form of Jesus Christ. Identifying dreams (an irrational, but regulated, part of nature) with prophecy (the social perpetuation of delusion) was integral to the progress of man. And it was here that Paine broke with Hobbesian absolutism and demanded that individual freedoms take precedence over social order: ‘This story of dreams has thrown Europe into a dream for more than a thousand years. All the efforts that nature, reason, and conscience have made to awaken man from it have been ascribed by priestcraft and superstition to the workings of the devil, and had it not been for the American revolution, which by establishing the universal right of conscience first opened the way to free discussion, and for the French revolution which followed, this religion of dreams had continued to be preached, and that after it had ceased to be believed. Those who preached it and did not believe it, still believed the delusion necessary. They were not bold enough to be honest, nor honest enough to be bold.’185 Their status as tools for diagnosis of soul or body denigrated or forgotten, dreams reappeared as social objects in the early nineteenth century. Dreams did not, as Augustine, Goodwin, and Tryon had suggested, indicate one’s state of grace. Nor, to judge by eighteenth-century hygienists’ silence on the topic, did dreams signify the state of the body. Poor digestion signified itself by a troubled sleep, not by some meaning inherent in the confused images of dreams. It was sleep, not dreaming, that, like all aspects of hygiene, demanded regulation on the part of the individual in the quest for health. Yet, at the social level, the regularity of dreaming required explanation. What made people’s minds work like those of madmen every night? This comparison between dreaming and

70

The Sleep of Others

madness had been proposed long before the eighteenth century, but it was only within the context of the revolutionary consequences of the Enlightenment in America and France that such an explanation became an imperative of social progress. The study of dreams could reveal the nature of belief, but it was their regularity, more than their content, that was at issue. Such a program of research required that investigators turn away from dreams as private knowledge or idiosyncrasies of the soul. Sleep, that part of the soul’s environment that fostered dreaming, needed to be rendered public, somehow.

2 Analogize and Experiment

By around 1800, dreams could no longer speak the truth about an individual’s body or soul. Their natural history was one of error, falsehood, illusion. Dream images reflected neither humoral balance nor a state of grace, because dreaming was nothing less than the product of a mind operating in the absence of reason. Hygienists like Cheyne and Burton had marginalized dreams, while political commentators like Hobbes and Paine had negated their once-positive message of transcendence. But even as dreaming lost its value as a guide to the self, it gained a new prominence as an investigative object. It provided an epistemic backdrop against which wakeful, public performances could be gauged. Such performances became all the more frequent over the course of the nineteenth century. In an effort to characterize the nature of the irrational, psychiatrists, neurologists, and psychologists carefully staged hypnotic experiments, which, some thought, were artificial analogues to the phenomena of sleep. Sleep itself began to attract greater attention from clinicians, who began to dismantle humoral analysis by describing sleep’s disruptions as diseases, such as insomnia or narcolepsy. Physiologists took up the experimental side of this task by framing the question of sleep in terms of the industrial problem of fatigue. In every instance, sleep was finally becoming visible.

Blood, the Brain, and the Mundane Pathology of Sleep The nineteenth-century origins of psychophysiology – the experimental study of relationships between physiological process and psychological phenomena – can be traced back to efforts to study circulating blood in the brain. If the brain was indeed the organ of mind, so the argument went, then the dynamic processes within must somehow reflect mental

72

The Sleep of Others

events. Anatomical approaches, such as those adopted by phrenology, could merely illustrate the material framework of such processes. The cerebral circulation, on the other hand, held out other possibilities. Blood was the nutriment of the brain, and the study of its movement was inspired by the hope that it could be linked to the brain’s role in sensation – the basis of all nineteenth-century empirical theories about mind. The study of sleep in the nineteenth century was closely linked to this ‘alimentary’ model of psychophysiology, which ultimately helped to sever the physiological problem of sleep from the domain of hygiene. Before the so-called ‘materialist revolt’ of the students of the Berlin physiologist Johannes Müller, experimental interest in sleep was limited to the demonstration that dreams did not have a supernatural origin but were instead caused by the suspension of the activity of the external senses.1 This limited purview of experimental research is not, in itself, particularly surprising: it was only in the nineteenth century that scientific research became definitively associated with experimentalism, often at the expense of natural historical or analytic ‘ways of knowing.’2 The origins of this experimentalist position on sleep can be traced back to one of the foundational texts of French physiology, George Cabanis’s On the Relations between the Physical and Moral Aspects of Man (1802). Cabanis (1757–1808) considered dreams to be the essential activity of the mind in sleep, which, he assumed, could be accounted for in empirical terms by studying cerebral circulation. For Cabanis, the study of the origin of ideas, or ‘ideology,’ was possible only through physiological investigation.3 Dreams indicated that the brain was active in sleep, but, Cabanis argued, the hallucinatory nature of dreaming implied that the situation was more akin to madness than to the normal processes of thought. Like the phenomena of delirium, dreams were the product of an imbalance of fluids in the brain. Citing the Edinburgh physician William Cullen (1710–90), as the first to recognize ‘the constant and definite relations between dreams and delirium,’ Cabanis extended Cullen’s original hint that sleep was not a unitary process. That is to say, the various organs of the body slept in different ways and at different times. It was the partial stimulation of the points in the brain that corresponded to the various organs that were responsible for the ‘irregular and confused images’ with ‘no basis in the reality of objects’ that characterized both dreams and delirium.4 This ‘partial stimulation’ was itself due to the incomplete contraction of nervous powers towards the brain that defined sleep for both Cabanis and for Cullen. The flow of blood followed the retraction of the ‘stimulating

Analogize and Experiment

73

causes,’ as they moved from the peripheral world of objects to the interior of the body.5 Ultimately, the ‘causes of excitation,’ he argued, concentrated to produce sleep in the brain. Cabanis had considerable difficultly convincing his critics in Paris and Montpellier that sleep was the product of an increase, rather than a decrease, of stimulatory activity in the brain: he included two lengthy footnotes explaining that, during sleep, there would be more blood in the brain, even as it acted less. His explanation for this paradox was obscure and made reference to ‘physiological laws on which it would be inappropriate to dwell at this point.’6 The second edition (1805) of On the Relations Between the Physical and Moral Aspects of Man contained an additional note that broached the ontological question of the very nature of sleep: ‘Some people appear to have misunderstood the meaning of this passage. I have not said that there is more action in the brain during sleep than during wakefulness, but that sleep is not merely a passive function ... [the brain] rests from wakefulness by sleep, and from sleep by wakefulness; but it is never in that inert state imagined by men who carry on the study of life notions only for a rough mechanism.’7 Cabanis, however, never seems to have actually attempted to measure blood flow to the brain during sleep. Even if he had, the fact that he never described any procedure of this nature in his most important work indicates that such experiments – if they indeed existed at all – would likely have had little meaning. Even for Cabanis, sleep remained little more than the temporary suspension of life. Sleep quite literally had no being and was thus incapable of being rendered into a proper experimental object. So Cabanis’s analysis relied on analogies drawn from pathological anatomy, then the dominant means of creating evidence for physiological arguments.8 The identity of the confused images of dreams and those of madness, argued Cabanis, depended on an alteration of the normal state of the brain. Even if structural defects were not directly visible, Cabanis insisted that the damaging effects of excessive and unequal blood flow left a visible trace: ‘The vessels of the ventricles [in cases of insanity] have been found to be sometimes stuffed with blackish, pitchlike, and deleterious matters. As in lesser degrees these organic disorders have been accompanied several times by corresponding and proportional disorders of the mental faculties, it is difficult not to attribute it to them when they are found in individuals affected by the maniac and furious madness.’9 Cabanis had placed dreaming on a continuum with other ‘disorders of the mental faculties,’ and thus it was to be expected that dreaming

74

The Sleep of Others

was caused by an imbalance, however transient, in cerebral blood flow. Such was the dominant medical view of madness in Napoleonic France, which also happened to be shared by many of Cabanis’s colleagues – most notably, the notorious François-Joseph-Victor Broussais, whose dedication to phlebotomy was unrivalled. Excessive blood flow to various organs of the body was thought to account for a huge range of illness, and it likewise underwrote the pervasive practice of bloodletting. Cabanis’s efforts to depict sleep as an active process fit uncomfortably with the clinico-anatomical methods that dominated academic medical thought at the beginning of the nineteenth century. How could sleep be represented as an activity when it consisted in the slackening of all motions? Cabanis had turned to the movement of sleepers to make his case, noting that the phenomenon of sleepwalking, which seemed to involve both judgment and the will, indicated that sleep was not entirely uniform. That is, at any given time, not all organs slept equally. ‘Asleep’ could describe both the active, but pathological, state of the entire body and the passive, restful, and normal condition of the organs. But the anatomical evidence for this position was not forthcoming. Pathological anatomy simply could not represent process. It could only present the reality of the organism’s state at the moment of death, hence the epistemological value of dissecting subjects who had died in a state of dementia. Dreaming seemed to present a set of phenomena similar to madness or dementia, and thus Cabanis attributed it, along with sleep, to a similar cause. The internal dynamics of sleep remained hidden from medical view. This intimate relationship between sleep and madness, which had been little more than a metaphor for Hartley, became an invaluable heuristic over the course of the nineteenth century. In the French case, links between dreaming, pathological anatomy, and the illusions and hallucinations of insanity were forged by writers like Cabanis in the early years of the republic.10 But outside France, medical interest in sleep was by no means limited to the explanation of its unique mental imagery. For practising clinicians, it was frequently the transformation of the sleep regimen and its association with mental disease that provided a focus for theories about the nature of sleep. Insomnia: The Nocturnal Face of Madness It is easy to think of insomnia as a universal experience impervious to historical examination. What could possibly have changed about a com-

Analogize and Experiment

75

plaint as pervasive and elementary as the loss of sleep? And where there is no difference, there is no history. Yet the current poverty of historical analysis on this topic hardly reflects the significance of the problem in the past, let alone the present. As I suggested in chapter 1, classical medicine conceptualized sleep as a regimental phenomena of the body. Like respiration, exercise, diet, evacuations, and sexual activity, sleep was a ‘nonnatural,’ in the sense that its routines were at least partly under the patient’s control. Disruptions of these routines were easily recognized by the well-informed patient, and questions of ‘how often?’ ‘how much?’ and ‘under which circumstances?’ dominated the medical discussion of the rituals and remedies that would restore the balance that defined the healthy individual. Sleeplessness, then, was not so much disease as disruption. Whereas gout or plague were exceptional events marked by symptoms that required interpretation, sleeplessness, like flabbiness or constipation, was merely a deviation from a regimen appropriate to the patient. The lack of a coherent definition of what constituted normal sleep encouraged an individualistic approach to the problem of its privation. Only the individual could perceive whether or not he or she was getting enough sleep, so medical intervention was largely limited to offering recipes that would appeal to the individual patient’s tastes and habits. This hygienic approach to sleeplessness appears to have dominated medical practice until the early nineteenth century. Even while Cabanis was beginning to localize sleep in terms of the brain’s activity, the experience of sleeplessness continued to be conceptualized in terms of hygienic routines, as habitus. Evidence for the value of such routines was typically drawn from clinical cases, which served to emphasize the individualistic nature of sleep. That is to say, sleeplessness still referred to the soul’s idiosyncratic control of the body, even as the universalizing brain was becoming the primary referent of sleep. Consider, for example, the role of the popular discourse of phrenology in The Philosophy of Sleep, written by the Glasgow physician Robert Macnish (1802–37), who had also penned a popular treatise on drunkenness. Macnish generated two versions of his book on sleep during the 1830s, and the differences between the them, as Macnish admitted in a later preface, were considerable. The original 1830 version made a few oblique references to phrenology in an effort to explain the phenomena of sleep. The revised version, on the other hand, was completely permeated by phrenological language insisting that dreaming, nightmares, sleepwalking, sleeptalking, and the like were determined by the varied and conflicting activities of the ‘organs of the brain.’ Yet, for all the phrenological hype, Macnish

76

The Sleep of Others

said little as to which organs of the brain were actually responsible for sleeplessness. Instead, he trotted out the standard platitudes of hygiene: avoid stimulating beverages, sleep on a firm mattress stuffed with hair rather than feathers, keep the chamber room dark, avoid excessive exercise or study late in the day, attempt to maintain a cheerful mood, and do not dwell on unpleasant emotions or fantastic images. He also recommended a few therapeutic preparations, including opium and its recently derived alkaloid, morphine. If these did not agree with one’s constitution, hyosciamus (a narcotic extract of henbane) or a ‘pillow’ of hops (a dose taken before going to bed) were prescribed. The latter, Macnish reminded his readers, had greatly relieved the prolonged wakefulness of the late King George III (who had been, no doubt, rather troubled by the loss of his American colonies). Macnish invoked phrenology in his revisions to The Philosophy of Sleep for rhetorical purposes. It coated the clinical anecdotes from which his theories were drawn with a glossy veneer of empirical, systematic investigation. This must have pleased the New York publisher who produced the revised version in 1834, since phrenology was an enormously popular doctrine of self-investigation in America at that time (the American edition of George Combe’s phrenological opus, The Constitution of Man, became an immediate best-seller when it appeared the following year). Yet it left the natural historical core of Macnish’s ‘philosophy’ intact. In Macnish’s time, sleep had not yet emerged as an investigative object; it was still defined as experience, the diversity of which defied the artificial conditions of experiment while it exemplified the private nature of consciousness. Sleeplessness was a desire for the reduction of consciousness – anticipated, but not yet arrived – and sleep itself, Macnish agreed, was somewhat akin to madness: ‘Restlessness,’ he noted, ‘when long protracted, may terminate in delirium, or confirmed insanity.’ The idea that a lack of sleep might somehow inhibit performance, however, was decades away. Short of this allusion to the privation of reason that was madness, what more could be said? Naming Insomnia Sleeplessness, not insomnia, was the topic of Macnish’s discussion. The distinction between these two terms is important, for insomnia’s history is rather different from that of sleeplessness. The latter term enjoyed a sustained usage in the English language that insomnia did not. Being ‘sleepless’ was a common enough theme in literature and poetry.

Analogize and Experiment

77

Indeed, one cultural historian has cogently argued that one form of sleeplessness, that of awakening in the middle of the night, was so common that the terms ‘first’ and ‘second sleep’ were spun around it.11 The intermediate period of wakefulness was populated by the most human of activities: arguing with one’s neighbours, sex with one’s bedmate, theft of another’s property, and the like. The expectations of a night of uninterrupted sleep appeared only with industrialization in the nineteenth century. Improvements in domestic and street lighting, evening employment and trade, and professional policing gradually rendered the night more and more like the day, and unbroken periods of activity, or its lack thereof, became increasingly more common. Insomnia reappeared in just this context. The word had long existed and was derived from the negation of the Latin somnus, sleep. Its first mention in the English language was in Cockeram’s English Dictionarie of 1623, where ‘insomnie’ is defined as ‘watching’ or ‘want of power to sleepe.’ ‘Watching’ – a term favoured by eighteenth-century hygienists – suggests a duty-bound vigilance or perhaps even a simple voyeurism, as suggested in Hamlet’s concluding line of ‘The Mousetrap,’ Shakespeare’s play-within-a-play: Why, let the stricken deer go weep, The hart ungallèd play, For some must watch, while some must sleep, So runs the world away.12

By mid-century, however, the usage of that rare term, ‘insomnia’ seems to have changed. ‘Insomnia’ is not present in either Philip’s (1658) or Coles’s (1676) dictionaries, but, in its place, we find ‘insomnious,’ defined as ‘wanting sleep, also dreaming much.’ How is it that the lack of sleep could also imply dreaming, which, by its very nature, seems to imply sleep’s presence? Perhaps these writers were simply confused. The Latin prefix ‘in-’ can serve to negate or to include. De insomniis is the Latin title of one of Aristotle’s treatises. It could mean either ‘on Sleeplessness’ or ‘on Dreams.’ The Greek title, however, is not ambiguous, nor is the substance of the treatise, which is most certainly about dreaming as something that occurs within sleep.13 Clearly, this sort of confusion was what the authors of the first version of the Oxford English Dictionary had in mind when, in 1901, they declared Blount’s (1656) and Bailey’s (1721) definition of ‘insomnious’ as ‘troubled with dreams’ erroneous.

78

The Sleep of Others

But put aside, for a moment, the notion that it even matters whether or not ‘insomnious’ was correctly used to describe sleeplessness in the seventeenth century. The dream, to this day, is as closely linked to wakefulness as was the Old English dréam, which functioned as both noun and verb and referred not to nocturnal dreams but to joy and mirth or the noise made by a musical instrument. This usage was so common that writers who wanted to refer to nocturnal dreams frequently used the word swefn, literally, ‘sleep.’ By around the fourteenth century, however, ‘dream’ began to obtain its current meaning, namely, visions witnessed in sleep or images of fancy appearing during wakefulness. Thus, the seventeenth-century writers’ definition of ‘insomnious’ may have simply been formalizing, in Latin, the understanding that the same relentless activity of the imagination that produced dreams in sleep could equally prevent sleep from coming at all. At any rate, ‘insomnia’ and its related terms could not have been common. Samuel Johnson’s 1755 dictionary – a landmark in the study of the English language – offers only silence on this point. Johnson surely would have mentioned the term had it been used, given that his 1753 ‘Ecomium on Sleep’ indicated the author’s considerable sensitivity to the fact that ‘so liberal and impartial a benefactor as sleep’ had been chronically ignored by historians and poets alike. Like the latenineteenth- and early-twentieth-century commentators on neurasthenia discussed in chapter 6, Johnson’s discussion of sleeplessness was part social critique. It was the idle rich, he argued, who suffered most from sleeplessness, while common labourers slept the best. British dictionaries generally followed Johnson’s lead for the next century. By the time of James Murray’s Oxford English Dictionary (1884– 1928), however, insomnia had made a definitive reappearance. To be sure, there was still confusion: the compilers of the Index Medicus, which began publication in 1879 under the auspices of the U.S. Army Library, continued to list classical or early modern treatises entitled ‘De insomniis’ under ‘sleeplessness.’ References to insomnia were scarce. But, by the mid-1880s, a flurry of articles appeared on the topic, and by the 1890s, the term was far more common in medical titles than ‘sleeplessness’ and ‘wakefulness’ combined. ‘Insomnia’ also began to appear in American dictionaries with a highly technical and medical meaning. Hunter’s American Encyclopaedic Dictionary (1894), for example, defined insomnia in both traditional terms, as sleeplessness, and as a pathological entity: ‘The disorder is of nervous origin arising from mental anxiety or overwork, and is one of the earliest and most marked features of

Analogize and Experiment

79

acute mania; the functions of the body are badly performed, and severe fever frequently accompanies it.’ In this perspective, it was the middle class, rather than Johnson’s idle rich, who suffered most from insomnia. While physical labour still promoted good sleep, excessive mental work and its consequent worry was the bane of many in America’s ‘Gilded Age.’ Whereas gluttonous luxury had once generated the individual experience of sleeplessness, insomnia was now the property of an urbanized population. In other words, nineteenth-century insomnia was caused by progress. The intimate relationship between insomnia and madness here is ambiguous: insomnia was simultaneously a sign and a cause of mania. Fatigue and emotional disturbance played key roles in insomnia’s aetiology, and their effects were determined in terms of physiological performance. By the end of the nineteenth century, insomnia had left behind the paradoxical and poetic connotations of vigilance, idleness, and dreaming once enjoyed by sleeplessness and exchanged them for a technical interface of psychiatric, psychological, and physiological dimensions. How had insomnia become such a dangerous aberration of night-time norms? Defining Sleep in the Clinic and the Laboratory The creation of insomnia as a symptom of an underlying mental disorder can be traced back, at least in the American context, to the work of William Alexander Hammond (1828–1900). Hammond, who was appointed surgeon general of the United States Army in 1862 (only to be dismissed sixteen months later), had an extraordinarily successful career as one of the first neurological specialists in America.14 His theory of sleep, which he began to formulate as early as 1854, proved to be a crucial element of his lucrative New York practice. His diagnosis of thousands of his patients with a condition he named ‘cerebral hyperaemia’ was the leading edge of a series of indeterminate, but unifying, diagnoses that characterized the nosology of mental illness in America after the Civil War and that ultimately converged on the concept of ‘neurasthenia’ by the end of the century.15 Cerebral hyperaemia was, as the name suggests, a condition involving the excess circulation of blood in the brain. It was a functional disease, meaning that, unlike the mania Cabanis observed in asylum patients, it could not be diagnosed post-mortem by a lesion in the brain. Hammond diagnosed cerebral hyperaemia clinically, by its symptoms, which

80

The Sleep of Others

were inevitably reported by his (mostly) private patients, rather than observed among asylum inmates. As such, cerebral hyperaemia was a disease of culture, civilization, and class. Its very existence depended upon the reliable testimony of its victims, who would report a litany of nervous complaints, from headaches, dizziness, forgetfulness, and hallucinations to ringing in the ears and back pain. But, for Hammond, its pathognomonic sign – the one symptom that served to distinguish this disease from all others – was insomnia. And just as Hammond’s patient population differed from that of Cabanis, so too did his theory of sleep upon which the concept of cerebral hyperaemia was based. These competing theories of sleep in some sense depended upon the different links these two investigators made between the clinic, experimental physiology, and pathological anatomy. In 1854, while serving as a physician to the U.S. Army in Kansas, Hammond saw a patient who had suffered a cranial fissure in a railroad accident, then an all-too-common source of neurological research material. While he was comatose, the patient’s scalp over the fissure appeared to rise. After he recovered, however, the skin normally appeared level with the rest of his scalp, except when he slept, at which point the skin was slightly depressed. What was the relationship between sleep and supposedly ‘allied’ states, like coma? While he was on medical leave a few years later, Hammond began to experiment on animals. He trephined the skulls of several dogs and rabbits, covered the holes with parts of a watch glass, and peered through these tiny windows to observe the movements of the dura mater that covered the cerebral cortex. Hammond subjected the animals to narcosis by ether, chloroform, and opium and compared these results to those observed during natural sleep. On the basis of observations of the colour and size of the brain, Hammond concluded that sleep was not on a continuum with any of these states. In sleep, blood flow to the brain was reduced. In cerebral hyperaemia, it was pathologically increased, which explained why insomnia was such a prominent symptom of the disease. The fact that Hammond’s diagnosis rested on a physiological theory that went against French authority fit well with Hammond’s emerging ideas about creating a ‘positive mental science’ based on the ‘facts of physiology’ rather than the ‘metaphysics’ of psychology.16 Laboratory-based, comparative physiology was beginning to uproot the French medical ideologues’ reliance on introspective evidence regarding the equivalent nature of hallucinations (on the part of the insane) and dreams (on the part of the medical philosophers and their reading audience). Or at least, this was one conclusion that Ham-

Analogize and Experiment

81

mond’s audience was encouraged to draw, forgetting, in the process, that his theoretical arguments relied just as heavily upon reported symptoms of nervous complaints as Cabanis’s depended upon the private recollection of the nature of dreaming.17 In his review of Henry Maudsley’s The Physiology and Pathology of Mind (1867), Hammond raised the materialists’ banner, demanding that ‘it is now necessary that the unholy barrier set up between psychical and physical nature be broken down.’18 Substance dualism and psychophysical parallelism were also dismissed. The brain was not the organ of some other thing called ‘mind.’ Mind was nothing more than the performance of the brain. Inspired by the evolutionary writings of Darwin and Spencer, Hammond attempted to interpret all neurological phenomena in materialist terms. As Lorraine Daston has pointed out in her study of British psychophysiology, sleep became a viable testing ground for physiologists wanting to turn mind into body.19 Because volition was suspended it sleep, it seemed a reasonable prospect that, if the natural ebb and flow of the will in sleep could be understood, other, more complicated mental phenomena would be illuminated in turn. Sleep offered a possible field of inquiry where the conflicts between free will as a moral imperative and mental science as a law-governed enterprise might be resolved. Of course, not all investigators agreed that the will was suspended in sleep. One notable exception was the Marquis Hervey de Saint-Dénis, who, in 1867, described the process whereby he claimed to be able to control the content and direction of his dreams.20 But the marquis was no medical man: he was a sinologist and historian at the Collège de France, with no links to experimental physiology. His introspective efforts thus fell outside the increasingly narrow confines of what it meant to generate experimental evidence. Following the British surgeon Arthur E. Durham, who first presented this method of visually inspecting the brain’s activity at the Oxford meeting of the British Association for the Advancement of Science in June 1860, Hammond took as a first principle that sleep could be defined in terms of its physiological function, which Durham had described as ‘that period of cerebral inactivity during which nutrition of the brain substance takes place.’21 This particular meeting, of course, became notorious for Thomas Henry Huxley’s confrontation with Bishop Samuel Wilberforce over the evolution of man. But Durham’s arguments were no less part of the growing strength of mid-Victorian materialism. His concept of ‘cerebral inactivity’ was crucial for Hammond’s understanding of sleep. During wakefulness, the brain’s activity

82

The Sleep of Others

was dominated by thought and the exercise of the will. Both were lacking in sleep, and thus the circulation of blood in the brain was reduced, as it would be in any organ at rest. Hammond expressed his dismay that so many investigators had been deceived on this point in the past: ‘It is well established as regards other viscera, that during a condition of activity there is more blood in their tissues than while they are at rest. It is strange, therefore, that, relative to the brain, the contrary doctrine should have prevailed so long, and that even now, after the subject has been so well elucidated by exact observation, it should be the generally received opinion that during sleep the cerebral tissues are in a state approaching congestion.’22 The source of this confusion, Hammond argued, was the ontological dualism that dominated most of the physiological thinking of the day. Mind and body were considered to be distinct and separable entities, and this contributed to the belief that thought, the activity of the mind, consisted primarily in picturing. Hammond’s vision of thought, on the other hand, was tied inescapably to the will. ‘Thinking,’ he insisted, ‘is an action which requires cerebral effort, and which is undertaken with a determinate purpose. We will to think, and we think what we please.’ ‘But,’ he continued, ‘it is very different with our dreams, which come and go without any power on our part to regulate or direct them.’23 Citing Locke’s Essay, Hammond argued that dreaming had no relationship whatsoever to thought, since the latter was always accompanied by the conscious awareness of oneself, which the images of dreams inevitably lacked. Dreaming was not, as Cartesian doctrine had it, the mode of the soul’s activity in sleep. Sleep involved the annihilation of the will and, as a consequence, the temporary abasement of mind. But what were dreams, if not some form of thought? Hammond drew upon associationist psychology, coupled with an image of the will as selfrestraint, to dismiss dreams as the activity of the imagination run riot. Dreams were ‘either impressions made upon the mind at some previous period, or produced during sleep by bodily sensations. These impressions, however they may be formed, are subjected to the unrestrained influence of the imagination.’24 Memory, which was to serve such an enormous role in Freud’s theory of dreaming three decades later, was almost completely ignored by Hammond, who instead relied on the quantitative testimony of his physiological instruments. One of his experiments, probably from the 1860s, involved a primitive manometer to detect changes in cerebral pressure in trephined animals. Although his ‘cephalohemometer’ offered only the fluctuations of coloured water

Analogize and Experiment

83

against an index, they nevertheless allowed him to measure, rather than simply observe, the degree to which the size of the brain was reduced in sleep. ‘Nothing can exceed the conclusiveness of experiments of this character,’ Hammond argued. ‘No mere theorizing can avail against them.’25 Hammond’s physiological claims about the nature of dreaming, however, should not be detached from his clinical studies of cerebral hyperaemia. The diagnosis unravelled during the 1880s and has hardly recovered since, even in the historical imagination. Edward Shorter, a historian of psychiatry, has been particularly harsh in his assessment of Hammond, whom he dubs ‘the dean of American reflex theorists.’26 ‘Few documents,’ Shorter says, reviving the views of several of Hammond’s contemporaries, ‘could testify more eloquently to the retarded state of American medical practise around the turn of the century than the work of William Alexander Hammond.’27 Bonnie Ellen Blustein has rehabilitated Hammond by emphasizing his work on sleep and its empirical grounding, rather than focusing on his notorious diagnoses.28 Blustein credits Hammond with introducing a number of items into American neurological practice, including the ophthalmoscope, Duchênne’s trocar (for muscle biopsy), and the dynamometer (to measure muscle strength).29 His work on poisons, which he conducted with Silas WeirMitchell (1829–1914) in the late 1850s, had led the French neurologist Charles-Édouard Brown-Séquard to describe Hammond as the ‘first original Physiologist in the United States.’30 Nonetheless, Blustein concedes that ‘by 1890 “cerebral hyperaemia” and the theory of sleep on which it rested had been seriously undermined.’31 By this time, American neurologists were beginning to strive for a higher degree of precision in their diagnostic criteria, an aspiration that corresponded to a professional scepticism about their ability to explain all aspects of mind (rather than merely manifestations of mental disease) by reference to physiological function. This self-doubt persisted into the early twentieth century and was to some extent perpetuated by the growing popularity of a psychological paradigm that traced the cause of certain mental illnesses back to the activity of the mind, and, in particular, the emotions.32 Hammond, whose work was steeped in the reductionist reflex theories pioneered by Marshall Hall in the 1830s, could have no truck with such ideas, and his work was subsequently discarded. Blustein concludes that ‘cerebral hyperaemia, along with the Hammond-Durham sleep theory, would disappear, not only from the medical textbooks, but also, for close to a century, from the histories of medicine as well.’33

84

The Sleep of Others

Stubborn Reflexes This may be an accurate assessment of Hammond from the perspective of the histories of neurology and psychiatry. But the fate of Hammond’s research in physiology was quite different. Although the names of both Hammond and Durham became detached from the theory that sleep was caused by a decreased cerebral circulation, their observations retained the status of empirical facts well into the twentieth century and were staunchly defended by some of the leading physiologists of the day. Echoing an experiment Hammond had once performed on himself, the French physiologist and future Nobel laureate Charles Richet described the disappearance of excitability in a dog’s cortex upon compressing the carotid arteries, and he, like Hammond, inferred that a similar process occurred in sleep.34 The Italian physiologist Angelo Mosso, whose 1891 book on fatigue was translated into several languages and went through numerous editions, produced an important book on the circulation in 1894, in which he argued that, in sleep, the blood pressure in the brain declines while that of the periphery increases.35 In the United States, the growth of physiology as a field independent of medical education meant that physiological facts no longer had to be harnessed to theories of disease.36 And so, just when Hammond’s reputation as a scientist seemed to be under attack by neurologists and psychiatrists, we find the notable Johns Hopkins physiologist William Henry Howell (1860–1945) confirming a decreased circulation in sleep.37 One of Howell’s students even extended this idea into the realm of hypnotic sleep.38 When Howell first published his American Text-Book of Physiology in 1896, a mere three pages was dedicated to the question of sleep, and Hammond was not mentioned at all. It was not until subsequent editions that Howell devoted a separate chapter to sleep, largely based on his own research.39 In later editions, Howell noted that his results had been challenged and that theories of sleep based on a diminution of blood flow in the brain had been ‘brought into question.’ Nonetheless, he simply cited these critical references and continued on, as though no evidence had been brought against his work whatsoever.40 In fact, it was not until the Yale neurophysiologist John F. Fulton reconstructed the entire textbook during the 1940s that Howell’s chapter on sleep was finally abandoned. Nathaniel Kleitman, the doyen of American sleep research during that period, noted the persistence of Hammond’s theory of sleep in 1939: ‘The idea of cerebral anemia as a cause or result of sleep,’ he complained, ‘refuses to die, and reviewers like Allen and Mott

Analogize and Experiment

85

seem to accept Mosso’s results as if nothing had been done on the subject since 1880.’41 Neurologists may very well have trashed Hammond’s nosology by the end of the century, as Blustein argues. They may also have rejected his ‘reflex theory’ paradigm, as Shorter suggests. But his ideas about the circulation decreasing in sleep persisted precisely where the locus of scientific medicine was supposed to be – in the physiological laboratory. How was this possible? The fact of the matter was that, unlike neurologists, physiologists had not abandoned the concept of reflex in their investigations. The focus of their interests simply shifted away from disease aetiology and towards the field of psychology, which underwent a dramatic expansion in the United States during the first decade of the twentieth century. The entry of psychology into industry, education, and the military meant that physiological knowledge now had a new conduit out of the laboratory. And, while the reflex paradigm was rejected by clinicians, it continued to be an indispensable part of how psychologists constructed the normal human mind.42 The work of Marshall Hall (1790–1857) and Johannes Müller (1801– 58) consolidated reflex action as the conceptual basis of materialistic physiology by the end of the 1830s. Hall proposed three different reflex systems: the ‘spinal system,’ which was unconscious and depended only on the presence of the spinal cord; the ‘excito-motory’ system, which involved conscious sensation; and the ‘sensori-volitional’ system, which involved the will.43 Müller took a more integrated perspective, arguing that, while the brain and spinal cord were anatomically distinct, their functions were not so easily separated. But, in spite of their differences, both agreed that it was the ‘reflex arc’ that translated sensation into movement. Hammond’s theory of sleep was thoroughly grounded in reflex theory. The lack of movement that characterized sleep was, he thought, the result of decreased stimulation owing to the onset of darkness, diminishing noise, and a general lack of intellectual and physical activity. Increased stimulation demanded a greater blood flow to the brain; when the stimulation diminished, so did blood flow, and sleep ensued. But such an account could not stand on its own, since lack of sensory stimulation was neither a necessary nor sufficient condition of sleep. Normal sleep was recurrent and reversible and was thus based on a cycle. To explain this cycle, physiologists had to have recourse to some regulatory principle that incorporated the instantaneous nature of the reflex into a longer-term explanation of periodicity. Hammond and his contemporary Durham described this periodicity in terms of a build up

86

The Sleep of Others

of ‘waste products’ in the brain on account of its increased activity during wakefulness. By the end of the century, this idea began to be described in terms of fatigue. Its early-twentieth-century counterpoint, which will be discussed in chapter 8, applied psychoanalytic concepts to argue that the origins of insomnia were not in the body but in the mind. A Sleep Disorder Stillborn: Narcolepsy In 1880 a new kind of disordered sleep – narcolepsy – began to take shape in Europe. Its discoverer, Jean-Baptiste Edouard Gélineau (1828– 1906), was an unlikely enough candidate for successfully cultivating a new disease within the hothouse of late-century French neurology. As a naval physician and general practitioner in Rochefort, a seventeenthcentury military base on the Atlantic coast, Gélineau had neither the central location nor the professional networks to be taken seriously by his Parisian colleagues.44 His development was not that of a universitytrained physician but more like that of a natural historian: he had studied surgery at the Rochefort Navy Medical School, but his inspiration came from collecting clinical and epidemiological data during his trips around the Indian Ocean, which earned him a doctoral degree in medicine from Montpellier in 1858. He left the navy two years later and set himself up as a private practitioner in Aigrefeuille d’Aunis, a small town near Rochefort. It was here that Gélineau made his reputation by peddling a pharmacological concoction called Dragées Gélineeau, which, he claimed, cured epilepsy and numerous other nervous disorders. The pills, which he introduced in 1871, contained a combination of bromide, antimony, and picrotoxin – essentially a combination of ‘uppers’ and ‘downers.’ The tablets were well known in French medical circles, and quite possibly made Gélineau a small fortune. He left in 1878 for Paris, where he set up a private neurological clinic. Relatively unknown and unaffiliated with Jean-Martin Charcot’s circle of neurological research at the famous Salpêtrière, Gélineau was not long in finding a star patient that would help launch his new diagnosis. The patient, ‘G.,’ was a male wine-barrel retailer, thirty-eight years old, who began to suffer from a series of ‘sleep attacks’ about two years before he came to Gélineau. According to Gélineau’s first report on the subject, the patient’s attacks were preceded by a feeling of ‘deep heaviness’ and of ‘a heavy load on the forehead and deep in the eyes.’ Seconds later, the patient fell into a deep sleep. One of the more notable features of the disease was that the attacks seemed to occur at any time: during a

Analogize and Experiment

87

meal, at the theatre, or even as the patient was attempting to have a conversation. Similar descriptions of this new disease had already been published by two German neurologists, who tended to view its phenomenon through the lens of epilepsy.45 Like epileptic seizures, sleep ‘attacks’ rendered the victim utterly helpless. The main difference between the two was in the different state of the musculature: epilepsy was characterized by muscular contraction, while sleep attacks involved a muscular relaxation typical of sleep. But in either case, the presumed cause was some sort of intoxication, in which the brain periodically became overwhelmed by a noxious substance. Gélineau, perhaps in the spirit of the nationalistic rivalries between the French and the Germans that dominated continental politics at the time, argued that his cases of narcolepsy were not simply unusual cases of epilepsy. For one thing, his bromides seemed to have no therapeutic effects whatsoever! More important, his patient could be awakened from his attacks as readily as he could be from normal sleep. But G. had also described an additional, and quite exceptional, symptom: his attacks often followed a strong emotional expression. As he burst into laughter after concluding a good professional deal, his legs would suddenly buckle beneath him, and he would fall dead asleep. Embarrassment would also provoke this cataplexy, which Gélineau dubbed astasie (the inability to stand up). The patient even reported being at the zoo, standing ‘around the monkey’s cage, rendezvous of the curious, the maids, the soldiers,’ and suddenly falling victim to an attack.46 Crowds proceded to gather around him, laughing, which of course only made the situation worse. This, thought Gélineau, was hardly reminiscent of normal sleep. Nor could such symptoms be mapped on to other diagnoses, such as syncope or agoraphobia. Following the suggestion of E.F.A. Vulpian, a pathologist at the Salpêtrière, Gélineau argued that there was a centre for emotional associations in the mid-brain, and that he had discovered a new disease that was the result of a shock to this region of the brain, where a ‘sleep centre’ was also located. He dubbed the new disease ‘narcolepsy.’ Sadly, for Gélineau, his diagnosis was soon rejected by the elite group of neurologists gathered around Jean-Martin Charcot. Even as they adopted Gélineau’s terminology, they dismissed the idea that Gélineau had described a new disease entity. ‘Narcolepsy’ came to describe all varieties of pathological sleep.47 Gilbert Ballet, who had recently replaced Charcot as chef de clinique at the Salpêtrière, set the tone in

88

The Sleep of Others

1882, declaring that those who, like Gélineau, claimed that narcolepsy was a ‘neurosis’ like hysteria or epilepsy were merely using the term to hide their ignorance.48 British neurologists were equally indifferent. In 1889 Richard Caton, a professor of physiology at Liverpool whose work will be discussed in chapter 7, described a patient whose ‘narcoleptic attacks’ seemed to originate with an obstruction in the larynx that effectively strangulated him and led to a build-up of brain toxins that caused him to doze off during the day.49 But this case had little to do with Gélineau’s version of narcolepsy. Nor, for that matter, did it offer a description of a new disease that would eventually be dubbed ‘sleep apnea.’50 Caton was quite clear that he felt himself to be describing an unusual phenomenon with a curious pathophysiology. Yet these odd symptoms were linked to character and individual idiosyncrasies; they did not provide the foundations of an entirely new category of disease (as they were to do during the 1970s and 1980s). For its part, the United States was then filled with Hammond’s ‘insomniacs,’ and neither narcolepsy nor Gélineau seem to have been very well known. The most prominent American neurologist of the period, Silas Weir-Mitchell, mentioned neither in his 1890 address to the Association of American Physicians, despite the fact that his talk attempted to offer a comprehensive overview of sleep disorders recognized at that time.51 The denigration of Gélineau’s work continued well into the twentieth century: Henri Piéron, whose efforts in establishing sleep as an experimental object will be discussed in chapter 4, rejected outright the idea that narcolepsy was anything more than a symptom.52 And, as we will see in chapter 7, it was not until after the 1953 discovery of REM had effectively created sleep as a viable object of laboratory research that narcolepsy finally stabilized as a specific (and highly significant) sleep disorder. The Physiology of Fatigue If sleep did not fare well as an object of clinical investigation during the nineteenth century, it certainly did no better in the physiological laboratory. In the early twenty-first century, it seems logical to link the exhaustion of mind or body with the phenomena of sleep. But this was not the case a century ago, when such problems were closely connected more to the study of the musculature than to that of the nervous system. Fatigue, not sleep, was the foundational object of laboratory analysis. In a provocative analysis of how the metaphor of the ‘human motor’ originated in materialist physiology and ended as an organizing principle

Analogize and Experiment

89

for human culture by the beginning of the twentieth century, Anson Rabinbach has argued that physiological interest in the problem of fatigue can be traced back to the work of Hermann von Helmholtz, a student at Müller’s laboratory in Berlin in the 1840s.53 Helmholtz provided the most consistent and clear statement of the law of the conservation of energy, which, when applied to mid-century physiology, suggested that all organic phenomena could be understood in terms of work. This analysis was soon extended to include mental processes. Thought was not, as Cabanis had suggested, something that was secreted by the brain, as bile was secreted by the liver. It was, like all the brain’s activities, the product of a transformation of energy, and was therefore, at least in principle, amenable to detection and measurement by physical methods. Fatigue, as a tangible experience of the exhaustion of energy, was an early target of such an investigative approach, and it was soon linked to the rise of the scientific study of work through both physiological methods and Taylorist time-management studies. But, in such a history of muscle and mind, there seems to have been little room for the physiology of a passive activity like sleep. Sleep was routinely depicted as simply the restoration of energy that naturally took place when work stopped. It was not considered to be a form of work in and of itself – in fact, sleep receives virtually no mention in Rabinbach’s narrative. In contrast to the clinically driven analysis of work done by Hammond or Gélineau, laboratory-based physiological research began to turn towards fatigue around the late 1870s. Between around 1845 and 1875, mechanistic physiology was dedicated to the examination of the microphenomena of animal nerve-muscle preparations. This experimental assemblage (frequently devised from isolating the sciatic nerve in a frog’s leg) could teach physiologists quite a lot about fatigue in terms of the rate and frequency of induced electrical shocks required to exhaust a muscle. But these models said little about how such phenomena could be translated into the subjective experience of fatigue, or how they could be usefully applied to the problem of work efficiency. This latter approach was taken up by the French physiologist Étienne-Jules Marey (1830–1904). Marey’s efforts were an important precursor to the biomedical holism that came to pervade much biomedical research in the decades following the First World War.54 Although Marey’s theories and tools were resolutely mechanistic and reductionist, his attempts, along with those of his student Angelo Mosso (1846–1910), to use graphical methods to capture and measure organic activity in its natural setting

90

The Sleep of Others

mark him as a methodological holist. Marey attempted to create an alternative to vivisection, then the mainstay of the physiological laboratory, and, to this end, he pioneered the ‘graphical method’ that eventually came to dominate sleep research. Yet it was fatigue, not sleep, that first attracted Marey’s attention. Although sleep pathologies offered some marginal clinical problems, sleep was still, from the perspective of the laboratory, a non-entity during the 1870s and 1880s. It was not until interest in fatigue disseminated a visual system of research that sleep became integrated, albeit slowly, into the new laboratory-based medicine of the late nineteenth century. Synchronization: The Kymograph The graphical method can be traced back to the work of one of Müller’s students in Berlin, Carl Ludwig. In 1847 Ludwig described a new device in the pages of Müller’s Archiv that would ultimately transform physiological practice (fig. 1). No longer limited to the rational reconstruction of motion through physical, chemical, or morphological analysis, some physiologists soon became actively engaged in the visual representation of biological time.55 Ludwig’s invention, the kymograph, or ‘wave writer,’ was the first self-inscribing physiological instrument, the origins of which have been traced, on the one hand, to similar instruments used in ballistic studies already in the 1830s, and on the other, to the graphical representations used to depict the efficiency of steam engines. Its invention, then, offers a poignant instance of how the seventeenth-century analogy drawn between organisms and machines became, in nineteenth-century physiological practice, an identity, through the intervention of representational instruments. Ludwig wanted to understand the influence of one vital system (respiration) on another (the circulation of the blood). To accomplish this, he needed to compare changes in the air pressure in the thoracic cavity to changes in blood pressure. The measurement of either phenomenon posed no difficulty, instruments then being available for both applications. It was the synchronization of two rapidly fluctuating measurements that was the central problem, since the precise linkage of these two simultaneous phenomena was beyond the perceptual skills of any experimenter. Ludwig’s solution, which was ultimately recreated in the discovery of the EEG and REM, was to historicize the phenomena in question. That is to say, he converted a chronicle (or table) of singular events (manometer readings) into a narrative (an unbroken trace). By

Analogize and Experiment

91

1 Late-nineteenth-century physiology textbooks typically introduced the graphical method with an image of Ludwig’s kymograph or ‘revolving cylinder.’ Many of the textbook writers had studied under Ludwig at Leipzig, and so the use of such images was as much an homage to their ‘beloved master’ (as Stirling described Ludwig in his dedication) as it was an instructional device for students. The surface of the cylinder (R) was covered with smoked glazed paper and driven by a clockwork mechanism (A). Were the inscribing stylus pictured here, it would be connected by a tube to the manometer on the left. The calibrating mechanism (not pictured) would have been a tuning fork or an electromagnetic timer. (William Stirling, Outlines of Practical Physiology [Charles Griffin: London and Philadelphia, 1888], 162)

92

The Sleep of Others

attaching a stylus to a rod-shaped float placed on top of the mercury column, Ludwig devised a means to trace a line on a rotating drum, driven at a constant rate by a clockwork mechanism, thus making possible their comparison in time. The physiological investigator’s labour now went into obtaining a proper record that could, when carefully edited and properly reproduced in scientific journals and monographs, allow the facts to speak for themselves. The rhetorical potential of this method was enormous, since it gathered experimenter and reader together around the same phenomenon. A large portion of the physiological field came to congeal around this one experimental practice: its technical solution to the problem of synchronization was soon transformed into a metonym for the discipline itself.56 The kymograph generated an entirely new mode of representation within the life sciences, in which the ‘work’ of the heart, lungs, and muscles could be converted into an unambiguous, measurable sign for the ‘vital force’ that animated matter. Ludwig’s invention offers an example of what the historian of science Nicholas Jardine has described as ‘calibration,’ a process that, in part, describes the ability of instruments to bring new questions into being by expressing data in a manner similar, but not identical, to that of older instruments.57 The former are ‘calibrated’ to the latter. The graphical trace that once accurately depicted changes in air pressure was rapidly adopted to express physiological changes in the body, ultimately transforming the entire discipline by drawing it closer to both physics and to industry. Helmholtz’s support for the graphical method appears to have been influenced at least as much by his participation in the Berlin Physical Society as it was by his use of the kymograph itself.58 His enthusiasm for this new mode of physiological investigation was supported by the German instrument maker Werner von Siemens and the arch-reductionist electrophysiologist Émil duBois-Reymond, both co-directors of the Society. The industrialized production of the kymograph by the Leipzig instrument manufacturer, Baltzar, ensured the central place of the device in physiological research.59 The graphical method is not of merely antiquarian interest: its progeny have helped transform modern biomedicine from a practice trading in patient-reported symptoms to one relying upon the analysis of mechanically generated signs. It has even become a method for resolving moral questions, such as establishing the time of death in order to utilize effectively organs for transplant.60 Yet there are clearly other ways of practising medicines and there are other means of determining the moment of death. As with any

Analogize and Experiment

93

technology, the success currently enjoyed by the graphical method in Western medicine cannot account for its original implementation. How did it become so popular? The Graphical Method: Étienne-Jules Marey This new tool for achieving practical objectivity in visualizing vital phenomena is probably best exemplified by the career of Étienne-Jules Marey.61 In 1857 Marey had completed his doctoral dissertation on the circulation of the blood. By studying the traces generated by arterial pulsation, and particularly the left and right radial arteries, Marey was able to diagnose various abnormalities in the heart’s rhythm and thereby helped move Ludwig’s kymograph out of the laboratory and into the clinic.62 From this point on, Marey dedicated the better part of his career to transforming the observational platform of experimental physiology. He founded the first laboratory for experimental physiology in France in his private residence on the rue Cuvier. His ‘physiological station’ soon received state sponsorship, and, by 1878, he had shaped a comprehensive methodology around his recording instruments. In place of the artificial pathologies created by vivisection, Marey proposed ‘la méthode graphique,’ which became an all-consuming, and infectious, passion. In the words of one colleague (in this case, the editor of Revue scientifique), ‘the graphical method, which in reality was and is only a means of study, became, through Marey’s scientific research, an end in itself.’63 In 1868, his analyses began to breach the walls of his physiological station. This was the year that Marey succeeded Pierre Flourens in the chair for ‘the natural history of organized bodies’ at the Collège de France. Their difference in investigative style was striking. Flourens (1794–1867), a protégé of the comparative anatomist Georges Cuvier, had based his entire career upon his skill in removing parts of the brain and observing the functional failures that resulted. Marey’s reputation lay almost entirely in his ability to construct devices that could detect and measure aspects of the organism without disturbing it. The most notable cultural impact of Marey’s efforts to reconstruct visually the passage of time was undoubtedly the birth of cinematography. But it was the problem of fatigue that provided him with the impetus to graphically display physiological time in the first place. The editorial preface to the first volume of his journal, Physiologie expérimentale, offered his readers a summary of his vision of how the graphic

94

The Sleep of Others

method would ultimately transform the science of life. The article that followed took the problem of fatigue to illustrate how the graphic method allowed experimental physiology to be applied to some concrete problems of human existence. The concepts of physiology, Marey complained, had been isolated from the physical sciences for too long.64 The study of organic motion had hardly changed since Albrecht von Haller had dubbed it Anatome animata over a hundred years earlier. It remained subservient to anatomical techniques and concepts. This was largely due to the lack of precision in physiological methods, a problem that was particularly pronounced in France, where Claude Bernard’s Introduction to Experimental Medicine championed the use of vivisection as the only possible means of studying life scientifically.65 Although Bernard (1813–1878) had insisted that all life was governed by the same physical and chemical laws that ruled over the inanimate world, Marey argued that Bernard’s reliance on vivisection continued to isolate physiology from the sciences that discovered those laws. Shortly after he became Flourens’s assistant in 1867, Marey began criticizing vivisection, describing it as a crude and destructive practice, incapable of overcoming the experimenter’s limited observational capacities. It ‘can do no more,’ he argued, ‘than lay bare the phenomenon simultaneously with the organ which is the seat of it; it reveals to our senses only what they are capable of perceiving.’66 In its place, Marey proposed that physiology be grounded in a new observational practice, one that would renounce vivisection and replace it with an investigational system capable of examining living organisms undisturbed. But this was no return to a vitalistic system that held there to be something unique about life: it was only through the application of ‘the great law of the conservation of force,’ Marey argued, that the last vestiges of vitalism could be definitively abolished from physiology.67 And this victory could be won only through a complete reconstruction of investigational practice around the use of inscribing instruments common to both the physical and the life sciences. The fact that these instruments were recent inventions offered an enormous incentive for physiologists to adopt their use: their dissemination had been so rapid and thorough that it would take little effort for the sciences of life to catch up to their physico-chemical counterparts. In fact, physiologists merely had to calibrate their concepts to the tools they were already using: ‘Inscribing instruments are found everywhere,’ cheered Marey, ‘in the observatories of astronomers and meteorologists, in physics laboratories, and in those of physiology.’68 The institutionalization of this

Analogize and Experiment

95

new medium soon became Marey’s personal campaign; the transformation of the life-world of the physiologist would, he argued, finally realize the long-standing analogy between machines and organisms by unifying the sciences in a shared phenomenology. Such ideas were not commonplace in French physiology, even in 1875. Not only were Marey’s heterodox ideas outside of laboratory-clinic relations as exemplified in the work of Bernard and Louis Pasteur, his physiological station was quite literally physically set apart from clinical medicine. Marey’s laboratory was originally located in the Sorbonne, until it was relocated to the Bois de Boulogne in 1881, and his journal, Physiologie expérimentale, was a not a product of the Paris Faculty of Medicine. It was instead the result of a new creature in the French academic system that appeared in the early 1870s: the École Pratique des Hautes Études. In the wake of France’s humiliating defeat by the Germans in 1871, the École Practique was designed as a distinctly French response to the perceived superiority of the German educational system. While its emphasis was on improving science education, its mission was to orchestrate all scholarly work in France by overseeing research in the faculties of law, medicine, theology, letters, and sciences, as well as that of the prestigious Collège de France, the Museum of Natural History (where Bernard’s laboratory had been relocated), the School of Pharmacy, the Paris Observatory, and various other institutions. Its architect, Victor Duruy (1811–94), considered this new system to be his greatest achievement. Duruy’s reputation as a liberal reformer had been seriously compromised by his reaction against the philologist and historian Ernest Renan, who, in 1863, was denied his elected post as professor of Hebrew at the Collège de France because his questioning of Christ’s divinity angered many powerful clerics. For his part, Duruy supported academic freedom, so long as it did not disturb the religious or political peace. His attempt to resolve the ‘Renan affair’ by relocating Renan to his old position at the Bibliothèque Impérial fell flat, and Renan was eventually given the college chair in 1870.69 In contrast, Duruy’s approach to scientific research pleased conservatives like Louis Pasteur, who had long complained of the chronic poverty of French science in comparison to its German counterpart. Duruy’s École Practique freed research from the entrenched strictures of French higher education, since the school was open to anyone, regardless of nationality or former education, and did not place particular emphasis on terminal degrees or examinations. The new faculty was generally quite young, and many were Germantrained. Thus, there was good reason for Duruy to describe his new insti-

96

The Sleep of Others

tution as ‘a germ which I am depositing in the cracked walls of the old Sorbonne,’ which, he hoped, ‘will crumble it.’70 So it was from the vantage point of a freshly minted research institution that Marey began to publish his studies on animal locomotion and the applications of the graphical method, caring somewhat less for medical therapeutics than for technological innovation for the perfection of society. After his 1863 monograph on the circulation of the blood, Marey began a series of studies of animal movement, which included the flight of birds, before publishing his definitive methodological statement in 1878. Duruy’s enthusiasm for Marey’s work reflected the latter’s obvious skills as an engineer, rather than his abilities as a medical practitioner. Duruy first met Marey in the mid-1860s, while touring the physics laboratory at the Sorbonne. Like his successor, Jules Ferry, Duruy was spellbound by Marey’s arsenal of recording devices, artificial organs, and mechanical birds, and immediately began to sponsor his research.71 This is not to imply, however, that Marey was uninterested in medical practice. On the contrary, the very title of one of his books – La méthode graphique dans les sciences expérimentales et principalement en physiologie et en medicine (1878) – suggests otherwise. But his focus was always on technological development, not the nature of disease. When cholera broke out in France during the summer of 1884, he quickly became part of a committee commissioned by the Académie de Médecine to investigate the matter.72 Like Pasteur, who was also on the committee, Marey supported the germ theory of disease and took a keen interest in the role of the water supply in the spread of cholera. But, in his response to the problem, he adopted a more typical hygienic approach, studying the flow of water supply and sending questionnaires out to doctors, asking them about the effects of weather changes, the disposal of faecal matter in the district, and local hygienic conditions. In short, he treated the cholera epidemic as a problem of civil and social engineering that could be solved by the visual analysis of motion, rather than the experimental study of bacteriology. Such an approach, which deployed technical measurements to evade the question of life, was well in keeping with Marey’s passion for tracing out the economy of living motion. His journal, which was abandoned after its fourth volume in 1879, covered almost all aspects of the imposition of the graphical method in the study of organic movement, but the very first article to appear in the journal set the stage for all that followed, insofar as it concerned the distinctly human problem of work, labour, and its economization through the control of fatigue.

Analogize and Experiment

97

The introduction to his article, which he had presented to the Académie des sciences in 1874, reiterated his rejection of vitalism and his fidelity to the fruits of German reductionism.73 The language of work and efficiency replaced all talk of the unique properties of life. All organs, he argued, could be analysed in terms of their ability to render motion regular and thereby increase the amount of ‘useful work’ they could do.74 ‘In the circulation,’ he observed, ‘the elasticity of the aorta and of the arteries does not merely transform the saccadic and intermittent movement of the heart into a continuous flow. To this well-known influence, one must add another, which is more important yet, but which has escaped the attention of physiologists to this point: the elasticity of the arteries economizes the work of the heart.’75 The motif of efficiency was the phenomenon of continuous, smooth motion. Elasticity allowed the rhythmic pumping of the heart to be communicated to the periphery, creating the relative stability of blood pressure. Marey used this image, grounded in the work of the organs, to depict the social work of human labour by shifting his analysis to the role of musculature in locomotion. It was the elasticity of muscle that allowed the successive ‘shocks’ to be transformed into the smooth motions of the limbs. From this analogy, Marey argued that ‘it is not much further to conclude that from the point of view of the utilization of work, muscular elasticity presents the same utility as that of the [blood] vessels.’76 Marey took the same approach to the problem of work as he had with the circulation: he set about creating an experimental system that could generate some sort of graphical trace. There was an important difference between these two problems, however. In his earlier work on circulation, Marey had built an artificial heart capable of generating traces that, he argued, were analogous to those that might be taken from the organ itself. An analysis of the former could thus help physiologists understand the heart’s natural efficiency without interrupting its processes (fig. 2). But, when it came to muscular work, the examples that surrounded Marey – human and animal labour – were all examples of inefficiency that needed improvement. So Marey constructed devices that could not only measure the work done but could improve the efficiency of the performance as well (figs. 3 and 4). Through such devices, he was able to consolidate the work performed by machines and that performed by bodies through the medium of the graphical trace. Marey had forged a mechanical identity out of a seventeenth-century mechanical analogy. No one was more explicit about this mechanical realization than Marey himself: ‘No

98

The Sleep of Others

2 Marey’s artificial heart, used to compare the dynamics of four different tracings recorded simultaneously. (Étienne-Jules Marey, ‘Mémoire sur la pulsation du cœur,’ Physiologie expérimentale 1 [1875]: 19–85, 71)

Analogize and Experiment

99

3 Readers of Marey’s journal, Physiologie expérimentale, were first introduced to his graphical method through its application to the physiology of work. The dynamograph pictured here transmitted changes in the force pulling a load that altered the pressure within the tambour (t), which, in turn, moved a stylus that inscribed a curve on a kymograph. Marey then analysed the curve to show how the addition of springs could ‘economize’ the work of pulling. (Étienne-Jules Marey, ‘Du moyen d’utiliser le travail moteur de l’homme et des animaux,’ Physiologie expérimentale 1 [1875]: 1–18, 11)

4 The tambour (T ) that drove the recording stylus (l) that inscribed the kymograph was at the very heart of Marey’s graphical method. Through its motions, Marey recorded a variety of physiological activities, all of which he interpreted under the rubric of work performed. (Étienne-Jules Marey, ‘Du moyen d’utiliser le travail moteur de l’homme et des animaux,’ Physiologie expérimentale 1 [1875]: 1–18, 23)

100

The Sleep of Others

doubt, the physiologists of old discerned levers, pulleys, cordage, pumps, and valves in the animal organism, as in the machine ... But these passive organs had need of a motor; it is life, it was said, which set all these mechanisms going, and it was believed that thus there was authoritatively established an inviolable barrier between inanimate and animate machines. In our time it is at least necessary to seek another basis for such distinctions, because modern engineers have created machines which are much more legitimately to be compared to animated motors.’77 The study of elasticity would ultimately benefit humanity by making labour more efficient: ‘To place elasticity between our muscular efforts and the masses they must move is to imitate the processes of nature for the better utilization of the intermittent action of our muscles.’78 The benefits of increasing elasticity went beyond the physiological domain: an elastic harness, Marey observed, helped ‘calm’ a workhorse by reducing the intensity of pulling a heavy load over a rough road.79 There was thus a psychological utility to making an organism’s ‘motor work’ more efficient: an offshoot of Marey’s application of the dynamometer was the reduction of the subjective experience of hauling a load to a quantitative problem capable of graphic illustration. Rather than offering tables of mathematical analysis, he simply reproduced the traces taken in two experiments – one generated by a dynamometer attached to a harness with an elastic connection, and one without. The area between the abscissa and the trace in the former instance was 26 per cent smaller than that of the latter, allowing Marey to conclude that ‘the economy of work and the diminution of fatigue obtained through this method of traction constitutes an important application of physiology to the amelioration of the fate of animals and of man.’80 This conclusion was in keeping with Marey’s thoughts about the improvement of the human race in general. In Animal Mechanism (1873), Marey indicated that the fate of humankind was actually a question about the relationship between the musculature and the will. Like many physiologists in France during the last quarter of the nineteenth century, Marey carved up ‘development theories’ into two major divisions: the ‘old school,’ whose members believed that species were inalterable forms that persisted, relatively unchanged, through time; and the proponents of the ‘new school,’ who argued that species evolved.81 He admitted that the modification of species seemed incapable of experimental verification, but he nevertheless insisted that Lamarck’s original insight of the inheritance of acquired characteristics was proba-

Analogize and Experiment

101

bly correct, as was Darwin’s more recent ‘addition’ of the phenomenon of natural selection (the difference between these two positions was, Marey felt, more rhetorical than substantial). He then presented his own position on development by describing the relative variability of the skeletal and the muscular systems. The structure of the skeletal system, was, it turned out, entirely dependent upon the muscles. ‘In the form of the bony structure,’ Marey argued, ‘everything bears the trace of some external influence, and particularly the function of the muscles. There is not a single depression or projection in the skeleton, the cause of which cannot be found in an external force, which has acted on the bony matter, either to indent it, or draw it forward.’82 In Marey’s internal economy of the body, tendons hollowed out bones, the curvature of the tarsus increased as the mobility of the bone decreased, and the shape of a bird’s wing depended upon how its muscles propelled it in flight. The analysis of muscular work offered the key to understanding the evolutionary process; but, since it was the will that ultimately directed the musculature, effort and desire were the engines of evolutionary change: It is understood that the skeleton, as it is modified, plays a passive part; that it is subject to the form imposed upon it by the muscle. But what gives to the muscle itself, an organ eminently active, and the true generator of the mechanical force by which the skeleton is in some degree modified, the particular form which is revealed to us by anatomy? We hope to demonstrate that the power to which the muscular system is subjected belongs to the nervous system. The nature of the acts which the will commands the muscles to perform, modifies the muscles themselves, in their volume and their form, so as to render them capable of performing these acts in the best possible manner. And, as this necessity which determines all the actions of animal life, governs the will, it is this, which, according to the external conditions under which every living being is placed, influences its form, and regulates it according to the laws which we must now endeavour to make known.83

Following this epistemological program, Marey began to outline his campaign for human improvement by making human labour more efficient. The constraints to transforming species under domestication, Marey acknowledged, were real and profound: ‘It would be necessary,’ he contended, ‘to do violence to the habits of animals, and to constrain them gradually to perform acts to which their organism is but slightly

102

The Sleep of Others

adapted.’84 Human beings, on the other hand, were capable of transforming such habits in themselves, which was why Marey, like Lamarck, placed humans at the forefront of evolutionary progress. His particular brand of experimental physiology, which was outlined in the remaining chapters of Animal Locomotion, would uncover the laws of efficiency that ran through the natural world of movement and show how they could be applied to the cultural world of human labour, ultimately leading to the improvement of the species itself. The Psychophysiology of Fatigue: Angelo Mosso The application of Marey’s time-and-motion studies to practical problems in the workplace, however, were still largely in the future in 1904, when, just before his death, he published his final paper on elasticity as the natural model of work efficiency.85 It was the investigations of his student Angelo Mosso that brought work physiology into the field of human productivity in the twentieth century by incorporating a deeper psychological dimension to the study of labour. This transition was itself the product of the graphical method: Mosso’s greatest invention, the ergograph, explicitly attempted to calibrate muscular exhaustion to the sensation of fatigue, thus uniting experimental physiology with the burgeoning science of psychology under the common goal of improving the performance of the human motor. Mosso’s ergograph (fig. 5) spread across Europe through the pages of his most popular work, Fatigue, which went through numerous German, French, and English editions in the years following its initial 1891 publication in Italian. Mosso claimed that his book – an eclectic amalgam of anecdote, image, and experiment – was the product of ten years’ worth of investigations. But, in fact, his interest in fatigue can be traced as far back to the early 1870s. Just as Duruy was centralizing the direction of scientific research in the Third Republic, and in the same year (1874) that Marey presented his dynamometer to audiences across France, Mosso appeared at the door of Marey’s laboratory on the rue de l’Ancienne Comédie. Mosso had recently completed a dissertation at the University of Turin on the growth of bones and then gone on to work under Moritz Schiff in Florence for two years. He then travelled to Leipzig, where he studied with Carl Ludwig for a year. The well-orchestrated, industrial precision of experiment in Ludwig’s laboratory evidently made quite an impression on the young student. Mosso later recalled that, arriving in Leipzig in 1873, he was ‘in time to be present at the

Analogize and Experiment

103

5 Mosso’s ergograph generated curves inscribed by a stylus, pictured here at the left edge of the table. Mosso argued that these ‘fatigue tracings’ depicted what had formerly been experienced only as a sensations and were, he claimed, unique to each individual. (Angelo Mosso, Fatigue, translated by Margaret Drummond and W.B. Drummond [London: Swan Sonnenschein, 1906], 88)

latest experiments made by Professor Kronecker in completing his researches upon the fatigue and the restoration of the striped muscles of the frog. It is a duty – and more than a duty, a pleasure – for me to avow that it was these experiments which first fired me with the desire of applying myself to the study of fatigue. The exactitude of the method, the elegance of the apparatus, the precision of the results, could not but charm a novice.’86 Hugo Kronecker was determined to understand the law-like properties of the physiology of fatigue; to this end, he removed the leg muscles from frogs, electrically stimulated them at regular intervals, and recorded their subsequent contractions on a kymograph. The resulting image looked much like the curves obtained earlier by Marey (fig. 6), which Mosso faithfully reproduced in his own book on fatigue. Marey’s figure illustrated the phenomenon of latent excitation (the increasing length of time the muscle remained in contraction as fatigue increased). Kronecker, however, wanted to demonstrate how the strength of the contrac-

104

The Sleep of Others

6 Variations of this image depicting the effects of a standard nerve-muscle preparation made from a frog’s leg were common in textbook and popular treatments of experimental physiology. The tracing is to be read from bottom to top, and the time of each curve (calibrated to the 100/sec. wave at the very bottom) increases with the number of contractions, thus illustrating the increasing recovery time of the preparation. (Angelo Mosso, Fatigue, translated by Margaret Drummond and W.B. Drummond [London: Swan Sonnenschein, 1906], 79)

Analogize and Experiment

105

tion decreased over time. After exposing the same muscle to up to 1,500 contractions, he concluded that the resulting ‘fatigue curve’ was always a straight line. That is to say, fatigue was a perfectly regular phenomenon, directly proportional to the time interval between equally strong induction shocks: the longer the interval, the slower the onset of fatigue. It was Kronecker who had recommended to Mosso that he travel to Paris to meet Marey. Although his stay in Paris was short, Mosso seems to have absorbed one crucial element of Marey’s experimental rhetoric: the need for physiologists to study things in their places, rather than submitting the vivisected remains of lower organisms to the artifice of induced shock. Mosso’s antipathy towards vivisection experiments was clearly derived from his work with Marey: ‘With frogs,’ Mosso argued in Fatigue, ‘it is impossible to reproduce the normal function of muscles and to imitate the action of a man who is doing mechanical work.’87 While the visual phenomena generated by the ergograph simulated the inscriptions Mosso had seen in Marey’s laboratory, its experimental context was quite different. Mosso’s ergograph could be operated only by human subjects who were both able and willing to follow instructions. Such an approach was in keeping with the general tone of Fatigue. The book began with a chapter on the migration of birds to illustrate the interplay of emotion, education, and muscular performance. Mosso offered a number of experiments demonstrating that the migratory instinct in pigeons was the product of education and was brought into play by emotion: ‘When they [carrier pigeons] are taken a long distance, the fatigue and toil which they endure in order to find their home once more is incredible. One might think they had become blind and had ceased to recognise danger; they care no more for their lives, they are infatuated with love.’88 Even the lowest animals incorporated emotion, and even music, in their movements: ‘Bees, which have been more minutely studied, furnish us with a very conclusive demonstration that emotion affects their flight, just as it does the gait of man. When startled or excited they emit a shriller sound. When in tranquil flight they seek honey from the flowers, they emit a la, and when in the evening they arrive wearied at their hive, the hum is on a lower note, namely, sol, just as we ourselves slacken our pace after a long walk.’89 Mosso’s attempt to bring emotion within the purview of the graphical method through fatigue was clearly evident in his choice of experimental subjects. In place of Kronecker’s twitching frog muscles, readers of Fatigue found ergograph tracings made by Mosso’s fellow professors at the University of Turin, whose individual records were frequently identi-

106

The Sleep of Others

fied by name.90 Mosso’s emphasis on the individual subject carried over to his criticism of Kronecker’s law. Fatigue, Mosso argued, was subject to individual variation. Individuals had their own unique, but rigorously consistent, style of fatigue that would produce a distinctive tracing. Some would slowly lose their ability to lift the three-kilogram weight, thus producing a long series of declining curves, while fatigue would descend suddenly upon others, causing their trace to drop off immediately. The ergograph, mused Mosso, ‘thus gives us a record of one of the most intimate and most characteristic features of our individuality – the manner in which we fatigue.’91 A remarkable experiment seemed to demonstrate that the basis of this individuality was located in the relationship between the nervous system and the musculature. Mosso isolated a group of muscles controlled by a single nerve in the middle finger, and then conducted two trials. In the first, the subject simply attempted to raise the weight as high as possible to the beat of a metronome. Then another tracing was taken from the same subject by electrically stimulating the nerve, causing the same group of muscles to respond involuntarily, thus eliminating ‘the mental element’ that Mosso felt influenced the fatigue curve.92 This technique, Mosso hoped, allowed him to distinguish between mental and physical fatigue and would eventually lead to an analysis of thought, creativity, and genius in the measurable terms of work. The distance Mosso had moved from Marey was evident in his emphasis on the role of the nervous system, long neglected by Marey. ‘The nervous system,’ Mosso concluded, ‘is the sole source of energy ... there exists only one kind of fatigue, namely, nervous fatigue; this is the preponderating phenomenon, and muscular fatigue also is at bottom an exhaustion of the nervous system.’93 While the ergograph was probably the best-known expression of the psychological dimensions of the graphical method by the end of the nineteenth century, Mosso was also known by physiological psychologists for his introduction of another instrument – the plethysmograph. He had been working on the device, which measured the change in volume in body parts, ever since he left Marey’s laboratory in 1874. After presenting a few papers on the subject, he eventually produced a book-length study on his experiments with the instrument.94 The book was dedicated to Kronecker and Marey, and it dealt primarily with the respective roles of the cardiac and arterial pulse in the blood’s circulation. There was very little psychology here, except when it came to the question of sleep. Happening upon two young subjects with small openings in their skulls,

Analogize and Experiment

107

Mosso created a simple device, built out of a Marey tambour, to record changes in cerebral blood pressure during sleep. At the same time, he used his plethysmograph to record blood pressure in the peripheral limbs. These few pages in Die Diagnostik des Pulses offered the first graphical evidence of a decrease in cerebral pressure. But, unlike Hammond, who argued that decreased circulation actually caused sleep, Mosso framed his observations in terms of the overall physiological economy.95 In his comparison of the two curves, Mosso observed that brain volume decreased in sleep while that of the arm increased. When a mild stimulus was applied, a slight rise in brain volume and a corresponding decline in arm volume resulted. Even though his subjects remained asleep, their brains seemed to be reacting to the outside world. The question of sleep haunted the pages of another of Mosso’s popular works of physiological psychology – Fear (1884). Here, Mosso used Marey’s sphygmograph to demonstrate that arterial tension increased during intellectual activity, as did blood supply. In contrast, fear was accompanied by a decrease in blood pressure. The exquisite regulation of the blood pressure, already established in his earlier works, took on a new laboratory aesthetic in Fear. Mosso had his subjects lie on a carefully balanced table that would tip at the head or foot end, depending upon where blood began to collect (fig. 7). He found that, with the onset of an emotion or any intellectual activity, the head of the table would begin to descend. But, despite the similarity of this apparatus to a bed, Mosso does not seem to have ever used it for the study of sleep. For such an investigation, Mosso, once again, drew upon pathological cases. In a few pages on night terrors (pavor nocturnis) in children, Mosso described his study of the changes in blood pressure that accompanied his subjects’ frightened awakenings. In the end, sleep was significant for Mosso only when it was pathologically disturbed, or when a subject with some sort of anatomical anomaly could be found, thus allowing an otherwise prohibited access to sleep’s physiology. While fatigue had clearly moved into the public world of experiment, sleep was still with few exceptions, a private phenomenon. As an investigative object, it remained hidden until its pathological alteration brought it (along with Mosso’s theories) to the attention of clinicians. The Function of Emotion: William James The invention, production, and dissemination of self-registering instruments such as the dynamometer, the ergograph, and the plethysmograph

108

The Sleep of Others

7 Although it looks as though this apparatus was designed to study sleep, it was in fact used by Mosso to examine changes in blood circulation during emotional states – fear in particular. Note the synchronous recording of blood pressure (from the plethysmograph on the foot) and respiration (from the respirometer around the chest). (Angelo Mosso, Fear [London: Longmans, Green, 1896], 96)

were a boon for psychophysiologists at the end of the nineteenth century, many of whom were beginning to take an interest in applying their psychological expertise to fields such as education and labour. This practical approach to psychological science has been characterized by historians as quintessentially Anglo-American, in contrast to the European tradition of conceiving of psychology as a way of doing philosophy.96 In the 1880s, however, the social utility of psychological knowledge was largely restricted to moral questions about the nature and limits of human freedom. To what extent did the body govern the mind? How did emotional states affect thought? Psychologists began to turn to evolutionary theory to respond to such questions, and one of the most influential representatives of such a turn was the American psychologist and philosopher, William James (1842–1910). James, like Herbert Spencer before him, made liberal use of evolutionary concepts in his claim that in order to be understood, consciousness must be interpreted in terms of its function. James was also one of the first to integrate Mosso’s work into his own argu-

Analogize and Experiment

109

ments. Several of Mosso’s experiments in Fear were cited by James in his Principles of Psychology.97 James’s essay ‘What Is an Emotion?’ appeared in 1884, the same year that Mosso’s Fear was published. James argued that the emotions were primarily affairs of the body, not of the mind; they were not quasi-intellectual perceptions of a fearful external object, but a mental interpretation of what the body did under threat. As such, sensations were prior to what were typically called ‘feelings,’ even though consciousness tended to reverse this priority: Our natural way of thinking about these standard emotions is that the mental perception of some fact excites the mental affection called the emotion, and that this latter state of mind gives rise to the bodily expression. My thesis on the contrary is that the bodily changes follow directly the perception of the exciting fact, and that our feeling of the same changes as they occur IS the emotion. Common sense says, we lose our fortune, are sorry and weep; we meet a bear, are frightened and run; we are insulted by a rival, are angry and strike. The hypothesis here to be defended says that this order of sequence is incorrect, that the one mental state is not immediately induced by the other, that the bodily manifestations must first be interposed between ... we feel sorry because we cry, angry because we strike, afraid because we tremble.98

This reversal of the fortunes of emotion fit in with James’s interest in function as an analytic axis for psychology. Emotions appeared in consciousness as a sensation of the body’s reflexive performance in any situation that was significant for survival. This reflex was, like all reflexes, a movement. And this movement must be detectable. The physiologist’s task was to enumerate visibly the subtle movements of the body that the mind understood internally as emotion. The crude morphological research for this project had already been done, and James approvingly cited Charles Bell’s Essays on the Anatomy of Expression in Painting as an example. But what remained was the study of the imperceptible changes that sensation provoked in our bodies, since James anticipated that this would furnish the proof that when we felt an emotion, what we were feeling was our body reacting. It was never the feeling that caused the reaction. ‘The researches of Mosso with the plethysmograph,’ James opined, ‘have shown that not only the heart, but the entire circulatory system, forms a sort of sounding-board, which every change of our consciousness, however slight, may make reverberate. Hardly a sensation

110

The Sleep of Others

comes to us without sending waves of alternate constriction and dilation down the arteries of our arms.’99 For James, the experience of emotion testified to the functional nature of consciousness. Mind was not an epiphenomenon, an incidental string of states that accompanied brain processes. Nor could it be explained by associationist psychology, which stipulated that every perception evoked a train of associated ideas. The body inevitably intervened, and Mosso’s work, thought James, could graphically reproduce this intervention, demonstrating that the work of consciousness had real and important consequences for human fitness and survival. Mosso was clearly thinking in similar terms. His understanding of fatigue as a sensation was inseparable from his teleological interpretation of fatigue. Fatigue had an evolutionary purpose: ‘What at first sight,’ Mosso argued, ‘might appear an imperfection of our body, is on the contrary one of its most marvellous perfections. The fatigue increasing more rapidly than the amount of work done saves us from the injury which lesser sensibility would involve for the organism’100 Psychology, physiology, and even historical analysis were beginning to converge around designs for the amelioration of human existence through scientific research, aspirations that were shared by the two educational psychologists (Margaret and W.B. Drummond) who first translated Fatigue into English. Social progress, as the gradual elimination of class difference, was a historical law for Mosso. But its conceptualization remained incomplete without an analysis of the fatigue that perpetuated such divisions. Calling for ‘fresh investigations ... made by independent men, by physiologists free from all preconceptions whether political, humanitarian, or social,’ Mosso described, as an example, the racial degeneration among Sicilian sulphur workers brought on by their endless fatigue. In their case, as in so many others, the development of industrial machinery had diminished the price of commodities but done nothing to relieve the misery of human labour. Yet Mosso had no truck with the politics of Marxist revolutionaries. The moral education of the masses through the dissemination of scientific knowledge would, Mosso hoped, prove a viable surrogate for political upheaval. It would increase the social sensitivity to the impending degeneration of a fatigued civilization, just as the sensation of fatigue in the individual inevitably prevented him from persisting in his destructive labour.101 By studying fatigue with graphical technologies and fusing the subsequent findings with evolutionary and even eugenic ideas, physiologists aspired to expand their domain beyond the medical realm to the classroom, the

Analogize and Experiment

111

factory floor, and even the political arena. But physiologists ultimately studied motion, not its absence. The muscular passivity that characterized sleep thus maintained sleep’s status as little more than an incidental diversion. Sleep in the Clinic? Liébeault, Bernheim, and Charcot Hypnotism offers a third example of how sleep was brought, albeit marginally, into the domain of late-nineteenth-century medical investigation. Sleep and hypnosis were generally thought to be virtually identical when A.A. Liébeault revived the clinical use of hypnotism in 1864. But, at that time, sleep was little more than a word that described a state in which the will languished. Hypnosis, then, was a state of sleep in which the hypnotizer’s will supplanted that of the subject through suggestion. Forty years later, Hyppolyte Bernheim, a devout follower of Liébeault, concluded that sleep and hypnosis were entirely unrelated: where true sleep was, hypnosis could not be, and where hypnosis was, there was nothing but suggestion. What remained of sleep after the debates over hypnotism subsided was dreaming, which will be the subject of chapter 3. The idea that pathological forms of sleep can suddenly intrude into the course of normal wakefulness is undoubtedly very old. In the late twelfth and thirteenth centuries, for example, the behaviour of ‘sleepers’ became an important problem for many theologians.102 Sleepers would fall into a trance-like state and act differently than they would when awake. They would perform violent or unusual acts, or suddenly became capable of doing things they could not ordinarily do. They also had little recollection of what they had when they returned to their normal state. Somnambulism, or sleepwalking, also took on a wide significance when A.M.J. Chastenet de Puységur introduced an induced form as a cure for nervous attacks. Puységur (1751–1825) was a disciple of Franz Anton Mesmer, whose work had launched a series of debates over the moral and therapeutic merits of ‘animal magnetism’ shortly before the French Revolution.103 Mesmer’s description of the therapeutic ‘magnetic passes’ said to cure his patients had much less to do with sleep than with what he took to be his discovery of a new natural force: a magnetic fluid that accumulated in animate and inanimate bodies and that could now be manipulated to restore and maintain health. Puységur, on the other hand, drew an analogy between the induced therapeutic state and natural sleep; indeed, he routinely described the cure as a magnetic ‘sleep’ or ‘artificial somnambulism,’ and compared it

112

The Sleep of Others

to the naturally occurring phenomenon of sleepwalking.104 Mesmer’s original appeal to physical forces had begun to give way. The title of one of the more notable treatises on magnetic cures indicates the new direction magnetic cures were taking: the miraculous cures of the Portuguese priest Abbé de Faria were described in his De la cause du sommeil lucide, ou Étude de la nature de l’homme (1819). Faria’s disciple, the French general F.J. Noizet, and his collaborator, the physician Alexandre Bertrand, continued and extended this analogy between sleep and magnetism in their own treatises. James Braid (c. 1795–1860), a Manchester surgeon, also rejected explanations based on magnetic fluids and forces. But his methods were also different from those of Faria and his followers. Instead of magnetic ‘passes’ over the patient, Braid had his subjects stare at bright objects until they entered a sleeplike state that, he claimed, was caused by a paralysis of the eye muscles.105 Despite the greater role that psychological or physiological explanations played in these various schemes, one thing remained constant: the intimate experience of sleep had replaced the objective notion of force as the dominant metaphor for understanding magnetic therapy. In turn, sleep was slowly refashioned as an investigative object as interest in hypnotism progressed over the course of the century. Scientific interest in hypnotism languished between the 1850s and the late 1870s, but its popular appeal, particularly in France, was maintained by Liébeault (1823–1904). Having taken his doctorate of medicine at Strasbourg, Liébeault opened a medical practice at Pont Saint-Vincent, just outside of Nancy. As a student, he had cultivated an active interest in animal magnetism, but his patients did not seem to respond to magnetic treatments. On 27 February 1860 he heard a lecture on ‘Braidism’ delivered to the Académie des sciences by Velpeau, one of several French surgeons interested in the phenomenon for its potential uses as an anaesthetic.106 Inspired, Liébeault began to experiment with the technique to cure disease by artificial sleep, rather than to block pain during surgery. In 1866 he published a monograph on the subject, which was reported to be so unpopular as to sell a single copy in ten years.107 Liébeault offered the foundations of a distinctively psychological approach to the study of the clinical phenomena of sleep. A comparison with the theories of Liébeault’s contemporary, William Hammond, serves to illustrate the differences between these two clinical approaches. Hammond’s theory of sleep, as we have seen, relied on accepted physiological principles – in particular, the reflex doctrine. Sleep appeared when brain activity, stimulus, and blood pressure all declined in concert. It was a pas-

Analogize and Experiment

113

sive, but necessary, withdrawal from the world. Disease appeared when sleep was neglected, often because of the nervous excitability that accompanied excessive cerebral blood circulation. Liébeault, in contrast, introduced the idea of ‘attention’ into his description of sleep, which he identified with ‘nervous force.’108 Attention was not diminished in sleep, but merely redirected, usually towards the idea of sleep. This redirection of attention had physiological consequences. When too much nervous force was located in one or another brain centres, various pathologies developed, as the force was conducted along the ‘grand sympathetic nerve’ to other organs of the body. Liébeault presented hypnotic therapy as a way to take advantage of the state of sleep, which was now provoked by a command from the hypnotizer, by correcting the pathological directions of the patient’s attention. In either instance, both doctors built a successful practice in which discipline played a crucial role. Liébeault enforced his from above, telling his patients, who were mostly peasants, how to think when their will was apparently suspended in hypnotic sleep. Hammond relied on his patients’ ability and desire to discipline themselves, providing his well-to-do New York clients with detailed regimens of electrotherapy, drugs, and suggestions regarding diet and hygiene.109 Liébeault would probably have been forgotten by medical history had he not received a visit from Hippolyte Bernheim (1840–1919) in 1882. A Strasboug physician, Bernheim, along with most of the university faculty, had been relocated to Nancy after the Germans annexed Alsace in 1871. He was convinced that hypnotic effects were not caused by a novel physical force or through normal physiological mechanisms. Rather, they were produced through suggestion, which relied on the transference of an idea from the experimenter to the hypnotic subject.110 To demonstrate the absence of truly physiological forces, Bernheim encouraged Liébeault to try his cures with ordinary water, which his patients were told had been magnetized. The cures worked, and the ‘Nancy School’ of hypnotism was born. Bernheim set his claims against those of the ‘Paris School,’ led by Charcot, who had been using hypnotism to illustrate his claim that the ability to fall into a hypnotic trance indicated that the hypnotized subject was actually a hysteric, even if other symptoms of hysteria were missing. Bernheim felt that anyone could be hypnotized: all that was required was a particular psychological relationship between the hypnotizer and the subject. This difference in approach could in some ways be linked to medical specialty. Diagnosis and prognosis were the first order of business for neurologists like Charcot. Actual cures were unlikely and remained so for decades there-

114

The Sleep of Others

after.111 Clinicians like Bernheim and Liébeault, on the other hand, worked primarily outside the asylum system and were much more likely to be inspired by the therapeutic potential of hypnotism. Although Bernheim initially drew the same parallels between hypnotism and sleep that Liébeault did, this soon changed. At first, he remained relatively faithful to Liébeault’s interpretation, arguing that hypnosis was a distinctive psychological condition that heightened the suggestibility of the subject. It could also produce sleep, which Bernheim viewed as a phenomenon of auto-suggestion.112 But, by the end of the 1880s, Bernheim’s success in producing all the key effects of hypnosis, such as hallucinations, by suggestion alone ultimately led him to develop a new theory. At the Twelfth International Congress of Medicine held at Moscow in August of 1897, Bernheim declared that hypnotism simply did not exist – there was only suggestion. The absence of some aspects of sleep in his hypnotized subjects had helped bring Bernheim to this conclusion. Some would fall into a deep state of sleep and forget they had been hypnotized. Some would even report having dreamed. But others remained fully awake and easily retained a complete memory of the events that had taken place while ‘hypnotized.’ The only thing these patients had in common was that they were all amenable to suggestion, something that the onset of true sleep, argued Bernheim, tended to eliminate. Bernheim was convinced that, while the images of dreams could indeed be manipulated by external stimulus, the classic signs of hypnotism (the sensori-motor phenomena of catalepsy, paralysis, and anaesthesia) could be created only when the subject was awake and capable of collaborating with the suggestions of the experimenter. Bernheim’s theory of suggestion was an expansionary one, in the sense that suggestion did not simply supplant hypnotism as a unique state of consciousness; it was a mundane part of life. The role sleep played as the normal analogue to the induced pathology of hypnotism shrunk in proportion. At the 1911 Congress of the International Society for Medical Psychology and Psychotherapy, Bernheim charged that ‘it would be an abuse of words to call it [suggestion] hypnotism, a word which involves the idea of sleep and the idea of a special state which would not be our ordinary state.’113 To confirm just how routine suggestion was, Bernheim turned to the very same methods that Marey and Mosso had used to render the experience of fatigue into an experimental object: ‘Without saying anything, I record [a waking subject’s] pulse on a cardiograph ... While the trace is forming I count the pulse loudly,

Analogize and Experiment

115

at first correctly. After a while I count more beats than there really are; and then less. Studying the trace later, I discover that the pulse rate increased during the accelerated counting, and diminished during the slowed counting ... without the knowledge of the subject.’114 Some participants at the Congress rejected Bernheim’s claims about the ubiquitous effects of waking suggestion. Others continued to insist that the phenomenon of post-hypnotic amnesia could still be used to distinguish hypnotism from suggestion. But, in truth, it did not really matter a great deal what anyone said about hypnosis by 1911. As a research program, hypnosis had been in decline since the 1890s. When confronted with a choice of either dispensing with hypnotic phenomena as illusory or confronting charges that they were morally manipulating their subjects, experimental psychologists opted for more conventional modes of investigation. But just as hypnotism, and with it, sleep, was becoming part of psychology’s past, physiologists began to take up sleep as an experimental problem that was distinct from fatigue and that could be studied with animal models. A popular review of sleep research illustrates once again the close correspondence between fatigue and sleep, particularly for reformers who (like Mosso) drew upon physiology to strengthen their social analyses. One of the first to make this shift was Marie de Manacéïne, a Russian educational reformer from St Petersburg. In 1890, she published a book on fatigue and its relationship to degeneration entitled Le surmenage mental dans la civilisation moderne. Two years later, she published a book on sleep, which, despite its title – Sleep: Its Physiology, Pathology, Hygiene, and Psychology – was overwhelmingly concerned with the importance of sleep hygiene and said little of its psychology. Variations of Manacéïne’s experimental methods, which will be examined in chapter 4, ultimately became the foundation of modern sleep research. Sleep and Memory: Delboeuf Sleep was, however, being fit somewhat uncomfortably into the discourse of memory through the study of dreams. As the nineteenth century wore on, memory was becoming an increasingly popular field for both physiological and psychological research, at both the professional and popular levels.115 ‘Organic memory’ was postulated as a support for eugenics and seemed to some to explain an array of physiological effects, from inheritance to habitual behaviours.116 Psychological memory, on the other hand, was rapidly starting to dominate the debates

116

The Sleep of Others

over hypnotism, hysteria, and the nature of mental illness. Joseph Delboeuf (1831–96), a professor at the University of Liège, was sucked into the vortex of hypnotism a few years after the ‘sciences of memory’ first made their appearance along with the revival of hypnotism in France. In 1886 Delboeuf visited Charcot at the Salpêtrière and Bernheim at Nancy to discover the truth about hypnotism.117 But a year before he made this trip, he published a book on sleep and dreams that featured an extensive review of some recent books on the physiology of sleep. The book’s curious title – Le sommeil et les rêves, considérés principlalement dans leur rapports avec les théories de la certitude et de la mémoire – indicated Delboeuf’s dedication to framing old problems of dreaming within a modern experimental context. In effect, Delboeuf championed the application of reason against (or better, in concert with) the psychological experimentalism that threatened to sweep away the old order of ‘armchair psychology.’ The book was an extension of a project that dated back to before 1879, when he published an analytic bibliography on theories of dreaming in the Revue scientifique. Delboeuf’s work was timely, since the intimate relationship between dreams, psychopathology, and identity was only just beginning to take shape around this time. One historian has even suggested that Delboeuf’s book had a substantial impact on Freud’s thinking, which will be examined in the following chapter.118 Nonetheless, it would appear that hypnotism supplanted Delboeuf’s interest in sleep. He published nothing further on sleep or dreams after his trip to the Salpêtrière, and instead became mired in the debates over the question of post-hypnotic suggestion and the moral dimensions of the use of hypnotic therapy in medical practice. In 1885, however, Delboeuf had physiologists in his sights. He argued that experimental physiology made a grave error by interpreting sleep in terms of fatigue. Sleep was not, he insisted, a state in which some unknown force appeared to ‘repair’ the body during the night. It was simply a passive state that followed the exhaustion of the senses: The function of nutrition in relation to sensation points us towards the cause of sleep and its periodicity. The nutriments that accumulate in the body serve or served, among other functions, to form the foundation of peripheral sensitivity. The latter loses its sensibility through its exercise; at last there comes a point where it can no longer contain the sensations and becomes, by consequence, incapable of reacting. Sleep seizes hold of us – sleep, the sign that there is a barrier between us and the external world. This period of torpor is used to reconstitute sensitivity, and, as this work

Analogize and Experiment

117

advances, sleep disappears, giving way insensibly to wakefulness. Sleep is not a function; it is a concomitant effect. It does not repair any force. The truth is that it appears when sensitivity is enfeebled, and it disappears when it is revived.119

The most interesting phenomenon of sleep, thought Delboeuf, was that it completely impaired the sleeper’s ability to doubt and simultaneously enabled the mind to evoke images taken from the past that had been completely abolished from waking memory. Experimental physiology, Delboeuf argued, could explain neither observation. In its place, he proposed an account of dreaming that was based on the notion that sensation fulfilled its function through a ‘fixation of force.’ An organism’s sensitivity was nothing more or less than its ability to transform external forces into memory traces that were ‘fixed’ in the organism and never could be erased. And, like all physical forces, the force that created memory tended towards a state of equilibrium. When an organism’s ability to fix sensation as memory became exhausted, a state of disequilibrium between the internal force of the organism and that of its environment ensued. This process ended in sleep, which featured a cessation of external sensation that caused the hallucinatory and memory-laden phenomena of dreaming. Despite the close linkage between Delboeuf’s analysis of sleep and the new science of energy, he drew evidence for his claims not from the study of musculature (as did Marey) but from a more traditional philosophical domain: sensation. ‘Natural or artificial,’ Delboeuf said, alluding to his interests in hypnotism, ‘sleep is always accompanied by an insensibility more or less extended, more or less profound. The cause of the one is the cause of the other.’120 Expanding his claims to encompass epistemology, Delboeuf concluded that dreams served to provide a native instance of ‘speculative doubt.’ On its surface, this doubt was insincere and superficial, since people seemed to believe in the reality of their dreams only while they were dreaming. But, in its social and historical context, doubt fuelled by dreaming proved persistent, and, by encouraging people to challenge accepted doctrine, promoted the health and well-being of the sciences. Delboeuf’s arguments indicate the extent to which dreams were gaining in philosophical and epistemological importance through their association with memory by the end of the 1880s. But was hypnotism – a word Braid had borrowed from Hypnos, the Greek god who brought sleep to humans – a kind of sleep or not? Yes and no, went the chorus.

118

The Sleep of Others

Bernheim’s final concept of ‘waking suggestion,’ which eliminated the hypnotic state altogether, smacked too much of irrationalism for most people. Albert Moll (1862–1939), a Berlin psychiatrist who, like Delboeuf, made pilgrimages to the Salpêtrière and Nancy, took a more tempered view of the relationship between sleep and hypnotism. But it turned on the question of memory, just the same. In 1889 Moll published a general survey of hypnotism – Die Hypnotismus – that was quickly translated into English, going through five editions by 1901. Moll is better known for his later work on sexual psychopathology, but he first made his mark by helping, along with Auguste Forel in Zürich, to disseminate hypnotic therapy among German clinicians. For the most part, Moll sided with the Nancy school. He rejected Charcot’s claim that hypnotism was a cryptopathology and insisted that the state was invoked by the hypnotist’s suggestion to concentrate on the idea of sleep. Moll disagreed, however, with Liébeault’s argument that the hypnotic state and normal sleep differed only insofar as one depended upon suggestion and the other did not.121 Moll found this misleading, because it assumed a unitary state when in fact there were actually two distinct stages of hypnosis: a light stage, which featured a loss of voluntary muscular movement, and a deep stage, which featured hallucinatory phenomena that were analogous to dreams. The lighter stage lacked the sense of fatigue and decreased mental activity that accompanied sleep; indeed, subjects frequently remembered their experiences while hypnotized. The deep stage, on the other hand, bore a much closer resemblance to true sleep, since external stimulus, be it in the form of physical events in a sleeper’s bedroom or suggestions from a hypnotist, both generated delusions. Here, Moll was simply following the path already laid out by Alfred Maury (1817–92), a professor of history and philosophy at the Collège de France, whose theory of dreams as the product of external stimulus had been all but codified by allusion to one of the many dreams he recorded in his Le sommeil et les rêve (1861): Dreaming himself a victim of the Terror in revolutionary France, he was chased, caught, tried, led to the guillotine, and executed. But when he awoke, he discovered that a curtain rod had fallen on his neck, which, he argued, had caused the chain of mental associations to unfold instantaneously, producing the dream-illusion that several horrific days had passed.122 As a psychiatrist (and not a historian), Moll worked not with the text of his own dreams but with his observation of patients in various states of hypnosis. He concluded that the sense delusions induced in deep hypnosis were analogous to the dreams that could be induced by physical stimuli

Analogize and Experiment

119

in sleep. The only difference was quantitative: suggestion in hypnotism was stronger than its physical counterpart in sleep.123 Moll did draw another important distinction between his work and that of the Nancy school: waking suggestion, he argued, was impossible. Bernheim had argued that hypnotism and sleep were effectively identical, and that the phenomena that had been wrongly attributed to hypnotism were rather the product of suggestion, which operated in everyday life just as it did in so-called hypnotic states. Moll, on the other hand, insisted that if hallucinations could be induced, the pathological state of hypnosis must be present. Hypnotism and sleep were analogical, not identical: Authors often confuse hypnosis with sleep in speaking of suggestions in the waking state. We have seen that the light hypnotic stages do not much resemble sleep; consequently we must not conclude that a state of contracture, &c., is, or is not, a hypnosis because it resembles sleep or not ... they [also] think that hypnosis is excluded in these cases of waking suggestion, because none of the usual methods of inducing hypnosis have been used. But the methods are not absolutely necessary for the induction of hypnosis ... If, then, we can do the same thing without any previous appearance of hypnosis, we must call the state a hypnosis all the same, particularly if there is subsequent loss of memory, which is generally the case in delusions of the senses. There has been a kind of hypnosis in both cases.124

Loss of memory had thus become the distinctive sign of the true (or deep) hypnotic state by the end of the nineteenth century. Its importance to the debates over hypnotism was so marked that people like Moll would even argue that, if amnesia were present, hypnotism must have taken place, whether or not anyone had intentionally induced it. Sleep, on the other hand, was retained as a sort of firewall against the artifice of hypnotism. Its very presence as a natural, but entirely unexplained, phenomenon provided the backdrop upon which a discourse of hypnotism and memory could be created. Moll and Bernheim both invoked sleep as a counterpoint to their ideas about hypnotism, as they elaborated them through the phenomena of hypnotic amnesia and waking suggestion, but neither attempted to investigate what sleep was, from either a psychological or a physiological point of view. When the use of hypnotism as an experimental technique imploded, the question of sleep became more attractive to those investigators who wanted to study the psychology of the unconscious from a physiological perspective.

120

The Sleep of Others

By the end of the nineteenth century, sleep was poised to become integrated with physiological investigation. Even while sleep’s clinical significance was in decline with the eclipse of Hammond’s nosology of ‘cerebral anaemia,’ it was brought, covertly, into the physiological laboratory through the growing epistemological power of the graphical method. Mosso’s study of fatigue had taken psychophysiological research to the borders of eugenics and social reform. Although sleep was only a marginal aspect of this research, it held a highly symbolic meaning for psychophysiology precisely because it offered a natural, mundane instance in which the body appropriated the mind with a lawlike regularity. The late-century revival of hypnotism and the corresponding interest in memory served to reinforce sleep’s symbolic status, but only by analogy. When hypnotism’s own significance dissolved, sleep’s importance was retained through the alliance forged between dreams and memory. Within a space of five years (1899–1904), the importance of both sleep and dreaming would finally be articulated in terms of function, in complete contrast to the position laid down by philosophers like Delboeuf.

3 The Ends of Darkness

Sigmund Freud was one of the first to argue that dreams served both a biological and a psychological function. His claim that the hallucinatory nature of dream images actually protected sleep inverted the convention that such instances of unreason were accidental or pathological but not purposeful. A neurologist by training, Freud rejected hypnotism as a method for studying madness, replacing it with a systematic interpretation of dreams in both a clinical and an autobiographical setting. Others offered alternative ways of rendering dreams into investigative objects. The French philosopher Henri Bergson, for example, rejected symbolic interpretation and argued instead that the disordered nature of dream images was the product of a mind detached from the pragmatic demands of waking life. Dreams did not serve a purpose in Bergson’s scheme, but they were an important source of evidence demonstrating that mental life was more than merely mechanical – it was itself purposive. A similar analysis was applied to sleep by Edouard Claparède, a Swiss psychologist. Sleep, he claimed, was not the mechanical product of the build-up of fatigue ‘toxins’ but rather an instinctive reaction of self-protection. Two Parisian psychologists, Alfred Binet and Nicholas Vaschide, championed their experimental efforts over the so-called ‘metaphysics’ of Bergson and Claparède. This dispute was ultimately resolved by Vaschide’s collaborator, Henri Piéron, who promoted sleep deprivation as a way of rendering sleep an independent experimental object while at the same time acknowledging its purposive nature. Sleep and dreaming had come of age.

From the Clinic to the Bedroom Despite its ambivalent reception when it was first published late in 1899, Freud’s most popular book, The Interpretation of Dreams, became an indis-

122

The Sleep of Others

pensable text for the study of dreaming within a decade. Books on dream interpretation had been a best-selling genre since the mid nineteenth century, but Freud’s was novel to the extent that his analysis incorporated biological principles in place of prognostication. The most important of these was teleology, a problem that dominated contemporary debates over the nature of life.1 Dreams, argued Freud, served a purpose. They protected sleep by disguising memories of desires that were normally aroused but that had been deemed illicit and repulsive by human civilization. Dreaming amounted to the necessary revival of such memories in a way that would not disturb the mind at rest. Although Freud’s work is often met with scepticism by neuroscientists today, the Freudian moment was absolutely crucial when viewed from the perspective of the rise of sleep research as an independent field. It introduced a functional account to sleep’s most mysterious mental phenomenon and thus established dreaming as a key problem for any physiological theory of sleep. After Freud, sleep researchers had to incorporate dreaming into their theories, rather than simply explain it away. Dreaming became an essential, rather than an accidental, aspect of sleep. But how did Freud start thinking about dreaming in the first place? Although he had received some training in physiology and anatomy as a medical student, Freud was no biologist. He was a neurologist, and, as such, he spent much of his time diagnosing patients and offering a slate of remedies, be they medicinal, behavioural, or any combination of the two. Many of his patients were well-to-do Viennese women suffering from some variant of hysteria. As discussed in chapter 2, this particular diagnosis had been linked to hypnotism by Charcot, but, like hypnotism itself, it had begun to fall into disrepute by the end of the century. Its symptoms were bewildering and protean, and many suspected that a large number of different diseases fell under the term ‘hysteria.’ Others felt that hysteria was little more than a word for a folie à deux that played out between a woman and her physician. Regardless of what hysteria actually might have been, it is clear that Freud’s ideas about dreaming were first formulated through listening to his hysterical patients describe their symptoms. This ‘talking cure,’ supposedly discovered by Freud’s colleague, Josef Breuer, was first presented in their joint 1895 project, Studies on Hysteria. Freud’s interest in the content of dreams must be set within his context as a clinician, listening to patients narrate their experiences. In his historical survey of the psychoanalytic movement, composed long after Interpretation of Dreams had gone through several editions,

The Ends of Darkness

123

Freud himself lashed his theory of dreams quite firmly to his clinical work with Breuer: ‘It was discovered one day that the pathological symptoms of certain neurotic patients have a sense,’ Freud declared. ‘On this discovery the psycho-analytic method of treatment was founded. It happened in the course of this treatment that patients, instead of bringing forward their symptoms, brought forward dreams. A suspicion thus arose that the dreams too had a sense.’2 But, in fact, neither Freud nor Breuer said very much about dreams in Studies on Hysteria. They said a good deal more about hypnotism. In perhaps the most famous case from Studies, that of Breuer’s patient ‘Anna O.’ (whose real name was Bertha Pappenheim), dreams played a quite minor role. Breuer described how his treatment of Pappenheim involved listening to her narrating ‘childish stories’ to him on a regular basis – usually one each day, two if she had missed a session. Breuer soon began to hypnotize Pappenheim, finding that, in this state, she would retell every event that had produced each one of her hysterical symptoms. After she related each story, her symptoms began to disappear, one by one, until she was finally cured. Unlike her earlier fantasies, these later stories were all verified by a meticulous diary that Pappenheim’s mother kept of her daughter’s behaviour. The cure, buttressed by the evidence of the diary, suggested that the cathartic action of remembering could cure hysteria. Psychoanalysis was born. The details of this case have been meticulously examined by Mikkel Borch-Jacobsen, a historian of psychiatry, who has found many of the claims to be suspect or even false.3 Pappenheim, it seems, was never cured of her hysteria through psychoanalysis. She actually returned to an asylum for treatment several times after Breuer’s treatment had finished. In the end, she seems to have successfully rehabilitated herself, eventually becoming a prominent philanthropist and social worker. Borch-Jacobson has also argued that Breuer cooked up many of the details of the case, recounted in print some thirteen years after the fact, on the insistence of Freud. Borch-Jacobsen’s interest is in the founding mythology of psychoanalysis as a therapy, so it is perhaps not surprising that he says nothing about Freud’s dream theories, which were then only tangentially related to treatment of neurotics. Yet the origins of these might well also have been misrepresented by Freud. In Studies, Breuer did record that Pappenheim’s ‘altered state’ in which she told her childish stories ‘may well be likened to a dream in view of its wealth of imaginative products and hallucinations, its large gaps of memory and the lack of inhibition and control in its associations.’4 But this was

124

The Sleep of Others

simply an analogy, and a commonplace one at that. Breuer said nothing about what Pappenheim herself actually dreamed. Freud’s own self-observations and self-analysis were a more likely source of his ideas about dreaming, and one that Freud himself acknowledged as significant. The same year that Freud and Breuer published Studies on Hysteria, Freud had a disturbing dream about one of his former patients – the famous ‘dream of Irma’s injection.’5 ‘Irma,’ described by Freud as a composite character, was a patient whom Freud had been unable to cure through psychoanalytic methods. In his dream, she appeared as a guest at a party held at Freud’s house. He chastised Irma for not accepting his proposed cure. She responded that her complaints, which included a pain in her throat, stomach, and abdomen, were more serious than Freud thought. Fearing he had overlooked some organic affliction, Freud proceeded (in his dream) to inspect his patient’s mouth. He found a white patch, along with greyish scabs formed in the shape of the turbinate bones (which lie just below the olifactory gland in the nose). Other physicians began to crowd around Freud, and all agreed that there was certainly an infection. Its source, Freud mused, was probably a dirty syringe used by his friend ‘Otto,’ who had given Irma an injection some time earlier. On its surface, the dream supported Freud’s theory that dreams were always the fulfilment of a wish, as Freud wanted to absolve himself of responsibility for Irma’s continued illness and ‘Otto’ appeared a convenient scapegoat. But, after a lengthy self-analysis, Freud became convinced that every aspect of his dream was the manifestation of a secret wish. Freud felt this dream to be so important that he returned to it over and over again in Interpretation of Dreams. In 1900 he openly fantasized, in a letter to Wilhelm Fliess (with whom he would soon break), that a marble tablet might one day be affixed to his house, stating, ‘Here, on July 24, 1895, the secret of the dream revealed itself to Dr Sigm. Freud.’6 Indeed, such a plaque did eventually appear in front of Freud’s residence, some eighty years later. Regardless of which of these two origin stories we follow, it is clear that Freud drew upon the well-established link between the delusions experienced in a hypnotic state and dreams. Memory played an integral role. Pappenheim’s cure was effected by her remembering, in her hypnotized state, all the details of each event that had then proceeded to generate each of her hysterical symptoms. Freud’s self-analysis equally relied upon memory, despite the fact that his dream was quite obviously the product of imagination. The dream made sense only when Freud related each element to people and events from his own past. As we

The Ends of Darkness

125

have seen in chapter 2, nineteenth-century investigators had related memory to dreaming and hypnotism in a variety of ways. The existence or absence of memory was frequently said to signify the existence of a particular state. Freud, on the other hand, did not so much treat these phenomena as signs of a particular state (of madness or of sleep) as analyse them as the products of a mental activity, that of remembering. And, since memory quite obviously served a purpose (we remember things in order to accomplish a goal), dreams, too, must have a function. ‘As we are accustomed to thinking teleologically,’ wrote Freud, ‘we are still likely to be attracted to those theories which include an insight into a function for dreams.’7 Purpose and Desire Dreaming and waking thought were intimately related for Freud. In fact, the dream itself was not even necessarily linked to sleep. This paradoxical position – that the dream’s function must be expressed in terms of its role in sleep even though the analysed dream was not necessarily the product of the sleeping mind – was forced upon Freud by the very nature of his object of study. Dreams narrated to others were stories of private experience. It was impossible to determining their veracity, or even to establish that they took place at any particular time, or in any particular state. The analyst simply had to assume that, whatever the dreamer said she had dreamed was, in fact, the dream itself: If anyone gives an account of a dream, has he any guarantee that his account has been correct, or that he may not, on the contrary, have altered his account in the course of giving it and have been obliged to invent some addition to it to make up for the indistinctness of his recollection? Most dreams cannot be remembered at all and are forgotten except for small fragments. And is the interpretation of material of this kind to serve as the basis of a scientific psychology or as a method for treating patients? ... We can help to overcome the defect of the uncertainty in remembering dreams if we decide that whatever the dreamer tells us must count as his dream, without regard to what he may have forgotten or altered in recalling it.8

Thus, Freud associated all dreams with the past, but as memory, not as event. Dreams were dynamic and changing, even (or perhaps especially) in their recollection. Yet they were always true, not because they recaptured, in some distorted way, an objective event, as did Maury’s

126

The Sleep of Others

dream of the guillotine, but because they expressed, through symbols, the permanence of hidden desire. They were true as meaning, not as record. Dreams were the active product of normal mental processes that went on all the time, not just during the night. Their function was to hide illicit desires from the sleeper’s mind by transforming these disturbing images – the ‘latent’ dream thoughts – into the more palatable scenes of the ‘manifest’ dream. Dreams demanded negotiation and interpretation. And, in this process of negotiation through free association, Freud argued that the dream process inevitably revealed itself, regardless of whether or not the story was truly a report of a dream. It was the narration of the dream as a process of self-representation that was paramount. Freud’s understanding of dreams hinged on the notion of interpretation. But, as he was at pains to point out in Interpretation of Dreams, his version of interpretation was quite unlike that of the ‘dream key’ books claiming to have discovered a code whereby each image of a dream held a distinctive meaning. Such books, following on the ancient practice of dream divination, usually took dreams to be portents of the future. But Freud’s target was the past. Why did dreams hide the truth about the past? And how did they accomplish this? Freud’s theory turned on the idea that every dream was the fulfilment of a wish. He had chosen the dream of Irma’s injection well – it graphically illustrated Freud’s wish to be considered a conscientious, careful, and successful physician. Dreams of falling, of failing an exam, or of close friends dying, however, presented a different situation. How could they be interpreted as wishes? All dreams, argued Freud, had a manifest and a latent content. The superficial appearance of dreams – the dream report – was not to be trusted. It did not represent the thoughts that gave rise to the dream, but instead was the product of the ‘dream work’ that served to repress those very thoughts which made up the latent content of the dream. The thoughts had to be hidden from consciousness because of their disturbing nature. They were typically (but not exclusively) expressions of illicit and immoral sexual desire that had their origins in the impulses of childhood. These impulses persisted in memory throughout adult life. Yet they were largely inaccessible to waking consciousness, since they had been completely repressed by an internal pattern of censorship that developed upon maturity. In sleep, this internal censorship was relaxed, and the pressure built up by repressed desire was allowed to escape, as a dream. There were actually two aspects of desire running through Inter-

The Ends of Darkness

127

pretation of Dreams. The first was the sleeper’s wish to continue sleeping, uninterrupted. The second wish was that of the unconscious, which wanted its urges to receive some sort of conscious representation. An expression of illicit desire might disturb the sleeper. Thus, the dream took unconscionable desire and cloaked it in the images of memory. Dreams thus served two closely related functions. On the one hand, they were the ‘guardians of sleep,’ in that they often took stimuli that might awaken the sleeper – a noise in the bedroom, or the feeling of hunger or thirst – and converted them into a story in which these sensations were either satisfied or portrayed as something that was part of the normal course of affairs, allowing the dreamer to continue sleeping. But they also protected sleep from a more threatening source of stimulus – the memory of repressed desire. By disguising objectionable desires in the neutral images taken from recent memory – the unnoticed events of the immediately preceding day, for example – sleep remained undisturbed even as the unconscious mind achieved a limited form of expression. Freud held fast to his functionalist account of dreaming, demonstrating how dedicated he was to adapting biological thought-styles to problems of mental representation.9 A teleological analysis of dreaming meant that ridiculous, bizarre, or frightening dreams could be analysed as wishes no less than dreams of explicit desire. In such instances, the desire was simply to continue sleeping, in spite of the outlandish images the unconscious mind brought to the fore. But such an approach implied clear, or even absurd, limitations to the dreamer’s testimony. In order to accept that the strange and sometimes frightening images of dreams actually encouraged sleep rather than inhibited it, the sleeper had to, in some sense, be aware that she was actually sleeping: What the Precon. often says to our consciousness when the dream goes too far, is: ‘Don’t worry! Go on sleeping. After all, it’s only a dream.’ Even though this is never uttered aloud, it is a good general description of the attitude of the dominant activity of our psyche towards dreams. I am bound to draw the conclusion that, throughout the entire state of sleep, we know just as certainly that we are dreaming as we know that we are asleep. We do not have to pay any attention to the objection that our consciousness is never directed to the one item – the knowledge that we are asleep – and that it attends to the other – the knowledge that we are dreaming – only on particular occasions, when the censorship feels as if it has been caught unawares.10

128

The Sleep of Others

Freud’s explanation of dreams laden with anxiety followed a similar pattern, insofar as he insisted that functional explanations were still possible in such instances where the dream awoke the sleeper. Dreams of falling or of loved ones dying, for example, were relatively rare – Freud insisted he had not experienced a real anxiety dream for decades and needed to reach back to a dream from his early childhood for appropriate analytic material.11 Rejecting the argument that anxiety dreams were a normal component of dreaming, Freud saw them as a pathological symptom. They indicated a lack of balance between the relative energies contained in the storehouse of sexual desire in the unconscious, and the preconscious mind’s ability to prevent this desire from erupting into consciousness. When the long work of the latter was overwhelmed by the former, anxiety dreams, which were equivalent to hysterical symptoms, were the result.12 Thus, the study of anxiety dreams belonged to psychopathology and was ‘entirely outside the psychological framework of dream-formation.’ By acknowledging that strange and frightening dreams could provoke a brief period of wakefulness and still retain their purposefulness, Freud successfully defused the argument that his theory could not account for all dreams because it could not explain anxiety dreams. The one belonged to normal psychology, while the latter served to indicate the presence of disease: ‘It is not actually the fault of the dream if, though normally the guardian of sleep, it has to appear as its disturber, and this need not make us doubt its fitness for its purpose. This is not the only instance in the organism where a useful arrangement becomes useless and disturbing as soon as there is some change in the conditions under which it comes into being; in these cases the disturbance at least serves the new purpose of signalling the change and summoning up the organization’s regulatory means against it.’13 All normal aspects of dreaming fell under Freud’s functional account. If, as in the case of anxiety dreams, the phenomena seemed to betray their assigned function, Freud simply declared the matter a problem for psychopathology. This was a remarkable shift from most earlier theories, in which dreams were taken to be degraded perceptions. Maury’s guillotine dream illustrated perfectly how the power of the waking mind was compromised in sleep. Such dreams clearly served no purpose, be it biological or psychological. No doubt, such a teleological perspective was probably far too reminiscent of the archaic mystical and religious rituals that occupied Maury when he wasn’t keeping his meticulous dream journal.

The Ends of Darkness

129

Freud’s Biology This is not to say, however, that Freud’s arguments were entirely without precedent. On the contrary, the last quarter of the nineteenth century saw several other biologically minded investigators getting into the game of dreams. The biological context of Freud’s ideas has been discussed at length elsewhere, and much of the evidence for the claims that he was a ‘biologist of the mind’ comes from his ‘Project for a Scientific Psychology.’14 Although Freud abandoned his ‘Project’ in 1895, his study of dreaming gave him the opportunity to revise some of the ideas first set out there. He eventually included many of them in the seventh chapter of later editions of Interpretation of Dreams. Repression, one of the most important concepts for Freud’s dream theory, had already appeared in the ‘Project.’ Repression drew upon the idea of energy, the science of which had been a Victorian invention.15 Maury had used the metaphor of the motor to frame his study of dreaming in 1865, arguing that ‘there is thus a motor that is only the vibrations of the nervous substance. This mysterious motor, in man, as in nature, unveils itself before experiment.’16 Freud tended to avoid such mechanical analogies and instead described the flow of psychical energy as a variant of the reflex concept. Every mental activity was accomplished through the transformation of energy that originated from outside the brain.17 Ideas became ‘invested’ or ‘charged’ with energy; memory, too, could be understood as an accumulation of energy. But memories differed from one another according to the amount of affect, or emotion, that was associated with them. Perceptions evoking very little attention or interest became memories with virtually no charge. Experiences or perceptions laden with emotion, on the other hand, held vast amounts of psychical energy. When these latter were repressed in memory, they represented a huge store of energy in the unconscious. The regulation of this energy was built upon the plan of a reflex apparatus, in which an excessive accumulation of psychic energy was experienced as pain and which therefore needed to be periodically released.18 Dreams provided a ‘safety valve’ whereby the energy accumulated in repressed memories could be released to relieve this built-up pressure. Freud justified his analytic turn from perception to memory by reference to the problem of the flow of psychical energy, which was always linked to purpose.19 Repression made consciousness possible by separating out memories from perceptions. The former were more subject to repression because,

130

The Sleep of Others

unlike the latter, they were unable to continually receive additional energy from the sensory organs. But, unlike animals, humans were able to invest memories with variable amounts of energy, thereby attracting conscious attention to them and changing the regulatory balance governed by repression. This ability could produce any number of neuroses, but it could also be harnessed for therapeutic means, through psychoanalysis. Normally, dreams were part of this regulatory balance, as they created meanings that were acceptable (if confusing) to consciousness while at the same time allowing for the release of pent-up energy. When this process went off the rails, the psychoanalyst was able to use the very same resource to create new meanings and restore a balance of psychical energy. Variations of Freud’s concept of psychical energy could be found in the work of many prominent nineteenth-century neurologists. Freud had, after all, studied under Theodor Meynert, one of the most important reflex theorists of the day, at his Institute of Cerebral Anatomy at the Vienna Medical School. Charles-Édouard Brown-Séquard, who succeeded Claude Bernard in the chair of experimental medicine at the Collège de France in 1878, was also spreading the concept of reflex theory among neurologists during the 1880s. We have already seen one application of the energy concept to the study of dreams in Delboeuf’s ‘fixation of force.’ Another can be found in a paper on dreams by the French biologist Yves Delage, whom Freud cited approvingly in Interpretation of Dreams. Delage (1854–1920) had started out as a zoologist, completing a thesis on the circulatory system of marine crustaceans in 1881,20 and he continued to conduct research at the Roscoff biological station in Brittany after becoming professor of zoology at the Sorbonne. Like many scientists of his day, Delage longed to make his work relevant beyond the strictures of experimentalism and the laboratory. To this end, he began to dabble in charity medical practice, published several novels under a pseudonym, and even tried his hand at poetry and philosophy. After experiencing a crisis in 1886, Delage began to use evolutionary concepts to expand the relevance of his biological research. In 1895 he founded a journal – L’Année biologique – to deal with just such questions of ‘general biology.’ Delage’s foray into the world of dreams appears to have been part of his attempt to render biological the complexities of human experience. In 1891 he published a paper on dreams in the Revue scientifique, a journal designed by its current editor, Charles Richet, for the sophisticated lay reader interested in science.21 In his article, Delage proposed that

The Ends of Darkness

131

dreams were in fact the nocturnal release of built-up sensory impressions had been received, but ignored, during the day. Dreaming thus allowed the mind to maintain an equilibrium of mental energy. For those who suffered from recurring nightmares or anxiety dreams, Delage suggested a concomitant therapy: such dreams could be eliminated by paying close attention to their every horrific detail, thus fixing them in conscious memory and thereby making them impervious to the dream process. Delage’s idea that unattended perceptions made up the bulk of the dream was welcomed (and adapted) by Freud, who also, as we have seen, emphasized the functional role of dreams in ‘discharging’ small amounts of energy as images. But Delage did not account for anything like the phenomena of free association that Freud had observed in his study of hysterics. The ‘talking cure’ brought these patients back to past events that were, in fact, filled with emotion: they were not the unattended perceptions of waking life. To explain the nature of this cure, Freud turned to another biological concept, the notion of recapitulation, which he described in terms of regression. Freud’s application of Ernest Haeckel’s ‘biogenetic’ law, which dictated that ‘ontogeny recapitulates phylogeny,’ has been discussed in considerable detail by others.22 In the wake of the defection of his star pupil, Carl Jung, from his close circle of disciples around 1912, Freud was obliged to begin to articulate a metapsychology in opposition to Jung’s theories.23 Like Jung, Freud used the notion of recapitulation to lend biological authority to claims that the normal psychological development of the individual was nothing less than an ordered progression, on a microcosmic scale, through the entire history of human civilization. But whereas Jung offered an interpretation of recapitulation in the tradition of German Romanticism, interpreting the symbols of dreams as archetypes ‘lodged in the soul,’ Freud tried to fit his ideas within a generalized physiological economy. He even went so far as to declare that the individual was nothing more than ‘an appendage to his germ-plasm.’24 As with many neurologists, Freud’s first serious introduction to evolutionary theory came through the work of the British neurologist Hughlings Jackson. Jackson had divided the brain into ‘integrative levels’ that represented evolutionary stages in the human nervous system: the cerebral cortex, which controlled language and thought, was the most recent development, while the brain stem, which regulated vegetative activities, was the most ancient cerebral formation. Neurological disease was described by Jackson as a process of ‘disintegration,’ in which

132

The Sleep of Others

damage to a higher, more complex system allowed the more primitive phenomena of behaviour to emerge uninhibited. Freud was enthusiastic about this convergence of evolutionary theory and neurological diagnosis; in 1891, for example, he used the concept of disintegration to explain the word-deafness suffered by aphasics.25 Freud’s work as a medical student in Ernst Brücke’s Physiological Institute in Vienna provided him with another venue to observe the intersection of the social and the biological. Freud met many people, like Breuer, who came to influence his thinking while at Brüke’s Institute. But it was Sigmund Exner, Brüke’s senior assistant and, ultimately, his successor, who probably exemplified the sort of biological reductionism to which Freud at one time aspired. Exner used evolutionary theory to shape social and cultural questions out of his neurophysiological investigations, and in 1891 he published his Outline to a Physiological Investigation of Psychical Phenomena. Like Angelo Mosso, Exner took fear as his paradigmatic example of a mental phenomenon that could be explained in physiological and evolutionary terms. He argued, for example, that fear of wild animals was an emotional reaction that, once learned in an effort to survive, could be inherited. But Exner went beyond an analysis of fear. He attempted to explain how the function of many emotions was not to protect the individual but to defend the group.26 In 1892 he published an essay entitled ‘Morality as a Weapon in the Struggle for Existence,’ in which he argued that moral sentiment had a physiological basis that always acts in favour of the community. The feeling of pride associated with the image of heroic action, for example, created a sense of duty, the sole funtion of which was to protect the group. This communal protection thus capitalized on the physiology of an individual’s emotions, even at the expense of the individual who might one day be encouraged to sacrifice his own life. Freud’s book on aphasia offered little in the way of teleological explanations, and the first edition of Interpretation of Dreams did not say much about evolutionary theory. But, when Freud began to embark on his metapsychology in the seventh chapter of the second edition (1909) of Interpretation of Dreams, he added some neurophysiological ideas that likely originated in his abandoned ‘Project.’ One of these was the idea that the sleep of the preconscious, which exhausted itself by performing its various functions of generating motor action, movement, attention, and thought during the day, needed to be protected. ‘Protection’ naturally invokes the idea of ‘defense’ – a term that Freud had been struggling with since 1894, when he began to approach the psychoneuroses

The Ends of Darkness

133

from the perspective of Abwehr, or ‘defence.’27 The biological need to protect sleep fit perfectly into his scheme to explain how dreaming helped to maintain the equilibrium of psychical energy. The concept of organismic defence, championed by prominent evolutionary theorists across Europe, allowed Freud to offer a functional account of dreaming in neurophysiological, psychological, and physiological terms.28 The proposal that dreams represented the fulfilment of a wish was by no means new; indeed, Freud acknowledged that at least six other authors had suggested such an interpretation.29 But the idea that every dream could be understood in such terms was quite novel. Such a generalization could be had only by grounding an analysis of dreaming in functional terms. To accomplish this, Freud borrowed concepts in neurophysiology and evolutionary biology. Drawing upon the former, he suggested that dreams defended the continuity of sleep that was necessary for the proper functioning of the nervous system. Taking from the latter, he depicted dreams as the residuum of earlier developmental periods, both of the individual and of the species. Freud’s approach was also innovative in his refusal to grant external stimulus an important role in the formation of dreams. Invoking function, of course, implied a goal. And the realization of a goal requires time. Thus, Freud needed also to draw upon the notion of memory as a means of characterizing the phenomena of dreaming as a temporal project. Of course, Freud was not the only biological theorist to conceive of dreams in terms of memory. But his quest to trace all dream elements back to associated memories offered a remarkable contrast to earlier physiological accounts, such as Maury’s, which explicitly eliminated (or at least minimalized) the significance of the passage of time in the process of dreaming. By relying on memory, Freud was able to construct an entire theoretical and therapeutic structure out of interpreting the contents of individual dream reports. This was starkly different from the approach of Freud’s contemporary, Henri Bergson, who took the form of the dream, rather than its content, as its most instructive element. The Threshold of Knowledge When Carl Jung attempted to disseminate Freud’s theory of dreams among a French audience through the pages of Alfred Binet’s l’année psychologique in 1909, Henri Bergson (1859–1941) was undoubtedly the most famous living philosopher in France,30 and arguably the most influential intellectual figure in Europe. Following the publication of

134

The Sleep of Others

his Creative Evolution in 1907, political, religious, artistic, literary, and philosophic circles in France, Britain, and the United States celebrated ‘Bergsonism’ as a liberation of subjectivity from the strictures of materialism. Among biological and psychological scientists, his work was typically received with caution, and even suspicion. But he could not be ignored, and numerous philosophers and scientists alike publicly condemned his work as anachronistic, delusory, and even dangerous.31 Despite his interest in the nature of the dream, Bergson was no psychoanalyst. Indeed, his 1901 essay on dreaming almost completely ignored Freud’s work.32 Yet even the subsequent translation of Interpretation of Dreams seemed to do little to change Bergson’s mind on the subject, since his 1919 revision of the essay, which included some substantial changes, included only a footnote stating that the tendency to dream about insignificant events of the preceding day was more or less what Freud had argued about ‘repression.’33 Although Freud and Bergson shared many interests in common, I will instead emphasize their differences. As far as the problems of dreaming and sleep are concerned, Bergson, I will suggest, is best read as a counterpoint to Freud, one that uses quite opposite arguments to forge a harmony of sorts around the novel problem of dreaming’s purpose.34 Although he was neither a clinician nor an experimental psychologist, Bergson, like Freud, arrived at the problem of dreams by first tinkering with hypnotism. He seems to have taken an active interest in this phenomenon around 1883, while teaching at the École normale.35 In 1886 he studied two boys who had been credited with impressive telepathic powers, only to conclude that the boys had actually experienced a heightened state of sensitivity that could be replicated under hypnotic suggestion. But he soon came to the view that such psychological experimentation was of limited value in understanding the nature of mind. Indeed, Bergson contended that the role of psychological experiment was largely to establish the limits of a science of mind, rather than to supply its data. Time and Free Will (1889), his first major publication, featured an extended polemic against experimental psychology in general, and psychophysics in particular. Psychophysics attempted to establish the law-like relationship between stimulus and sensation using the method of ‘thresholds.’ Experimenters varied the intensity of a stimulus (a light or colour, the distance between two points of a compass applied to the skin, or a weight held in the hand) and then asked their subjects to judge the minimum amount of change they could perceive in the stimulus. At a certain point, subjects were unable to discriminate

The Ends of Darkness

135

between two physically different stimuli, and the resulting difference, after being averaged over a large population of experimental trials and subjects, was taken to be a fundamental unit of sensation. A personal experience – sensation – could now be expressed as an objective and measurable quantity. Bergson claimed that the entire program of psychophysics was in error, because it confused magnitude with intensity. The former was a measure of some thing, the units of which were all identical to one another. The tone of a sound, as determined by its frequency, was, for example, always the same, regardless of its volume. The latter category, however, involved the comparative strength of subjective, experienced qualities, for which there was no universal measure, even in the same subject. Feelings, be they joy, sorrow, fear, pain, or muscular effort, did not have a measurable minimum difference, argued Bergson, because there was no ‘identical residuum’ remaining after their qualitative differences had been eliminated.36 Bergson did not deny that psychophysical experiment successfully established relationships between physical stimuli and subjective reports. What he did argue was that such reports were not indicative of the nature of lived experience. Rather, psychophysics exemplified a more comprehensive and pervasive epistemology that treated knowledge in its practical dimensions only. Bergson’s criticism of psychological knowledge, then, was also a critique of the expansive tendencies of modern science, which Bergson codified as the persistent attempt to represent time as space.37 Time, Bergson insisted, is perceived directly by the soul as ‘real duration.’ It is a quality that mediates the intensity of sensations. Space, on the other hand, is pure quantity, the essence of which is objectification. Space enables subjects to arrange the world as a collection of things that stand in relationship to each other. Psychological science taught that these ‘things’ included ideas that followed each other in temporal succession and that could be analysed as though they were equivalent to the objects in space that were manipulated (and so successfully exploited) by Victorian science. Bergson argued that ideas were not so separable. Even the simply act of rising to open the window, and then, upon standing, suddenly forgetting what one wanted to do, could not be successfully analysed into two discrete ideas – one of an end to be attained and the other of a movement to achieve this end. The idea of the end had itself ‘tinged with a certain colouring the mental image of the intended movement.’38 In other words, the original intention persisted in the movement itself. The two were bound up together as a purposeful act.

136

The Sleep of Others

Ideas, then, needed to be understood and analysed in their dynamic temporal context, rather than depicted as discrete bodies occupying a unique space. Dreaming, it seemed to Bergson, provided a ready example of the importance of such a metaphysics. Evocative dreams, he argued, were like passionate feelings, in that they ‘coloured’ all aspects of memory.39 They were closely related to the physiological state of sleep, which ultimately shaped the very essence of dream images. The relaxation of organic functions that took place in sleep initiated two related phenomena: the suppression of external sensation forced the mind to draw upon memory to supply it with images, and the organic separation from the world of activity prompted the transformation of the experience of time. The rigid, homogeneous, and successive instants that made up the time of waking life became reworked in sleep as the fluid, interdependent, and intertwined order of events that characterized dreams. Dreaming became, for Bergson, a primitive experience of ‘real duration.’ These ideas were substantially expanded in Matter and Memory, Bergson’s 1896 critique of neurophysiology and its concomitant doctrine of brain localization. Did memories reside in a particular place in the brain? Once again, Bergson challenged a scientific project by arguing that the question upon which it was based was in error. The sensori-motor responses that neurophysiologists and neurologists relied upon in their experiments could not, Bergson argued, signify the presence or absence of memory, since the two fields were radically different facets of human existence. The ontological status of memory and (organic) matter were incommensurable; that is to say, memory existed independently as image, liberated from all physical constraints, while matter was completely governed by physical and vital forces. The relationship between the two was mediated by the necessity of engaging the world through purposeful activity. Bergson visualized the structure of memory as ‘an inverted cone,’ an image he would use throughout much of his later work (see fig. 8). At the top of the figure (the base of the cone), mental states were made up of an ‘expansive memory’ filled with images and completely removed from the world of sensation and activity.40 At the bottom of the figure, the cone of memory was focused to a point, representing how memory became enrolled in the service of perception, sensation, and muscular response. This was the ‘plane of action.’ For a history of sleep and dreaming, the point here is not to defend, challenge, or even explain the details of Bergson’s arguments, which were sometimes complex to the point of obscurity. The real problem at

The Ends of Darkness

137

8 Bergson depicted the phenomena of memory as an inverted cone with its apex touching a plane. The cone (SAB) represented the totality of one’s recollections fixed in memory. One’s immediate representation of the world (the plane P ), on the other hand, was dynamic, always moving away from the base of the cone. The two met in the phenomena of sensation and movement (S), the point at which memory was focused upon the pragmatics of perception and action. For Bergson, detatching oneself from this sensori-motor engagement with the world constituted dreaming. (Henri Bergson, Matière et mémoire: essai sur la rélation du corps à l’esprit [Paris: F. Alcan, 1896], 165)

138

The Sleep of Others

hand is to explain how and why he used dreams to make his claims comprehensible to the vast lay audience that he ultimately engaged. We need to remember that, just as Victorian scientists aspired to professional status by adopting various strategies of popularization, so, too, did philosophers justify their existence by wading into mass culture.41 It was not, after all, a scientist, but this particular philosopher whose popular lectures at Columbia University in 1913 were credited with having caused the world’s first traffic jam in New York City.42 An appeal to dreaming was a fundamental component of Bergson’s popularity. The ‘expansive memory’ entirely divorced from sensation and activity was dubbed by Bergson ‘the plane of the dream.’ Bergson thus contrasted dreaming to what he called ‘attention to life’ in the ‘plane of action.’ Memory played a dramatically different role in each plane. In the dream, memory supplied contemplative images, entirely separated from all movement. These images were thus divorced from the measurable experience of time. ‘A human being,’ declared Bergson, ‘who should dream his life instead of living it would no doubt thus keep before his eyes at each moment the infinite multitude of the details of his past history.’43 The plane of action, on the other hand, featured the body dictating the use of memory in turning sensation into perception. Contiguity (the ability to perceive the continuity of phenomena over time) and similarity (the ability to recognize an object as being ‘like’ another) relied upon memory, but without it being recognized as such. Choice and action were inconceivable without the body. ‘Our body,’ Bergson wrote, ‘with the sensations which it receives on the one hand and the movements which it is capable of executing on the other hand, is then, that which fixes our mind, and gives it ballast and poise.’44 Attention was the result of the sensori-motor system focusing memory. Sleep represented a relaxing of this tension, thus engendering the random memory images of dreams. To defend this notion of sleep as relaxation Bergson appealed to the theories of sleep advanced by the Spanish neuroanatomist Santiago Ramón y Cajal. This theory, also known as the ‘amoeboid theory of sleep,’ argued that the synapse – the tiny distance between one neuron’s axon and another’s dendrite – grew larger during sleep, which accounted for the diminished sensori-motor activity of this state.45 Bergson recognized that this theory was controversial, but he was nevertheless fascinated by the idea that the nervous system might consist of ‘everywhere conducting lines, nowhere any centers.’ It suggested that sleep could be defined as ‘a functional break in the relation established in the nervous system between stimulation

The Ends of Darkness

139

and motor reaction.’ He then combined this theory with two well-worn ideas: that sleep was caused by fatigue; and that dreams were akin to insanity. The result, he thought, confirmed his opposition between dreams and action. I quote Bergson at length only to demonstrate how closely related these associations were in his work: it appears more and more probable that this relaxing of tension in the nervous system [sleep] is due to the poisoning of its elements by products of their normal activity accumulated in the waking state. Now, in every way, dreams imitate insanity. Not only are all the psychological symptoms of madness found in dreams – to such a degree that the comparison of the two states has become a commonplace – but insanity appears also to have its origin in an exhaustion of the brain, which is caused, like normal fatigue, by the accumulation of certain specific poisons in the elements of the nervous system. We know that insanity is often a sequel to infectious diseases, and that, moreover, it is possible to reproduce experimentally, by toxic drugs, all the phenomena of madness. Is it not likely, therefore, that the loss of mental equilibrium in the insane is simply the result of a disturbance of the sensori-motor relations established in the organism? ... If our analyses are correct, the concrete feeling that we have of present reality consists, in fact, of our consciousness of the actual movements whereby our organism is naturally responding to stimulation; so that where the connecting links between sensations and movements are slackened or tangled, the sense of the real grows weaker, or disappears.46

Philosophers have occasionally defended Bergson against the charge that he ignored psychoanalysis.47 But, given Bergson’s deep interests in dreaming during the late 1890s, as well as his enormous popularity in France and abroad, we might just as well ask why it is that Freud says nothing about Bergson. Freud, of course, has many detractors, but a historically minded philosopher has argued that Freud’s theories about the mind were not ‘pseudoscientific,’ because they rested upon a number of theories, drawn from several disciplines, which were acceptable science in Freud’s day. From psychology, Freud borrowed ideas about ‘associationism, developmental psychology, reflexology, evolutionary and functionally distinct faculties, and the project of tracing human mentality and civilization to its prehistoric origins.’48 Yet Bergson was equally engaged with many of these fields between 1895 and 1907. His work also seems to have been well known to Carl Jung.49 It is beyond the scope of the present study to examine the reasons why Freud might have ignored

140

The Sleep of Others

Bergson. My point is simply that, around the turn of the century, Freud’s was not the only, or even the most prominent, theory to deploy an analysis of dreaming as the basis of a critique of modern culture. Dreaming and Spiritualism To this point (1896), Bergson had made substantial use of dreaming as an introspective vehicle that could conveniently convey his ideas to his reader. Regardless of whether his readers had any experience or knowledge of madness, hypnotism, or neurophysiology, they doubtless knew something of dreaming. But, although he had incorporated it into his philosophy, Bergson had yet to make a thorough analysis of dreaming. This was to come in the wake of a growing interest in occult phenomena, not just on Bergson’s part, but within much of Western Europe. Bergson made his first formal and public commitment to psychical research by joining the Institut psychologique international and the Institut général psychologique in 1900.50 Of course, the institutionalized study of ectoplasmic apparitions, table-turning, telepathy, mediums, and mind-reading considerably predates Bergson’s entry into the field. In Great Britain, the Society for Psychical Research (SPR) had been formed in 1882 by a group of scientists and philosophers connected with Trinity College, Cambridge. Their membership included several who had been part of the many spiritualist societies that had sprouted up across Britain since the 1850s. The question of survival – not that of the species, but of the soul after death – had loomed large in the SPR’s research agenda since its inception, but by the 1920s attention began to turn towards more ‘respectable’ pursuits that, it was hoped, would be more amenable to conventional laboratory techniques of measurement and intervention.51 Dreams were just such a target. In France, the systematic study of occult phenomenon arrived much later. Charles Richet, the doyen of physiological knowledge in France, helped create the Institut psychologique international in 1900. Aside from Bergson and Richet, its membership also included Pierre Janet, Cesare Lombroso, and William James.52 Bergson was a close friend of Richet, and both shared an enthusiasm for the experimental and graphical demonstration of occult phenomena. Indeed, the enduring champion of the graphical method, Marey, had been, with Bergson, a member of a special group of the Institut général psychologique commissioned to study psychical phenomena.53 In fact, at one seance that Bergson attended, an examination of Eusapia Palladino, a Neapolitan

The Ends of Darkness

141

medium famous for her supposed telekinetic powers, a recording device was apparently put to use. The description of the seance is brief, but its results – the destruction of the recording instrument in the middle of the experimental session – seems to strangely reflect Bergson’s own ambivalence about Marey’s research: ‘Séance 1905–V–10–11. Rupture of a rubber tube connecting the Marey balance to a recording tambour, just as Eusapia broke a red pencil. (Controllers: on the left, M. Chapentier; on the right, M. Bergson).’54 Bergson was not an experimentalist by trade. He was a philosopher and thus relied more on logical and linguistic analysis than experimental demonstration. Dreams were thus a natural point of entry for a study of the occult that would, at the same time, allow Bergson the opportunity to defend the traditional introspective methods then threatened by experimental psychology. Bergson’s interest in dreams was brought to a focus at the Institut psychologique international. He was asked to compose an address on this topic, and he presented it before a crowd of more than five hundred on 26 March 1901.55 Bergson took this opportunity to clarify his position on the relationship between sensation and memory. Citing Hervé de Saint-Denis, Alfred Maury, Philippe Tissié, and the American psychologist George Trumbull Ladd, Bergson contradicted what he had said in Matter and Memory, and argued that all dreams originated in sensation. Dreams were fabricated out of vague sensations experienced in sleep and given distinct form by images drawn from memory.56 This process was not substantially different from that of waking perception, which invoked the clarity of memory to fill in the vagueness of sensation. Unlike his analysis of dreams in Matter and Memory, Bergson’s new position entirely abandoned any close-knit relationship between dreams and madness. The psychology of sleep thus took on a new significance in Bergson’s analysis, but the physiology of sleep was not of any great moment to this cause.57 Rather, the reconstruction of the mind’s activities upon awakening was the key to understanding the nature of dreaming, and this could be accomplished by traditional philosophical introspection.58 To this end, Bergson described a dream of standing at a podium, addressing an assembly. The noise of Bergson’s fantasized audience grew louder and louder until the room filled with the rhythmical cry of ‘Out! Out!’ He then awoke, only to hear a dog barking in his neighbour’s garden.59 Naturally, the dog outside Bergson’s window spoke French, and its yelps of ‘Ouâ, ouâ’ appeared in Bergson’s dream as ‘A la porte! à la porte!’ The felicitous English translation rendered

142

The Sleep of Others

‘arf! arf!’ into ‘out! out!’ Aside from Bergson’s unusual interest in dogs (he actually preferred the company of his two cats), the dream was so self-referential that it seemed almost wholly contrived.60 But such a dream was well in keeping with the French tradition, which stretched at least back to Descartes, of deploying dreams in the service of philosophical points regarding the nature of experience. From this perspective, Bergson launched into his analysis, which, in a veritable mockery of some strains of experimental, objective psychology, took the form of a waking consciousness interrogating a dream consciousness.61 The only difference between the two, Bergson concluded, was the amount of energy the former expended in making the perceptions that were the ‘perfect fit’ between sensation and memory. This energy, which Bergson described as the attention that held back the floodgates of memory, began to lag over the course of the day, producing the fatigue that ended in sleep: ‘Your life in the waking state [argued the ‘dream consciousness’] is, then, a life of toil, even when you suppose you are doing nothing, for at every moment you must choose and at every moment you have to exclude ... You choose among your memories, since you reject every recollection which does not mould itself on your present state. This choice which you are continually accomplishing, this adaptation ceaselessly renewed, is the essential condition of what you call common sense ... But it fatigues you in the long run. Common sense is very fatiguing.’62 The dream consciousness, on the other hand, ‘did nothing,’ because it was completely detached from waking life. ‘Dormir c’est se désintéresser’ (to sleep is to lose interest) quipped Bergson – a phrase, he noted in later editions of this essay, that had been picked up by other theorists, including the Swiss psychologist Edouard Claparède, discussed later in this chapter.63 The idea that sleep was identical to a withdrawal from life explained, thought Bergson, a number of distinctive features of the dream. The instability of its images indicated that the strength of association made between two ideas was correlated to the strength of the will. Without interest, associations between ideas were little more than random, as they seemed to be in dreams. The rapidity with which the dream unfurled was further evidence of the mind’s detachment from one’s environment and the ‘homogeneous’ time of the world. Finally, Bergson adopted Delage’s theory that the dream consciousness had a marked preference for the insignificant, unnoticed details of the past day. If dreams were about events of grave concern, this sleep tended to be very fatiguing. Normal sleep, on the other hand, typically featured

The Ends of Darkness

143

images that had passed through waking consciousness and aroused little or no attention.64 Like many of his contemporaries, Bergson had begun to take a deep interest in applying evolutionary concepts to questions about the mind. But, unlike Freud, Bergson never seriously applied such concepts to the problem of dreaming. Whereas Freud regularly invoked ideas of recapitulation and regression in his later writings on dreams, Bergson made only incidental references to dreams in his biological opus, Creative Evolution (1907). Why this lacuna in the midst of the growing popularity of psychoanalysis? Why didn’t Bergson tie the problem of dreaming to biological theory? For Bergson, such an alliance was a non-starter: dreams could play no role in evolutionary theory, simply because dreams served no biological or even psychological function. His critique of associationism – a position that was picked up by several psychologists studying sleep – also made it impossible for him to accept the psychoanalytic method of free association. Dreams were merely the random selections of the mind taken from memory. They were useful insofar as they could help the introspective philosopher illustrate the nature of time and memory, but they served no organic purpose. As a consequence, dreams could have no place in evolution; nor could they have any significance for functionalist psychology. They were the marginalia of human existence, valuable only for their ability to fix the modern and scientific vision of time against a backdrop of the ‘real duration’ that is the experience of time. For Bergson, dreams represented the limits of scientific explanation; they were not its future frontier, waiting to be conquered. It was as a limiting case for science that Bergson’s position on dreaming, and his philosophy in general, achieved its popularity. His lecture remained popular long after it was first published in 1901. Several months after Bergson began his series of lectures at Columbia University in February of 1913, Edwin Emery Slosson, a chemistry professor and science popularizer, translated and published the essay on dreams in his daily, the Independent.65 The New York Times carried numerous feature articles and editorials on Bergson, particularly after the pope denounced his philosophy in August 1913.66 The Times took a rather sceptical position towards the French philosopher, fearing American readers might be taken in by Bergson’s continental charm.67 Slosson, on the other hand, insisted that the importance of Bergson’s work far exceeded that of Freud. The question of dreams was in vogue again, argued Slosson, who had toured Europe and interviewed Bergson in 1910, and ‘the cause of this revival of interest is the new point of view

144

The Sleep of Others

brought forward by Professor Bergson.’68 Slosson set Bergson beside Maurice Maeterlinck, Henri Poincaré, Élie Metchnikoff, Wilhelm Ostwald, and Ernst Haeckel as one of the most important ‘living prophets.’ A devout Congregationalist, he was at pains to show that Bergson did not in any way reject the validity and importance of scientific knowledge. On the contrary, he portrayed Bergson as a spiritualist for a scientific age, offering an approach to dreaming that both avoided the ‘excesses’ of psychoanalysis and would appeal to the ‘pragmatic genius’ of the American people.69 There was even room, Slosson noted, for the survival of the soul after death in Bergson’s philosophy, something that Bergson himself had indicated in his presidential address to the SPR in London in 1913. After the English translation of Creative Evolution appeared in 1911, Bergson was reputedly selling more books in Britain than he was in his native France.70 Whether his ideas were accepted or not was another matter. Charles Sherrington (1857–1952), the most important neurophysiologist in Britain at the time, offered the following assessment in a letter written to his former student Alexander Forbes: ‘As to that book on Bergson I thought the introduction by Ray Lankester so good that you might enjoy looking at it. But the actual text itself was much inferior to the introduction. What I have read of Bergson I find rather nebulous – the problem of the mind & matter nexus seems so insoluble that it seems for our epoch unattractive. One thing appears clear I think, viz. mind is not an epiphenomenon & no science – least of all physiology of nervous system – should treat it as such.’71 Sherrington was undoubtedly referring to E. Ray Lankester’s introduction to Hugh Elliot’s 1912 demolition of Creative Evolution, entitled Modern Science and the Illusions of Professor Bergson. Elliot, a former military man, wrote popular tracts on evolutionary theory and had produced a biography of Herbert Spencer and a translation of Lamarck’s Philosophie zoologique. Lankester, on the other hand, was a serious morphologist and evolutionary theorist. Sherrington, who would have been the last to argue that there were no limits to what experimentalists could reveal about the mind, seems to have felt that Bergson’s work amounted to little more than meaningless excursions into metaphysics. Doubtless, many biologists and physiologists shared this opinion, but some did not. Shorn of its affinities with madness, Bergson’s vision of dreaming illustrated the need for a psychology of sleep. Indeed, it was Henri Piéron, a young psychologist, who pursued the physiological research necessary to assess the value of Bergson’s depiction of sleep as a disinterested state

The Ends of Darkness

145

(see below). Freud’s influence on the physiological study of sleep actually arrived much later, once psychoanalysis began to gain influence in the United States during the late 1920s and 1930s. This irony – that the ‘anti-scientific’ Bergson should shape the direction of scientific research, while Freud, the ‘biologist of the mind,’ failed to make an immediate impact – has been missed by historians who have failed to grasp the context in which these figures were writing. Bergson’s theory appealed to the crypto-religious spirit of his age, which wanted to preserve a sense of spirituality and freedom while at the same time pursuing a pervasive analysis of the mind as a pragmatic phenomenon oriented towards organismic survival. For Bergson, dreams were the soul’s last stand in the midst of la belle époque. He drew upon the distinctive flow of their images as an exemplary form of subjectivity, an existential state that could not be grasped by a psychology that took the function of consciousness as its object. Sleep and the Concept of Function in Psychology The question of sleep had been left mainly to physiologists and neurologists throughout the nineteenth century. Neurologists studied sleep as a crypto-pathology, while physiologists depicted sleep as the mechanical result of an exhausted body. Neither discipline took sleep to be anything other than the passive consequence of fatigue. Sleep was not a biological object in the sense that it represented an active, organic process. It was merely the absence or diminution of such a process. Delboeuf said as much in 1885, when he rejected the idea that sleep could even have a function (see chapter 2). Such a definition of sleep left only dreams for psychological research. The most influential experimental psychologist of the century, Wilhelm Wundt, devoted a special section of his Principles of Physiological Psychology (1873–4) to dreams. But Wundt treated dreams as a species of hallucination, not as a phenomenon dependent for its existence upon a normal physiological state. One of Wundt’s many disciples, Paul Radestock, published a monograph on the subject five years later. But despite its title – Schlaf und Traum (sleep and Dreams) – only one chapter out of ten attempted to provide a detailed account of sleep. The Anglo-American heir of the Wundtian experimentalist tradition, Edward Bradford Titchener (1867–1927), narrowed the psychological study of sleep and dreams even further. Titchener’s numerous textbooks hardly mentioned dreaming at all, other than to include it, along with hypnotism and insanity, as a curiosity of abnormal

146

The Sleep of Others

psychology. Like Delboeuf, these psychologists took sleep to be little more than the physiological backdrop for dreaming. Apart from dreaming, the ‘psychology of sleep’ held little meaning for a science with experimentalist pretensions. Dreaming and sleep were subsequently excluded from Titchener’s treatment of the psychological structure of the normal adult mind, which was typically studied through some variety of self-observation. This situation slowly began to change as psychologists began to incorporate evolutionary concepts into their research. The tradition of describing psychological phenomena, such as emotion and sensation, as purposeful phenomena that contributed to the survival of the species can be traced back to Charles Darwin’s The Descent of Man and Selection in Relation to Sex (1871) and The Expression of the Emotions in Man and Animals (1872). But, in the former book, Darwin made only a few cursory remarks about the dreams of animals, which, he thought, might indicate their capacity for imagination and perhaps even a rudimentary form of self-reflection. In the latter work, he said nothing at all of dreams. Social scientists inspired by Darwin’s theory of natural selection, on the other hand, had a great deal to say. In Researches into the Earliest History of Mankind (1865) and Primitive Culture (1871), the British anthropologist Edward Burnett Tylor identified dreaming as one of the origins of the human belief in the soul, which he dubbed ‘animism.’72 The sociologist Herbert Spencer made similar claims in his popular Principles of Psychology (1870–2). In both instances, dreams were invoked as a way of locating the origins of the religious sentiments that many Victorian scientists hoped to purge from their professional and national culture. Animism, revived in Tylor and Spencer’s age as ‘spiritualism,’ was an example of what Tylor called ‘a survival,’ or a relic of archaic and primitive forms of thought that inhibited social and intellectual progress. Given such an account, the depiction of dreams as functional was hardly possible. Their persistent cultural status as mundane manifestations of the spirit world was thus seen by aspiring professional social scientists as nothing more than evolutionary residue that would soon be expunged. A thoroughgoing commitment to functional explanation – the notion that psychological phenomena should be explained in terms of their role in human survival and evolution – has traditionally been recognized as a distinctively American contribution to psychology. Edwin Boring, in his classic History of Experimental Psychology, offered the commonplace that the United States was naturally receptive to functionalism, because

The Ends of Darkness

147

it was still a new country in the late nineteenth century, filled with the ‘pioneers’ spirit’ and ‘the belief that usefulness is the chief good.’73 The rapid development of psychology as an independent discipline in the United States is probably a more important factor than national character, however. As numerous historians of psychology have argued, the practical import of psychological phenomena was an integral feature of American psychologists’ defence of their disciplinary identity, which had its basis in applications in business management, industry, education, and the military.74 Many American psychologists, even those who were not actively engaged in applied research, would describe themselves as ‘functionalists,’ insisting that consciousness was not an epiphenomenon but rather the practical invention of evolutionary progress. The utility of consciousness reflected American psychologists’ own belief in the practical value of their professional knowledge. Functionalist thinking about sleep, however, did not begin in the United States, despite the astonishing level of professional and academic development that psychology had reached there by the first years of the twentieth century. Nor did it begin in Germany, the ancestral home of experimental psychology. It began, instead, in France, where psychological research was comparatively cut off from the laboratory and dominated by clinical studies of hypnotism and psychopathology. France was undergoing enormous political and social change in the first years of the twentieth century. The Dreyfus Affair had split the nation into two camps – those who supported the idea of France as a modern, liberal republic, and those who wanted to return France to the glorious position it had occupied before its defeat at the hands of the Germans in 1871. The formal separation, in 1905, of church and state brought with it a series of radical reforms to the French educational system, but it also sparked a conservative Catholic revival. Internally, the nation was plagued by massive strikes and widespread labour unrest. On the international scene, France appeared, after 1905, to be at risk of losing its colonial influence in North Africa to Germany. The malaise of the nation seemed to go hand in hand with interest in pathologies of the body. During the last quarter of the nineteenth century, numerous French scientific, medical, and literary observers used the biomedical concept of ‘degeneration’ to weld together a host of social problems, including sexual perversion, criminality, suicide, alcoholism and drug addiction, and (particularly in France) a chronically low birth rate.75 These pathologies, argued Théodule Ribot, who held the first chair in experimental and comparative psychology at the Col-

148

The Sleep of Others

lège de France, could be traced back to the problems of a diseased will.76 The ability to direct consciously the course of mental and physical events, Ribot argued, had been the last in a series of evolutionary advances that distinguished the civilized man from the savage. This ability was also the most susceptible to degeneration. Ribot’s fascination with the pathological, a tradition perpetuated by his successor, Pierre Janet, set the tone for psychology in France. It was in this context of the dominance of clinical investigations, evolutionary debates, and positivist rhetoric that sleep began to take shape as a scientific object. A Biological Theory of Sleep, and Its Critics The idea that the study of sleep could make a valuable contribution to understanding normal psychology originated on the periphery of French cultural influence, in Geneva. Here, the Swiss psychologist Edouard Claparède (1873–1940) first suggested that sleep was more than simply the defeat of consciousness by fatigue. Consciousness played a crucial role, Claparède argued, because sleep was actually the product of a sensation that carried with it an evolutionary advantage. It was the sensation of fatigue, not fatigue itself, that caused sleep, and this feeling ultimately prevented the physiological damage that would otherwise result from the build-up of fatigue toxins in the body.77 Claparède’s ‘biological’ theory of sleep soon became a commonplace among psychologists, through the aphoristic formulation: ‘We do not sleep because we are intoxicated; we sleep so as not to become intoxicated.’78 In future years, Claparède would become famous for his work in child psychology, or ‘experimental pedagogy,’ as it was then known. Indeed, his Institute Jean-Jacques Rousseau, founded in 1912, was later home to Jean Piaget (1896–1980), who became director in 1921 and whose work in ‘genetic epistemology’ eventually made him the most famous developmental psychologist of the twentieth century. But it was his novel approach to sleep that first made Claparède’s reputation among psychologists. The similarity between Claparède’s aphoristic position on sleep and the ideas of William James on emotion (see chapter 2) is no accident. Like James, Claparède attempted to bring together biological finalism and the psychology of sensation by depicting the work of consciousness as purposeful. Both were responding to what they perceived to be a misguided reliance on experimental analysis and a materialistic metaphysics in psychological research. Claparède, like James, began his career study-

The Ends of Darkness

149

ing medicine. But he turned to psychology under the inspiration of his cousin, Theodore Flournoy, whom Claparède heard lecture on ‘The Soul and the Future of Psychology’ in 1891.79 Flournoy (1854–1920) taught experimental psychology at Geneva and was heavily influenced by James, whose work he propagated in reviews and a book, Métaphysique et psychologie (1890). James himself had studied natural history under Flournoy’s father (Claparède’s uncle) in Geneva in 1859 and, around 1890, began to cultivate a close relationship with the son, whom he visited in 1892. Their correspondence indicates the extent to which they agreed on the proper direction of psychological science.80 James offered his unqualified praise of Flournoy’s Métaphysique et psychologie, declaring that ‘this is to be really “scientific” without being a bar[b]arian, as so many of our “scientists” are! ... It behoves all of us who on the whole agree in aims and methods to close up our ranks and give each other a helping hand, and perhaps our “School” will prevail!’ Claparède was undoubtedly being groomed for just such a school. He was just beginning a study of the phenomena of audition colorée (the perception of sounds as colours) when he first met James in Flournoy’s laboratory in 1891. James was clearly intrigued, and he suggested to Flournoy that Claparède study an American experimental subject a few months later.81 The same year that Claparède presented his biological theory of sleep (1904), James recommended to Flournoy that Claparède take over as director of Flournoy’s laboratory.82 By 1915, Claparède had become professor of psychology at Geneva, replacing Flournoy, who had been made professor of the philosophy of the sciences. Claparède’s description of sleep as an active, adaptive instinct fit nicely with the functionalism of James, John Dewey, and James Angell that dominated American psychology before the First World War. Such accounts described consciousness as nothing less than the internal expression of an organism’s self-interest, which Claparède eventually codified as the ‘Law of Momentary Interest.’ As the cutting edge of adaptive response, consciousness tended to follow whatever was advantageous at the time. Sleep provided a paradigmatic example: a feeling of fatigue resulting in sleep could appear even in the absence of muscular or intellectual exhaustion, just so long as there was nothing more important engaging an organism’s attention. Thus, fatigue and sleep were related, but not in any straightforward mechanical way. Their relationship was always governed by the organism’s immediate situation, which was inevitably perceived in terms of adaptation and self-interest. The French reaction to Claparède’s theory was rather different from

150

The Sleep of Others

the American response. Teleological thinking was warmly embraced by the American proponents of functionalist psychology, who had set themselves up in contrast to their ‘structuralist’ counterparts, led by E.B. Titchener at Cornell University. Titchener argued that psychology’s goal was akin to that of chemistry: reduce the object (consciousness) to its elements, and then study the way these elements combined with each other.83 Likewise, he argued, psychology was more than simply an applied science; it could not be reduced to its technical applications any more than could physics or chemistry. Psychology’s experimental methodology and its agnostic approach to metaphysical systems also served to separate it from philosophy. Comprehending mind in terms of purpose, Titchener suggested, committed both these errors. The Contexts of French Psychology Titchener’s arguments were a minority view among the practically oriented psychologists in the United States. In France, however, psychology was a much different affair. There, psychologists were almost always trained first as philosophers.84 Their authority came not from their ability to experiment, but to reason; their end was not to effect a practical transformation of society, but to reveal the nature of mind. The scale of their enterprise was also dramatically different: by 1903, there were at least forty psychological laboratories across the United States, and they awarded more doctorates than all other fields except chemistry, physics, and zoology.85 In France there were but three laboratories, all of them in Paris, and only one was found outside a hospital: Pierre Janet’s was at the Salpêtrière asylum; Edouard Toulouse’s was at the Villejuif asylum; and Alfred Binet’s was located at the Sorbonne. Psychology was thus a relatively minor concern in France, and it was distinguished primarily by the clinical observation of psychopathological phenomena. French philosophy and physiology, on the other hand, was a battleground of metaphysical arguments and teleological premises. Bergsonism was on the rise, and Richet had insisted that all physiological analysis had to be conducted from the perspective of organismic defence.86 Functionalism could not possibly supply French psychologists with a disciplinary identity distinct from either philosophy or physiology, and the clinical experimentation championed by Janet had reached an impasse by the first years of the twentieth century (see chapter 2). This left the psychophysiological experimentation upheld by Binet and Toulouse, whose efforts (at least in the case of Binet) involved the explicit rejection of Berg-

The Ends of Darkness

151

sonism as a threat to the project of establishing psychology as a science. Their work was marginalized by the Parisian intellectual elite, but it was ultimately these two laboratories that successfully adapted Claparède’s theory to experimental intervention. When Claparède’s theory first appeared in France, reaction to it was ambivalent and guarded. Binet set the tone in the pages of his journal, L’année psychologique, in which he agreed that fatigue theories could not explain the periodicity of sleep. But, Binet argued, Claparède’s ‘interest’ theory of sleep offered little more since it provided no physiological mechanism.87 Claparède’s claims, he suggested, amounted to a restatement of Bergson’s aphorism, dormir c’est se désintéresser. In 1906 this last comment was tantamount to a charge of metaphysics – a charge that carried a great deal more anti-scientific weight in France than it did in the United States at the time. To many French social reformers in the wake of the Dreyfus affair, metaphysics in any form, but particularly as religious belief, threatened the very existence of the republic. In 1894 Alfred Dreyfus, a Jewish military officer, was charged with selling military secrets to the Germans. Although Dreyfus was officially pardoned in 1899, the affair persisted until 1908, when the ashes of one of Dreyfus’s most outspoken champions, Émile Zola, were transferred to the Panthéon, where Dreyfus had almost been assassinated.88 The affair utterly polarized French society. Dreyfus had been convicted only because his handwriting resembled that found on documents associated with the espionage. Many young republicans were convinced of Dreyfus’s innocence and argued that he had been the victim of the anti-Semitic sentiments of those who supported the glory of Old France, with its deep traditions of military and clerical order. The ‘antiDreyfussards,’ supported the conviction, even arguing that the truth of Dreyfus’s innocence was irrelevant: the psychiatrist Joseph Jules Dejerine, for example, reportedly stated that if Dreyfus really was a good patriot, he would spare France the turmoil and stop protesting the charge.89 When Dreyfus was retried and pardoned in 1899, the antiDreyfussards became enraged. Fearing a rightist coup, the leftists galvanized under the rubric of ‘republican defence’ and swept the elections of 1902, inaugurating a new period of reform that was to make France ‘safe’ for republican values. This translated into an almost institutionalized anti-clericalism, since anti-Dreyfussards were, more often than not, Catholic conservatives. In 1905 Napoleon’s Concordat of 1801 was revoked, effectively ending the role of organized religion in state affairs. Clergymen no longer received a state salary, and the church’s

152

The Sleep of Others

role in educational affairs was severely curtailed. The Inventory followed, in which sacred objects in cathedrals and monasteries across France were handled and counted by taxmen in the service of the state. Riots broke out in many provincial areas, and a Catholic revival emerged in response. It was thus in this context of ‘republican defence’ that Binet described Claparède’s theory in terms of Bergsonian metaphysics, which Binet felt were both conservative and dangerous.90 By 1906 Alfred Binet (1857–1910) had been closely implicated with republican reforms for a little less than a decade. He had begun his education studying law but eventually abandoned it for medicine. He failed to complete his medical studies, however, and began, around 1880, to read psychology. Though he received no formal instruction in the field, he started to publish papers on the subject in Ribot’s journal, the Revue philosophique. Around 1882, he met a former classmate, Joseph Babinski, and was drawn into Charcot’s circle at the Salpêtrière. Binet passionately defended Charcot’s vision of hypnotism as a symptom of disease, but his claims were effectively demolished by Delboeuf. Like so many others, Binet abandoned hypnotism as an experimental method, publishing his last book on the subject in 1892. Binet thus began his search for other empirical means of establishing the ‘laws of character.’ But, given that he lacked both a laboratory and a teaching appointment, it could just as well be said that Binet went searching for subjects. His two young daughters were some of the first. After leaving the Salpêtrière in 1890, and before joining Henri Beaunis at his psychological laboratory at the Sorbonne in 1891, Binet, inspired by the graphical method, studied developmental changes in the reaction times and bilaterality of hand movements in his children using a Marey tambour and a kymograph. He combined these with cognitive tests, in which his children’s ability to recognize small differences in length between lines drawn on different cards turned out to be almost as good as that of the adults he studied. Once Beaunis appointed him director of the laboratory at the Sorbonne, Binet had access to a larger group of subjects, which included schoolchildren. It was, of course, his work with this particular population of human subjects that would eventually generate his greatest legacy to psychology: the IQ test. In the spring of 1895, Binet travelled to Bucharest to lecture at the university there. He still lacked a teaching position in France but was nonetheless greeted in the newspapers as ‘a representative of modern science,’ and enthusiastic crowds attended his numerous lectures. In

The Ends of Darkness

153

the process he acquired a young disciple, Nicholas Vaschide, who followed him to Paris to study in his laboratory at the Sorbonne. By this time, Binet had become consumed by the idea that psychology should focus on the study of individual differences. Knowledge of the psychological variations that existed within a population would, he argued, ‘have practical importance for the pedagogue, the physician, the anthropologist and even the judge.’91 To that end, Binet and his collaborator, Victor Henri, outlined a number of categories in which individuals could be tested: memory, mental imagery, imagination, attention, comprehension, suggestibility, aesthetic sensibility, moral sensibility, muscular force, willpower, and motor habits. Binet immediately set Vaschide to work on this project upon his arrival in Paris in 1896. It is tempting, given the current importance of standardized tests in educational psychology, to depict its most important progenitor as having nursed a deep interest in the field from the outset. But Binet’s decision to use schoolchildren as subjects for his study of individual differences was deeply caught up in the intellectual, technological, political, and professional circumstances in which he found himself. As Binet himself portrayed it, he and Vaschide came to study large groups of schoolchildren simply because no other avenues of research were open to them.92 Binet was smitten with the potential uses of Mosso’s invention. The device ‘realized a new principle,’ in that it could isolate and graphically depict the work done by a single group of muscles, which could, in turn, express the state of physical, intellectual, and emotional fatigue in the whole organism.93 The ubiquity of the instrument in psychological laboratories at the time meant that its use played a prominent role in Binet and Vaschide’s early studies.94 But its results needed to be expressed, Binet thought, in terms of individual differences. Differential conditions and tendencies to fatigue, not fatigue itself, were the object of interest here. But isolating a cooperative group of subjects proved exceedingly difficult. The problem, thought Binet, was partly structural. Wundtian laboratories formed the basis of psychology’s success as an academic discipline, but these laboratories had also deformed the science by their reliance on experimental introspection. Students would participate uncritically in introspective experiments ‘to sanction a good grade in their exams’ and thereby earn their doctorate.95 Wundt’s disciples then went on to found laboratories elsewhere, perpetuating the same meaningless exercise in professional legitimation. The British eugenist Francis Galton had successfully created a psychological labora-

154

The Sleep of Others

tory that took individual differences as its object, but, Binet and Vaschide observed, such laboratories did not offer degrees. They thus lacked students and were frequently denied routine access to a large pool of subjects, making the spread of such laboratories unlikely. To solve the problem, Binet and Vaschide first turned to amateur sporting societies and private gymnasiums. But they found the subjects there utterly undisciplined and uncooperative: experiments quickly degenerated into sporting contests. The use of public figures, who were often motivated to participate out of sheer self-love, was an alternative with which Binet had had some experience. But such people could not possibly represent the wide spectrum of differences Binet hoped to discover. So Binet turned to the schools, where he encountered further difficulties. Administrators of colleges and lyceums would not allow psychological studies, because, as Binet and Vaschide candidly put it, ‘the students of these institutions have parents who know about the journalists and the disputes; one fears that the parents, ignorant of the nature of the research to be conducted on their children, suspect some form of hypnotism or some study of materialism, and raise complaints that resound in the newspapers or in the courtrooms.’96 In desperation, they settled on a study of the children at the primary schools, where they found the administration much more open to experimental investigations. Binet’s interest in pedagogy increased dramatically after 1898, when he published the studies he and Vaschide had conducted, filling hundreds of pages of his L’année psychologique. Théodore Simon, an intern at a colony for retarded children at Perray-Vaucluse, was so impressed by Binet’s work that, in 1899, he arranged for Binet to supervise his MD thesis. That same year, Binet was asked to join the newly founded Société libre pour l’étude de l’enfant. Under his direction, the Society petitioned the Ministry of Public Education to do something about the education of retarded children. Proper identification of such children was in order, and he and Simon quickly developed a series of tests that differentiated children who learned normally from those who did not. Binet still lacked an official teaching position, but he now enjoyed state sanction, routine access to a large group of subjects, and a practical justification for his work. In Binet’s estimation, Bergson’s philosophy threatened the rational, technocratic, and secular educational system he was then helping reform. Indeed, the two men’s intellectual networks were completely opposed. In 1902, when Binet made a bid for Ribot’s chair in experimental psychology at the Collège de France, Bergson supported Binet’s

The Ends of Darkness

155

rival, Pierre Janet. Binet was endorsed by fellow experimentalist Marey, but Janet won out. Several years later, Binet was again disappointed when he tried to succeed Janet at the Sorbonne, where he lost out to George Dumas. Binet’s failures seemed to push him towards philosophical heterodoxy, since his increased passion for intelligence testing and his attack on Bergsonism both occurred during this difficult period in his career.97 In 1907, Binet published a further extension of his reformist attempts to transform French education – the so-called ‘Binet Report.’98 His choice of targets was again strategic. Instead of attempting to aid ‘imbeciles’ in the schools, Binet went to the other extreme, analysing the prize of the French intellectual tradition – the teaching of philosophy. The philosophy curriculum in the lyceums and colleges had been revised as part of a series of anti-clerical educational reforms mounted in 1902, in the name of ‘republican defence.’ Religion and metaphysics had lost their status as mandatory parts of the curriculum in the process, and Binet was curious as to what, if anything, had replaced it. The answer had considerable relevance for the status of scientific psychology in France, since practically all French psychologists during this period first trained as philosophers. Binet proceeded to send out an eighteen-question form to three hundred professors across France, and when he presented his report to the Société de philosophie in 1907, he summed up his findings in a single phrase: ‘The philosophy of Being, like State religion, is on the run because the number of believers has diminished.’99 Many students, Binet was pleased to report, were demanding that the new psychology be taught in place of ontology or metaphysics. But progress was impeded by the fact that most instructors lacked the competence to demonstrate even the simplest experiments. This accidental obstacle was furthered by a more sinister design: Bergsonism was being rapidly and uncritically embraced by many of these same instructors. Binet fell just short of calling Bergsonism a fashionable surrogate for religious metaphysics, but his conclusions were unmistakable. Despite all republican efforts, faith had snuck back into the minds of the French intellectual elite. Putting Sleep to the Test Binet’s reduction of Claparède’s sleep theory to Bergsonian metaphysics was thus part and parcel of his more general attack on Bergson’s philosophy. But Binet did not himself devote any effort to an empirical study of sleep. Perhaps, having embraced and then rejected hypnotism as an experimental method, he felt he had already dealt with whatever ques-

156

The Sleep of Others

tions could have been raised by the problem of sleep, which, as we have seen, lacked the status as an object of experimental psychology for investigators of Binet’s generation. In any event, such tactical battles were better left to followers, even though Binet’s tended to be few and far between. He lost his Rumanian student Nicholas Vaschide (1873–1907) in 1898, just as he was redirecting his interests away from psychopathology and towards pedagogical reform. It is not known exactly why Vaschide left Binet’s laboratory. One historian, in noting that Binet broke with Vaschide following the American criticism of their mishandling of statistical data, suggests that Vaschide ‘seems not to have shared Binet’s rigor as an experimentalist.’100 But, given the continued frustration of Binet’s hopes at gaining a teaching post, he might well have simply needed someone to blame, and Vaschide proved a convenient scapegoat. In any event, it is clear that even if Vaschide did not share Binet’s rigour, he certainly approved of his rhetoric. ‘Experiment’ became Vaschide’s watchword at his new post as chef-de-travaux in the psychological laboratory at the asylum at Villejuif. From 1900 until his untimely death in 1907, Vaschide collaborated with Henri Piéron and Edouard Toulouse, creating a textbook that would eventually define the nature of experiment among French psychologists. Vaschide, unfortunately, is a veritable non-entity in histories of French psychology. Attention has instead been focused on his colleague, Henri Piéron (1881–1964), who is traditionally seen to be a key figure in directing French psychology away from pathological psychology and towards an experimentalism characterized, above all, by stimulusresponse experiments (fig. 9).101 He has been described as ‘the true founder of scientific psychology’ in France, and the impact of his science of comportement is in many ways equivalent to that of the ‘behaviourism’ of his American counterpart, J.B. Watson.102 Certainly, Piéron played a pivotal role in establishing psychology as an empirical discipline in twentieth-century France, but such stories are unable to account for sleep’s persistent ability to transcend its reduction to a purely psychological object. Piéron himself was unable, or unwilling, to configure sleep in such a manner, and he more or less abandoned the topic after the First World War. But he was inspired to frame sleep in terms of experimental psychology precisely because of his exposure to the work and ideas of Claparède and Vaschide. An examination of Vaschide’s work, then, will provide the context in which Piéron first solidified his reputation as an experimentalist by rendering sleep as an investigative object.

The Ends of Darkness

157

9 Henri Piéron (1881–1964) was one of the first investigators to promote the method of ‘prolonged wakefulness’ as a way of framing sleep as a physiological problem. (E.G. Boring et al., A History of Psychology in Autobiography, vol. 4 [New York: Russell and Russell, 1952], 55)

158

The Sleep of Others

In 1900 Vaschide found himself in the newest laboratory for experimental psychology in France. It was the creation of Edouard Toulouse (1865–1947), a medical doctor who first gained prominence in 1896 with a psychological study of Émile Zola entitled Les rapports de la supériorité intellectuelle avec la névropathie.103 In 1898 Toulouse was appointed médecin-chef at a very new asylum located on the southern outskirts of Paris. Unlike the Salpêtrière, which traced its history back to the seventeenth century and featured an illustrious clinical tradition, the Villejuif asylum was a thoroughly modern creation, having been completed only in 1889. In 1900 Toulouse created a laboratory at Villejuif under the auspices of the École Practique des Hautes Études. Despite the fact that his new Laboratoire de psychologie expérimentale was located within an asylum, Toulouse had no intention of restricting his experimental inquiries to the problem of madness. This is not to say that Toulouse was not interested in mental illness. On the contrary, he was an active and outspoken eugenist, keenly devoted to the prevention of mental illness through the early identification of pathological individuals using psychopsychological tests the likes of which were being concurrently developed by Binet and Vaschide.104 Toulouse wanted such testing to pervade every aspect of the young republic, where he eventually hoped to find ‘a psychophysiological laboratory acting as an organ of selection and classification’ that could direct citizens to their appropriate stations in education and industry.105 The wave of strikes that shook France around 1905 were, Toulouse thought, partly the result of workers suffering from performing tasks unsuited to their personality. Toulouse argued that his testing laboratories could set standards for these ‘human machines’ exactly as laboratories established standards for industrial products. Toulouse, who later took over as editor of the popular Revue scientifique from Charles Richet, wrote so widely on his rationalized approach to psychiatry and psychology that ‘Eh, va donc chez Toulouse!’ became a popular expletive among irate Parisian cab drivers.106 One of the first publications to emerge from Toulouse’s laboratory, however, emphasized this practical orientation in a rather different way. In 1902 Vaschide and Piéron, then a young préparateur and another refugee from Binet’s laboratory (he spent only a few months there in 1899), produced a curious monograph entitled La psychologie du réve au point de vue médical.107 The book’s premise – that dreams could be used as an effective diagnostic tool – was a strange antiquarian throwback in a selfconsciously modern age, invoking the Aesculapian cults as a precursor to their study. Yet it fit well with Toulouse’s mandate to create a form of

The Ends of Darkness

159

psychological knowledge that was based in psychopathology but had practical applications beyond the clinic. The book, which was little more than a review of the literature, concluded that dreams did indeed come from organic sensations and thus should be exploited as such in clinical medicine outside of psychopathology.108 But the authors made the recommendation with more than healing in mind. The more extensive use of dreams in clinical medicine and the rational analysis of dreams in general would also have positive social ramifications, since these practices would tend to ‘project the light of science on these obscure regions, making radical superstitions vanish.’109 In a later work, Le sommeil et les rêves, Vaschide continued this theme of using science to combat the persistent superstitions surrounding sleep and dreaming. Indeed, superstition and the occult fascinated Vaschide, as it did a number of prominent French scientists of the day, the most notable of them being Charles Richet. Vaschide made a study of chiromancy while conducting his research on dreams, and apparently had his own death foretold by a palm reader named Mme Fraya.110 Fate conspired to cut short his research program, and Vaschide died suddenly of pneumonia in 1907, only thirty-three years of age. His book on dreams was published posthumously, in 1911. In it, Vaschide took careful note of Freud’s study of dreaming, devoting an entire chapter to the wish-fulfilment theory.111 But he disagreed with Freud’s interpretation of anxiety dreams, and he rejected any distinction between the manifest and latent content of the dream. Instead of interpretation, Vaschide opted for meticulous observation verging on the obsessive. He made recordings of the pulse and respiration and coupled these with his observation of his subjects’ motor activity in sleep. He even watched the faces of his sleeping subjects, hoping to observe an emotional expression that could then be correlated to a dream. Putting together his physiological observations and his subjects’ dream reports, Vaschide concluded that all dreams had one element in common: emotion. Dreams were never, or even essentially, purely ideas or images. They were inevitably cloaked in emotion that tended to augment the intensity of the dream image, thereby convincing the dreamer that the dream was actually part of waking experience. But there were, of course, serious limits to such introspective evidence. Indeed, the emotional character of dreams was oftentimes entirely hidden from the dreamer, and it was only through physiological surveillance that the ubiquitous emotivity of dreams could be made manifest. The sleeping body’s responses to dream images were identical to the physiological

160

The Sleep of Others

changes that characterized emotional states. This inevitable emotional response to images rendered the dreaming process a kind of ‘spiritualization,’ in which dreams took on a supernatural guise ‘unknown to waking life,’ and which fuelled belief in their divine origins.112 Vaschide went further, suggesting that he had located a physiological basis for the belief that ‘abstractions’ – a word he routinely used to describe dream images – could somehow reflect an underlying reality.113 He was effectively taking the first steps towards a physiology of metaphysics. This use of experiment to undercut philosophy comes as no surprise, given the tenor of the experimentalist critique against Bergsonism. Vaschide rejected Claparède’s theory of sleep along similar lines. Such theories, Vaschide chided, merely ‘transported our systematised ignorance into another domain,’ and while he agreed that one could derive a concept of inhibition from the phenomena of sleep, such a concept added nothing at all to what was known about sleep itself: ‘The hypothesis of sleep as a biological instinct reminds me of those classroom demonstrations that are meant to strike at the passive intellects of students and docile readers. They appear to explain certain phenomena, but only if certain initial conditions are met. The why of sleep remains as mysterious as ever.’114 Depicting sleep as an active instinct, Claparède had postulated that some sort of ‘sleep centre’ must exist in the brain. This, charged Vaschide, was evidence that Claparède had been unduly influenced by the ‘mania’ of neurologists for finding ‘centres at every turn,’ which was not, he suggested, so very different from the phrenological search for centres of emotion by ‘cranial inspection and percussion.’115 Like the spiritual forces that supposedly lay beneath the emotionally laden dream images, such centres were simply abstractions and had no foundation in empirical evidence. In place of Claparède’s theoretical musings, Vaschide instead called for facts. ‘Hypotheses,’ he complained, ‘have had bizarre and diverse fates in the history of the sciences ... The belief in words invented to explain what is yet inexplicable makes me appreciate all the more the facts we must endeavour to multiply in profusion.’116 The graphical method, had, at least in Vaschide’s laboratory, finally trumped false belief. Even the privacy of dream worlds could now be rendered accountable to public inspection. Claparède responded in kind, albeit after Le sommeil et les rêves was published and some five years after his detractor had died. It was not, he argued, a question of whether or not his biological theory rested on facts, but of which facts one chose to accommodate in constructing a

The Ends of Darkness

161

theory. Vaschide’s description of sleep as a negative phenomenon had relied on the extant fatigue theory, which could not explain many facts about sleep. For example: How was it possible that sleep often appeared in the absence of fatigue? Why did sleep tend to come in one long period, rather than in several short ones? Why was sleep often brought on by hypnotism or through the association of ideas? It was not the facts themselves, Claparède insisted, that lay at the core of their dispute. It was rather the origins of those facts. Which required explanation, facts derived from lived experience and introspection, or those generated in the laboratory? Vaschide’s obvious and exclusive preference for the latter constituted, in Claparède’s view, a curious kind of pathology in itself, one that was certainly interesting for the psychology of the professional deformation of certain individuals given to experimentation: these persons have lost the habit of seeing what goes on around them in real, concrete life to such an extent that, for them, in order to have the right to the title of ‘fact,’ a phenomenon must have occurred within the four walls of a laboratory or a hospital room, or have been recorded on a registering cylinder or by means of some other apparatus! The facts, as one can observe them in everyday life, count for nothing! My theory of sleep rests precisely on these mundane facts, which have been completely neglected up until now, counting for nothing among people like Vaschide and others who, without a doubt, think like him.117

Recalling the overwhelming significance of the concept of ‘defense’ in evolutionary biology and psychology (never mind republican politics or the international conflicts of the day), Claparède reminded his readers that ‘if we prevent sleep, the individual will soon be in a state of complete exhaustion. But, assuredly, it is always the physiologist, who, interpreting the result of the experiment, draws from it the notion of defence or protection. This notion, of course, will not inscribe itself on a registering cylinder!’118 The End of Darkness: Henri Piéron Claparède’s rebuttal appeared just as Henri Piéron was completing his doctoral dissertation, ‘Le problème physiologique du sommeil,’ which ultimately became one of the twentieth century’s most influential scientific studies of sleep. The impact of Piéron’s work, however, was less in

162

The Sleep of Others

France than it was in the United States, where Nathaniel Kleitman used it to shape his own inquiries a decade later. ‘One of the best books ever written on the subject of sleep,’ wrote Kleitman in 1923, ‘is Piéron’s Le Problème Physiologique du Sommeil, which besides embodying the results of original research on sleep from the histological, biochemical and physiological standpoints (this by a psychologist!) contains a valuable and exhaustive bibliography brought up to 1912.’119 Kleitman’s surprise that a psychologist would have approached sleep from a physiological perspective indicated his lack of understanding of the context in which Piéron was working. Sleep had become a somewhat controversial topic in the wake of Claparède’s biological theory, and Piéron’s dissertation was an attempt to straighten out the differences between Claparède’s approach and that of his former collaborator at Villejuif, Vaschide. Like Vaschide, Piéron railed against the speculative excesses of functionalist thinking. Teleology needed to be defeated, in all its guises. What was needed were more facts. But Piéron was equally intrigued by the question of memory. He disagreed with Bergson’s divorce of memory and matter, and for good reason: Piéron was a zealous reformer who felt that the immense popularity of Bergson’s introspective philosophy was yet another example of France’s failure to keep pace with modern life. Introspection was, like religious and metaphysical belief in general, on its way out. Piéron thus endeavoured to invent a way to study sleep that separated it from consciousness but retained the idea of memory. He found it in his technique of ‘enforced wakefulness.’ Ribot, Pierre Janet, and the neurophysiologist Louis Lapique must have seen the writing on the wall when they recommended that Henri Piéron serve as préparateur in Toulouse’s freshly minted laboratory in 1901, for here was one of the few places in France where the young philosophy student could satisfy his desire to study the nature of mind in a laboratory, rather than in a clinic.120 Piéron’s father, who had taught mathematics for seventeen years at the Lycée Saint-Louis and become inspecteur-en-chef of public education in 1894, insisted on a scientific education for his youngest son, and by opting for experimental psychology, Piéron seemed to be satisfying his father’s demands as well as his own curiosity about the nature of the relationship between mind and body. Piéron had immersed himself in all the battles being waged over the soul in France at the turn of the century. He studied under Charcot’s successor at the Salpêtrière, Fulgence Raymond, who advocated a strict psychological determinism. Janet also held a clinic there, and Piéron had served as his secretary before visiting Binet’s laboratory

The Ends of Darkness

163

at the Sorbonne, which introduced him to the practices of psychological testing. Piéron’s greatest sources of inspiration were outside neurology and psychopathology. Piéron wanted to develop a theory of mind capable of integrating the phenomena of consciousness into a theory of life, so, like many other French biologists of his day, he turned to evolutionary theory as a way of making psychology scientific. He was particularly smitten with the work of one of his father’s former students, a ‘most precocious mathematician side-tracked to biology’ named Félix Le Dantec (1869– 1917).121 Le Dantec was the most prominent popularizer of the debates between neo-Lamarckians and neo-Darwinians that raged in France from the 1890s until the Great War. But he was no champion of either side, since he insisted upon an evolutionary theory that was empirically confirmed, completely explicable in mechanistic terms, and devoid of the Malthusian implications of natural selection.122 He favoured neoLamarckism, if only because he felt that the question of inheritance could be framed in strictly materialist terms of physico-chemical determinism. Acquired traits had to be inherited through some sort of biochemical or biophysical mechanism, and the job of the biologist was to discover this mechanism. In contrast, he dismissed neo-Darwinian claims about selection as the primary (or even exclusive) mechanism of evolution as metaphysics. August Weismann’s statement regarding the continuity of the ‘germ plasm’ was, for example, rejected by Le Dantec as ‘a revival of the spiritualist error’ because it effectively made the germ plasm immortal.123 It was Le Dantec who introduced Piéron to Alfred Giard (1846–1908), the grandmaster of evolutionary theory in the Third Republic. The first chair devoted to the topic at the Sorbonne, for the study of the ‘Evolution of Organized Bodies,’ had been awarded to Giard in 1888. The very existence of this position was the subject of violent debate in the Parisian municipal council, for its creation seemed expressly dedicated to the overthrow of the eclectic ‘spritualist’ thought that had dominated French philosophy for much of the century. Those who supported the creation of the chair pointed out that, in Great Britain and in Germany, evolutionary theory had long provided the epistemological foundations of the moral sciences, but that France continued to lag more than a century behind its European competitors.124 Piéron thus entered the field of scientific psychology at a time when Binet and Toulouse were attempting to render it more practical, and when Le Dantec and Giard hoped evolutionary theory could refabricate the moral fibre of the republic. Not surprisingly, Piéron adopted the

164

The Sleep of Others

posture of a materialist reformer early in his career. He passed his agrégation in philosophy in 1903 only through the intervention of a sociologist, Lucien Lévy-Bruhl, who opposed Alphonse Darlu, a philosopher who suggested to Piéron (who had already failed twice) that he had no hope of passing. Piéron later labelled Darlu a ‘narrow-minded moralist’ who was ‘violently opposed to the scientific method in psychology,’ and this low opinion was shared by other prominent social scientists of the day, including Émile Durkheim.125 But rather than teaching philosophy, Piéron chose instead to pursue his doctorat ès sciences, a research degree. He chose sleep as his topic. Piéron was already intimately acquainted with many of the difficulties and iconic importance contained by such an immediate yet obscure topic. He had conducted considerable research on the clinical use of dreams under Vaschide’s direction in Toulouse’s laboratory, and quickly made himself familiar with Claparède’s work. Like Vaschide and Toulouse, Piéron dedicated himself to the creation of a scientific psychology that was psychophysiological, analytic, and with practical applications in education and industry. The three consistently worked together at hammering out a manifesto of experimental psychology from their earliest collaboration. In 1901 they presented their first version at the Physiological Congress in Vienna, which was published as a book a few years later and reworked into a second edition after Vaschide’s untimely death.126 This became a foundational text for French psychology, in part because of Piéron’s rapidly rising stature in the field. Piéron assumed Vaschide’s role of chef-de-travaux of the laboratory at Villejuif in 1907, and when Binet died in 1911, he beat out Janet to head the laboratory at the Sorbonne. Janet had been thrown out of his laboratory at the Salpêtrière by one of his rivals, Dejerine. Thus, Piéron effectively amalgamated two formerly competing sites under his charge, while his main competitor, Janet, was left with virtually nothing. He had even inherited the main organ of psychology in France: Binet’s old journal, l’année psychologique. A close relationship had evolved between Claparède and Piéron during this same period. They exchanged articles as early as 1904, and Claparède even suggested that their families vacation together the following year.127 Although he challenged Claparède’s biological theory as vague and incomplete, Piéron was at least as sympathetic to a teleological reading of sleep as Claparède was enthusiastic about having some physiological flesh put on his rather skeletal psychological theories. Utilizing the concept of ‘interest,’ Claparède had hoped to redescribe sleep as a posi-

The Ends of Darkness

165

tive phenomenon that was an end unto itself, and he chastised his friend for suggesting that normal sleep could be the product of the circulation of some sort of internal toxin.128 Nonetheless, he respected Piéron’s penchant for experiment, and praised the manuals he had created in collaboration with Toulouse and Vaschide. For his part, Piéron surely lent a sympathetic ear to Claparéde’s complaints about Vaschide’s ‘bizarre analysis’ of the new biological theory of sleep.129 Unlike his former collaborator at Villejuif, Piéron was no iconoclast. His study of sleep was clearly designed to frame sleep as an experimental and theoretical object worthy of investigation in its own right, independent of any links it might have to religious sentiment. Indeed, Piéron’s efforts, literally and figuratively, were set on the task of draining sleep of such metaphysical associations. Organic Memory Such a transformation is never proposed or accomplished in a vacuum. Simply put, Piéron was able to divorce sleep from questions of the soul not purely by virtue of devising an experimental technique but because such a technique resonated with current pursuits by other researchers. In this case, the prize was memory. How did memory function? What was its purpose? Were the phenomena of memory entirely circumscribed by individual consciousness, or could material bodies also be said to remember? For his first major study, published while he was still completing his doctorate in the natural science, Piéron chose a topic dear to the heart of his masters – organic memory.130 This concept suggested that heredity and psychological memory were essentially the same. Its proponents insisted that the site of memory was the body, and it could be investigated through the study of natural history and developmental biology, in effect putting a materialist spin on the debates over the psychological nature of memory so well described by Ian Hacking.131 Both scientific and literary sources indicate that the idea of organic memory arose simultaneously with the nationalist fervour that swept through Europe in the last quarter of the nineteenth century.132 A nation quite literally was what its citizens could remember. Such a doctrine further reflected European anxieties about race relations and the processes of degeneration that were supposedly at the core of numerous social problems across the continent. This was particularly the case in France, where the idea of degeneration exposed the immense insecurity the nation felt at losing its leading cultural and scientific role after its

166

The Sleep of Others

fateful loss to the Germans in 1871. Several prominent French scientists advocated one or another concept of organic memory: it had long been a favourite theme of Théodoule Ribot’s, while Richet used it to defend his racist eugenics.133 This new interest in memory was not just rewriting the soul. It was also redefining the body. Piéron contrasted his analysis of memory to those of Richet and Bergson. Memory was neither separable from matter, as Bergson had argued, nor the unique property of human beings, as Richet maintained. Indeed, in Piéron’s view, memory was hardly a property at all. It was rather a description of a historical process that could be identified in all stages of matter. Noting the similarities between the curves representing Hermann Ebbinghaus’s findings regarding the speed of learning and those representing autocatalytic monomolecular reactions, Piéron argued that memory was nothing more than ‘a persistent influence of the past on a present state,’ or the ‘consecutive effect of events, which have since disappeared, on current phenomena.’134 Rhythm, argued Piéron, was one of the most basic manifestations of memory, so his book began with a discussion of the various rhythms that appeared in the lowest forms of organized matter, such as the influence of light on the rhythmic movements of plants, and that of the tides on various sea creatures. Piéron characterized such phenomena as ‘anticipation’ and argued that what distinguished the anticipatory reflex from that of the anticipatory rhythm was that only the latter could appear spontaneously, in the absence of any external excitation. Such rhythms were adaptive, and Piéron insisted they be classed under the rubric of memory because of their function: they indicated ‘a use of the past to determine a future that was advantageous to the organism.’135 Anticipatory reflexes were pervasive, and were integral for both nature and culture. They could be found ‘in every degree of differentiation and evolution, from the nervous cell of the Actina [sea anemone] up to the cerebral centres of man, from the reaction to the tides up to our own scientific forecasting.’136 Organic memory was thus relevant to Piéron’s republican vision of social progress. Against Richet, he insisted that memory was not an acquired form of intelligence that had evolved only in humans and was inherited by a yet unknown cellular mechanism.137 Such an analysis limited the concept of memory to the ability to represent ideas to oneself, which was nothing more than a holdover from pre-scientific psychology derived from introspection. Memory was rather a form of repeated activity, which could be objectively observed in the laboratory. Memory was as much social practice as it was an inherited characteristic.

The Ends of Darkness

167

As an investigative model, Piéron preferred the work done by comparative psychologists in the United States to the introspective methods of Wundt and his disciples. But, in any event, French psychology languished behind most other nations in terms of the progress of psychological knowledge. There was, Piéron thought, a perfectly good sociological reason for this. American centres of psychological knowledge were widely diffused across the country, thus forcing psychologists there to find ways of cooperating across vast distances. One of the most effective means of accomplishing this was by standardizing subjects that were portable – in other words, by using animals rather than introspective observers (a method that was then championed by, among others, Robert Yerkes at Harvard). In France, on the other hand, psychology, like all scientific research, was completely centred in Paris, promoting an intensive bickering among practitioners and preventing new ideas from being accepted and quickly disseminated.138 Curiously, France did have an intellectual precursor to a style of psychological research that focused on what people did, rather than on what they thought. French Catholicism’s greatest apologist of the seventeenth century, Blaise Pascal, Piéron argued, was the true forerunner of the revolt against introspection, for he had prepared the way for a study of man based on the observation of behaviour alone. Indeed, Piéron cited a letter to this effect, in which Pascal insisted that ‘in order to determine whether or not it is God who makes us act, it would be better for us to examine our external behaviour [comportement], rather than study our internal motives.’139 In this way, Piéron effectively nationalized what was then emerging in the United States as the study of behaviour, culminating in J.B. Watson’s famous manifesto against experimental introspection.140 Equally significant was his appeal to Pascal, which circumvented the criticism that the study of comportement inherently threatened religious belief, as more radical exponents of experimental psychology, like Vaschide, had maintained. By extending the domain of memory, Piéron effectively delimited the purpose of scientific psychology, which both Richet and Bergson had assumed was to explain human consciousness. A comparative methodology that studied behaviour, argued Piéron, eliminated this problem entirely. Piéron’s analysis of behaviour was also linked to practical concerns. Clearly, the progress, or the ‘evolution,’ as Piéron dubbed it, of memory was relevant to social improvement.147 Richet had identified memory with intelligence and argued that the exercise of mental powers, coupled with the practice of artificial selection, would inevitably improve

168

The Sleep of Others

the lot of mankind by developing superior human beings. The human mind was, after all, the most precious instrument in the pursuit of scientific knowledge. And, for Richet, science was the only road to progress. ‘One cannot have scientific conquests without new instruments,’ Richet argued in his eugenist manifesto, ‘and the instrument of all science, one more necessary and more powerful than any telescope or galvanometer, is the human spirit. Thus, the first condition, indispensable to all progress, is the progress of the human mental machine.’142 But such an analysis seemed misguided to Piéron, since it presupposed intelligence to be an exclusive characteristic of individuals. And besides, he mused, ‘[t]he biological evolution of memory appears to have been terminated,’ and human breeding programs were unlikely to do anything to change this. Progress was still possible, but at an entirely different level: ‘Mental progress is incontestable,’ Piéron conjectured, ‘but it concerns knowledge; this knowledge, however, is no longer individual. It is not at all hereditary – it is social.’143 Piéron thus invoked exactly those structures that he and Toulouse had set out to improve: institutionalized vocational guidance, a rationalized educational system, and the abolition of metaphysics and speculative philosophy in the curriculum. Such efforts were on a historical continuum with much earlier developments, such as the invention of the alphabet and printing. It was progress in these areas, rather than the improvement of the individual human body, that would secure happiness and social stability in the future. The greatest obstacle to such progress was in overcoming tradition. History, which seemed to have provided eugenists like Richet with evidence of the biological superiority of the white race, had to be overturned: There remains the danger of a wilful preoccupation with things of another time ... history, this recapitulationary form of social memory, can occasionally raise legitimate concerns ... but it is not good to always look behind, to absorb or to hypnotize oneself in contemplating the road that has been taken. Happy, in a sense, are the people who are without a history, who can look only to the present and to the future. All their efforts are fecund, and the great heights currently achieved by science and industry in America are, to a great extent, possible only in the absence of the burden of history. In France, on the other hand, the predominance of historical study certainly appears to be one of the principal causes of our relative decadence. It is through science that social progress is created, and it is sterile to give oneself up to the vain knowledge of the past. To see better what has been done, one forgets to do something. Greece, which today lives entirely in its memory, still believes itself to be a great nation.144

The Ends of Darkness

169

Sleep as Rhythm and Reflex The question of memory provided Piéron with a unique topic through which he could weave a unified tapestry of scientific psychological research that touched on several popular themes, while at the same time overcoming seemingly intractable debates regarding the relationship between the individual and society, the body and the mind, religion and science, and even the old and new worlds. It is within precisely this framework that Piéron simultaneously attacked the problem of sleep. Claparède had suggested that sleep was an instinct of defence, invoking the same teleological principles that Richet relied upon in his physiology. Piéron, on the other hand, rejected the idea that this defensive purpose somehow caused sleep. Instead, he described sleep as a form of memory. Sleep’s rhythmical form provided clear evidence of adaptation, but this periodicity did not arise out of any need for defence.145 It was rather a case of the force of history acting on the present. To distinguish between the chemical mechanisms that caused sleep and the organic memory that dictated its periodicity, Piéron devised a new method of investigation that was all the things he had advocated for a scientific psychology. His method of ‘enforced wakefulness’ was everything that the philosophic-introspective or clinico-pathological study of sleep was not: it was objective, laboratory-based, standardizable, and portable. Its evident brutality was not, at least in Piéron’s estimation, worthy of comment. Yet the fact that the majority of photographs included in Piéron’s thesis on sleep were of dogs desperately staving off their own strangulation gestured towards a new vision of sleep as not merely an object unto itself but an ahuman one at that (see figs. 10 and 11). The study of sleep had definitively been divorced from that of the study of dreams. Or so it seemed. Piéron’s investigational method followed a well-worn formula borrowed from French physiological science: induce a pathological state in the organism, observe the results, and infer from these the nature of the normal state.146 Claparède explicitly rejected this type of experimental analysis, arguing in a letter to Piéron that ‘You feel that sleep also has organic or toxic causes. I certainly agree; but we are here no longer concerned with normal sleep – the frontier into pathology has been crossed, and one is now dealing with exhaustion.’147 Piéron paid little heed to his friend’s criticism. Although Claude Bernard had codified this means of investigation for physiology, it was by no means novel for French psychological science by the time Piéron began his studies. As early as 1870, Théodule Ribot had proposed an identical method of

170

The Sleep of Others

10 Piéron’s work inaugurated a century of studying normal sleep by its abnormal extension. When this photograph was taken, Piéron informed his readers that the dog, Tunis, had been awake for 248 hours. The dog would still respond to its name, but any substantial movement would put pressure on its collar, leading to strangulation. (Henri Piéron, Le Problème physiologique du sommeil [Paris: Masson, 1913], 268)

11 Tunis, on the verge of complete exhaustion and death. As horrific as they seem, such images represented Piéron’s hope that sleep could be studied rationally and systematically as a phenomenon of observed behaviour. (Henri Piéron, Le Problème physiologique du sommeil [Paris: Masson, 1913], 269)

The Ends of Darkness

171

12 Clignard, another of Piéron’s dogs, asleep following an injection an hour and a half earlier. The injected substance, which Piéron dubbed ‘hypnotoxin,’ could not be isolated by other investigators. (Henri Piéron, Le Problème physiologique du sommeil [Paris: Masson, 1913], 316)

research, arguing that illnesses, be they artificial or naturally occurring, were ‘natural experiments,’ and could be used to illustrate the normal state of the mind or brain.148 Thus, hysteria, somnambulism, and other psychological diseases could be paired with the induction of hypnotism as appropriate investigative tools. To a similar end, Piéron, in 1907, described his method of ‘enforced wakefulness,’ as a pathological mirror of sleep that could help illuminate the normal mechanisms of this mysterious state.149 Piéron was not, as we saw in chapter 2, the first to propose preventing experimental subjects from sleeping as a primary means of studying sleep. Manacéïne, had described a similar method as early as 1894.150 Richet was a great supporter of Manacéïne’s work, and Piéron cited it numerous times in his dissertation. But such efforts were inevitably tied to sleep as a medical problem, not as a physiological state with its own peculiar natural history. Piéron – for the historically contingent reasons outlined above – was the first to attempt such a comprehensive construction of sleep as an independent object of research. It was in part by virtue of his efforts that sleep research ultimately took the peculiar forms it did in Europe and North America. His methodological ‘predecessors’ might well have provided useful inroads to a (much-

172

The Sleep of Others

needed) comparative history of sleep, but they played little role, if any, in the creation of sleep as an investigative object. Piéron’s outlook was a pragmatic one, and his commitment to experimentalism was by no means exclusionary. He was equally impressed by American comparative psychology, so, combining the two approaches, his dissertation offered a phylogenetic as well as a physiological description of sleep. Indeed, he took the historical origins of sleep as a starting point for his dissertation, which was a particularly effective way of both showcasing the study of comportement and circumventing the spiritual questions that had surrounded sleep for so many centuries.151 Piéron, for example, could not have considered the issue of whether or not plants ‘slept’ as a valid problem had he taken consciousness, rather than behaviour, as his starting point. And, while the sleep of plants might well be depicted as a ‘real question,’ their religious sentiments were obviously beyond the pale of serious scientific research. Once he had framed the question, Piéron’s response to it relied upon an application of his experimental method. It certainly seemed, he argued, as though some sensitive plants, such as Mimosa pudica or Hedysarum gyrans, slept, because they periodically entered a state of rest. But this periodicity was not, in itself, enough to determine the presence of sleep. To be defined as sleep proper, the phenomena had to indispensable to the survival of the organism, which was itself demonstrable by experimental intervention: ‘We consider sleep to be a necessary periodic state, this periodicity being relatively independent of exterior circumstances, and characterized by the suspension of the complex sensori-motor relations that unite the organism to its milieu and enable it to continue its life, and, in particular, its means of nutrition. It remains for us to see if the states considered as states of sleep do or do not possess these diverse characteristics, and whether or not the common name of sleep is justified.’152 Piéron found that, even if sensitive plants were deprived of sleep by placing them in continuous illumination, their rhythms of motion would disappear after a few days and they would continue to thrive. Their ‘sleep,’ such as it was, was thus almost entirely dependent upon external, rather than internal, circumstances. Their periodicity was an example of an ‘anticipatory reflex,’ but it was not true rhythmic sleep. Extrapolating upon his work with plants, Piéron made a similar argument concerning the hibernation of certain mammals, which he described as an adaptation to external conditions of cold and lack of food. It could be eliminated if these extreme conditions were removed, and was therefore not an example of true sleep.

The Ends of Darkness

173

Piéron’s main experimental subjects were mammals, since their physiological behaviour displayed the greatest amount of variation in the concomitants of sleep, which included changes in circulatory activity, respiration, digestion, and secretion. Unlike Mosso or Hammond (discussed in chapter 2), Piéron did not feel that such concomitants were necessarily causally linked to sleep. These rhythms were, like organic memory, simply expressions of the remarkable adaptive ability of the organism. They often persisted in the absence of the phenomena with which they were associated, and were thus of interest to Piéron in their own right. But altering sleep patterns was sometimes a convenient way of studying such alterations. In 1907 he and Toulouse studied the inversion of the diurnal temperature rhythm among night-shift nurses at the Villejuif asylum.153 The study was part of a series of investigations into the physiology of work that were becoming an integral part of the human sciences in France and that were typified by the discovery, in 1905, that female typists were far more efficient at their work than their male counterparts.154 It was already known that body temperature was slightly higher during the day than at night. But the availability of a group of subjects with inverted sleep/wake routines allowed Toulouse and Piéron to see if this rhythm could be changed. They discovered that their subjects actually maintained higher temperatures during sleep when they first changed shifts. The rhythm’s normal relationship to sleep and wakefulness was fully restored only five to six weeks later. It was the study of experimental insomnia in dogs, however, that Piéron felt to be his most important contribution to the problem of sleep. The technique was simple enough. Piéron attached a dog’s collar to a leash that was hung at a point on a wall above the dog. The leash was long enough to allow the dog to sit or move around, but only just. If the animal tried to lie down, the collar would tighten around its neck, and begin to strangle it. Thus, the dog was obliged to maintain consciousness for as long as possible, but its relative inability to move implied (at least to Piéron) that it was subjected to little stimulus. Just as Mosso and Binet had felt that the ergograph effectively isolated fatigue in a single muscle group, so, too, did Piéron argue that he had found a way of isolating sleep by controlling for the effects of fatigue. His reliance on animal subjects also meant that he had eliminated the problems that had plagued Binet and Vaschide in their studies. Piéron’s limited concern for either his subjects’ emotional state (a prerequisite for conducting such experiments) or their willingness to cooperate in the experiment meant that he could be assured that each subject would

174

The Sleep of Others

perform to the best of its ability, since its very survival was at stake.155 Responses were thus standardized in a way that had been impossible for the competitive athletes or distracted students used by Binet and Vaschide. And, unlike Claparède’s analysis of interest and its relevance to sleep, it was the necessity of sleep, not the desire for it or the feeling of sleepiness, that was the object of Piéron’s investigations. More important, all sources of fatigue were internal, and purely physiological. The dogs were not, Piéron argued, exhausting themselves by motor activity or by responding to external stimuli. Their only expenditure of energy was in attempting to stay awake as long as they possibly could. Thus, Piéron had discovered a way to make the minimum physiological requirements of consciousness visible. Remarkably, some of Piéron’s animals were able to stay awake for up to five hundred hours. When they eventually strangled, or slipped into a coma, Piéron immediately set to work studying the physiological transformations they had undergone. Or rather, his assistant, René Legendre, seemed to have conducted most (if not all) of the pathoanatomical work from which Piéron drew his conclusions.156 Just as his neo-Lamarckian masters had struggled to identify a chemical mechanism for heredity, so, too, did Piéron hope to discover the chemical basis of sleep. His methodology fit his research into two important aspects of early-twentieth-century physiology – immunology and fatigue research. His attempt to transfer the pathological state of the exhausted animal to a normal, or ‘fresh,’ animal parodied both Richet’s successful transfer of immunity from one animal to another three decades earlier and the contemporary anti-diptheria serotherapy ubiquitous throughout North America and Europe. Similar experiments had been conducted independently by Mosso and by Wilhelm Weichardt, a physiologist at the University of Erlangen, but both had studied subjects exhausted by fatigue, not those that succumbed to the natural sleep that Piéron felt he had generated in his laboratory.157 After examining the animal brains for aberrations (he found some minor changes in the pyramidal cells of the frontal cortex, but Piéron was no anatomist, and this played little role in his argument), he took various parts and fluids from his exhausted dogs and transferred them to ‘fresh’ animals. After injecting whole blood, serum, and cerebral pulp into the peritoneal cavities, veins, and brains of normal dogs, Piéron finally discovered an active substance in the cerebro-spinal fluid. When he injected this material into the fourth ventricle of a normal dog’s brain, the animal presented ‘a more or less irresistible need to sleep.’158 This effect, he

The Ends of Darkness

175

insisted, could not be due to the increased pressure resulting from a greater amount of fluid now in the animal’s brain, because he had first withdrawn a like amount of fluid from the animal before the toxic injection. To characterize his newfound ‘hypnotoxin,’ he followed textbook immunochemical practice. Its effects were diminished when the fluid that contained it was heated to around 65oC, or when it was exposed to oxygen. It could be precipitated by alcohol, and could be extracted from this precipitate by water, although this extract was somewhat less powerful than the original substance. In brief, hypnotoxin was a unique substance generated not by muscular exertion but through the normal activity of the nervous system. Its effects were specific to this system. Where muscular fatigue produced a marked hypothermia, for example, Piéron found no such phenomena in his sleep-deprived dogs. And, in contrast to hypnotoxin, fatigue toxins were produced in the blood and destroyed there, leaving no trace of damage to the nervous system.159 The effects of hypnotoxin, Piéron argued, could be demonstrated only when it was injected directly into the brain. So, while Piéron continued to speak of sleep in terms of fatigue, he now described sleep as a form of nervous fatigue. It was engendered not by the labours of thought (he refused to use this word to describe the mental activity of animals) but by that of sensori-motor activity. The production of hypnotoxin ebbed and flowed according to the rhythms of the wakeful exercise of the sensori-motor apparatus and its repose. This rhythmic regulation, in fact, was the very phenomenon that provided the physiological basis of all behaviour – the phenomenon that Piéron, the psychologist, wanted to investigate. Without it, the work of the sensori-motor system could not exist. His colleagues seemed to concur. In a long and enthusiastic review of Piéron’s published thesis, Yves Delage noted that Piéron had demonstrated how ‘the work of the nervous system, and in particular the sensori-motor work necessary for the maintenance of equilibrium and the motor reactions appropriate to all the diverse excitations that can appear,’ was in fact impossible without sleep.160 This physiological definition of sleep effectively eliminated, thought Piéron, the idea that hypnotism was a kind of sleep. It seemed clear that the hypnotic state was in no sense necessary to the organism’s survival, so, from a biological perspective, it was unworthy of attention.161 Of course, the hypnotoxin that Piéron had isolated could not explain sleep on its own. The endless days of wakefulness that were necessary to generate enough hypnotoxin to put a fresh animal to sleep were merely a simulacra of the sleep that occurred naturally. Clearly, hypnotoxin was

176

The Sleep of Others

but a mechanical trigger and needed to work in concert with something else in the organism to generate normal sleep. It was at this point that Piéron’s reticence to speculate broke down. In a lengthy concluding chapter, Piéron paid tribute to Claparède’s idea that sleep could be explained in psychological terms, as interest. While he avoided the idea of organismic defence so dear to Richet, Piéron acknowledged that it was ‘Claparède who, in his theory of the sleep-instinct, has found a seductive formula, and diffused a “biological” conception of sleep, the success of which has not yet been determined.’162 Claparède’s depiction of sleep as a state of disinterest had, of course, been proposed by many others, including Leibniz, Liébeault, and Bergson. But it was incomplete, argued Piéron, because it postulated no mechanism. But, if it was rewritten in physiological terms as the manifestation of an ‘inhibitory mechanism,’ in the brain, conditioned to be triggered by an accumulation of hypnotoxin, Claparède’s biological theory was indeed the most ‘complete’ theory of sleep available.163 The difficulty here concerned the explicit teleology of Claparède’s idea. Sensori-motor fatigue did, eventually, seem to cause irreparable neurological damage, but did this necessarily imply, as Claparède had suggested, that protecting the organism against such damage could be invoked as a cause of sleep? Piéron felt this formulation to be meaningless: by the same token, one could say that organisms breathed to prevent asphyxiation, or ate to prevent starvation. Talk of final causes was sterile and misleading, and Claparède’s proposal was, in fact, beyond the pale of experimental investigation. In place of Claparède’s finalism, Piéron proposed an explanation of sleep in terms of ‘anticipation.’ Hypnotoxin was but an internal contribution to the appearance of sleep – a physiological trigger. Sleep’s periodicity, on the other hand, was an example of the persistence of anticipatory rhythms – rhythms that had been established in the organism through the repeated association of sleep with relaxation, darkness, and muscular fatigue.164 Piéron thus offered a description of sleep that was more synchronic than diachronic. Following the style of his positivist, neo-Lamarckian forerunners, he devised a method that separated the humoral (physicochemical) aspects of sleep from its habitual rhythms. But, while the latter offered Piéron the potential of aligning psychology with biology under the rubric of organic memory, it was the former problem that was the focus of his experimental investigations. Whereas Claparède had formulated the question of sleep as a question about how sleep had evolved, Piéron was more modest, content to add yet another factor,

The Ends of Darkness

177

hypnotoxin, to the growing list of concomitants of sleep. Sleep was not an active phenomenon for Piéron as it was for Claparède: the inhibitory mechanism he proposed could not be said to have its function as its cause. His hesitancy to accept such an approach cannot be separated from the anti-clerical and anti-metaphysical sentiments that permeated the Third Republic in the first years of the twentieth century. Thus, Piéron preferred to describe sleep exclusively in terms of its mechanical causes, which, though they might not always be present, might be capable of acting at a distance, through the phenomena of anticipatory rhythms (a concept which bore a substantial similarity to Ivan Pavlov’s idea of ‘conditioned reflexes,’ discussed in chapter 4). The onset of hostilities in the summer of 1914 reduced sleep to relative insignificance once again, even as psychoanalytic approaches to dreaming thrived in the war’s aftermath. Efficiency and performance continued to be dominated by concepts of fatigue, but little attention was paid to sleep until the outbreak of a strange new disease, ‘sleeping sickness,’ began to appear in epidemic form in Europe and North America just as the war was drawing to a close. Despite this brief hiatus, both sleep and dreaming had arrived as autonomous scientific objects by the dawn of the twentieth century. Definitive, unique methods of investigation had been construed around each, and they were at least nominally capable of being pursued in the absence of the traditional medical or religious concerns that had dominated their history to that point. Darkness had come into its own as something other than the absence of light and life.

4 Inhibition and Disease

The Russian physiologist Ivan Pavlov began studying sleep around the same time as Henri Piéron, but the origin of their interests and their methods were very different. Whereas Piéron turned to sleep as a way of sorting out problems of physiology from those of psychology, sleep quite literally interrupted Pavlov’s experimental study of conditioned reflexes. He was thus forced to incorporate sleep into his theory of nervous function as a balance between excitation and inhibition. And, while Piéron’s methods were eventually adopted as the key mode of performing sleep in the laboratory, Pavlov’s work came to dominate the clinical contours of sleep. The appearance of a new epidemic disease, encephalitis lethargica, extended sleep’s significance in the clinic. This strange disease also helped localize the control of sleep in the mid-brain, which investigators argued had been damaged during the course of infection. Ultimately, attention turned away from sleep’s relationship to fatigue and sensation, and moved towards an analysis of how the brain regulated sleep’s periodicity.

Myths, Inhibition, and Accident Just as Piéron, the young psychologist, was moving towards physiology, Ivan Pavlov, a Russian physiologist well into middle age, was making gestures towards psychology. Sleep proved a useful meeting ground for each. Pavlov (1849–1936) began his investigations of ‘higher nervous activity’ around 1901 and two years later, when he was based at the Military Medical Academy in St Petersburg, presented his first paper on conditioned reflexes, at the XIV International Congress of Physiology in Madrid.1 In that paper, he described the concept of ‘conditioned reflexes’ as an outgrowth of his work in the physiology of digestion, for which he was to win

Inhibition and Disease 179

a Nobel Prize in 1904. Pavlov divided the digestive process into two phases. The first involved salivation, which in turn triggered the second phase of nervous and chemical secretion that helped break down the food once it entered the stomach. Pavlov and his collaborators dubbed the first phase ‘psychic secretion,’ because they noticed that their experimental dogs began to salivate even before the food was placed in their mouth. The sight or smell of meat, or simply placing the dog in its harness, would cause it to salivate. When one of his students, A.T. Snarskii, attempted to explain this phenomenon in terms of the dog’s subjective sensations and desires, Pavlov objected and insisted that the phenomenon must be described in the classical physiological language of reflex. This story of the ‘birth’ of the method of conditioned reflexes has often been repeated but does not bear close examination. As it turns out, Pavlov misrepresented his student’s work, which argued that psychic secretion was a low-level, habitual association that had nothing to do with problems of the will or judgment that were the stock-in-trade of much introspective psychology.2 In contrast, Pavlov harboured aspirations to enter the psychological field and turned to the work of another student, I.F. Tolochinov, to do so. Tolochinov’s 1902 thesis argued that psychic secretion had to be understood as a reflex that, unlike those reflexes studied by physiologists, had been formed by experience. Pavlov appropriated Tolochinov’s thesis and linked it to I.M. Sechenov’s famous 1863 essay on the ‘reflexes of the brain,’ an old touchstone for liberal progressives who set their materialism against the Orthodox tsarist regime that ruled Russia. Pavlov’s method of conditioned reflexes, which he announced in 1903, gave experimental flesh to Sechenov’s ideas. Initially, the notion of a conditioned reflex was rather simple. Pavlov divided all stimuli into two groups: unconditioned and conditioned. The former were innate and would elicit a response without any training. Chewing food, for example, would always provoke salivation. Conditioned stimuli provoked a response only when repeatedly associated with unconditioned stimuli. Pavlov demonstrated, for instance, that the mere sound of a metronome could make a dog salivate, provided it had been presented with food in conjunction with this sound numerous times in the past. From this straightforward paradigm, Pavlov and his many collaborators developed a complex and sometimes unwieldy series of experiments that purported to explain psychological phenomena in physiological terms. Within a few years, Pavlov had ceased to describe conditioned reflexes in terms of excitation only. He began to describe his experimental results in terms of ‘inhibition.’

180

The Sleep of Others

This was a felicitous choice of terms on Pavlov’s part. The word ‘inhibition’ had a rich history over the course of the nineteenth century, and it played a crucial role in tying together arguments about moral conduct, self-control, mental order, and physiological function.3 The German term (Hemmung) first gained popularity during the 1820s and 1830s, when J.F. Herbart, a professor of philosophy at Königsberg, used it to describe mental order and controlled conduct. Herbart had rejected Kant’s assertion that psychology could never itself become a true science because its objects, introspective observations, were related only in time, not in space, as was the case with the objects of the physical sciences. According to Herbart, the nature of mind could be rationally deduced from general principles which in turn could be empirically tested. To this end, he proposed a system of mental dynamics involving the relative strengths of mental elements: a stronger ‘presentation’ would fill consciousness by ‘inhibiting’ weaker ones, which would then become unconscious.4 Thus, Herbart married British associationist psychology with the language of dynamics. Wilhelm Griesinger introduced this concept of inhibition to German psychiatry during the 1830s, and it also became part of reflex physiology through E.H. and E.F.W. Weber’s 1845 discovery that the stimulation of a frog’s vagus nerves slowed, or ‘inhibited,’ its heartbeat. By the end of the century, inhibition (from the Latin root in + habere, ‘to hold in’) had crossed several disciplinary boundaries, as it came to signify any regulatory force that opposed the free reign of excitation and thus forged order out of chaos. Psychologists, neurologists, psychiatrists, and physiologists alike contributed to inhibition’s expansive meaning. H.S. Curtis, for example, titled his 1898 study of restlessness in children ‘Inhibition.’ The author, a student of American psychologist G. Stanley Hall, worried about the term’s ambiguity: ‘The title under which this article appears may prove deceptive … those who would not be interested may be led to examine it, while those who might find in it something of interest will pass it by unheeded.’5 He feared his audience would think that his paper dealt with the finer points of physiology, when it was in fact based on his observations of children’s behaviour in school. But Curtis turned the word’s indeterminism to his advantage by linking it to evolutionary theory (a favourite strategy of Hall’s). Curtis, who also campaigned for the incorporation of play in education and the creation of school playgrounds, argued that educators who used strict discipline to prevent children’s nervous, purposeless, and distracting

Inhibition and Disease 181

movements in the classroom had the wrong approach. They should instead work harder to attract their pupils’ attention, thereby harnessing the natural power of ‘inhibition,’ which he took to being ‘nearly equivalent to natural selection.’6 Curtis developed this curious analogy by arguing that the brain’s attempt to exercise continual restraint over the reflex activity that helped to develop the musculature paralleled the ‘purposeful struggle’ within and among species that maintained order in the wild.7 Clearly, inhibition signified order in an enormously wide variety of contexts. Pavlov first mentioned inhibition in 1909, and he interpreted it primarily in motor terms. The English neurophysiologist Charles Sherrington described inhibition as a necessary feature of the integrative action of the nervous system around the same time.8 But, while Sherrington eventually came to define the current idea of inhibition as a force that modulates the transmission of nervous signals at the level of the synapse – the chemical junction between two different neurons – Pavlov’s use of the word was different. He conceived of inhibition in strictly systemic, rather than morphological, terms.9 Inhibition was a force that was in every way the mirror of excitation. The two forces opposed and balanced each other; whereas excitation provoked movement and response, inhibition prevented it. It is perhaps not surprising that Pavlov found the most dramatic manifestation of inhibitory force in sleep and hypnotism, where movement and responsiveness appeared to be severely restricted. Pavlov was not alone in linking sleep to inhibition. The French neurologist Charles-Édouard Brown-Séquard proposed a theory of dynamogénie, which depicted all diseases in terms of an imbalance of excitatory and inhibitory forces within different organs of the body. His doctrines were well received by neurologists, who proceeded to construct a therapeutic system based on the idea that disordered reflex activity could be carried by the nervous system and spread to other organs.10 BrownSéquard’s ideas were grounded in his 1856 discovery that the removal of the adrenal glands quickly killed experimental animals. The function of these glands, Brown-Séquard suggested, was to control and regulate, not to excite. The study of endocrinology blossomed, as investigators uncovered numerous and previously unforeseen connections between various systems in the body. In 1889 Brown-Séquard tried to fit the contentious problems of sleep and hypnotism within this schema, under the rubric of dynamogénie, and as part of his criticism of Charcot’s penchant for

182

The Sleep of Others

attributing control of a variety of different mental functions to specific regions of the brain.11 The phenomena of sleep and (more important) hypnotic trance were global results of the inhibition of mental activity. They were not the products of a stimulated brain centre. Pavlov might equally have taken his initial inspiration for applying the thesis of inhibition to sleep from debates surrounding hypnotism. But his biographers and historians of conditioned reflexes have been largely silent on the subject, perhaps, in part, because of the incomplete state of Pavlov’s archive.12 It is clear that his theory was very similar to that of Rudolf Heidenhain (1834–1897), who taught Pavlov physiology at Breslau in the late 1870s and early 1880s and who was then claiming that hypnosis was the product of a cortical inhibition arising from overstimulation of the visual, tactile, or acoustical senses.13 Without citing his former teacher, Pavlov proposed a similar theory in 1910, just a year after he began describing conditioned reflexes in terms of a balance between inhibition and excitation. But perhaps hypnotism was not the real target of Pavlov’s arguments. As pointed out in chapter 2, hypnotism was in decline by the first decade of the twentieth century. Why would Pavlov raise the question anew, and then only to trot out a wellworn theory? It might have been a generational issue: Pavlov was over sixty, and his career spanned a period during which many physiologists used hypnotism to frame questions about the nature of the will in experimental terms. Or perhaps Pavlov simply wanted to assimilate this old problem, hypnotism, and incorporate it into his theory of ‘higher nervous activity,’ thereby making his new investigative method seem that much more comprehensive. One possibility that seems not to have been considered is that sleep actually played a pivotal, rather than incidental, role in Pavlov’s experimental practice around 1910, and that this importance persisted throughout the remainder of his career. Pavlov himself offered a strikingly simple explanation for his interest in the problem of sleep – it held up the work of his laboratory. In a lecture entitled ‘Some Fundamental Laws of the Work of the Cerebral Hemispheres,’ read before a meeting of the Society of Russian Physicians in 1910, Pavlov remarked that ‘for many years we noticed that our dogs became sleepy; this interrupted our work, for the conditioned reflexes weakened and disappeared.’14 Pavlov’s notebooks from 1911–13 indicate that the experiments of M.N. Erofeeva, one of his first female praktikany or coworkers, were particularly important in this regard.15 He repeated the story in 1915, at a meeting of the St Petersburg Biological Society.16 It

Inhibition and Disease 183

must be recalled, however, that this research was not operating under the same conditions as physiological experiments conducted by Bernard or even Piéron. Whereas these researchers worked independently or in very small groups composed of a leader, a technician, and at most a few students, Pavlov’s laboratory was quite literally a factory dedicated to the production of both knowledge and useful products.17 Yet Pavlov’s students, despite being physicians, had little knowledge or appreciation of physiology. It was the need to redress precisely this sort of ignorance that sustained Pavlov’s Physiology Division of the Imperial Institute for Experimental Medicine in St Petersburg. The Physiology Division was a place for physicians to gain the ‘scientific’ training that had recently been made mandatory for military doctors. The students were expected to choose a research topic, investigate it, write it up and defend it in a mere two years. Developing methods for the formulation of research questions, surgically preparing dogs for observation, keeping experimental animals healthy for as long as possible, and preparing products for market (particularly gastric juice sold as a treatment for dyspepsia) were thoroughly routinized. They had to be, for around one hundred of these workers passed through Pavlov’s laboratory between 1891 and 1904, and about three-quarters of them successfully defended their theses. Any obstacle to this progress needed to be dealt with quickly and efficiently: sleep quite literally shut the assembly line down. So Pavlov dealt with the issue of sleep by building his theory of inhibition around it, thereby incorporating it into the production of knowledge within his laboratory-factory. Hypnotism provided a means of linking the problem of sleep (which was also a labour-production problem) to a long-standing, but recently abandoned, problem in psychological science. This interpretation of events is further strengthened by the fact that, around 1910, Pavlov was engaged in another attempt to refine his experimental assemblage. He hoped to redesign his laboratory in order to eliminate all possible sources of uncontrolled stimuli.18 Street noise had to be muted, vibrations dampened, illumination made uniform, and climate controlled in order to minimize any interference with the process of forming conditioned reflexes. This new motif of isolation applied equally to the relationship between experimenter and animal subject. To avoid experimental animals picking up unintended behavioural cues from investigators, the two had to be somehow separated (fig. 13). Although Pavlov’s new laboratory was constructed only after he obtained renewed state support under the ‘market communism’ of Lenin’s New

184

The Sleep of Others

Economic Plan during the mid-1920s, his description of such a laboratory in 1910 indicates that sleep was just one of many factors Pavlov was attempting to control for in his experimental program. But, if sleep first appeared to Pavlov as an accident of his experimental practice and institutional design, he nevertheless considered it an integral part of his research program. A 1923 letter from Pavlov’s most dedicated American disciple, W. Horsley Gantt (1892–1980), described his master’s research to the Harvard physiologist Walter B. Cannon in the following terms: ‘He [Pavlov] told me that he considered his most important work that on sleep and the conditioned reflexes; he is also doing some showing [sic] that acquired characteristics can be inherited and still a great amount on digestion, although he does not feel so much interest in this.’19 Gantt highlighted Pavlov’s work on sleep but said nothing of any interest in hypnotism. That same year, Pavlov set off on his first tour of the United States, a tour that had been organized in part by Cannon because of his fear that Pavlov’s work was suffering greatly in the Soviet Union (fig. 14). Sleep figured prominently in both of the published papers that appeared in America following his lecture tour.20 Pavlov’s interest in sleep endured until the very end, as indicated by the title of one of his very last lectures presented a year before he died in 1936: ‘The Problem of Sleep.’21 Roger Smith’s historical analysis acknowledges how important Pavlov’s work on sleep was to his concept of inhibition, but says nothing as to its fate.22 Sleep is treated as a ‘case study’ of something greater, not an object of interest in its own right. Pavlov’s contemporaries were not so taciturn. His views netted considerable attention, for example, in Piéron’s contribution to a French psychological encyclopedia, which also featured analysis of Claparède’s discussion of sleep and wakefulness.23 Pavlov claimed to have isolated three kinds of inhibition describing different relationships between the conditioned and unconditioned stimuli encountered in the dog’s environment, all of which served to eliminate or reduce the strength of the conditioned response, which he and his workers measured by meticulously collecting drops of saliva and other body fluids following each experimental trial. ‘External’ inhibition referred to the effect a powerful new stimulus had over another weaker one. A dog, for example, could be conditioned to reject food altogether if the food was repeatedly accompanied by an electrical shock. The reflexive response to food, signified by drooling, could be thus inhibited. ‘Internal’ inhibition described the varying fate of conditioned reflexes over time and in novel situations. If the unconditioned

Inhibition and Disease 185

13 Photographs illustrating the separation of animal and investigator in experiments in a Pavlovian laboratory. (Ivan Pavlov, Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex, translated and edited by G.V. Anrep [New York: Dover Publications, 1960], 33)

186

The Sleep of Others

14 Pavlov and W.B. Cannon met for the first time at the Carnegie Nutrition Laboratory in Boston on 26 June 1923. (Courtesy Harvard Medical Library in the Francis A. Countway Library of Medicine)

stimulus, for example, did not consistently accompany the conditioned stimulus, the conditioned response could become completely inhibited and disappear. A similar phenomenon could result when a new stimulus was presented alongside an established conditioned stimulus, but without the original, unconditioned stimulus. In such situations, the new stimulus could inhibit the older conditioned reflex. But this ‘extinction’ of responses was not always complete, and Pavlov noted that conditioned reflexes could often be revived, provided the conditioned stimulus was once again accompanied by an unconditioned stimulus.

Inhibition and Disease 187

Through this scheme, Pavlov attempted to render the concepts of associationist psychology amenable to experiment. Sleep, Pavlov argued, represented a third kind of inhibition.24 It was not, strictly speaking, ‘external,’ for its source was not always found in the organism’s environment. Nor was it strictly ‘internal,’ because it was not merely the triumph of one stimulus over another. Most important, sleep was a kind of inhibition that encompassed not just the awareness of something but awareness itself. Pavlov’s explanation of sleep’s appearance was clearly derived from the original frustrations that he and his praktikany had had with their experimental animals: in a normal situation, organisms would fall asleep following the repeated presentation of a single stimulus, just as they did in Pavlov’s laboratory. Sleep (and hypnotism) thus illustrated a particular principle of internal inhibition – the law of irradiation – which Pavlov’s contemporary, Vladimir Mikhalovich Bekhterev, described as the most important law of the activity of the nervous system.25 Excitation and inhibition, Pavlov argued, irradiated across the surface of the cerebral cortex, forming a ‘mosaic’ of points that guided the behaviour of the animal at every turn.26 Many Pavlovian experiments were committed to localization, in the sense that they assumed that the surface of the skin had an analogous ‘cortical projection,’ with each point on the skin corresponding to a unique point on the cortex. But stimulation was not simply a matter of excitation, and, in his explanation of the reciprocal influence of excitation and inhibition, Pavlov abandoned localization in favour of a more holistic approach: ‘I shall not commit a great error,’ Pavlov said in 1910, ‘if I liken these two phenomena [excitation and inhibition] to positive and negative electricity.’27 Excitation could not exist without a corresponding inhibition, just as electricity could not flow without a positive and negative charge. Thus, every excitation involved a wave of inhibition that irradiated across the cortex. With the repeated presentation of the same stimulus, this inhibition became more and more generalized, finally ending in a state of sleep, the complete irradiation of inhibition across the cortical surface, the function of which, Pavlov repeatedly noted, was to protect the cortical cells from the damage that would otherwise result from extreme fatigue. Such repetitive stimuli, as Pavlov discovered in 1910, were everywhere in his laboratory. Even something as simple as the harnesses that restrained the animals during experiments could become an unintended part of the dogs’ repertoire of sensations that created a conditioned reflex, and that could accidentally provoke an inhibitory response which rendered an active, responsive animal into a sleepy liability.

188

The Sleep of Others

The Irradiation of Sleep in Psychiatry If the origins of Pavlov’s theory of sleep were primarily practical, rather than theoretical, it is perhaps no surprise that it was in the practical domain that they were most welcome. Positivist psychologists like Piéron generally dismissed Pavlovian terminology as unnecessary.28 Predictably, Claparède suggested that Pavlov’s theory of inhibition explained nothing at all about normal sleep, which was not induced by prolonged excitation (as in Pavlov’s experiments) but which was instead a periodical phenomenon.29 Psychiatrists and neurologists, on the other hand, had already started to work with ideas similar to those of Pavlov. Treating mental illnesses with hypnotic drugs long predated Pavlov’s work; sodium bromides, for example, were used to treat ‘hysterical epilepsy’ as early as the 1850s.30 But the interwar period saw a renewed effort to treat schizophrenia and psychoses with barbiturates, a new group of compounds that began to be mass-marketed as a treatment for insomnia around 1905. Similar drugs continued to appear on the market, and in the early 1920s Jakob Kläsi, a psychiatrist working under Eugen Bleuler at the Burghölzi clinic in Zurich, began attempts to treat schizophrenia with a new hypnotic drug called Somnifen.31 The renewed popularity of various forms of ‘sleep therapy’ coincided with a number of other physical cures for mental disorder, including the induction of malarial fever in neurosyphilitic patients by Julius Wagner-Jauregg in Vienna (1917), insulin coma (1933), Metrazol convulsion (1934), and electroshock (1938) therapy for schizophrenics. These therapies, along with the wildly popular psychosurgical procedure of lobotomy, pioneered by the Lisbon neurologist Egas Moniz and popularized by the American neurologist Walter Freeman and his neurosurgeon sidekick, James Watts, were a common response by clinicians to the growing number of incurable schizophrenics being ‘warehoused’ in asylums during this period. Such therapies were, at one and the same time, a potential cure, a convenient way of conducting research by exploiting institutionalized populations whose consent was largely inconsequential, and a potential solution to the administrative difficulties of managing a growing asylum population with dwindling resources.32 In psychiatric clinics, as in Pavlov’s laboratory-factory, interest in and application of sleep was not independently pursued but followed the convergence of intellectual, technological, ethical, and administrative issues. This is not to say that Pavlov’s ideas prompted the explosion of sleep therapies in psychiatry. There is no indication, for example, that Kläsi

Inhibition and Disease 189

ever invoked Pavlovian theory to justify his practice of putting patients to sleep for as much as eleven days at a stretch. But, during the 1930s, Pavlov did begin promoting sleep therapy in the Soviet Union by arguing that, if hypnotic symptoms were indeed a natural reaction to the neuroses, the induction and extension of the same phenomena might provoke recovery in severe mental illness. Pavlov promoted the use of some of the older bromides in psychiatric therapy, and sodium amytal, a barbiturate, was introduced to Soviet psychiatry by Pavlovian psychiatrists in 1937.33 In the wake of the often brutal implementation of Josef Stalin’s first two five-year plans, Soviet asylums were, even more than their Anglo-American counterparts, filling up during the 1930s.34 The trend continued in Soviet asylums following the Second World War, when patients were often narcotized for a period of one-and-a-half to two days and then treated with an equal length of induced convulsions or fever. One of Pavlov’s students, Ezras Asratian, used sleep therapy to treat phantom pains and traumatic shock among the wounded. Such therapies were part of the enduring Pavlovian legacy, which included not only the spread of the method of ‘classical’ conditioning among American psychologists but also the use of behaviour therapy in American psychiatry.35 In contrast, those few physiologists or psychologists interested in sleep found little of value in Pavlov’s method of conditioned reflexes. Pavlovian experimentation relied on the induction of sleep, not its deprivation, as Piéron had proposed. This method of substituting an artificial form for what was perceived to be the deprivation of what came naturally perhaps seemed too similar to the use of hypnotism as an investigative tool for the study of sleep, which had by then been widely rejected. Of course, one of the greatest differences between the method of conditioned reflexes and the hypnotic studies of the 1880s was in Pavlov’s reliance on animal, rather than human, subjects. This naturally invoked a radical shift in the language used to describe the relevant mental phenomena, but it maintained, as Claparède pointed out in his encyclopedic review of the field, an almost complete disregard for the periodic, rhythmical nature of sleep. Whereas Piéron had approached sleep as a problem demanding the creation of a unique methodology, Pavlov had run into sleep as an obstacle that interrupted his experimental labours. He accounted for it by simply assimilating it into his theory of inhibition. This difference between the two investigators is further illustrated by examining the legacy of the graphical method. Piéron’s work at Villejuif had relied on the devices that Marey and Mosso had contrived

190

The Sleep of Others

to observe the temporal course of phenomena without disturbing them. His approach to sleep generally (though not exclusively) relied upon just this sort of ‘non-interventionist’ thinking. Pavlov on the other hand, saw the graphical method as little more than an afterthought. Drops of saliva measured the relative strength of a reflex, with little regard for the nature of this strength as a function of time. This is not to say Pavlov’s position on the graphical method did not change. Indeed, he offered the following intriguing comments in a lecture delivered at the Military Medical Academy in Petrograd in 1924, just a year after he presented his new ideas on sleep to an American audience: ‘I believe the aggregate of facts given in the present lecture can be taken as sufficient proof of the view that sleep and internal inhibition are fundamentally one and the same process. I personally do not know, up to the present, of a single fact in all our researches which contradicts this conception. It is to be deplored, however, that we have as yet no reliable graphic method of registration of sleep. On some occasions we tried to apply for this purpose a graphic registration of the position of the head of the animal. A perfection of some such method for the graphic registration of sleep is greatly to be desired, so that the whole evidence regarding sleep can be expressed in an exact quantitative manner.’36 Despite such sentiments, Pavlov continued to assume that inducing sleep in the laboratory through conditioning was an appropriate and comprehensive substitute for natural sleep. Just two years before Pavlov delivered this lecture, however, a young physiologist in Chicago was taking up the graphical method in his own study of sleep. Nathaniel Kleitman, whose work will be discussed in the following chapter, had already been in communication with Pavlov by this time and was busy positioning his own research against that of the master of the conditioned reflex. If Pavlov’s theory of sleep had little impact among psychologists, the same certainly could not be said for his methodology, which clearly influenced the behavioural studies that burst onto the American scene in the years just before the United States entered the First World War. Behaviorism would likely have blossomed in the United States with or without the Pavlovian contribution to the field. But Pavlov, a Nobel laureate, lent a substantial amount of scientific credibility to this enterprise, just at the time when American psychologists were reconstructing their field towards a professional and experimental orientation. Robert Yerkes, who was struggling to legitimate his comparative psychology within the Department of Philosophy at Harvard, first introduced

Inhibition and Disease 191

Pavlov’s method of conditioned reflexes to an American audience in 1909.37 Pavlov’s work in digestion was well known to Yerkes’ ally, Walter B. Cannon, and the physiological origins of the method of conditioned reflexes proved a potent weapon against Yerkes’s introspectionist overlord, Hugo Münsterberg, who was reluctant to promote him.38 Although Yerkes’s own research made little of the method, J.B. Watson’s 1913 call for psychologists to adopt behaviour, rather than consciousness, as the proper object of a scientific psychology was followed by his endorsement of the method of conditioned reflexes in his presidential address to the American Psychological Association two years later.39 But, while functionalist psychology had at least at least provoked some controversy over the question of sleep in its analysis of the biological significance of consciousness, behaviourism simply ruled consciousness out of court. As we have already seen in past chapters, sleep had long been framed in terms of a loss or diminution of consciousness.40 Traditional conceptions of sleep thus had very little to offer to the experimental study of behaviour. As a result, the question of sleep as an experimental problem was taken up not by psychologists but by those few physiologists interested in understanding the biological foundations of conscious experience. A Novel Pathology: Sleeping Sickness, 1917–29 The appearance, however, of a series of epidemics of sleeping sickness, dubbed ‘encephalitis lethargica’ by the Viennese neurologist Constantin von Economo in 1917, offered an exceptional opportunity for medicine to take up the question of sleep anew.41 The irony that sleep appeared as an epidemic disease just as it was being proffered as a psychiatric cure should not be lost here. It aptly illustrates the ambivalent nature of sleep: a sweet restoration of life’s energies on the one hand, and a frighteningly close relative of death on the other. And, like sleep therapy, the outbreaks of encephalitis lethargica opened up possibilities for the application of Pavlov’s theory of sleep, as neurologists struggled to frame this new disease in the language of inhibition. These epidemics brought sleep to the attention of neurologists, bacteriologists, physiologists, psychologists, and psychiatrists across the world. Massive amounts of funding and biomedical resources were turned towards the investigation of sleeping sickness and the treatment of the devastating effects it often had on its victims. Encephalitis lethargica earned sleep a place in the biomedical science of the twentieth century. It also gave sleep a

192

The Sleep of Others

physical location in the subcortical brain through Economo’s neuropathological research. Historians have generally failed to investigate the role of encephalitis lethargica in the growth of neuropsychiatry, preferring instead to focus on the worsening institutional conditions in asylums, the spread of psychoanalytic theory, and the emergence of a new wave of organic therapies for mental illness in the years between the wars. A possible reason for this exclusion might be encephalitis lethargica’s relative lack of rhetorical value for historical debate. It was neither a glorious triumph of modern medicine nor a poignant moment of insight for psychoanalysis. Its epidemic origins were never satisfactorily explained, its status as an independent disease entity was frequently challenged, and its role as an epidemic infectious disease paled in comparison to the millions who perished in the wake of the influenza pandemics of 1918. Yet, nonetheless, the disease proved to be an important touchstone for neurologists who, particularly in the northeastern United States, were then striving to make their speciality relevant in the midst of the triumph of laboratory-based public health. Concerns about another influenza pandemic, perhaps wrought first by outbreaks of avian flu, have recently brought sleeping sickness back into view during the early twenty-first century.42 Historical research on the topic, however, has largely been left to neurologists with a keen clinical interest in encephalitis lethargica but who are generally uninterested in how investigators’ understanding of the disease was shaped by the context of 1920s biomedicine, or in the impact the disease had on the study of sleep.43 When such problems are addressed, they are inevitably framed in negative terms that highlight the ‘deficiencies’ (in the words of one author) of neurology and psychiatry of the 1920s that prevented the advance of knowledge.44 Granted, investigators during this period clearly suffered from a lack of consensus regarding viral epidemics. It was not yet clear that viruses were not simply submicroscopic bacteria but an entirely new kind of infectious particle altogether. The general (but by no means universal) agreement that the outbreaks of encephalitis lethargica were in fact viral epidemics thus suffered from a surfeit of meaning.45 But, as we have seen in the case of inhibition, a term’s multiple meanings can render it a strategic tool. Interwar research in encephalitis can be considered ‘deficient’ only if we insist upon adopting the standpoint of early-twentieth-first-century diagnosis and therapeutics. Other narratives are certainly possible. If the history of the disease is written with the disciplinary development of neuropsy-

Inhibition and Disease 193

chiatry and sleep research in mind, it is clear that the actors themselves agreed that encephalitis lethargica presented an extraordinary opportunity. More important, assuming the presence of a context rather than the absence of scientific knowledge allows this interesting and largely unwritten episode in the history of disease to emerge in full relief. A curious form of sleeping sickness appeared among some of the patients in Julius Wagner-Jauregg’s neurological clinic in Vienna early in 1917. The hospital was filled with soldiers, whose brain injuries or neurosyphilis represented enormous potential for clinical research. These patients proved to be an excellent source of material for investigating neurological function. Shrapnel injuries left a visible trace on the skull and brain, so such wounds were a viable means of pursuing the program of brain localization. The treatment of insanity by organic methods, in particular the use of induced malarial fever to treat tertiary syphilis, was also taking shape in Wagner-Jauregg’s clinic. Soldiers, then, as now, were a pliable and accessible population of experimental subjects. The seven patients who appeared in January 1917, however, were unusual in several ways. First, they were civilians, with no traumatic brain injury. Secondly, they all displayed a series of symptoms, including fever, delirium, vision disorders, and convulsive movements, that did not fit well within any current diagnosis. A talented young neuroanatomist, Constantin von Economo (1876–1931), who had recently returned to Vienna after serving in the air force in the south Tyrol, took up the problem. Economo had been working under Wagner-Jauregg since 1906, but the outbreak of war interrupted his research.46 Economo soon identified excessive sleep as the common element to all these patients’s behaviour. They could drop off while sitting up, or even while standing. Some remained asleep for weeks at a time. This was not a case of coma, however, since the patients could be roused with little difficulty. Their thoughts did not seem particularly irrational or disordered to Economo, and, after the non-specific, ‘flu-like’ symptoms wore off, his patients were left with a variety of motor symptoms, including rigidity, motor paralysis, and ophthalmoplegia (involuntary movements of the eyes). But excessive sleep was the most pronounced symptom, and it was on this basis that Economo had christened the mysterious disease ‘encephalitis lethargica’ and wrote up seven cases for a Viennese medical journal.47 Sleeping sickness was not a new term. It was well known around the world to both medical practitioners and the general public as an epidemic disease that caused a morbid sleepiness among central Africans and that had claimed hundreds of thousands, if not millions, of victims,

194

The Sleep of Others

since the 1880s. Its medical name was trypanosomiasis, named after the organism, Trypanosoma gambiense, which had been identified in 1905.48 The disease had been well publicized in Europe, because it presented ongoing concerns for the ruling colonial powers in Africa, who feared that the disease would cause social unrest and, more important, devastation of their indigenous labour force. Environmental measures to eliminate the tsetse fly, which was the main disease vector, had been taken since around 1900. But a treatment was not forthcoming until the development of Bayer 205 in 1922, at which point the German Colonial Society made headlines by suggesting that the rest of the world could benefit from the new drug, provided that Germany could have some of its African colonies returned.49 As Maryinez Lyons has demonstrated in her engaging social history of the disease, trypanosomiasis was an important geopolitical, economic, epidemiological, and bacteriological concern. In neurological terms, however, the disease seems to have counted for little. On the one hand, there were not likely many neurologists in central Africa at the time interested in pursuing such a topic of investigation. It is also clear that the nature of public-health investigation and experimentation played out differently in Africa than it did in Europe or North America. Whereas neurological disease in the latter areas featured brains or minds whose agency had been compromised, in the African colonies, trypanosomiasis was depicted in terms of its adverse effects regarding the ongoing exploitation of an indigenous labour force. Experimental drug trials, for example, that would have been considered ethically suspect in Germany were given free rein in the colonies.50 This lack of concern for psychological questions made physical incapacity and death, rather than sleep, the intriguing feature of African sleeping sickness for Euro-American investigators. The arrival of encephalitis lethargica to Europe redirected the lay and medical concept of ‘sleeping sickness.’ Mysterious epidemics of sleeping sickness had been reported in Europe before: strange cases of Schlafkrankheit had struck Tübingen in 1712, and the curiously named nona that struck northern Italy in 1890 had been widely reported in the medical and popular press.51 During this latter outbreak, Ludwig Mauthner, a Viennese ophthalmologist, had noticed that opthalmoplegia seemed a constant symptom of the disease. He argued that the disease was almost certainly a polioencephalitis, similar to epidemic meningitis, and that it should also be considered a pressing public health concern.52 Economo agreed with Mauthner that nona and encephalitis lethargica were likely one and the same disease. He also presented his clinical observations in

Inhibition and Disease 195

conjunction with his neuropathological examinations and argued that the disease featured damage to a specific area in the mid-brain. But, while Mauthner’s arguments had little or no bearing on the fortunes of sleep physiology, Economo’s certainly did. If encephalitis lethargica (or nona) had not changed between 1890 and 1917, scientific conceptions of sleep definitely had. We have already seen, in chapter 2, how Mosso used the graphical method in his cursory discussion of sleep in terms of the physiology of sensation during the 1890s. His investigation, which relied on human subjects, was merely an extension of his study of another sensation, that of fatigue. By 1917, however, investigators had begun to impose a much more robust separation between sleep and questions about the nature of mind. Hypnotism had been scrapped as an experimental tool, and the widespread use of animal models had helped transform the problem of sleep into a question about brain function. Piéron had forged a chemical basis for fatigue theories of sleep. Claparède had propagated the claim that sleep was not the annihilation of consciousness but its active defence of the brain. Pavlov had defined sleep in terms of a cortical inhibition, and his American counterparts explicitly rejected any consideration of introspective questions about nature of the will. By the time Economo began to frame encephalitis lethargica in terms of sleep, then, sleep itself had changed. Periodicity had begun to replace fatigue as one of the phenomena of sleep that demanded explanation. The epistemological value of this new disease to sleep physiology was that it provided, in epidemic proportions, a human analogue to the animal model of sleep that had only recently been established. There were no ‘deficiencies’ here but rather a plethora of new ideas and experimental assemblages. Localizing Sleep Economo’s early studies of encephalitis lethargica were first and foremost works of clinical pathology. He was a skilled neuroanatomist and he used his talents to defend the status of encephalitis lethargica as an independent disease by distinguishing it from the encephalitis that was merely an occasional symptom of influenza, food poisoning, tuberculosis, and meningitis. He did not immediately position his work as a contribution to any particular theory of sleep. Rather, his first paper was filled with descriptions of symptoms, coupled with post-mortem anatomical observations. In particular, he observed that certain areas of the midbrain in his patients, including the grey matter of the third ventri-

196

The Sleep of Others

cle, the area of the ocular nuclei and the aqueduct of Sylvius, and the floor of the fourth ventricle, had been infiltrated by small, abnormal cells.53 It was only in his second article, published in 1917, that he began to take a position regarding the existence of a subcortical sleep centre.54 ‘The explanation of the symptoms of the strange desire to sleep,’ he admitted, ‘is difficult.’ It could be due to a specific toxin produced by the still unknown virus, or perhaps it was a result of increased brain pressure, as is seen in hydrocephalus, or the byproduct of an inflammatory reaction. But Economo argued that the peculiar way encephalitis patients slept ruled out such explanations. In fact, it was not so much that his patients’ sleep was peculiar as that it was almost normal. Economo’s patients could usually wake up and respond clearly when aroused – they just simply fell back asleep when given the least opportunity. The sleep due to a toxic infection, increased brain pressure, or inflammatory reaction, he argued, would not be so reversible. It would be more like a coma. For Economo, his pathological observations quickly carried over into a critique of theories of normal sleep – particularly those based on the circulation of some sort of ‘sleep toxin.’ In this shift from the pathological to the normal, he was following Mauthner, who had suggested that normal sleep was caused by a functional break between the cerebral cortex and the lower regions of the nervous system. Mauthner’s evidence was strictly analogical: the droopy eyelids and occasional paralysis of the eyes experienced by normal sleepers were chronic symptoms of nona, and ocular movements were then thought to be controlled by certain subcortical brain centres. The pathological eye movements and constant sleep must therefore be the result of damage to the same region of the brain. These speculations amount to little more than an aside in Economo’s second paper. It was not until nearly a decade later that he began to articulate publicly a formal theory of sleep based on his observations of encephalitis lethargica. The gap between 1917 and 1926 was an important one. In the interim, encephalitis had spread throughout Europe and North America. Accurate morbidity rates varied widely, but the Matheson Commission for the Study of Epidemic Encephalitis (MC) noted that, by 1928, 85,000 cases had been reported worldwide.55 Commission members argued that the disease was actually far more prevalent than such numbers indicated, since it was both novel and difficult to diagnose. Indeed, one of the primary goals of the MC, which had been forged out of the New York Academy of Medicine’s Commission on Public Health, was

Inhibition and Disease 197

to gather and communicate knowledge about the disease to further its diagnosis and assist in research. New York City neurologists, who were then struggling to refashion their field around a more rigorous scientific basis and a deeper social relevance, perceived a great opportunity in studying the disease thoroughly. Although the MC and its researchers did generally pursue the study of normal sleep, their work, along with encephalitis outbreaks in the United States, coincided with Pavlov’s EuroAmerican tour, in which he publicized his theory of sleep to any physiologist who would listen. Economo’s own reputation as a neuroanatomist had been consolidated by the 1925 publication of his major work on the cellular structure of the cortex, which was quickly translated into French and English. Economo was thus in a good position to speak on a timely topic. In a 1929 lecture delivered to the College of Physicians and Surgeons at Columbia University, he described his new theory of sleep, which had also been laid out in his book on the same topic.56 He began by noting that ‘the extinction of consciousness, this most striking symptom of sleep of man and of higher animals,’ had ‘appeared until recently as the essential characteristic of sleep and as the only one which demanded explanation.’57 Dominant theories had thus explained sleep as the result of a ‘lack of stimuli,’ the textbook evidence for which had been provided by Adolf Strümpell, a professor of psychiatry at Leipzig. In 1898 Strümpell reported his observations of a patient who suffered from an almost complete cutaneous and sensory anaesthesia: only his left ear and right eye remained responsive. As soon as Strümpell closed the one good eye and plugged the ear with cotton, his patient immediately fell asleep, thus demonstrating his theory. Economo, however, insisted that Strümpell’s patient was merely acting on suggestion. Besides, he argued, ‘lack of stimuli’ theories of sleep were unsatisfactory in other ways: they could not explain why sensation was undisturbed in pathological sleep; they described sleep as a change of consciousness, which could not explain the sleep of certain plants; they could not account for the alternating sleep and wakefulness of animals without a cerebrum or in anencephalic babies; nor could they explain the qualitative changes that went on in sleep, such as the changes in sugar and calcium content of the blood and the narrowing of the pupils. Most important, however, these theories could not account for the periodicity of sleep. Sleep’s regular, diurnal rhythm was left unexplained, as were the many variations in organic functions (respiration, heartbeat, bodily secretions) that followed this same pattern. It was

198

The Sleep of Others

hardly coincidental that Economo emphasized that it was precisely this rhythm that was often inverted or even completely destroyed in encephalitis lethargica, regardless of whether or not the dominant symptoms were pathological sopor or chronic insomnia. The ‘toxic fatigue’ theories of Piéron, Economo concluded, were no better in explaining sleep’s reversibility, since they relied completely upon the initiation of sleep as a kind of fatigue, albeit one limited by the functioning of the nervous system. Economo was considerably more enthusiastic, however, about the concept of hormonal regulation, which had been proposed by the Italian neuropathologist Giovanni Mingazzini, around 1920. Economo approved of Mingazzini’s idea that one type of hormone dominated during sleep and another dominated during wakefulness, because it redefined sleep as an active state of the body, rather than as a period in which activity was eliminated.58 But despite such exceptions, the idea of a sleep centre deep in the brain went generally unrecognized by neurologists and psychiatrists alike. When they described sleep at all, it was typically in terms of cognitive function; they therefore looked to a modification in the activity of the cerebral cortex as its origin. The advent of encephalitis lethargica, argued Economo, had changed this perspective, because its pathognomonic symptom – the destruction of a normal sleep/wake rhythm – was inevitably accompanied by brain damage. There was little consensus on the cause of encephalitis lethargica, however, and its symptomatic classification fared little better: the Matheson Commission, for example, adopted a conservative approach, and collected a wide spectrum of symptoms reported from encephalitis cases while refusing to identify any one of them as absolutely characteristic of the disease. Indeed, some critics of Economo even collapsed trypanosomiasis and encephalitis lethargica into a single category, arguing that there were many instances of both that featured no subcortical damage whatsoever.59 In any event, Economo proposed that there was a regulatory centre for sleep deep in the brain, lying close to other centres that controlled vegetative functions such as respiration (see fig. 15). Economo dubbed this centre, which coordinated the various physiological and psychical changes brought on in sleep, the Schlafsteuerungszentrum, or ‘sleep steering centre.’ Given Economo’s passion for flying and driving motor cars, this was perhaps not such an odd choice of terms. To explain how the centre functioned, Economo returned to the concept of inhibition, which he borrowed directly from Pavlov, completely ignoring the fact that Pavlov had focused exclusively on the activity of the cerebral cortex in his work on conditioned reflexes.60 Instead of Pavlov’s

Inhibition and Disease 199

15 Based on his experience dissecting the brains of encephalitis lethargica patients, Economo identified a region in the midbrain (surrounded here by a dotted line) responsible for regulating sleep. The patient’s eye paralysis corresponded to the posterior part of the area (the right of the region, just touching the oculomotor nuclei), while the anterior limit was found at the grey walls of the third ventricle (V3). (Constantin von Economo, ‘Sleep as a Problem of Localization,’ Journal of Nervous and Mental Disease 71 (1930): 249–59, 257)

wave of inhibition spreading across the surface of the cortex, Economo argued that inhibition actually originated deep within the brain, in the sleep centre, which then spread outwards to the thalamus and the cerebrum. There was also, he argued, a corresponding ‘waking centre,’ the task of which was to maintain the cortex in a tonic state. Admitting that his theory was ‘somewhat similar to the hormonal explanation,’ Economo also proceeded to incorporate elements of the fatigue theory into his argument, suggesting, like Claparède and Piéron, that small quantities of fatigue toxins triggered this inhibitory centre, thus preventing the brain from becoming more intoxicated. Despite the seemingly ad hoc nature of aspects of Economo’s theory, he had successfully shifted the focus of the problem of sleep. ‘Inhibition’ was no less vague a term than ‘fatigue,’ but the former seemed to explain the rhythms of sleep in a way that had been almost completely

200

The Sleep of Others

ignored in the older fatigue theories. The brain was not simply a transfer point, where one form of energy could be converted into another. It was a self-regulating, automated organ that relied upon fatigue toxins to govern its overall performance of inhibition and excitation. Sleep, Economo implied, demonstrated how the brain’s work was administrative and organizational, rather than translational. His centres for sleep and wakefulness governed the states of consciousness. Their significance for determining consciousness’ contents subsequently receded into the background. Sleep Comes to America Economo concluded his 1929 lecture by pointing towards the potential therapeutic value of locating a sleep centre in the brain. Once such a centre had been clearly identified, he argued, therapeutic efforts could be directed towards this area, making it possible ‘to treat insomnia and other sleep disturbances in a better and more active way than by drugs or by the roundabout way of hydrotherapy and psychotherapy.’61 This notion surely resonated in New York City, the site of Economo’s lecture, since psychotherapy there had, in some corners, begun to merge with neurology in the analysis and even treatment of encephalitis symptoms. Indeed, Economo’s lecture was published in Smith Ely Jelliffe’s Journal of Nervous and Mental Disease, then the most important forum for what was becoming known as ‘psychosomatic medicine.’62 Psychosomatic medicine was a curious mix of organic and psychoanalytic theories of disease, and Jelliffe (1866–1945) was one of its main advocates. Jelliffe had studied under Emil Kraepelin in Munich and Dejerine in Paris, and he was, along with A.A. Brill and J.J. Putnam, an early American champion of psychoanalysis. He also published several articles and books on encephalitis lethargica.63 Proponents of psychosomatic medicine did not, it should be noted, deny the physiological origins of mental illness. Indeed, their approach was strongly holistic, in the sense that they argued that every disease had a psychological component and that, by using psychotherapeutic techniques to analyse and treat symptoms, health could be restored. Their search for causation, as one author has put it, was gradually replaced by a quest for correlation – a transition that was well in keeping with the rise of the ‘risk factors’ approach to chronic disease discussed in chapters 7 and 9.64 Medicine and psychology readily mingled during the 1920s and 1930s, particularly when it came to the problem of encephalitis lethargica, which

Inhibition and Disease 201

featured movement disturbances and tics that had previously been associated with catatonic schizophrenia. Following the lead of Adolf Meyer, one of the most influential figures in American neuropsychiatry in the early twentieth century, many psychiatrists and neurologists were beginning to depict major mental illnesses in psychological terms. The epidemic nature of encephalitis lethargica made it impossible to deny its biological etiology, but this in no way prevented many neurologists and psychiatrists from treating the disease as a model for studying the formation and possible resolution of psychological symptoms as well.65 In his classic 1945 textbook, for example, the psychoanalyst Otto Fenichel encouraged his readers to understand ‘Jelliffe’s attempts to “psychoanalyze” encephalitic symptoms ... as a study of the ways in which the personality reacts to or makes use of the symptoms rather than as a belief in the “psychogenesis” of encephalitis.’66 One might expect that the repeated outbreaks of an epidemic disease sharing a number of symptoms with psychotic illness would have encouraged neuropsychiatrists to reject psychodynamic theories of mental illness as irrelevant. But this was not the case. Instead, the disease provided an opportunity for psychoanalysts and sleep theorists alike to advance their theories. More important, encephalitis lethargica drew attention to the question of sleep. When Economo delivered his lecture in 1929, his American audience was probably better prepared to think about encephalitis lethargica than their European counterparts. The disease had arrived on American shores in early 1918, shortly after it had first appeared in Europe. Over the following decade, it struck thousands of Americans, and in global terms, its focus by the early 1920s was New York City. The mode of transmission remained completely unknown, and it was frequently remarked that the disease affected the wealthy and the able as well as the poor and the infirm. Even after the epidemics subsided a few years later, the philanthropic support and scientific interest that had begun early in the decade continued, unabated. Simon Flexner, the first director of the Rockefeller Institute for Medical Research, took up research on the disease around 1921, shortly after the New York Academy of Medicine’s Commission on Public Health began pursuing the same problem in earnest.67 In May of that same year, the New York Times, in reporting on the completion of the Mount Sinai Hospital (at a cost of $4 million), noted that one of the hospital’s main features was the consolidation of the neurological and medical departments, making it possible for the same physician both to diagnosis and to treat. The only disease deemed worthy of mention in the article was epidemic encephalitis.68

202

The Sleep of Others

The long duration of the disease sometimes kept its more famous victims in the news for months. In June 1925 Mrs J.P. Morgan, wife of John Pierpont Morgan, Jr, the American banker who had helped to finance nearly half-a-billion dollars’ worth of loans to France and Britain during the war, fell ill with epidemic encephalitis.69 She died the following August. In 1927 Morgan donated $200,000 to equip a floor of the Neurological Institute of New York, then associated with Columbia University, for the study of the disease. That same year, William J. Matheson (1856–1930), a successful industrial chemist who was eventually diagnosed with encephalitis lethargica, decided to turn his own illness into a major biomedical project by creating the Matheson Commission for the Study of Epidemic Encephalitis. Matheson had his own rather unorthodox ideas about the disease, but he nonetheless allowed a panel of neurologists, bacteriologists, and public-health officials to draw upon his wealth to finance a project run jointly out of the Neurological Institute of New York and the New York Academy of Medicine.70 Its first real fruit was an 850-page report reviewing all the work that had been done on the disease to date. When Matheson died a year after the report’s publication, his $400,000 endowment for encephalitis research was administered by Columbia University. Such philanthropy both drew upon and fuelled civic interest in the disease. Like Economo, writers for the New York Times were fascinated by the transformation of the sleep regime that befell the victims of epidemic encephalitis. A favourite editorial focus was the enormous length of time that encephalitis victims sometimes spent incapacitated, a phenomenon that was persistently described in terms of ‘sleep,’ even as its very description emphasized how very much unlike sleep it actually was. One child in New York slept for a month.71 A man in Arkansas slept for three years, awoke for a moment at 8:30 in the morning, yawned, and fell back asleep.72 Another newsworthy feature was how the patients emerged from sleep, if they were indeed lucky enough to do so at all. One woman, a Mrs Fred Tracy of Oxford, Chenango County, New York, achieved ‘the record sleep of the year.’ She was eventually awakened by a ‘talking machine,’ brought in by her neighbour as a last resort. But she fell back asleep almost immediately.73 Another woman in Wisconsin had been asleep for nearly two years before the ‘spell’ was broken by the appearance of her six-year-old son at her bedside, which had by then been moved to the county asylum. Upon awakening, she recalled being perfectly aware of all that was going on around her the entire time she was asleep, but she was unable to open her eyes, move, or even speak.

Inhibition and Disease 203

Her memory, however, appeared to be perfectly intact. She was able to recall events of the Great War, as well as several family members who had fallen on the field of battle.74 Just as epidemic encephalitis seemed to attack both rich and poor, it also appeared equally incapable of discriminating between young and old. Children were frequent victims of the disease, but, whereas persistent somnolence was the most striking symptom in adults, children seemed to suffer developmental problems following a ‘bout’ of encephalitis. Even if it had not been diagnosed as such, epidemic encephalitis was frequently blamed, retrospectively, for behavioural problems in children that ranged from inattention in the classroom to outright delinquency. As early as 1921, reports began to emerge about restless and hyperactive children who had suffered from encephalitis lethargica and now were unable to behave properly at school or at home.75 If a child suddenly began to display abnormal behaviour, brain damage due to a ‘mild encephalitis’ was often cited as the cause, even if the initial ‘flulike’ symptoms had been missed. In 1924 Earl D. Bond, a neuropsychiatrist at the Pennsylvania Hospital at Philadelphia, helped to set up a small boarding school for the ‘re-education’ of delinquent children. He emphasized the use of intensive, individualized attention and positive reinforcement, which he advocated for ‘normal’ children as well. Bond published a book on the subject in 1931, and, in a Times article, he argued that epidemic encephalitis was responsible for ‘the flood of juvenile delinquency’ that had been described in one of the many reports by the Wickersham Commission on Law Observance and Enforcement that were published that same year.76 The link between encephalitis lethargica and delinquency played upon the general perception that encephalitis was highly underreported, despite the fact that it had been made a reportable disease in New York City since the beginning of 1921. It was not only pediatricians and educational psychologists who adopted such a diagnosis. Some children also appeared to have a keen awareness of the cultural significance of this new disease. In the summer of 1925, for example, a ten-year-old boy had been caught trying to steal a woman’s purse in Luna Park, on Coney Island. The boy was too slow, and, upon feeling a tug at her handbag, the woman whirled around and captured her tiny assailant.77 The boy later confessed to police that he was a chronic purse snatcher – he had stolen twenty-six of them to that point. Each time, his frustrated mother had returned the purse to its rightful owner. Naturally, the police inquired as to why the boy would behave this way. Sleeping sick-

204

The Sleep of Others

ness, the boy replied. It had apparently left him with a mania for taking purses. The physiology of sleep was obviously not the central concern for the neurologists, psychiatrists, child psychologists, medical practitioners and public-health officials who took an active interest in encephalitis lethargica. They were much more interested in the nature of the disease itself – its viral origins, its means of transmission, and possible therapeutic interventions. The disease, however, disappeared before any of these questions could be answered definitively. Mosquitoes were identified as a disease vector in several encephalitis outbreaks that took place during the 1930s, but the symptoms of encephalitis lethargica were so protean that it was difficult to assess whether or not investigators had ever been dealing with a unified disease entity in the first place – or such was the conclusion of some prominent neurologists (such as Tracy J. Putnam), as well as of the final Matheson Commission report, published in 1939. Yet such opinions counted for rather little by this time, since all agreed that, whatever it was, encephalitis lethargica had almost completely disappeared. Still, the emergence of epidemic encephalitis marked an important episode in the development of sleep physiology. Economo’s study of the disease had provided anatomical evidence of what Claparède and Piéron had tentatively suggested a decade earlier: that sleep was the product of a regulatory centre located deep within the brain. Economo’s theory of sleep spread across the Western world, right alongside the disease that had spawned it. Encephalitis lethargica also helped to bring sleep to the forefront of neurological research, and move it out of the field of fatigue physiology, which had, by the late 1920s, receded to the backwaters of medical science. It helped to sever sleep from the physiology of sensation and unite it with the study of neurophysiological organization. The circulatory images of the fatigue theories and the reflexive images of stimuli theories were giving way to a picture of sleep as a self-regulatory phenomenon, with its periodic ebb and flow as one of its most striking characteristics.

5 Performing Sleep

The encephalitis lethargica epidemics of the early 1920s brought the concept of a ‘sleep centre’ to the forefront of neurological research, emphasizing that sleep had both a structure and a function. Such a perspective took hold among a small group of investigators at the University of Chicago during the interwar period. While American behavioural psychologists tended to ignore sleep, Nathaniel Kleitman, a physiologist who had studied under Piéron, successfully depicted sleep as a state characterized by a set of organic activities whose dynamics could be captured in the laboratory. Although Kleitman’s early attempts to link his work to neuropsychiatric problems failed, he did successfully correlate these dynamics to behaviours and to conscious states, ultimately producing an evolutionary theory of sleep. His colleague, Edmund Jacobson, proved more successful in translating the practices of sleep physiology to clinical practice, and his understanding of the nature of sleep was transformed in the process.

Picturing Sleep in a Research Field Nathaniel Kleitman (1895–1999) is a pivotal figure in the history of twentieth-century sleep research in North America. Before Kleitman, sleep was an interesting but incidental topic of research for physiologists and psychologists alike. Claparède, who first proposed a biological theory of sleep, actually conducted little research in this field. Likewise, Piéron more or less abandoned the problem of sleep for the physiology of sensation shortly after publishing his dissertation in 1913. Pavlov came to the study of sleep only late in his career, and Vaschide’s premature death cut short his well-developed interest in the subject. Economo also died at a relatively young age, just as he began seriously pursuing

206

The Sleep of Others

the question of a sleep centre. Kleitman, on the other hand, clung to the problem of sleep with an unparalleled tenacity. He had an extraordinarily long career, and the problem of sleep was always at its centre. His first publication on the subject appeared in 1923; his last, a summary of his theory of a ‘Basic Rest-Activity Cycle,’ came out almost sixty years later, in new journal entitled, appropriately enough, Sleep.1 The very existence of such a specialized journal owed a great deal to Kleitman’s efforts. His laboratory at the University of Chicago – the first to be organized around the study of sleep – was at the heart of American sleep research for decades. Many of the most prominent sleep researchers in the United States, including William Dement, David Foulkes, Allan Rechtshaffen, and Eugene Aserinsky, studied there and went on to establish sleep laboratories in other universities and medical centres across the country during the 1960s and 1970s. Yet, although the story of sleep research is in many ways caught up in Nathaniel Kleitman’s own story, it would be wrong to think that a study of his biography could somehow provide an authoritative route to the history of sleep research in the twentieth century. Sleep did not hold a great degree of internal coherence when Kleitman began to study the topic. It was little more than a baroque collection of facts which he organized and pursued in a way that very much depended upon the context in which he laboured. Until a research community consolidated around the phenomenon of rapid eye movement, sleep remained fractured as an investigative object. For their part, sociologists have already examined the growth of sleep research after the Second World War as a case study in how the choices available to researchers and their determination of risk and rewards influence the rise of scientific subdisciplines.2 This approach offers considerable advantages, since it relies on a number of different perspectives gathered by interviews with several researchers, rather than depending on a singular biographical perspective. The result is more than the sum of its parts, for the authors ultimately provide an analysis of how scientists embarking on research in a new field make their decisions. But this emphasis on ‘choice’ also has its limitations. It assumes not only that scientists’ behaviour is inevitably rational (an assumption that might not be so easily made for artists or politicians) but that the retrospective accounts that scientists offer are the most reliable sources of information. This assumption may be helpful, if the goal is to determine the nature of scientific decision making.3 Yet it tends to relegate the history of the object of scientific research to near insignificance, since the scientists’ testimonials inevitably take their

Performing Sleep 207

objects as already existing, out there in nature, waiting to be discovered. The role that institutions, instruments, physical sites, and cognitive approaches played in creating those objects remains unanalysed. This is a particularly inappropriate means of approaching the history of sleep research, which, as we shall see in the following chapters, relied upon the instrumental visualization of phenomena and its functional explanation in order for the field even to appear as a option for investigators to choose. Scientific objects and experiments have lives of their own that call for historical analysis.4 The graphical method had an enormous impact on how sleep was visualized, both in the laboratory and in the clinic. Without an analysis of how such technologies combined with methods and theories to create a stable investigative platform for sleep researchers, it is impossible to understand, for example, how a crucial phenomenon such as the burst of REM occurring during discrete periods of sleep remained largely outside the realm of scientific observation until the early 1950s. It was not the case that opportunities to observe were lacking. As discussed in chapter 2, Nicholas Vaschide watched the faces of his sleeping subjects throughout the night. Yet he offered no observation of eye movements that can be easily correlated to REM. Nor was it the case that eye movements in sleep went unnoticed. In the present chapter, I will discuss the example of Edmund Jacobson, a clinician who suggested that eye movements and dreams were related but who ultimately concluded that the relationship was unimportant. So what sort of cognitive, instrumental, and institutional elements needed to be in place before this particular observation could mean anything at all? How did REM become a scientific fact of such significance that entire research program could be organized around it? Both the initial analysis of eye movements observed in sleep-laboratory subjects and the discovery that these movements could reliably be related to dreaming emerged out of the University of Chicago between around 1920 and 1950. Although there were certainly other places where such questions could have been, and probably were, asked, a study of the research school located there will move us towards an answer.5 Nathaniel Kleitman: The First American Sleep Researcher Four weeks before Ivan Pavlov delivered his lecture on sleep at the University of Chicago in July 1923, a young physiologist named Nathaniel Kleitman submitted his first publication to the American Journal of Physi-

208

The Sleep of Others

ology. The paper was the first of a long series of articles that Kleitman was to publish in this journal over the next several years and that would ultimately form the core of the book that came to define him as the world’s leading sleep researcher: Sleep and Wakefulness as Alternating Phases in the Cycle of Existence. Kleitman had long been aware of Pavlov’s research in sleep, and he indicated in his 1923 paper that he had in fact been communicating with Pavlov on the topic. It is not clear whether Kleitman actually attended Pavlov’s July lecture at the University of Chicago. But this hardly mattered, since Kleitman, who was fluent in both Russian and German, would probably have got as much information out of a letter from Pavlov as anyone who listened to Pavlov’s son deliver the brief translated version of his father’s lecture. The high visibility of Pavlov’s work in Kleitman’s 1939 textbook is further testimony that Pavlovian ideas played an important role in Kleitman’s thinking about sleep. The only individual cited more frequently in the first edition of Sleep and Wakefulness was Henri Piéron. (Curiously, it is Piéron’s intellectual lineage that has endured in historical accounts of sleep research, while Pavlov’s name has been virtually erased from the annals of the field.6) Kleitman was not only aware of Pavlov’s research, but he approved of it at the outset of his studies and pursued a line of inquiry that was, in effect, a hybrid of the experimental styles of Pavlov and his other mentor, Piéron. Kleitman’s ready embrace of such a marginal topic of physiological interest as sleep was perhaps in part a function of his tumultuous personal experience before arriving in the United States. He was born in 1895 in Kishinev, the capital of Bessarabia (now Moldavia), and had been in the United States a mere eight years before Pavlov arrived. One commentator, who interviewed Kleitman when he was ninety-five, noted that ‘the story of his life exemplifies every word – written or related – about the fate of the Wandering Jew.’7 The Jews of Bessarabia suffered greatly during the last years of the nineteenth century and the early years of the twentieth. The reign of Tsar Alexander III (1881–94) saw the extension of ‘Russification’ throughout the Russian empire, and this program had devastating effects among the non-Russians living in the peripheral states of southeastern Europe. The extreme social agitation that marked Russian history during the three decades that culminated in the Revolution of October 1917 included an exceptionally high degree of anti-Jewish sentiment among the peasantry. This was keenly felt in Bessarabia, where Jews quickly became a target of officially sanc-

Performing Sleep 209

tioned prejudice.8 Kleitman surely remembered – if he did not experience its effects himself first-hand – the pogrom that resulted in the death of forty-five Jews and the beating of hundreds in his hometown during Easter of 1903, and again in October 1905. By 1912, labour unrest and strikes were sweeping across the industrialized cities of Russia, and Balkan nationalism was continuing to have negative repercussions among the Jewish communities in the area. Like many others, Kleitman decided to leave. At seventeen years of age, he headed for Palestine to study medicine at the American College in Beirut.9 But, with the outbreak of war, he became increasingly concerned that, as a Russian subject, he would fall under the suspicion of the Turkish authorities in Palestine. So he fled to Rhodes, where he boarded the first ship that would take him out of Europe. The freighter happened to be American. Thus, with no express intention of going to America, Kleitman soon became one of the almost eighteen million immigrants who arrived in the United States during the period 1890–1917. When he landed in New York City in 1915, he spoke little English and had practically no money, but he nonetheless was able to earn an undergraduate science degree from the City College of New York in 1919, and taught chemistry there while completing his MA at Columbia University. Rootless but talented, and with his medical career foiled, he was now ready to embark on a career in physiological research. Anton J. Carlson and Physiology at Chicago In 1922 Kleitman began working on his PhD under A.J. Carlson at the University of Chicago. Anton Julius Carlson (1875–1956) had headed the physiology department there since 1904, and, along with William H. Howell at Johns Hopkins, Walter B. Cannon at Harvard, and Joseph Erlanger at Washington University in St Louis, he was a major force then directing the course of physiological research in the United States. It was during this period that physiology began to develop as a discipline with greater independence from the medical practice that had underwritten its growth during the nineteenth century. Carlson, who had himself immigrated from Sweden in 1891, surely sympathized with Kleitman’s situation. When Carlson first arrived at Chicago in 1904, a ‘research imperative’ had become entrenched in the American university system, and the small number of full-time faculty in physiology departments no longer divided their time between clinical practice and teaching medical students.10 They were now expected to devote all their

210

The Sleep of Others

time to research and teaching. Carlson’s research school at Chicago was thus a first-generation affair, and he might well have wanted to extend the opportunities he had had to a bright young immigrant, much as he had been, three decades earlier.11 Experimental practice in physiology was also becoming more dependent upon technological innovation, a tendency that was particularly pronounced in the United States.12 American science was transformed during the first decades of the twentieth century, as industry, government, and private philanthropists invested in research as they had never done before. The models for return on such large-scale investment were frequently taken from industry, which had itself been recently transformed by innovations in managerial and engineering techniques, such as those pioneered by Fredrick W. Taylor.13 It was thus during Carlson’s tenure at Chicago that American physiology moved from a marginal position to one of world leadership. Within five years of Carlson becoming a full professor and head of the Department of Physiology at Chicago in 1914, a new era for American physiological research began, as the Guggenheim and Rockefeller foundations, along with the National Research Council (NRC), began to support post-doctoral fellowships in the field. Producing physiologists was itself becoming an industry of sorts, and Carlson oversaw the completion of the enormous six-storey Hull Biological Laboratories, designed to rival the equally massive labs at Michigan, Harvard, and St Louis. There, he trained most of his more than one hundred and seventy students, and he ultimately produced as many post-doctoral fellows at Chicago (thirty-one) as did Cannon at Harvard (thirty-two).14 This situation could hardly have been more different from the one that Carlson encountered when he first arrived in the United States For Carlson, at least, physiology was more a calling than a career. The profession, such as it was, featured few links to industry and continued to be dominated by the need to teach medical students how to think and act scientifically by training them in the laboratory. Research in the field was conceived, executed, and even funded by the individual. Carlson himself came to physiology by default, not design. By working as a carpenter’s assistant for several years in Chicago after first emigrating from Sweden, he saved up enough money to go to the Augustinian Academy and College at Rock Island, Illinois, with the intention of becoming a pastor in a Swedish Lutheran church. He earned an MA in philosophy, but his time behind the pulpit was brief, since he soon lost his faith. His thinking had taken a decidedly materialist turn (which expressed itself in his politics as

Performing Sleep 211

well as in his science), and he decided to begin the study of the physiology of the nervous system. Carlson borrowed enough money to attend Stanford and worked there under O.P. Jenkins, studying the rate of nervous conduction in slugs. From there, he went to the Marine Biological Station at Woods Hole, Massachusetts, and examined the role of the cardiac nerves in the coordination of the heartbeat in the horseshoe crab. His work, which was published in the American Journal of Physiology in 1904, helped demonstrate that the rhythmic behaviour of the heart was due to the action of the autonomic nervous system, not the muscle. The publication of this work brought a job offer from Chicago that same year, and Carlson remained there for the rest of his career. His time was divided between teaching medical students physiology and overseeing experimental research. Carlson was directly responsible for the new national and international prominence of physiological research coming out of Chicago. One estimate suggests that he taught at least 5,000 medical students, participated in 151 MSc degrees in physiology, and personally supervised 112 PhDs.15 The number of papers that had his name attached to them was greater than any of his American competitors, and he generated on his own more publications between 1898 and 1918 than did most physiological institutions.16 His productivity was reflected by his stature within the profession. He served as the president of the American Physiological Society from 1923 to 1925 and as the chairman of the board of editors of Physiological Reviews from 1932 to 1950. Carlson lived and died by his experimentalism, and, while this approach obviously served him well when it came to increasing productivity within his department, it ultimately proved to have a dampening effect on his legacy. He derided the notion that biology could, in any way, be a mathematical or theoretical pursuit.17 And his hard-nosed positivism prevented, rather than promoted, a major experimental discovery by one of his students. He either rewrote, or demanded the revision of, a paper by a student, Ernest L. Scott, for publication in the American Journal of Physiology. The paper had been taken from Scott’s thesis on the effects of injecting a pancreatic extract in diabetic dogs, and Carlson had it rewritten to underline the speculative and tentative nature of the conclusion that the injected substance actually lowered sugar excretion and the dextrose-to-nitrogen ratio observed in the animal’s urine. This effectively eliminated any priority claim Scott might have had against the physiologists at the University of Toronto who announced their discovery of insulin almost ten years later.18

212

The Sleep of Others

When Kleitman came to work with him, however, Carlson was well known for his work on the physiology of digestion, which was itself an extension of his earlier research on the role of the nervous system in generating and sustaining the body’s rhythmic behaviours. In 1916 he published The Control of Hunger in Health and Disease, a book that focused on the sensation of hunger as a function of the nervous impulses that arose from the gut. His experimental studies included both animal and human subjects, despite Carlson’s complete lack of training in clinical medicine. His star subject was a young man, Fred Vleck, who had drunk a solution of caustic soda at the age of six, which eventually led to the closure of his esophagus. He underwent surgery six years later that enabled him to feed himself through a fistula in his stomach. When more surgery was planned, the boy took flight, but Carlson nonetheless was able to take advantage of this ready-made bit of experimental apparatus, the human equivalent to the ‘dog technologies’ that Pavlov had taken years to perfect (fig. 16). His method consisted primarily of taking recordings of Vleck’s stomach contractions by inserting a balloon, attached to a flexible tube, through his subject’s fistula and into the stomach. The balloon was then inflated, and the other end of the tube attached to a manometer and kymograph. The result was a unique experimental assemblage that allowed Carlson to merge physiological and psychological styles of investigation. ‘Our gastric fistula man, Mr. V,’ observed Carlson, ‘offers an exceptional opportunity for studying the relations of certain conscious states, particularly those associated with foods and with eating, on the activities of the empty stomach.’19 Like Pavlov’s dogs, Vleck was incapable of digesting the food that he was nonetheless able to chew. Carlson could thus use his subject to study the ‘psychic secretion’ of digestive fluids in the stomach without contamination from the masticated food or from saliva secreted in the mouth. But the great additional benefit to Carlson’s human subject, of course, was his ability to introspect. Thus Carlson’s experimental assemblage enabled him to synchronize body and mind in a way that Pavlov could not. Against both Pavlov and Walter Cannon, Carlson argued that both the tonus of the stomach and gastric secretion were not triggered by the appearance of food, or of any stimulus that had been conditioned to be associated with food. Instead, the activity of the stomach, like that of other organs, followed a rhythmical pattern yet unexamined: ‘The extrinsic nerves to the stomach,’ Carlson insisted, ‘play a rôle similar to that of the nerves to the heart in the regulation of the heart rhythm.’20 The rhetorical force of Pavlov’s rejection of mental states as investigative

Performing Sleep 213

16 Fred Vleck, A.J. Carlson’s experimental subject. Carlson recorded Vleck’s gastric contractions (with a vasomotor) and cardiac changes (with the plethsmograph on Vleck’s arm) to correlate the sensation of hunger with physiological phenomena. (A.J. Carlson, The Control of Hunger in Health and Disease [Chicago: University of Chicago Press, 1916], 34)

214

The Sleep of Others

objects depended upon the notion that salivation was an unambiguous sign for what was traditionally regarded as a conscious state – namely, the feeling of hunger. The experiments conducted by the Pavlovians seemed to demonstrate that this sign could be easily manipulated, thus emphasizing the pliable nature of consciousness and its more appropriate description as a bundle of conditional reflexes. But Pavlovian investigations typically depended upon invoked responses; indeed, it was by establishing strict routines by which those responses could be invoked that his army of students was able to get its work done in his laboratory. Pavlovians perceived themselves as being in the business of building up conditional reflexes from the bottom up. There was little or no room here for finding out about the natural schedules and periodicities of organic behaviour. Carlson, in contrast, emphasized just this approach which had been celebrated by Marey and other exponents of the graphical method described in chapter 2. Against Pavlov’s claim that psychic secretions instigated the processes of digestion and were thus governed by the activities of the cerebral cortex, Carlson insisted that the stomach went about its business of contracting and secreting quite apart from any sensory stimulus or cortical activity. The responses Pavlov had isolated were merely the phenomena of ‘central reinforcement.’ In other words, they were artifice. The precious drops of saliva that Pavlov had so carefully collected and measured were merely the product of attention and could not be reliably linked to the normal operations of digestion. They had nothing to do with the regular and automatic behaviour of the digestive apparatus, the rhythmic contours of which Carlson had captured on his kymographic drums: ‘In the normal individual the empty stomach exhibits periodic hunger activity, and there is no evidence to show that this primary automatism of the empty stomach is in the least influenced by eating one or by eating five meals a day ... the milder hunger contractions do not enter consciousness as pangs of hunger if the individual’s attention is directed into other channels. They are felt as hunger pangs if the individual’s attention is directed toward food and eating. The attention is thus directed, consciously or subconsciously, about the time the individual is accustomed to eat. The periodicity of this subjective attention to the milder hunger cravings can probably be altered by training.’21 Carlson’s work on digestion emphasized the interaction between consciousness and the rhythmic autonomy of the body. He considered mind and body to be two separate entities. Pavlov’s work, on the other hand, always invoked the immense flexibility of the cerebral cortex, which

Performing Sleep 215

seemed capable of associating practically any stimulus to an unconditioned one through training. One of the more infamous experiments to come out of Pavlov’s laboratory, for example, involved conditioning a dog to salivate while an experimenter burnt its skin and flesh with an electrode. In this depiction, mind was nothing more than trained reflex. Carlson did not share Pavlov’s rejection of introspective evidence, and his own use of human subjects (including himself) underwrote his attempts to reveal the limits of conscious awareness in physiological function. His work was the result of a search to identify the parallels between mind and body by articulating the mutual influence between the mind’s perceptions and the body’s rhythms. Experimenting with Sleep at Chicago Kleitman’s study of sleep followed this same path, as he attempted to sort out the psychological and the physiological aspects of sleep through the common currency of rhythm. Like Carlson, Kleitman lacked a background in clinical medicine, and the practical trajectory of his research was not aimed at generating a theory of disease. Surrounded by functionalist psychologists who drew their scientific legitimacy from physiology, Kleitman approached sleep as a question of biological rhythms, perception, and social influences conjoining to create the phenomena of sleep and wakefulness. Kleitman’s first paper on sleep encapsulated the way in which he felt his topic might capture the psychological, physiological, and social aspects of the phenomenon under experimental cover. When it appeared in the American Journal of Physiology in 1923, Kleitman was on the verge of completing his PhD under Carlson and was getting ready to study in Europe by virtue of an NRC fellowship he had recently won. His paper began with a self-conscious, almost ironic tone, as he invoked the standard refrain of those who studied sleep – namely, that it was perennially ignored by physiologists. But Kleitman was an adept reader. Even in making the claim, he recognized its distinctive rhetorical form: ‘Most investigators of the physiology of sleep, in reporting their findings, remind their readers, by way of apology, of the tremendous importance of the subject for the advancement of our knowledge of physiology as a science, as well as for the rational treatment of insomnia.’22 He went on to suggest that a study of the topic, which held such innate interest for him, also had practical consequences, in the form of improving time management and industrial efficiency, that might, in some sense, be

216

The Sleep of Others

unique to the twentieth century. Thus, the explicit goal was to determine if a reduction of sleep was possible or advisable. His Taylorist ideal of improving productivity contrasted sharply with Mosso’s vision of ameliorating the burdens of fatigue for the common labourer: ‘They [sleep investigators] also like to record the fact that the average individual spends more than a third of his life in sleep, has been doing this from time immemorial, and raise the question whether eight hours or more of sleep a day really constitutes the minimum penalty for keeping awake the rest of the time. It seems reasonable to suspect that as in the case of protein consumption, there is a large “factor of safety” in the amount of sleep we are getting, and that it could be considerably reduced without impairment of health or loss of efficiency. This and other questions related to industrial physiology, especially physiology of fatigue, can be answered only by a thorough systematic study of the subject.’23 This was not to imply that sleep was a purely physiological problem, but only to emphasize that physiologists, with their emphasis on improving human performance, might have a unique perspective on sleep that had thus far been ignored by those psychologists who had dominated the field to that point. The reign of the psychologist over sleep, argued Kleitman, came by virtue of their ability to create animal models for the study of human mental operations.24 Physiologists’ disciplinary dedication, on the other hand, was to animal models as representative of somatic illness and performance only. And, since human vivisection was disallowed, the study of objects that were simultaneously physiological and psychological was simply impossible. This was, of course, mere preamble to Kleitman’s proposed solution, of which Pavlov’s method of conditioned reflexes appeared as particularly noteworthy: ‘Pavlov and his co-workers, in their study of the conditioned reflexes, found that their animals frequently fell asleep during the experiment. They found that as a result of prolonged action of a uniform excitant their animals invariably fell asleep. The complete data of their experiments have not been published as yet, but in a personal communication to the writer Pavlov states that their results indicate that sleep and the so-called internal inhibition of a conditioned reflex are identical phenomena, the former being diffuse and the latter localized.’25 But, although Kleitman was prevented from expressing any discontent with Pavlov’s unpublished research, it was clear that he was much more interested in adapting Piéron’s method of experimental insomnia to human subjects. Piéron’s approach would allow Kleitman

Performing Sleep 217

to frame sleep in terms of states of consciousness, just as Carlson had done with digestion, and it would provide him with a means to wrest sleep from the psychologists, many of whom had adapted Pavlovian methods themselves. In fact, Kleitman’s express criticism of Pavlovian methods came a few years later, in 1925, when he happened to witness an experiment at the Laboratory of Pharmacology at the University of Chicago in which an injection of morphine, rather than feeding, was used as the unconditioned stimulus to train dogs to salivate to conditioned stimuli, such as the experimenters’ entrance into the dogs’ room.26 ‘It occurred to us,’ Kleitman and a colleague later remarked, ‘that this reflex might be used to test Pavlov’s theory of sleep.’27 Kleitman and Crisler took the experimental context itself as the conditioned stimulus. After several days of placing their dogs in their stands and injecting them with morphine, Kleitman and Crisler observed that they could make the dogs drool for fifteen to twenty minutes without the injection simply by putting the animal in its stand. One dog in particular had a tendency to salivate excessively and continuously at the beginning of the experiment, even after all effects of its morphine injection had worn off. This same animal also fell asleep in its stand frequently, both before, and long after, its morphine injection. It also continued to drool, even after it had fallen asleep. The most salient point the authors drew from this study was that Pavlov was probably incorrect to argue that sleep must be a generalized inhibitory state; for, if this were true, the dog’s salivation would have been inhibited during sleep, which was not the case at all. On the contrary, the authors suggested, sleep was itself an independent inhibitory state. Kleitman presented this material at the XII International Congress of Physiology, held in Boston in 1929, which Pavlov also attended. Although there appears to be no record of their meeting on this occasion, it does seem likely: Kleitman had communicated with Pavlov several years earlier, and the former’s presentation on sleep was the only one of its kind at the conference that might have satisfied Pavlov’s express desire to understand the nature of sleep. Pavlov certainly would have had good reason to pay attention to Kleitman, for the latter’s experiments threatened the epistemic status of salivation as a functional, multifaceted index of inhibition. Kleitman was refashioning sleep as a unique object of physiological research. He had no doubt that sleep was in some sense a product of conditioning. But sleep was not readily subsumed under the rubric Pavlov had tried to establish for it.

218

The Sleep of Others

Experimental Sleep and the Clinic: Biomedical Holism and the Chicago Neuropsychiatry Project In the late 1920s and 1930s, some American physiologists took Pavlov’s work as a model of how body and mind could be incorporated into experimental research. Pavlov shunned psychological terminology in favour of a nomenclature of brain activity, yet his method of conditioned reflexes attempted to analyse the way the organism and its environment interacted. Pavlov worked on a macro-level of behaviour, with organisms (usually dogs) that were entirely intact. His work was later criticized by some philosophers, who argued that the study of conditioned reflexes was reductionistic and could not capture the entirety of an organism’s behaviour, which must include the appearances of phenomena to the self.28 Nonetheless, holism, that inevitable counterpoint to reductionism, has its own historical trajectory, and Pavlov’s work could fit into the peculiar holistic rhetoric that permeated much of American biomedicine and physiology during the interwar period. Pavlovian research can be contrasted, for example, to Charles Sherrington’s work on spinal reflexes. Sherrington assumed the classical physiological position that the function of the nervous system would be best understood by tearing it down to its most basic element, testing it, and then rebuilding the whole in these elemental terms. Pavlov, on the other hand, started with the entire organism as his experimental object, albeit a modified version that was itself a form of technology. This approach offered a reasonable, if at times uncomfortable, fit with the holistic axiom that the whole could not be reduced to the mere sum of its parts.29 Pavlov’s methods were attacked by self-proclaimed holists, like the German émigré psychiatrist Kurt Goldstein (1878–1965), who felt that the study of conditioned reflexes had no relevance outside the phenomena of ‘training’ and ‘drill.’30 And self-described mechanists approved of Pavlov’s neurophysiological explanations of psychological phenomena. One such devotee, Lancelot Hogben, the chair of zoology at the University of Cape Town, South Africa, declared in 1929 that ‘in the light of Pavlov’s work we can now envisage the possibility that the methods of physical science will one day claim the whole field of what can properly be called knowledge.’31 Hogben was taking aim against his fellow countryman, Jan Christiaan Smuts, the sometime philosopher of science and South African statesman. Smuts (1870–1950) had coined the term ‘holism’ in 1926, in his attempt to describe how the synthetic drive to create ‘wholes’ was an innate part of the natural order. Yet even

Performing Sleep 219

this holist could cite Pavlov’s work approvingly: Smuts argued that Pavlov had shown that the synthetic work of consciousness was amenable to experimental investigation, and that such synthesis could be inherited, effectively creating a novel and dynamic whole out of individual experience.32 If the project of the method of conditioned reflexes appeared reductionistic, its practice was, at the very least, ambivalent on this point. For his part, Kleitman did not describe himself as a holist. Nor was he a Pavlovian, since he expressly rejected Pavlov’s theory of sleep as ‘generalized inhibition.’ In fact, Kleitman, like Claparède, eventually abandoned all of Pavlov’s theoretical conceits as comprising ‘more of a problem in semantics than in physiology.’33 But Kleitman found himself in the midst of an institutional push for holistic approaches to the medical and biological sciences that certainly bankrolled much of his research, if it did not determine his direction altogether. As the editors of a recent study of holism point out, ‘philosophers can provide formal and rather elaborate definitions of the term, but in the world that historians inhabit, holism is essentially relational; it constitutes a rhetorical claim made in opposition to other approaches that are characterized as excessively narrow or reductionist in focus. Indeed what is holistic for one individual is frequently perceived as reductionist by another.’34 The organism’s need to adapt to ‘natural forces and rhythms’ in order to harness the healing power of nature was a fundamental tenet of many interwar holists. Although Kleitman did not explicitly adopt the rhetoric of holism, his efforts to understand sleep in terms of adaptation as rhythm put his work firmly within the holistic biomedical discourse that surrounded him during the 1920s and 1930s at the University of Chicago. Medical research at the University of Chicago was in a unique position during this period. It was, in many ways, the fulfilment of Piéron’s dream of a scientific environment almost entirely void of historical pretensions and tradition. The university had originated as a small Baptist institution but had been completely transformed through an enormous endowment by John D. Rockefeller in 1890. It was unusual among the major American universities in that it had no medical school. All clinical instruction took place at Rush Medical College, which became affiliated with the university in 1898. When, in 1927, a four-year medical program was finally created through the opening of the University of Chicago Clinics, its founders and faculty adopted a self-consciously modern approach: clinical medicine at Chicago was to be identified with biological science.35 As several historians have shown, variations of a holist theme permeated the

220

The Sleep of Others

changes transforming medical and social science at Chicago in these years, which attempted to orchestrate research on issues of society and health from combined perspectives of multiple disciplines.36 In the neuropsychiatry project at Chicago, the epistemic role of the clinical case report, the mainstay of medical communication well into the twentieth century, was to be replaced by controlled experiment. As Franklin McLean, a physiologist by training and the associate dean of biological sciences, remarked in 1930, Chicago needed ‘to recognize Medicine as Biology rather than to incorporate various biological sciences within Medicine as is usually done.’37 Turning the clinic into a laboratory, rather than simply applying laboratory research to clinical medicine, implied an explicitly holistic and interdisciplinary approach to medical research. This problem was particularly acute in the psychiatric research of the period, which frequently treated asylum patients as little more than experimental material. In 1929 a committee was created with a mandate to study the possibility of organizing a department of psychiatry at Chicago. In their proposal to the General Education Board of the Rockefeller Foundation, the committee explicitly took on the rhetoric of holism and interdisciplinarity. They argued that the resolutely individualistic approaches of physiologists, psychologists, neurologists, and psychiatrists had to be coordinated with those of social scientists, who ‘have recognized that society as a whole is more than the sum of its constituent parts: – i.e., that society itself has its own peculiar characteristics not predictable on the basis of the study of the individual alone.’38 The new project in neuropsychiatry was to ‘bridge this gap’ by assuming, at the outset, that the ‘so-called functional disorders, including the psychoses and psychoneuroses, far from being in a class by themselves, are ultimately to be understood only through an integration of anatomy, physiology, psychology, social behavior, chemistry, and medicine.’ But, as one historian has observed, this attempt by the members of the Committee on Psychiatry to institute such a program at Chicago was ill-timed. The General Education Board at the Rockefeller was in the process of being dismantled, and its Progressive-era rhetoric of social amelioration was being traded in for an emphasis on fostering scientific research as an end in itself.39 There was also internal conflict among the committee’s members, particularly over the role of the founder of the Chicago Psychoanalytic Institute, Franz Alexander, whose appointment in the medical faculty infuriated some of the biologically oriented neurologists even as it charmed others. The original committee quickly ‘went to pieces on the rocks of psy-

Performing Sleep 221

choanalysis,’ according to a 1931 letter from the Office of the President of the University of Chicago to Max Mason of the Rockefeller Foundation.40 But a new committee was formed under Franklin’s McLean’s auspices that same year. Earlier in 1931, Alan Gregg, the director of the Rockefeller Foundation, had met with McLean to discuss the organization of psychiatry at Chicago. The problem, thought McLean, was the extreme variety of approaches to mental illness that would ultimately hamper the psychiatry department’s development. The neurophysiologists Percival Bailey and C.J. Herrick were, along with the psychologist Karl Lashley, identified as being against psychoanalysis, while Franz Alexander had made plans to leave for Germany because of the hostility towards him on the part of the medical faculty. Gregg himself was plagued by doubts about the direction of the new medical school at Chicago and recommended that the second committee make no formal proposal for the time being.41 Despite Gregg’s concerns, the interdisciplinary aspects of the new psychiatry program were highlighted in the second committee’s proposal. This was likely due to the fact that psychiatric issues were already being studied in the existing departments of physiology, neuroanatomy, and psychology, and treated in the new medical facility. The decision before the committee, then, was how it could best institute holistic approaches in practice: Should they opt for rigid centralization and create an independently powerful department of psychiatry? Or should a tolerance of various approaches prevail? The committee chose a middle road: they suggested that a department should be established but that appointments should be drawn from the full spectrum of approaches to mental illness, including psychoanalysis. The assumption underlying this interdisciplinary approach was that it would eventually yield a unified vision of mental illness and treatment. This was McLean’s vision as well. In an address delivered at the meeting of the American Orthopsychiatry Association in New York in February 1931, McLean began by drawing a distinction between the social and biological sciences but moved quickly to the ‘psychological and strictly bio-medical aspects of mental disorders.’ While both approaches were ‘necessary for continuing progress’ in the field, he concluded, ‘they must eventually converge.’ Psychoanalysis had made considerable progress over the past few decades, but this did not ‘relieve us of the necessity of carrying on appropriate studies of a more strictly biological nature.’42 Behind closed doors, McLean and his colleagues on the new committee were equally circumspect, but in a different way. In what may have been an oblique reference to Alexander’s

222

The Sleep of Others

provocative work in psychoanalysis, the new committee resolved that ‘in the absence of a Department of Psychiatry there is a tendency for these independent efforts to become firmly established as independent units and that unless some temporary means of coordinating these efforts is established they may prove to be a source of embarrassment to a future Department of Psychiatry.’43 Biological, sociological, psychoanalytic, and psychiatric perspectives were thus to be brought together at Chicago in an effort to understand disease at a meta-level. These changes also brought with them transformations in professional organization. Clinical faculty members were no longer able to teach part-time and supplement their income through private practice: they were now hired by the university as full-time researchers and obliged to give up their private (and often lucrative) careers. Kleitman thus found himself working in the midst of some vast institutional reforms at Chicago. After receiving his PhD in physiology under Carlson in 1923, he spent two years abroad, studying under Rudolf Magnus (1873–1927) at Utrecht and under Piéron and Louis Lapicque (1866–1952) in Paris. Piéron was an obvious choice for anyone interested in sleep. Magnus, who was nominated for the Nobel Prize in Physiology or Medicine in 1925 for his work on posture and muscle tonus, was also a world-renowned pharmacologist. He was keenly interested in physiological rhythms, as was Lapique, whose research focused on the comparative study of nervous excitation times in various tissues, which he dubbed ‘chronaxie.’44 Kleitman was clearly preparing himself to study sleep as movement and rhythm. He was immediately hired as assistant professor at Chicago upon his return in 1925 and spent the next few years publishing a number of papers on the physiology of sleep. After Kleitman was granted tenure in 1929, the next step was to secure the financing of his research. Carlson was an effective fund-raiser and often managed to procure gifts for the department from wealthy friends, as well as arranging toxicology and drug tests for manufacturers. Funding physiological research during the Depression was a difficult prospect, but Carlson proved both innovative and generous. In one instance, graduate students were hired to test the effects of non-alcoholic beer for a major American brewer by a series of blood tests after drinking a bottle every fifteen minutes, with or without a liverwurst sandwich.45 Women, who were probably more welcome as researchers in Carlson’s lab than in any other in the country, also participated, although they imbibed somewhat less of the ‘near beer’ than the men. Carlson sometimes paid his students out of his own

Performing Sleep 223

pocket and even allowed one student to live in the laboratory off the food meant for the lab animals while he pursued his PhD. In 1934 the Wander Company, a food manufacturer, gave Kleitman a grant of $3,500 for one year, which was subsequently renewed the following year.46 One of the products Kleitman was required to test was Ovaltine, a drink mix that was promoted as (among many other claims) a remedy for sleeplessness. After a few trials, Kleitman tentatively concluded that three tablespoons of the drink mix might enhance sleep, although there was no evident effect with only two tablespoons. The manufacturer immediately initiated an advertising campaign, complete with individual testimonies, to the effect that researchers at the University of Chicago had confirmed that Ovaltine brought on sleep. Carlson objected and the advertisement was pulled. But this did not prevent Carlson and Kleitman from becoming the butt of good-natured jokes around the department for sometime thereafter. A cartoon depicting a newly wed bride gazing at her sleeping husband on the bed with the caption ‘Damn that Ovaltine!’ ultimately became a permanent feature on the departmental bulletin board.47 Carlson’s support of Kleitman’s research took a unique form in the mid-1930s, when he made a concerted effort to get Rockefeller Foundation funding for his student. The neuropsychiatry project was, for the second time around, beginning to take shape, having found both a prospective leader and an institutional home. The program was to be located in a twelve-bed unit at Albert Merritt Billings Hospital run by psychiatrist Roy Grinker (1900–93), who was scheduled to return from two years of psychoanalytic training in Europe in 1935.48 That same year, Carlson forwarded Kleitman’s name as someone who could contribute physiological expertise to the new program. Kleitman was already known to Alan Gregg, since the two had met in March of the previous year during a conference in New York City.49 As a prelude to the meeting, Kleitman had sent Gregg an outline of the work he had conducted on sleep since 1922, and the document offers a unique window on how Kleitman depicted his work to a growing bureaucracy of funding agencies.50 Specifically, Kleitman emphasized the importance of two problems he deemed to be central to the field – namely, the question of the immediate cause of sleep, and the cause of sleep’s diurnal (twenty-four hour) rhythm. Kleitman framed each problem as ultimately having physiological solutions; indeed, the role or cause of dreaming in sleep, which, as we have seen, had been a central part of the study of sleep for centuries, hardly figured in his account. But he

224

The Sleep of Others

nonetheless refused to abandon psychological measures and methods entirely, as had Piéron. Rather, he used them as a way of creating a set of data that required correlation with observed physiological transformations taking place in sleep. To answer the first question, Kleitman had, in his 1923 paper, adapted Piéron’s method of ‘experimental insomnia’ for use with human subjects.51 Depriving six young male students of sleep for 40 to 115 hours, Kleitman then submitted them to a battery of physiological and psychological tests. Physiological measures indicated considerable changes. Heart rates slowed and blood pressure dropped over the course of the experimental trials, even as core temperature remained stable. Mental tests revealed that subjects’ ability to conduct routine calculations remained constant but their attention frequently strayed. Subjects reported a ‘slight buzzing in the head,’ or a ‘sensation of emptiness,’ as the period of insomnia wore on. Kleitman was himself a subject, and his commitment to the experiment was perhaps demonstrated by the fact that he was the only one able to stay awake for 115 hours. His behaviour, framed by the artifice of a traditional psychological experiment involving both a subject and an observer, was a crucial part of the experimental data, which suggested the intrusion of delusional thoughts over the course of the induced insomnia. While conducting a routine mental test, the observer of the experiment found Kleitman looking up and calling out logarithms in their proper order, and then suddenly declaring, ‘It is because they are against the system.’ Upon questioning, Kleitman revealed that ‘he had been under the impression that he was having a heated argument with the observer on the subject of labor unions.’52 Such commentary was much more than a curious rhetorical flourish. One of Kleitman’s most important observations had come from just this sort of introspection. Subjects found that sleep became irresistible when they were allowed to relax. Sleep was most easily brought on not by depriving the subject of external stimuli, such as light and sound, but by eliminating proprioception (the feeling of one’s own musculature) by encouraging or allowing subjects to relax. Although he did not claim to have isolated any single cause of sleep, Kleitman concluded that muscular relaxation was an essential factor for the appearance of sleep; conversely, muscular tension was necessary for wakefulness. After all, Kleitman noted, no one fell asleep at dance marathons! Fatigue encouraged relaxation, which in turn generated sleep. Extreme fatigue, however, caused pain, thus increasing the afferent stimuli from the muscles

Performing Sleep 225

and ending in insomnia. With this relocation of the cause of sleep from external stimulus to proprioception, Kleitman perpetuated the functionalist tradition, initiated in the American context by William James, of assigning purpose to feelings, in particular, the sensations that came from one’s own musculature. At its core, however, Kleitman’s focus on the physiological parameters of sleep encouraged him to describe consciousness in primarily passive terms; it appeared when there was a requisite amount of stimulus, and disappeared when this amount slipped below a required threshold. Such a mechanistic perspective could not, Kleitman admitted, account for the diurnal periodicity of sleep, and at this point, he fell back on the language of conditioned reflexes.53 Pavlovians had agreed that conditioned reflexes originated in the cerebral cortex, which Kleitman fit within the British neurologist Hughlings Jackson’s ‘doctrine of levels,’ which stated that the higher, cortical functions of the brain were the most recent, in evolutionary terms, and were also the first to break down with the onset of neurological disease. The lower, subcortical regions were the source of vegetative functions, such as the regulation of blood pressure, respiration, heart rate, muscle tone, and so on. Kleitman had observed that the diurnal rhythms of these latter functions persisted throughout periods of experimental insomnia, even as the psychological phenomena of attention and association could, with effort, function normally at any time. In his letter to Gregg, Kleitman suggested that his dualistic approach put him in a unique position to ‘settle the controversy concerning localization of a sleep “center” in the cerebral cortex or in the subcortical structures,’ a question that had been rendered more acute by the recent outbreaks of encephalitis lethargica.54 Kleitman’s hypothesis was that there were actually two centres for sleep. The first was subcortical – it worked on a diurnal basis, was not linked to environmental changes such as the presence or absence of daylight, and could be found both in lower animals, decorticated dogs, and the very young of higher animals, including man. The cortical mechanism, on the other hand, was generated in higher animals through individual experience. It was this mechanism that was responsible for adapting sleep and wakefulness to the twentyfour-hour cycle of day and night. Kleitman’s experimental program of sleep research was simultaneously physiological, psychological, and sociological, and as such, it resonated with the interdisciplinary and holistic tenor of the proposal put forward by the Committee on Psychiatry. Sleep was an internal physiological necessity, but its rhythm, Kleitman argued, was learned. It was a

226

The Sleep of Others

psychological adaptation to the social and cultural conditions that demanded long periods of wakefulness, and, as he suggested in his 1923 paper, there were prospects that it could be adapted to new regimens of sleeping and waking. Kleitman described his research as a close fit with the new unit at Billings Hospital, where he hoped to study ‘the sleep and diurnal variation in performance of mental defectives, from the lowest imbecile to the highest grade moron, to determine the extent to which the cortical center of sleep dominates the subcortical one.’ In addition, he wanted to study the sleep disturbances in cases of encephalitis lethargica, ‘the sleep habits of psychopathic individuals,’ and ‘the various types of insomnia met with clinically.’55 Gregg’s initial response was one of caution. In their initial interview, Kleitman emphasized that his research could continue to be based out of Chicago’s Department of Physiology but that he required more funding to expand it to the point where could achieve clinical relevance. The greatest obstacle to this expansion was not the one typically encountered by the Rockefeller Foundation – the development and creation of scientific instruments on a large scale. The central difficulty was instead the problem of finding sufficient numbers of human subjects willing to participate in such long-term experiments. This problem could be avoided, argued Kleitman, if Gregg would integrate sleep research with the psychiatry project at Billings Hospital.56 Kleitman must have told Carlson of Gregg’s hesitant reaction to his request, for, a few months later, Carlson pleaded for Gregg to grant Kleitman a second audience. He defended Kleitman’s work as ‘a significant part of the problems in neuro-psychiatry that are being pursued on this campus,’ but that were currently under threat as a result of recent reductions in the Physiology Department’s budget.57 The request seemed to fall on deaf ears. After returning from a summer in Europe, Kleitman tried to meet Gregg again that September.58 He managed to see only Gregg’s secretary, who assured him that he had not been forgotten and that Gregg was planning on discussing it with his colleagues in the Foundation sometime in the fall. But winter soon arrived, with no word from Gregg. In January 1935 Carlson crafted a stronger plea on Kleitman’s behalf. He reminded Gregg that he had initially expressed interest in the clinical extension of Kleitman’s work and had then framed the question in terms of how it could be best supported: as an independent project in the study of sleep, or as part of ‘the larger project of psychiatric research’ at Chicago.59 ‘Dr Kleitman and I,’ he continued, ‘are naturally anxious for some word from you concerning this matter.’

Performing Sleep 227

Gregg arrived in Chicago a few weeks later, prepared to hear Carlson and Kleitman out, this time not in the Foundation’s offices but in Kleitman’s own laboratory. Despite Carlson’s continued ‘characteristic turmoil’ in his struggles with anti-vivisectionists protesting experiments conducted in his department and across the United States, he was, Gregg noted, absolutely steadfast in his enthusiasm for Kleitman’s work.60 Gregg was equally impressed. Kleitman, he noted, presented himself and his work much differently in this second interview. Whereas he had earlier been formal and awkward, Kleitman, to Gregg’s surprise, was now animated, precise, and clear. He seemed at home in his laboratory, and able to communicate his ideas and hopes more effectively when surrounded by the various devices – many of which he had invented himself – that provided experimentalists with a visual picture of sleep (see figs 17, 18, 19). ‘Interview is remarkable illustration of advantages of seeing a man in his own laboratory as contrasted with interview at RF [Rockefeller Foundation] offices,’ noted Gregg, who ultimately decided to fund Kleitman’s project. The idea, so common to many ethnographically based studies of science since the late 1970s, that one could not grasp the true nature of a scientific project by merely listening to a scientist’s description of it became entrenched in Gregg’s mind through his encounters with Kleitman. Five years on, his belief in the importance of the ‘laboratory tour’ remained: ‘It is dangerous to believe that the presentation of a request in our offices is likely to be about the same as the presentation in the laboratory of its origin. Kleitman had begun his scientific career as a diener [a medical laboratory assistant] at the Rockefeller Institute. He must have reverted to his old mood in his first presentation in our office, for only on second presentation in his laboratory did he do himself justice. My conclusion is never to judge a man out of his laboratory if you can avoid it.’61 Given the fact that Gregg’s initial optimism regarding the value of Kleitman’s research had almost completely waned by the time he made this comment, it is perhaps surprising that he did not take the opposite tack and suggest that such interviews should be made outside the laboratory, to avoid becoming spellbound by the inevitable arrays of instruments and graphical displays. But Gregg was hardly the only one to be charmed by these trappings. Contemporary press coverage of Kleitman’s work, as well as some more recent biographical sources, frequently made reference to the fascinating laboratory devices he had invented for the study of sleep.62 Clearly, agencies like the Rockefeller Foundation were recognizing and even reshaping the way scientific

228

The Sleep of Others

17 Kleitman was keenly interested in infant sleep, since it was an important component of his evolutionary theories. Pictured here is his motility recorder, attached to a crib. (Courtesy Special Collections Research Center, University of Chicago Library)

Performing Sleep 229

18 Kleitman’s motility recorder was probably the most important of the graphical devices he invented to study sleep. Through the use of such instruments (which have renewed use in the study of sleep disorders today), Kleitman characterized sleep as the diminution or cessation of the activity of the skeletal musculature. (N. Kleitman, N.R. Cooperman, and F.J. Mullin, ‘Studies on the Physiology of Sleep. IX. Motility and Body Temperature during Sleep,’ American Journal of Physiology 105 [1933]: 574–84, 576, used with permission)

work was related to instrumental design and performance. Sleep quite literally emerged as an object of research in precisely this context, in part because the graphical technologies that enamoured the popular press allowed this private phenomenon to be visualized for the first time. Sleep was entering the era of mechanical reproduction. From April 1935 until October of 1938, Kleitman received funding totalling just under $16,000 from the Rockefeller Foundation.63 The money was clearly linked to the neuropsychiatric project at Chicago, but there were problems from the very beginning. Grinker returned from Europe to start up the department in the spring of 1935, and, in a letter to Gregg in October of that year, he offered an almost idyllic description of the twelve-bed unit. There was a hydrotherapeutic room, a psychology laboratory for the study of aphasia and the behaviour of patients

230

The Sleep of Others

19 Kleitman’s twin daughters Esther and Hortense frequently served as his experimental subjects. (Courtesy Special Collections Research Center, University of Chicago Library)

Performing Sleep 231

with frontal-lobe damage, and the rudiments of a chemical laboratory for drug studies. Ninety per cent of the patients were schizophrenics, noted Grinker, who was busy conducting pharmaceutical, metabolic, and ‘electrical’ studies in order to determine ‘what differentiates schizophrenics biologically.’64 But ‘modified analytic methods’ were also used in such cases, and these were applied as well to the study of phobias in children in the pediatric unit. Clinical conferences, held four mornings every week, were well attended. Links with the Department of Sociology were expected in the near future, and the Department of Education had also expressed considerable interest in the project. A program of sleep research had not yet been established, but Grinker’s interest in Kleitman’s work was emblematic of his optimism for future interdisciplinary relationships: ‘Cooperation with other people on the campus has been much better than I had anticipated,’ he wrote. ‘We are arranging so that Kleitman can do shis [sic] work on “Sleep” on our unit, and will admit patients especially interesting to him.’65 The fact that Grinker felt obliged to place the word sleep in quotation marks is indicative of just how unfamiliar such research must have been to psychiatrists during the 1930s. For his part, Kleitman was taking a long time setting up his laboratory. At the end of March 1936, the better part of his funding remained unspent, and he had to pen an apologetic letter to Gregg for not moving more quickly.66 By this time, things were beginning to fall apart. Kleitman’s letter came just around the same time as a divisional committee on neurobiology was struck, presumably to protect the biological orientation of the psychiatry unit. Grinker had already expressed his dissatisfaction with the fact that medical interests seemed to dominate the unit, and, when he discovered that he was being paid less than his clinically oriented colleagues, he revolted and took up a position at the Michael Reese Hospital.67 Grinker was not alone in his complaints, for several other members, in particular Bailey, were unhappy with their situation. The problem was simultaneously institutional and economic.68 Biologists did not then recognize psychiatric research as being at all related to their science, and this tended to diminish the authority of biologically based researchers, such as Bailey. This situation did not begin to resolve itself, at least in the United States, until the 1940s, when neuropsychiatry came to play a more important role in the cognitive foundations of neuroscience and cybernetics.69 The unit’s clinicians, on the other hand, were, like all other clinicians at Chicago, obliged to pool the fees received from the private patients they saw at the university hospitals.

232

The Sleep of Others

Unlike most other medical schools in the United States, the University of Chicago did not have access to the charity patients at a public hospital, so private patients were effectively being charged for being used as ‘clinical material.’ The university’s level of remuneration to the clinicians was not deemed comparable to that available in private practice, with the result that research-oriented neuropsychiatrists quickly discovered that their clinically oriented colleagues were better paid, and the latter group turned more and more towards private practice outside the university altogether. The full-time system of research and teaching that had marked physiology’s independence from medicine had forced investigators to choose between the two alternatives. In July 1936 Grinker was replaced by David Slight, a psychiatrist from McGill University. His research interests were diverse and included migraine, endocrine function in emotional states, the physiological effects of psychotherapy, and the role of allergies in mental disease. But he was unable to maintain the interdisciplinary structure of the neuropsychiatric unit. In fact, quite the opposite occurred, for he seemed to impress no one. One early, colourless assessment described him as ‘a competent teacher and an able administrator,’ but ‘not a brilliant investigator.’70 Perhaps more important, his background and the direction of his research were primarily, if not exclusively, clinical. This proved to be a problem for Kleitman, who had forged an alliance with Grinker on the basis of their common biological interests, only to see it shattered upon the latter’s departure. Kleitman published only four papers during his three-and-a-half years at the Albert Merritt Billings Hospital, but none of them was in a medical journal, despite the fact that the motion for his original grant cited ‘catatonic states, post-encephalic sleepiness, catalepsy, insomnia, and narcolepsy’ as the important medical questions that would be illuminated by his work.71 The fact that his funding continued until 1938 was probably a product more of accident than of design. Less than a month before Grinker left, Gregg extended Kleitman’s grant to October 1938.72 Kleitman informed Gregg on several occasions that he was working on a monograph on sleep – the massive tome that would become Sleep and Wakefulness in 1939. His continuing biological orientation was revealed most clearly to Gregg when he made a request, in early 1938, for additional funding to travel to Norway in order to conduct research on the stability of the diurnal rhythms of sleep in the continuous summer light of the Arctic summer.73 This project was clearly on a par with his famous ‘Mammoth Cave’ experiments, which took place that same year. He and a colleague, Bruce Rich-

Performing Sleep 233

20 Kleitman utilized his motility recorder in his 1938 experiments at Mammoth Cave, Kentucky. The buckets were filled with water, presumably to discourage insects from climbing into bed with Kleitman or his co-investigator, Richardson. (Courtesy Special Collections Research Center, University of Chicago Library)

ardson, spent six weeks in a cave in Kentucky trying to maintain nontwenty-four-hour sleep/wake schedules in the complete absence of daylight (fig. 20). Kleitman was unable to adjust to any of the attempted routines, but his younger colleague succeeded in maintaining a twentyeight-hour rhythm. Despite his own failure to adapt, Kleitman nonetheless concluded that the the diurnal sleep/wake rhythm was, indeed, pliable, given the right circumstances. The experiment, held in public in a popular tourist destination and a national historical site, garnered considerable publicity.74 Gregg, however, was clearly unimpressed by this sort of research and turned down Kleitman’s proposal for funding of his Norway project.75 When his original funding expired in October 1938, Gregg was in no mood to resuscitate any part of it. His final appraisal was harsh, and stemmed from an eminently practical standpoint. ‘It is not evident,’

234

The Sleep of Others

declared Gregg, ‘that the grant for Kleitman’s work on sleep has provided results proportionate to the total expenditure of $15,583.74.’76 In Gregg’s mind, Kleitman’s research had been a dead end, despite its empirical productivity. ‘There is little to report,’ he wrote, ‘beyond saying that the grant produced facts about sleep which are not of much significance for all that they are conscientiously planned and accurately obtained.’ The results of the research were ‘negligible without being false or useless.’ Worse still, Kleitman’s project had failed to encourage the sort of internal cohesion that Gregg had hoped to foster. In his support of the study of sleep, which he depicted as ‘a condition intimately related to exhaustion and to the recuperation of the activity of the nervous system,’ Gregg had clearly hoped to bring psychiatry and physiological research closer together. But these links had failed to materialize. This was due in part, Gregg thought, to Kleitman’s character as a ‘thoughtful, hardworking, unobtrusive subordinate’ who was unable to capture Slight’s attention following Grinker’s departure. In the future, Gregg decided, interdepartmental ties should be forged between departmental heads, not junior members. But, for all his negative personal judgments and sociological theorizing on the opaque nature of inter-faculty dynamics, Gregg was adept enough to recognize that the core of the problem had to do with the nature of the subject matter itself. ‘Sleep,’ he noted, ‘remains a most common phenomenon, marvellous to the few and attractive as a major interest to almost no one. It is possible that this grant illustrates the error of attacking a large task in a small way.’77 There simply was no community of sleep researchers during the 1930s – anywhere. And scientific research, if it was to survive during the twentieth century, had to be a communal enterprise. Financing a research project of no consequence to either psychiatry or physiology could not possibly help unite those fields, particularly if the project itself was part of an utterly marginal field almost devoid of investigators. Sleep had been successfully detached from its traditional linkages to dreaming and to fatigue. It had its own diseases. It even had a certain degree of theoretical structure surrounding it. But there was, as yet, little at its core. It remained, not quite nothing. An Evolutionary Theory of Sleep, 1939 Despite the failure of his work within the neuropsychiatric project, Kleitman seemed well aware of just how unique the situation had been, for it

Performing Sleep 235

was almost certainly the first time that sleep had been funded as an object of research, as opposed to an incidental part of a researcher’s wider career. Unaware of Gregg’s disdain, he wanted to acknowledge Gregg by name in his upcoming book on sleep but was encouraged not to by his colleagues at Chicago. So, much to Gregg’s relief, he sent a letter of thanks instead.78 This was a pity, since a published acknowledgment might have brought to the attention of the book’s admirers (of which there are now many) the unique institutional and social conditions under which sleep had emerged as a thing worth studying. Kleitman’s Sleep and Wakefulness represented the culmination of Kleitman’s experience under Carlson. It was also a testament to the early failure of the Rockefeller Foundation to establish an interdisciplinary neuropsychiatric unit at Chicago. Kleitman’s precise experimental style, his characterization of physiological activity as inherently rhythmical, and his quest to design and construct instruments that could visualize sleep can all be traced back to Carlson’s influence. Kleitman’s ‘evolutionary hypothesis,’ which argued that sleep’s diurnal rhythm was not inherited but conditioned, reflected the Rockefeller Foundation’s attempt to bring social and cultural factors under the purview of biomedical research. Pavlov’s work furnished the key intellectual context. Like Carlson, Kleitman was both inspired by and critical of Pavlov’s approach to physiology. Pavlov had constructed an elaborate theoretical structure and nomenclature to explain how the cerebral cortex functioned in conditioning. Carlson and Kleitman, on the other hand, relied on self-inscribing instruments to make their case about the periodicity of digestion on the one hand and sleep on the other. The graphical method they used proved a far more robust tool than the collection, measurement, and tabulation of drops of saliva. It offered a medium of communication that could easily inscribe, calibrate, and synchronize an array of wellknown physiological phenomenon. Its reliance on instrumental performance meant that it was portable and, therefore, universal. Pavlov’s methods, in contrast, were unwieldy, easily contaminated, and difficult to standardize; thus, they appeared local and contingent, their applicability questionable outside Pavlov’s Institute. The method of conditioned reflexes also suffered from shortcomings that hearkened back to the heyday of hypnotism. Like hypnotism in the nineteenth century, Pavlovian experiments induced sleep. They had no pretences to studying the phenomenon in situ. Although the creation of phenomena is quite obviously at the core of much twentieth-century experimental sci-

236

The Sleep of Others

ence, it fit uncomfortably with the subjective experience of sleep as a rhythmic routine of fatigue emerging at the end of a long period of wakefulness. Claparède had expressed such concerns in 1934, and Kleitman had embraced them in his adaptation of Piéron’s method to the study of human subjects. In Kleitman’s hands, the graphical method became a mechanical and visual analogue of the embodied experience of sleep. Yet Kleitman did not entirely abandon Pavlov’s conception of the conditioned reflex. By combining it with a Jacksonian perspective on hierarchical cerebral organization, Kleitman proposed an evolutionary theory of sleep in Sleep and Wakefulness that could account for both the physiological mechanism of sleep onset and the diurnal alteration of sleep and wakefulness. This mechanism functioned in a manner similar to that which drove the autonomic stomach contractions of Carlson’s subject Vleck. Its effects, Kleitman argued, were best observed in naturally undeveloped organisms or their pathological or experimental equivalents: newborns, anencephalous babies, and decorticate dogs. All these subjects tended to remain awake only so long as wakefulness enabled them to meet their internal demands of nutrition and excretion. This Kleitman dubbed the ‘wakefulness of necessity.’ It was phylogenetically primitive, and was indentified with the damaged subcortical brain structures that Economo had found in his encephalitis patients. ‘Wakefulness of choice,’ on the other hand, was an example of the vast evolutionary advances higher organisms had achieved over the restrictive demands of subcortical structures. It enabled these organisms to maintain a state of full awareness even when there were no internal bodily demands to be satisfied. This was a phylogenetically young system that Kleitman attributed to the powerful influence of the cerebral cortex on behaviour. The twenty-four-hour rhythm of sleep and wakefulness, Kleitman argued, was not caused directly by the alternation of day and night. Rather, its causal influence was mediated through enculturation. The sociocultural practices and disciplines surrounding sleep meant that the diurnal rhythm would triumph over the more frequent subcortical rhythms. This ensured longer, ‘consolidated’ periods of wakefulness that were, in every instance, of benefit to the species. Wakefulness was no longer the default state assumed by experimental psychologists, physiologists, and biologists. It was an evolutionary achievement, one that, Kleitman hoped, had substantial social consequences, and could therefore provide the basis for research in the flexibility of the sleep/

Performing Sleep 237

wake schedules in fields as disparate as work physiology and child rearing.79 The very local context of the neuropsychiatry project at the University of Chicago, which attempted to integrate sociological, physiological, and psychological phenomena into a holistic and interdisciplinary program of research, had clearly left an imprint on Kleitman’s thinking about sleep. Edmund Jacobson and the Imperative to Relax Kleitman was not alone in pursuing this connection between psychology and physiology, even within Carlson’s Department of Physiology. During the 1920s and 1930s, Edmund Jacobson (1888–1983), who had trained as a clinician and as a psychologist, had created a form of relaxation training that would eventually lead him to examine some of the same phenomena of sleep that interested Kleitman. Although Jacobson’s work culminated in a clinical therapy, rather than in an investigative technique, he, like Kleitman, invoked an image of sleep that emphasized the concepts of efficiency and purpose, as well as the important role played by internal sensations. There is no question as to whether or not Kleitman was aware of Jacobson’s work. Carlson’s department was one of the larger units in the United States at the time, numbering about nineteen researchers in the early 1930s. Kleitman was the only one expressly interested in sleep, but he and Jacobson, despite their different orientations, knew and cited each other’s work. Kleitman, for example, referred to Jacobson’s work on relaxation several times in a 1928 public lecture on the ‘Problem of Sleep.’80 Jacobson’s work, however, was animated by a problematic that was much different from Kleitman’s. Sleep’s function, its rhythm, and its evolutionary development were largely irrelevant for him in comparison to the more pressing clinical problems of insomnia and its relationship to hypertension. Jacobson was the early-twentieth-century heir to neurologists such as Hammond and Silas Weir-Mitchell, rather than a direct descendent of physiologists such as Bernard, Marey, or Mosso. Despite these differences, an examination of Jacobson’s work provides unique insights into the nature of sleep research at Chicago during the 1920s and 1930s. The interdisciplinary bent of the university during this period brought him and Kleitman together, both on problems of sleep and relaxation within Carlson’s department and in the ill-fated neuropsychiatry unit at Billings Hospital. Perhaps most important, they shared a similar devotion to graphical methods in their investigative practices. Despite their

238

The Sleep of Others

differing orientations, the success of their work hinged on developing and exploiting new ways to represent sleep visually. Jacobson’s work has been almost completely ignored in sociological and historical accounts of sleep research, but it has received some minimal attention among historians of psychology.81 His method of ‘progressive relaxation,’ still in use today, is certainly his best-preserved legacy, and I have discussed its origins in detail elsewhere.82 But, given Kleitman’s own emphasis on relaxation as an internal sensation and its role in sleep, Jacobson’s efforts are worth examining here. They provide a striking counterpoint, as well as a more successful clinical analogue, to Kleitman’s offerings on the subject in Sleep and Wakefulness. Jacobson’s interest in the relationship between mind and body has been systematically woven into the fabric of his life story.83 He was born in Chicago in 1888, the only child of a Jewish-American entrepreneur. When he was ten, a fire broke out in a hotel that his father owned, and in which the family lived. The young Edmund was surprised to see how the residents of the hotel, many of whom he had come to know well, behaved like completely different people when confronted with this terrifying situation. His interest in the psychological and biological effects of nervous tension was inaugurated. Of course, Jacobson’s interest in nervous illness was as much a reflection of his age as it was a product of this shocking childhood event. The study of nervous excitability was a prominent aspect of psychological and neurological research in the United States around the turn of the century, spurred on by George Beard’s popular 1881 book, American Nervousness. Jacobson chose to pursue the subject through experimental psychology, then the fastestgrowing discipline in America. He completed his undergraduate degree at Northwestern University in 1908 and immediately began graduate studies at Harvard under William James and Josiah Royce (1855–1916). After completing his doctoral thesis on inhibition, he received his PhD in 1910 and went on to Cornell University to study ‘systematic experimental introspection’ under Titchener, then the most outspoken proponent of experimental introspection in America. Titchener was not, despite his training in physiology at Oxford, sympathetic to functionalist explanations of any sort; nor did he approve of the decidedly practical bent of his younger behaviourist colleagues. In fact, his passion for the study of the normal human mind never really fit in to the mainstream of American psychology, which was rapidly turning towards the study of mental performance and individual differences. Just as American psy-

Performing Sleep 239

chologists were touting their ability to test and classify populations of subjects as their most important professional and scientific attribute, Titchener remained mired in the practice of training students in the rather obscure art of experimental introspection. Jacobson arrived at Cornell just as Titchener’s laboratory was becoming embroiled in an international debate that would eventually contribute to the collapse of introspection as an investigative method in experimental psychology.84 At the centre of the dispute was a ques- tion about reductionism and the nature of thought: Were thoughts merely composites of sensations, feelings, and images, or were they non-reducible, independent components of mental activity? How, for example, was the meaning of a statement achieved? Potential solutions to this problem would have a considerable impact on how psychologists went about their research, which would, in turn, affect the authority of their field by drawing it closer to physiology on the one hand and philosophy on the other. A number of psychologists led by Oswald Külpe at the University of Würzburg held to the position that thoughts were non-reducible and independent. By deploying a flexible method of introspection (Ausfragemethode) that allowed for a free and sympathetic communication between the experimenter and observer, the Würzburg group claimed to have discovered thought elements that had no relationship to perceptions coming from sensory organs. But such ‘imageless thoughts’ could be detected only if the psychological observer had the proper introspective training, and if the experiment was conducted in the right manner. Titchener was the most outspoken critic of the ‘Würzburg School’ in the United States. He insisted that the interview-like approach proffered by the Würzburgers allowed too great a role for suggestion. A more rigid and repetitive approach to introspection was required, with much less independence and freedom of expression afforded to the experimental observer. Between 1909 and 1912, his laboratory at Cornell produced a number of studies that directly addressed the existence of the ‘imageless thoughts.’ Jacobson made an important contribution to this debate, since his technique of examination included an explicit request that, when the observers went through transcripts of the experiment, they identify which of their statements directly described ‘conscious processes,’ and which referred to ‘statements concerning meanings, objects, stimuli and physiological occurrences inside.’85 To the twenty-first-century mind, this probably seems like little more than an exercise in copy editing. And it was. But, within its historical context,

240

The Sleep of Others

it was a form of introspective training. Its purpose was to translate the Würzburgers’ insistence that psychological experiments include an empathetic relationship between the experimenter and the observer into an explicit set of rules that observers were obliged to follow. The results were perhaps predictable: the observers failed to detect the existence of any thoughts that could not be traced back to sensations, feelings, or images in the mind. Despite his adept handling of experimental introspection at the hands of its American master (or perhaps because of this), Jacobson was not particularly impressed by the state of scientific psychology. He left Cornell after only a year of working with Titchener. The problem of imageless thought was never resolved; rather, psychologists were abandoning the issue altogether and arguing that behaviour, not consciousness, was the proper object of a truly scientific psychology. Jacobson was thus left with a set of investigative skills that were quickly losing their relevance to psychological research. He also had more pressing difficulties. His father’s business ventures were generally unsuccessful, and he was often absent for long periods of time. It was up to him to provide for himself and his mother, who had helped put him through university by working as a typist. An additional factor might well have been the rather dim prospects for Jews in American psychology during this period. Discrimination against Jews, as well as many other groups, was pervasive in American society during the early twentieth century, and many universities established an informal but rigid system of ‘quotas’ limiting access to both education and academic positions.86 Medicine almost certainly held out greater attractions for Jacobson than did psychology. The Clinical Implications of Introspection Following his year with Titchener, Jacobson returned to Chicago to study medicine at Rush Medical School, which was then affiliated with the University of Chicago. He joined Carlson in the Physiology Department, teaching there for one quarter before beginning his studies at Rush. After finishing his courses, he planned to split his internship between the Phipps Psychiatric Clinic at Johns Hopkins University in Baltimore and the Cook County Hospital in Chicago. The Phipps Clinic was run by Adolf Meyer (1866–1950), one of the most influential psychiatrists in America. He was also one of Titchener’s most outspoken critics. Meyer was deeply engaged in bringing psychological instruction into the medical curriculum, so the Phipps Clinic was an obvious place to go

Performing Sleep 241

for someone with Jacobson’s talents. Indeed, the behaviourist J.B. Watson was already working there. Meyer welcomed Jacobson’s application, but the plan was foiled. A month before he was to leave for Baltimore, another prospective intern drowned while swimming, thus forcing Jacobson to take up his internship at Cook County six months earlier than he had originally planned. The opportunity to work with Meyer never repeated itself. The experience at Cook County proved difficult. It was a large institution of some 2,000 beds, and Jacobson was apparently the subject of discrimination there after his arrival in the autumn of 1915. Initially, the hospital refused to allow him to intern there at all, simply because of his Jewish background. After he threatened to sue the hospital, they accepted him, but the administration refused to assign him to any particular ward. Jacobson’s resourcefulness was practically unbridled, so he took to ‘making his own assignments,’ treating whichever patients would allow him to (and many refused). This experience, he later related to his long-time assistant, was what ultimately turned him into a proficient diagnostician.87 After leaving Cook County, Jacobson attempted to set up a private practice. But his disastrous internship proved fateful, since it meant that he was left with virtually no referrals from clinicians at Cook County. Frustrated, he obtained a part-time instructorship in Carlson’s department in 1919, and he also began clinical work at Michael Reese, a Jewish hospital in the Chicago suburbs. He also worked closely with Harvey A. Carr (1873–1954), the chair of psychology at Chicago and one of the most prominent functionalists of the day. Jacobson quickly merged these seemingly disparate interests by building both his clinical practice and a research program around a technique he dubbed ‘progressive relaxation.’ This was nothing less than the practice of systematic experimental introspection translated into clinical terms. Between 1917 and 1924, Jacobson conducted a series of studies at the University of Chicago on the effects of relaxation training on gastric hyperacidity, esophageal irritability, arterial hypertension, and nervous excitability. Using X-rays and electrophysiological recording of muscle currents (electromyograms, or EMGs), Jacobson demonstrated that relaxation training could have a substantial impact on each of these disorders. The origins of the technique had come from Jacobson’s earlier work on the psychology of inhibition. At Harvard, and later at Cornell, Jacobson had noted that sudden stimuli did not inhibit the perception of continuous stimuli, as he had expected. The reverse occurred, as subjects reported an increased intensity of smells, such as that of India rubber or beeswax,

242

The Sleep of Others

when a klaxon was suddenly sounded.88 Upon interrogating his observers, Jacobson discovered that they were actually making greater efforts to attend to the odours whenever they heard the sudden sounds. As he was to do for much of his future medical career, he counselled his subjects to stop making such efforts and to remain as passive as possible. And relax they did; or, rather, they tried. Since they were still obliged to give an exhaustive inventory of what passed through their minds, Jacobson’s subjects reported their difficulties in becoming utterly passive while at the same time attending to their every sensation. And yet they slowly achieved a state of ‘mental relaxation,’ which did not, for all that, serve to increase the inhibitory power of the sudden sounds but which did lead to a state that generally diminished all sensations, provoking feelings of reverie and detachment. Jacobson interpreted this discovery in terms of proprioception. A decrease in muscle tension brought about through the practice of relaxation, he argued, effectively inhibited thought. Jacobson’s definition of ‘inhibition’ separated the data of psychology and physiology along the fault line of fatigue. Inhibition was ‘the reduction of any conscious activity while the stimulus is in operation and [which] undergoes no corresponding diminution,’ but the reduction ‘must not be due to fatigue.’89 In other words, the practice of relaxation produced inhibition, without either causing fatigue or being caused by it. It was, like experimental introspection, a product of a mental discipline that had to be learned. In effect, Jacobson had, like Pavlov, stumbled upon a variant of sleep as something that prevented (or quite literally, inhibited) the completion of his experimental labours. He then turned the situation around and made the problem itself an object of research.90 His medical technique of progressive relaxation involved an initial patient interview, careful observation by the physician, and a systematic practice of record keeping that would chart the patient’s progress in learning how to relax properly.91 The technique, like that of psychoanalysis, was in many ways a tacit one that served to break down the physician’s absolute authority over the patient. Relaxation was a skill that needed to be mastered before it could be properly taught: ‘Like any new procedure, for instance a surgical operation, the present method is best learned by the physician who sees what is done rather than from a written description. The aim is to train the patient by his own metal [sic] – his own initiative. He learns to localize tensions when they occur during nervous irritability and excitement and to relax them. It is a matter of nervous reëducation.’92 But, for all its reflexivity, Jacobson insisted that the technique was

Performing Sleep 243

nonetheless scientific. It was due to a combination of intensive introspective training accompanied by the graphical technologies that populated the physiological laboratory and that made possible the distinction between ‘ordinary’ and ‘progressive’ relaxation. The former involved the mere reduction of activity, while the latter incorporated a vigilant surveillance rendered habitual with the help of machines. Jacobson taught his patients to recognize the feeling of muscle tension in their own bodies by concentrating their attention on their muscular sense. They were instructed to attend to specific muscle groups (such as the flexors of the right forearm) until they could easily locate the distinctive feeling of each. At the same time, they were taught to relax each group, ‘progressively,’ until they reached the point that they could no longer sense any tension at all. The goal was to create a habit out of relaxation that could be practised anywhere. Despite the fact that Jacobson was selling progressive relaxation as a form of introspective training that could both prevent and cure certain forms of disease and reduce hypertension, the question of sleep was never far from his mind. Indeed, in his first monograph on the topic, published by the University of Chicago Press in 1929, Jacobson included a personal anecdote, describing ‘a nightly insomnia which persisted for hours, while mental activity continued regardless of need of rest.’ He had suffered from this insomnia in 1908, while working on his dissertation under James.93 The clinical use of sleep was featured throughout Jacobson’s book, which began with his criticism of the famous ‘rest cure’ of Silas Weir-Mitchell. The cure was an extended period of inactivity and virtual captivity, primarily for hysterical and neurasthenic women. It had been popularized by Charlotte Perkins Gilman in her 1892 short story, ‘The Yellow Wallpaper,’ in which the patient’s imagination runs riot as her body endures a forced rest. Jacobson found it strange that WeirMitchell had clearly recognized the importance of nutrition in health and disease by advocating an active choice of diet in his 1879 text, Fat and Blood and How to Make Them, but was unable to frame relaxation in similar terms. Weir-Mitchell’s ‘rest cure,’ he noted, advocated a kind of passivity of the body that would eventually transfer its effects to the mind, thus hindering, rather than helping, the neurasthenic patient. Weir-Mitchell thought of relaxation as something that happened when bothersome and distractive stimuli were eliminated. Jacobson, in contrast, described the hygiene of relaxation in terms of an active training of the will. Learning to relax was part of a program of self-improvement, and its perfection included an enhancement of what was normally expe-

244

The Sleep of Others

rienced as natural sleep, which Jacobson was at pains to distinguish from other forms of artificially induced quiescent states: ‘Subjects independently agree in reporting that this resulting condition is pleasant and restful. If persistent, it becomes the most restful form of natural sleep. No university subject and no patient has ever considered it a suggested or hypnoidal or trance state or anything but a perfectly natural condition. It is only the person who has read a description without witnessing the actual procedure who might question this point.’94 Instruments, Popular Successes, and Biomedical Failures Like Kleitman, Jacobson tied sleep to the subjective experience of relaxation. Their respective methodologies were almost perfectly symmetrical: Kleitman’s subjects did everything they could to prevent relaxation and extend wakefulness, while Jacobson’s patients spent endless hours trying to relax. Neither was able to successfully incorporate their work into the neuropsychiatry unit at Chicago, however. In Grinker’s first report to the Rockefeller Foundation in October 1935, he depicted Jacobson’s efforts in a poor light: ‘We have had no real problem arise except one in which Doctor Edmund Jacobson became practically psychotic over a patient of his who entered our unit, and unfortunately exposed the weak status of his own medical work by doing so. I have no doubt that you have heard reverberations of this through his patients among the officers of the Foundation. I regret to say that he has utilized every possible means to hurt us but at the same time has exposed himself.’95 Given Grinker’s earlier comment that 90 per cent of the patients in the small ward had been diagnosed schizophrenic, it is certainly possible that Jacobson might have encountered some serious difficulties in teaching his patients his peculiar take on experimental introspection. But the fact that some of the officers of the Rockefeller Foundation were also his patients betrays the growing success of his private clinical practice, regardless of Grinker’s negative assessment. He treated a number of prominent and wealthy people over the course of his long career, including Mrs John D. Rockefeller, Jr, the Chicago meat packer Oscar Meyer, members of the Vanderbilt family, and even the Harvard physiologist Walter B. Cannon.96 The ‘weak status’ of Jacobson’s medical work notwithstanding, Simon Flexner, the director of the Rockefeller Institute for Medical Research, met Jacobson in New York in 1930. He was also pleased to hear later in the year that his nephew’s wife had come under Jacobson’s care, having suffered ‘a mild encephalitis.’ When Flex-

Performing Sleep 245

ner inquired as to her health, Jacobson cheerfully responded ‘she promises to continue to co-operate and if she does so, it would seem as if the outcome should be favorable.’97 Such successes eluded him at the University of Chicago. Like his onetime critic, Grinker, Jacobson left the university for good in 1936, in a dispute over the fees he was earning from his now highly successful private practice. Finding himself with numerous patients and no research facilities, he proceeded to build his own in downtown Chicago, calling it the Laboratory for Clinical Physiology. There, he continued his investigations until he closed his practice in the late 1970s. Jacobson’s Laboratory was the realization, on a somewhat smaller scale, of what the University of Chicago had apparently promised to build for him in the 1930s: a laboratory expressly designed for precise electrophysiological measurement, complete with several rooms shielded from electrical interference by thin sheets of copper, exceptionally sensitive galvanometers, and a central control room where the experimenter could monitor numerous electrical recordings taken from several subjects simultaneously.98 Although Jacobson indicated that he incorporated graphical recordings into his technique of progressive relaxation, such instruments do not appear to have been an integral aspect of his practice at the outset. Jacobson’s method was rather an amalgam of experimental introspection and therapeutic technique. The central focus of progressive relaxation was the training of the experimental subject or patient (the two categories were equivalent for Jacobson). His papers thus retained the somewhat outdated style once demanded by experimental introspection: verbose descriptions of the experimenter’s commands and the subject’s self-observations, followed by an assessment of the subject’s progress in learning the technique and the experimenter’s description of the subject’s transformed behaviour, and finally buttressed by some physiological measurements (blood pressure, pulse or respiratory rate, and the like). But graphical traces played little role here and were not reproduced in his early articles. Shortly after Jacobson became a research associate under Carlson, however, graphical evidence took on a much more significant role. In 1926 he wrote Alexander Forbes, a neurophysiologist at Harvard, inquiring about the design of some of his instruments. He wanted to build a device sensitive enough to demonstrate the presence of the least amount of current: ‘I have been wondering whether your apparatus described in July 1920 would meet my purposes. I wish to try to amplify action currents from extremely slight movements of the (human) eyes and

246

The Sleep of Others

some larger muscles. I presume that the condenser method with a high resistance string would be appropriate. I presume also that a camera such as you describe would be required. I have been considering the purchase of a large Hindle galvanometer. Any suggestions or advice from you will be appreciated, and I should like to have any of your reprints that may be available.’99 Forbes responded a few days later, sending Jacobson reprints of his paper, as well as some new circuit diagrams that had improved on his original work from 1920.100 The following year, Jacobson published the first of a series of papers that indicated his new devotion to electrophysiology. He wanted to use the new generation of vacuum-tube amplifiers to drive extremely sensitive instruments that could indicate the presence of very small action currents associated with mental activity.101 Borrowing seasoned introspectionists from Carr’s Psychology Department, Jacobson asked his subjects to relax and then made them perform various tasks involving association or reflection. They found that relaxation and mental activity were incompatible. Their eye and facial muscles, which had been photographed while the subjects relaxed, lost their flaccid appearance when their minds became engaged. ‘We are reminded,’ said Jacobson, ‘of the assertion of Hughlings Jackson that a motor element is involved in every conscious activity.’102 To demonstrate the parallel activity of mind and body, Jacobson turned to the graphical method. Using a string galvanometer connected to a vacuum-tube amplifier, he recorded the action currents of various muscle groups in subjects adept at progressive relaxation while performing various mental tasks.103 Subjects were asked, for example, to imagine or to remember bending their left arm, without actually moving it. The string in the galvanometer, calibrated to deflect one centimetre per millivolt, would move while the winding roll of bromide paper in the camera dutifully recorded the event. Jacobson published his results in a series of seven articles in the American Journal of Physiology between 1930 and 1931, followed by a summary that appeared in American Journal of Psychology in 1932.104 Such experiments demanded precise and sophisticated instrumentation. Although Jacobson claimed that he had considered using the string galvanometer for such work as early as 1912, it was not until the invention and mass manufacture of vacuum-tube amplifiers that deflections as small as a microvolt could be measured. Jacobson clearly revelled in the technical details of his new device, and he included in his articles a schematic diagram of the amplifier and calibration potential of his instrument, which had been custom-built for him

Performing Sleep 247

by Bell Laboratories.105 He also recounted his sources of error and frustration in considerable detail. The Edison storage battery would lose its charge. His carefully designed electrodes would become polarized, sometimes because of temperature difference between them, causing the string to fluctuate. Electrical motors in the Hull Biological Laboratories (which had naturally been outfitted with an elevator) would start and stop, thus disturbing the string. An AC light circuit had to be removed and replaced by a direct current system. Even after shielding the room, the subject’s couch, and all the apparatus with metallic screens and sheets of galvanized iron, Jacobson still found that ‘the experimenter needed to remain as quiet as possible while the record was being taken,’ in order to prevent an accidental inscription or, worse, a broken string. Subjects were also a potential source of error. If they moved when the circuit was closed and the specially designed shunt had been set to provide the greatest sensitivity, the string might snap. Emotional reactions also created problems. They could create polarizing effects in the skin (the galvanic skin response, or GSR, a mainstay of lie detectors) and distort the string readings. This issue had come to preoccupy a number of physiologists conducting animal experiments in the early twentieth century, who began to incorporate such responses into their theory and practice.106 Jacobson was no exception to this trend, and he turned to the value of his method of relaxation as a solution. His subjects apparently needed to be as finely tuned as the apparatus, although Jacobson stopped just short of saying that they had to have been trained by his own method: ‘In order to study electrical effects arising from peripheral nerves and muscles during a particular type of mental activity, it is necessary to have a state of mental rest in the preceding and succeeding moments. This can be accomplished in persons trained by the method of progressive relaxation and is found in some persons who have no such training. An individual, with or without such training, who persisted in unabated mental activity throughout the period would obviously be unsuitable as a subject.’107 Experimental subjects, hypertensive and insomniac patients, and graphical apparatus came to share a similar status in Jacobson’s quest for ever-greater precision. The difference between subjects and patients was merely quantitative, since even those who suffered no overt health problems did not know how to relax properly. The differences between instruments and humans were, in many ways, equally irrelevant: humans, like machines, were detection devices, capable of improve-

248

The Sleep of Others

ment, error, and failure. But what, exactly, did they detect? Relaxation, Jacobson repeatedly noted, was an imprecise term. Movement, however, was not; and movement could be detected equally well with machines, properly built, as with humans, properly trained. Although Jacobson claimed, in 1925, not to be testing a ‘motor theory of consciousness,’ it was clear that his experimental efforts can be organized under this rubric. The motor theory of consciousness was an attempt to resolve disputes over the epistemological value of introspection and mental imagery by calibrating it to the objective observation of behaviour as movement. It held that all mental images were accompanied by a movement in the corresponding sensorimotor system. A mental image of running, for example, would inevitably include slight movements of the legs. The classic statement of the theory had been produced in 1916 by Margaret Floy Washburn (1871–1939), a student of Titchener’s.108 With the advent of tube amplification and a corresponding increase in the ability to detect small bioelectric potentials, Jacobson re-engineered the theory to suit the new electronic age. Every task Jacobson examined – imagining, abstract thinking, or remembering – was accompanied by a graphic trace taken from some part of the body, indicating that all these mental activities involved the appropriate sensory organ or organs – in particular, the throat (for ‘inner speech’), the eyes, and the skeletal musculature. His work implied that the internal life of the mind was nothing more than the inhibition of the movements of the appendages and speech organs. This was a twentieth-century extension of James’s theory of emotion into the realm of thought. Whereas James argued that emotion could not exist without proprioception, Jacobson’s work demonstrated that thought could not occur without an accompanying action current in the muscles. If visible movement itself was not essential to mental activity, as behaviourists argued, Jacobson’s ‘graphical microscope,’ when hooked up to an appropriate subject, proved that at least the electrophysiological essence of movement – the action current – accompanied every mental activity. He dubbed his device the ‘integrating neurovoltmeter’ because it demonstrated the close alliance between the musculature and the brain. Always working on the margins of psychological and physiological science, Jacobson published his findings at exactly the time that the Jamesian theory of emotions was beginning to unravel. In 1927 Walter B. Cannon argued that emotions were impulses that originated in subcortical centres of the brain – in particular, the thalamus – rather than from sensory impulses from the viscera that were registered

Performing Sleep 249

in the cortex.109 Cannon’s theory thus turned the emotions into a question about the relationship between various parts of the brain, rather than the ‘mind and muscles’ concept that Jacobson relied upon. But the link between the musculature and self-identity had a long history, particularly in the West,110 and Jacobson successfully capitalized upon this popularity as he moved further and further from academic science during the 1930s. You Can Sleep Well, 1938 By the end of the 1920s, progressive relaxation had become the basis of a successful clinical practice, pursued by Jacobson in both Chicago and in New York City.111 But, while he had written Progressive Relaxation (1929) expressly for clinicians dealing with hypertensive patients, his efforts during the 1930s were directed towards a popular audience. In 1934 he published You Must Relax, an enormously successful introduction to his technique for lay people. The book went through numerous editions over the next forty years and was almost certainly aided by Jacobson’s popular radio show that ran through the 1930s.112 Shortly after leaving the University of Chicago, Jacobson published a popular book on sleep, entitled You Can Sleep Well: The ABC’s of Restful Sleep for the Average Person. The book was written from the perspective of a friendly and paternalistic family doctor advising the quintessential 1930s American Everyman: a hard-working salesman with a wife and children, struggling to make ends meet in a tough economy, surrounded by disease and misfortune on all sides, and suffering from a chronic insomnia. Jacobson was careful to outline the laboratory origins of his relaxation technique and to link the ideas expressed in You Can Sleep Well with past and current physiological research (see figs 21 and 22).113 But this was Jacobson’s first publication after leaving the University of Chicago, and he did little to hide his disdain for academic science, which, he insisted, was becoming increasingly detached from the concerns of ordinary people. He flattered his readers by encouraging them to believe that they had the democratic right to learn about the latest advances in medicine: ‘Who but the public must finally benefit from scientific work well done, or suffer from lack of it? On whose shoulders does the daily expense for this work ultimately rest in a democracy, where research depends upon many donors and upon taxes for that purpose? Shall we keep our facts to ourselves or, mindful of what has occurred in certain foreign lands when science has fallen before political outbursts, do what we can to

250

The Sleep of Others

21 To facilitate his patients’ mastery of ‘Progressive Relaxation,’ the rooms of Jacobson’s Laboratory for Clinical Physiology were designed to look like bedrooms. The rooms were entirely shielded in copper plate to prevent electronic interference. Everything was a trompe l’oeil : the plants were artificial, and the windows were false. (Edmund Jacobson, You Can Sleep Well: The ABC’s of Restful Sleep for the Average Person [New York: Whittlesey House, 1938], facing 188, used with permission)

22 The cupboard above the bed in the patient rooms of Jacobson’s laboratory contained the leads for his integrating neurovoltmeter. Note the lead for detecting action potential, which signified muscular tension in the eyes. (Edmund Jacobson, You Can Sleep Well: The ABC’s of Restful Sleep for the Average Person [New York: Whittlesey House, 1938], facing 189, used with permission)

Performing Sleep 251

share our secrets, so that a greater number will take interest in them and, deriving profit, may support and sustain our further scientific ventures? An increasing number of physicians today favor the latter course, because they believe that the public has a right to be educated.’114 To this end, Jacobson offered a detailed description and numerous photographs depicting the work done at his Laboratory for Clinical Physiology. The utility of the pictures was twofold: they helped the layperson recognize and understand progressive relaxation by depicting a subject practising its various stages, and they visually linked the practice with advanced precision technology. The graphical tracings described in the book had originally been used by Jacobson to verify that his relaxation technique actually did serve to inhibit reflex activity; now they were incorporated into his therapeutic technique, as he routinely showed them to his patients and archived them as a matter of course. Despite the fact that he no longer worked in a university laboratory, the scientific validity of his efforts was thus preserved for public consumption. The graphic method still held its charms, even though seven decades had passed since Angelo Mosso was so deeply moved by the inscriptions he saw in Carl Ludwig’s laboratory. Educating the public was a serious task for Jacobson, and, when his fictional ‘family doctor’ seized the opportunity to take his patient on a tour of his laboratory, the reader of You Can Sleep Well ended up with a brief survey of the history of sleep research, which included a reference to Kleitman’s famous Mammoth Cave experiments.115 Naturally, a popular work on sleep, published just as Freudian theories were beginning to take root in American psychiatric practice, had to say something about dreaming. Jacobson, however, stuck to his motor theory of consciousness and insisted that one could reliably estimate the beginning of a dream by observing the movements of the string galvanometer, which recorded the action currents in the jaw and the arms. When the record showed movement after being relatively quiet for a period of time, a dream was almost certainly taking place.116 An even easier way of determining the start and end points of a dream, however, could be had without the precision equipment used to measure action currents and, more recently, brainwaves: ‘When a person dreams,’ [said the family doctor to his patient], ‘he is tense in some locality. Most often his eyes are active. Watch the sleeper whose eyes move under his closed lids, but be very quiet as you do so. Awaken him in some unobstrusive [sic] way, such as clearing your throat, rustling a

252

The Sleep of Others

paper or whatever slight disturbance you need to create; but be sure not to startle him. After you have done this a few times, if he is willing to answer your questions, he will tell you whether or not he has been dreaming. Dreams are quickly forgotten as a rule, but if you and the sleeper go at the matter seriously, you are likely to find after awakening him that he has seen something in a dream. You are less likely to get a report of some visual picture if you watch him closely and arouse him at a time when his eyes have been completely relaxed or approximately so.’117

Dreams, argued the good doctor, sitting on the porch with his patient one warm summer’s eve, were little more than the product of a troubled mind. Progressive relaxation would probably not eliminate dreaming altogether, but it would certainly diminish their frequency and their vividness – something quite positive for the patient in question, who suffered from disturbing dreams almost every night. Worse, the patient had a passing familiarity with Freudian theory, which rendered his dreams even more distressing. Jacobson rejected psychoanalysis as mere suggestion; progressive relaxation, in contrast, was based on laboratory science. The evidence that dreams could be effectively eliminated or at least reduced lay in the fact that the doctor (that is to say, Jacobson) had successfully detected the complete rest of the eyes in subjects both awake and asleep. Peace of mind was possible, provided one worked hard enough at it. More important, progressive relaxation would provide its practitioner with the right tools to succeed in life in a way that mere psychoanalytic self-knowledge could not: ‘“If you and I assumed,” [said the doctor to his patient], “that you would feel better after having your dreams analyzed, I still would not be in favor of it, any more than I would be in favor of reducing the burdens on your nervous system by having you leave your work, and rest at home. You should learn to shoulder normal burdens and should not desire to be relieved of them. Relaxation aims to make you fitted for life in this sense.”’118 Jacobson effectively wrote dreaming out of his model for the laboratory-based study of sleep. His response to the triumph of the Freudian portrayal of dreaming as the staging ground for anxiety was to change the terms of the debate. Within Jacobson’s technological adaptation of the motor theory of consciousness, dreaming lost its status as a normal phenomenon, becoming instead a notable and near-ubiquitous instance of the failure of people to develop and exploit their introspective abilities. Given such an outlook, it was impossible for Jacobson to articulate the rhythmic dynamics of normal sleep, because his therapeutic

Performing Sleep 253

techniques were expressly designed to redefine supposedly normal experiences like dreaming as instances of residual tension that were potentially pathological. Thus, even though Jacobson was well equipped to discover rapid eye movements in sleep, and had even gone some distance to describing them, they remained utterly outside his cognition. His mission was to perfect the control of sleep by intervention, not to establish its normal parameters. Despite their common reliance upon graphical technologies, Kleitman’s vision of sleep was remarkably different from Jacobson’s. Kleitman approached sleep as both a naturalist and an experimentalist. On the one hand, he built devices like the motility recorder to be portable, in order to capture rhythmic phenomena in their natural places, be they an infant’s crib or a cot in Mammoth Cave. On the other, his work in sleep deprivation relied upon a quite dramatic form of intervention, in which sleep was artificially interrupted to determine its normal function. In either case, sleep was something social, in the sense that it was performed and sustained by populations and its change could ultimately be explained in terms of benefits for a group or species, not for an individual. Jacobson, on the other hand, used similar technologies to perfect individual performances. His experimental efforts were invariably directed towards transforming the self through harnessing the power of the will, not locating the regulation and control of autonomic functions by the brain. The functionalism espoused by Kleitman in his evolutionary hypothesis of sleep and the pragmatism of Jacobson’s arguments for the utility of physiological knowledge came together in their common emphasis on the role of muscular relaxation in sleep onset. The sensation of the activity of the voluntary muscles, or proprioception, was the very essence of will, a notion that had been well propagated in the United States through the writings of James. Questions about the will had been carefully elided in Piéron’s work on sleep, and Kleitman’s adaptation of the method of enforced wakefulness for use with human subjects did not include the will’s return as an active force. Kleitman noted the role of a sense of relaxation in sleep, but he refrained from building an entire empirical structure around this passive observation. In contrast, there were no passive observers in Jacobson’s scheme: if you practised progressive relaxation, you were experimenting. Had Jacobson’s work been adopted as the template for sleep research, the field would have taken a direction very different from the one it eventually did. But he moved almost entirely outside orthodox scientific

254

The Sleep of Others

research, particularly after he left the University of Chicago. At precisely the same time – the mid-1930s – the study of brainwaves through electroencephalography promised to bring together mind and brain in a definitive and novel way. It was this instrument, more than any other, that ultimately created the modern sleep laboratory as a place where sleep was observed and measured, rather than performed. The advent of EEG framed the phenomena of sleep in new ways that eventually gave way to the re-entry of the very observation that Jacobson had once derided: the movement of the eyes during sleep.

6 Sleep Finds a Groove

The discovery that bursts of rapid eye movements followed a predictable rhythm throughout the night transformed the scientific study of sleep. Reluctant to maintain the divide between research on sleep and the study of dreams, investigators during the 1950s and 1960s strove to reunite the two around the phenomena of REM. In the process, they created the modern sleep laboratory, with its array of graphical recording techniques, its emphasis on observing human subjects, and its intimate links to clinical medicine. Instruments made possible this reunion between sleep and dreams. Recording the brain’s electrical activity by electroencephalography helped frame the question of sleep as a process regulated by the brain. The very idea that such an activity could be characterized in terms of a distinct and discrete series of rhythms or ‘sleep stages’ implied that sleep was an active process. Given this complex periodic structure, sleep had to be more than the mere absence of wakefulness. The existence of sleep stages suggested that sleep served a variety of hitherto unknown functions that could be elucidated only by experimental research. It also treated the question of whether or not dreaming had its own rhythm as one with a realizable answer. When eye movements were observed by Chicago researchers for the second time during the early 1950s, they were better prepared to turn this into a meaningful fact than had been the case fifteen years earlier.

The Shifting Ground of Observation in Sleep Research Neither Nathaniel Kleitman nor Edmund Jacobson can be given exclusive credit for discovering rapid eye movements. Despite their mutual interest in the nature of sleep and their familiarity with the latest biophysical technologies, Kleitman and Jacobson both turned a blind eye

256

The Sleep of Others

to the prospect that dreams might have a rigid regularity about them. This was not a real question for them, because, as we saw in chapter 5, the idea that sleep was best represented as an experience remained at the core of their investigational projects. This is not to deny that animal experimentation was inconsequential for Kleitman, or to dismiss as irrelevant the emphasis Jacobson put on observing and training distinct populations of patients, such as hypertensives. By virtue of such strategies, Kleitman moved further from Pavlovian methods and Jacobson drifted away from classical introspection. But, in spite of these techniques of objectification, both investigators continued to depict sleep in terms of subjective experience. In particular, they relied upon the characterization of sleep as the product of relaxation, which was itself a sensation of one’s own musculature. Internal sensations, be they a ‘craving’ for sleep after a hundred hours of wakefulness or an incessant feeling of tension that prolonged insomnia, remained integral to their discussions of sleep. The array of instruments and techniques they developed to assay sleep certainly helped depict sleep in more positive and active terms. How could it be otherwise? The relentless accumulation of facts about sleep could hardly serve to make sleep appear more passive. Facts required some theoretical account of their existence, and such theory inevitably put sleep in an active light, as both a goal to be achieved and as an agent creating effects. But neither Kleitman nor Jacobson overcame their tendency to script their physiologies of sleep around a firstperson perspective, as the progressive elimination of wakefulness to the detriment of efficient productivity. Sleep needed to be controlled more than it needed to be understood. This passive depiction of sleep left dreaming as unworthy of study and ultimately extended the view that sleep was as it was experienced – as nothing. No amount of eye movements observed in the night could change this. The arrival of electroencephalography in the mid-1930s, however, proved itself capable of transforming the study of sleep. EEG (the abbreviation simultaneously refers to the machine, its inscriptions, and its practive, demonstrating the close relationship between the three) offered a source of information on brain states that could be correlated to an experimental subject’s introspective reports and behavioural observations. Sleep stages, the description of different brain states within sleep, revealed an internal rhythm of sleep that was unrelated to sensation and response. Like other facts about sleep, the existence of sleep stages demonstrated that sleep was an active process. What was different in the case of sleep stages was that they were direct records of an intact brain’s orga-

Sleep Finds a Groove 257

nizational activity. They circumvented the need to obtain such information second-hand, through bodily excretions or introspective accounts sampled over time. Thus, the EEG fuelled the depiction of sleep as an active process by providing an objective, automated, and continuous visual trace of the brain’s organization in time. Sleep no longer was as it was experienced; sleep was as it had been inscribed by a machine. This transformation does not, as the title of an important paper on the history of sleep research would suggest, raise the problem of charting the changing definitions of sleep.1 As one French physician adeptly noted nearly two centuries ago, ‘it is no less difficult to give a good definition of sleep than it is to provide a proper definition of life.’2 The transformation rather involves the problem of understanding how particular configurations of instruments, methodologies, and institutions come to have the power to create definitions in the first place. The EEG’s authority to define the nature of sleep appeared only within the context of a re-evaluation of the nature of organic life itself that started to take shape in the United States during the 1930s, financed by philanthropic organizations like the Rockefeller and Josiah Macy, Jr foundations. Even as their directors were pulling the rug out from under Kleitman (as we saw in the previous chapter), agencies like the Rockefeller Foundation were financing work that applied the EEG to problems as diverse as epilepsy and psychological personality. Whereas Kleitman had constructed his instruments around a limited set of classical problems of sleep, American scientific research was then being reordered around instruments and their capacities. It is to such instruments that we must now turn. EEG and the Reorganization of American Science The first English-language review of electroencephalography indicated clearly the remarkable new directions brain researchers were pursuing in the late 1930s.3 The review was written by Herbert Jasper, who was then working on EEG as part of a joint effort between the departments of psychology and psychiatry at Brown University (1932–8), shortly before he went to the Montreal Neurological Institute to work with the neurosurgeon Wilder Penfield.4 Jasper had spent three years studying neurophysiology under Lapicque and Alfred Fessard in Paris and was excited about the EEG’s potential for overcoming the narrow strictures of behaviourist psychology. The crucial link to be made, he argued, was between observed behaviour and the ‘rhythmic electrical impulses’ that arose from ‘the gray matter within the central nervous system.’

258

The Sleep of Others

Although the physiological origin of brainwaves was as yet unclear, their graphic depiction through EEG would open up the black box of the brain by visualizing its activity as it happened. Neurophysiologists could seemingly transcend the current practice of inferring the role of brain structures by observing behaviour before and after surgical intervention. What they could now perceive, thought Jasper, was the core organizational activity of the brain: its ‘centrally maintained activity which “selects” those afferent stimuli to which it “will” respond.’5 The neurological organization characterized by the dynamics of the EEG would ultimately demonstrate that the cortex did more than facilitate or mediate sensation and response. Its internally regulated state determined or defined what, in fact, could count as sensation and response.6 Jasper’s emphasis on the brain as a site of ‘centrally maintained activity’ could have been written by any number of enthusiasts of electroencephalography in the late 1930s. But the idea that there had to be a centralized system that supported the ability to ‘select’ from incoming stimuli and enable a ‘willed’ response was more than mere description. It was a metaphor that described the structure of many aspects of American life during the interwar period. A shared passion for articulating the role of centralized systems among scientists and the organizations that support their research has been noted in a historical analysis of the ‘protein paradigm’ that dominated experimental biology before 1953.7 In both this case, and in the rise of EEG, such research can be traced back to the Rockefeller Foundation’s original mandate to investigate and orchestrate forms of social control. This was a period of considerable transition in American life, as Progressive-era precepts regarding the value of scientific solutions to social problems were melding with a recognition that giant industrial interests provided the wherewithal to effect the ‘Americanization of the world.’ President Theodore Roosevelt’s 1902 revival of the discredited Sherman Antitrust Act had started the process of forcing the reorganization of some of America’s largest corporations, including, among others, the Northern Securities Company, E.I. Dupont, and American Tobacco. The results of Roosevelt’s efforts, which included the dissolution of Standard Oil in 1911 and the subsequent creation of the Rockefeller Foundation two years later, made the enormous economic and political power of corporations more palatable to the sensibilities of the American republic. The Rockefeller Foundation was an example of this new hybrid of policy-driven scientific research, industry, and philanthropy. Its officers ‘regarded medicine, education, and public health as part of a larger

Sleep Finds a Groove 259

process of enculturation leading to social control and economic stability.’8 To this end, it promoted a new kind of eugenics, one based less on supposed racial difference than on the industrial fitness of the American population – an idea that, as we have seen in chapter 5, resonated in studies of sleep. By 1920, this concept had been taken up under the rubric of ‘social control,’ which promoted the applied knowledge of the social and behavioural sciences in the organization of all aspects of American life. The end of the decade witnessed the emergence of the largest philanthropic merger in history: in 1929 the Laura Spelman Memorial Fund and the Rockefeller Foundation were combined to create a $260 million behemoth. Herbert Hoover, who would soon mobilize science and industry in the creation of vast engineering projects, had just been elected president the previous year. Four years later, in the grips of the Depression, Hoover was replaced by Franklin Delano Roosevelt. Roosevelt’s ‘New Deal’ represented the American response not only to the Depression but to the rise of the centrally planned economy in Soviet Russia through Joseph Stalin’s ‘Five-Year Plans,’ the first of which had begun in 1929. In contrast to Stalin’s violent expropriation of goods and property from those who refused to participate in the Great Soviet Experiment, Roosevelt asked for, and received, vast executive powers to implement a series of institutional controls encouraging the cooperation of business, labour, and government.9 The great era of bureaucratic centralization had begun. The Rockefeller Foundation’s program to orchestrate American scientific progress in the life sciences and, in particular, its work with the EEG must be read in this context. Indeed, one historian has noted that the era of an ‘activist system of sponsorship’ at the Rockefeller began only in 1932, with the appointment of Warren Weaver as director.10 Weaver was a new breed of philanthropic manager: he had a singularly deep scientific expertise as an applied mathematician, yet he was encouraged to define a program of sponsorship that was focused in its intent and all-embracing in its reach. Working under the shadow of his old mentor and colleague, Max Mason, who then headed the foundation’s Division of Natural Sciences, Weaver turned towards biology as a focal point for his efforts. He initially relied heavily upon Gregg, a neurophysiologist by training, who had been named director of the Medical Sciences Division in 1930. As we have seen in chapter 5, Gregg was dedicated to merging psychiatry and psychology with the biological sciences. Weaver borrowed Gregg’s idea of ‘psychobiology,’ but quickly extended it to encompass the physical and chemical sciences as well. Weaver

260

The Sleep of Others

called this fusion the study of ‘vital processes,’ but it was, in essence, a managerial, rather than conceptual, approach to scientific research. ‘Vital processes’ referred less to what was being studied than to how these investigations were conducted. Under this standard, Weaver attempted to bring the work of all natural-science disciplines to bear on a single problem, be it the study of the effects of radiation on growth and development or the spectral analysis of organic compounds. Typically, this involved formulating problems around the construction and performance of a particular kind of instrument. In comparison with the ‘grants-in-aid’ concept, which allowed a large degree of freedom to the individual researcher, the team-based research that Weaver supported obliged scientists to formulate problems around the data that their ultracentrifuges, electron microscopes, spectroscopes, and cyclotrons produced.11 This was an eminently pragmatic strategy, one that provided enough managerial control to guide research without stifling it and that avoided funding interdisciplinary feuds between biologists and physicists over the right to study the ‘vital processes.’ It also allowed Weaver to encourage research that elided, or completely ignored, ‘biophilosophical’ or metaphysical debates that might have served to structure the knowledge of a given field. The EEG was rapidly incorporated into American medical science along these same organizational lines. Although the instrument was initially developed by others working outside the Rockefeller, Weaver’s patronage was integral in getting the EEG into the mainstream of biomedical research. But the EEG did not thrive simply because of Rockefeller support. Indeed, some instruments that were backed by massive support from the foundation, such as the spectroscope (used in the analysis of biological materials), simply flopped.12 One of the reasons the EEG succeeded where the spectroscope failed was because the former was set within the context of the American hospital, which was being rapidly restructured around technological innovation between 1900 and 1925.13 Whereas the hospital had once focused on the housing and care of patients, by 1925, the bulk of hospital resources were directed towards the efficient diagnosis of disease. The focal point of this transformation were assay technologies. X-rays and blood exams, for example, were originally introduced as aids to clinical judgment regarding the nature of a fracture or to determine whether a patient had syphilis. But, by the 1920s, the use of such tests had come to constitute medical judgment itself. To practise scientific medicine now meant more than being trained in the basic sciences. It meant using instru-

Sleep Finds a Groove 261

ments and assays both to diagnose and to study disease. In the American context, scientific medicine also invoked specific kinds of managerial practices that made both new and old technologies routine standards of practice. The X-ray, for instance, did not become a routine part of medical care simply because it revealed the truth of whether or not a bone was fractured. If that were so, this technology, which was relatively inexpensive and fairly simple to operate, would have been immediately adopted in hospitals when it was first introduced. Instead, as one historical study has revealed, leading American hospitals did not make the Xray a routine part of diagnosis until after the First World War.14 In order for X-rays to become part of medical practice, diverse bureaucratic structures, including fee arrangements, a specialized profession of radiology, and standardized techniques for making and reading an image, needed to be in place. This happened only when hospitals began to be managed according to the dictates of industry. As middle-class patients started to make up more and more of the hospital population, hospitals increased in size. Their finances became more complicated as they were obliged to attract paying patients through the use of new technologies, such as X-rays and the electrocardiogram (ECG). The management of such institutions became scientific as well and they typically embraced Taylorist ideas about how to achieve more efficient production. Cost accounting was adopted to determine which hospital services were the most profitable. Standardized records, filled out by specialists and subspecialists frequently linked to the operations of a specific diagnostic test, became an integral part of medical practice: rubber stamps depicting a generic human thorax, for example, replaced hand-drawn images. Individual, handwritten comments were replaced by typewritten reports and Hollerith punch cards to better accommodate the ever-growing number of medical personnel serving the individual patient. Health care became a truly social phenomenon as staff and machines united to communicate consistent information about a disease that struck a patient, rather than describing a patient afflicted with an illness. When electroencephalography first appeared in North America in 1934, this scientific and bureaucratic transformation was nearly complete. The new research-driven hospital was set to welcome any technology that could produce a record capable of being efficiently processed. In the case of EEG, room had been prepared by the advent of the electrocardiogram, a detailed description of which had been provided by Willem Einthoven in 1902, just a few years after Wilhelm Röntgen announced his discovery of the X-ray.15 The ECG seemed to offer great

262

The Sleep of Others

promise for the diagnosis of myocardial infarction (heart attack). Yet hospitals were slow to adopt the technology: the New York Hospital, for example, purchased an ECG machine in 1914, while the Pennsylvania Hospital in Philadelphia did not buy one until 1921.16 In the latter case, the widespread application of the ECG followed the bureaucratic trail blazed by the use of X-ray machines, which brought not only technological sophistication but a system of fee-retainment by specialists, a standardized means of reporting tests, and a corresponding ideology of efficiency that emphasized the mechanical production of precise, quantitative measures of vital processes. Electroencephalographic technology was almost identical to that employed in ECG. But, in the 1930s, there was an additional reason for hospitals to welcome the introduction of a machine that seemed capable of providing reliable, standardized information about the mind: asylum reform. The 1930s saw a massive reorganization of the asylum that was patterned upon the reforms that had already been instigated in hospitals around 1900. Reformers were trying to bring American psychiatry into mainstream medicine during the 1930s. Historians of psychiatry have long noted that these reform efforts brought with them a number of new organic therapies, including insulin coma, metrazol shock, electroshock, and lobotomy, and discussion has generally focused on whether or not such therapies actually helped patients.17 The question of how technology affected psychiatric diagnosis during this period has, until recently, gone largely unnoticed. A long-standing exception to this rule has been the comparatively extensive study of the role of EEG in the diagnosis of epilepsy, an object of historical attention for more than fifty years.18 The discovery, made during the mid-1930s, that a three-persecond ‘spike and wave’ EEG pattern was a characteristic sign of epilepsy confirmed the disease’s status as a brain pathology, rather than a mental disease. This was heralded as an accomplishment of scientific medicine applied to psychiatry, insofar as it demonstrated how the application of standardized, instrumentally based tests could sort the mad, who were subject to incarceration in asylums, from the neurologically defective, who required medical treatment. The EEG could help manage the diagnosis and treatment of patients in an efficient and scientific manner that could easily exploit existing administrative structures and medical practices. It was on this basis that funding agencies began to support EEG research in several centres almost immediately after the technology first appeared in the United States in 1934. Stanley Cobb and William

Sleep Finds a Groove 263

Lennox at Harvard received Rockefeller Foundation funding, as did Frederic A. Gibbs at the University of Illinois, and W. Grey Walter at the Burden Neurological Institute in England,19 while the Josiah Macy, Jr Foundation supported Hallowell Davis’s research at Harvard. But this funding was linked to the capacities of the machine, not to its application to the diagnosis of epilepsy. The focus on this one central triumph of EEG, then, serves to obscure the initial reception of the instrument, which was both diverse and expansive. Many researchers, for example, assumed that the EEG would uncover an enormous pool of latent and hitherto unnoticed ‘epileptoid’ conditions that required treatment and control.20 Others aspired to apply the instrument in an extraordinarily range of situations, such as the determination of personality type or the potential compatibility of romantic partners.21 This extraordinary optimism regarding the potential applications of EEG proved central to the creation of modern sleep research. Yet the history of EEG has not been subject to such scrutiny, even in sociological accounts of the field.22 As one sociologist has recently noted, EEG’s success can in many ways be credited to its ability – highly prized in twentieth-century science – to rapidly generate novel data in existing fields of inquiry.23 Modern sleep research evolved around this instrument through a similar dynamic. The funding that EEG researchers received from the Rockefeller Foundation and other agencies ultimately helped entrench sleep research in the biomedical laboratory. With the discovery of sleep stages and the routinization of all-night electronic surveillance of sleeping subjects’ brains, periodic eye movements during sleep and their link to dreaming achieved a new importance. Data about dreaming and its ties to psychopathology could be now be rapidly and reliably generated by an instrument. Despite continued ignorance as to how, exactly, the brain generated the EEG trace, the instrument held out the potential for the wide-scale rationalization of psychiatric diagnosis. Such a perspective was impossible so long as investigators like Jacobson or Kleitman ignored questions of brain organization and instead focused on the relationship between sleep and the surveillance, by mind or machine, of peripheral movement in the musculature. Alpha ‘Electroencephalogram’ refers to a graphical trace that reflects the voltage change taking place between two electrodes. The trace can appear in a variety of guises: on a kymographic drum, a sheet of paper, photo-

264

The Sleep of Others

graphic film, a cathode ray oscilloscope, or (more recently) a computer screen. Likewise, the electrodes can assume a variety of forms, from extremely fine platinum-iridium needles piercing the scalp to metalsurface electrodes ten or twelve centimetres in diameter. The electrical arrangement of the electrodes (mono- or bi-polar) and their positioning on the scalp are further variables that electroencephalographers have had to deal with in the past. The most important phenomenon of early EEG work was the ‘alpha rhythm.’ This is a regular, sinusoidal waveform of about ten waves per second that appears in most (but not all) subjects when they are in a state of relaxed wakefulness. The neuroanatomical origins of this ‘synchronized’ rhythm are still unknown, and its central importance has been as a demonstration that the brain is itself a source of electrical energy, regularly discharging accumulated electrical reserves in an orderly manner and in the absence of external stimuli. When similar patterns were first recorded from the brains of sleeping subjects, they were considered in the same light as the alpha rhythm. Although their origins were obscure, their rhythmic pattern were taken as a sign that sleep followed an orderly course throughout the night – a course that could be charted only by the EEG. The discovery of the alpha rhythm and the discovery of the EEG are generally taken as synonymous, and credit is almost always given to a psychiatrist from Jena, Hans Berger. Although Berger (1873–1941) himself did not believe that all brain functions had specific locations in different regions in the brain, there is a close link between the doctrine of localization and the use of electricity to study the brain.24 The doctrine drew its inspiration from the announcement, in 1871, by Gustav Fritsch and Eduard Hitzig that the electrical stimulation of parts of the cerebral cortex produced specific motor reactions. Richard Caton, a professor of physiology at University College, Liverpool, pursued a similar line of research as he tried to map functional brain areas by detecting and measuring the electrical activity that resulted from a natural movement, rather than either stimulating the cortex directly or surgically removing a portion of the brain and observing the results. In the process, Caton observed what he described as ‘continuous spontaneous electrical activity’ on the surface of the dog brains he was studying with his reflecting galvanometer, which measured the slight changes of angle of a reflected beam of light when its source – a negatively charged mirror suspended within a coil of wire – was subjected to changes in voltage.25 But this hardly amounted to a discovery, since Caton himself took little interest

Sleep Finds a Groove 265

in the observation: it was not at all obvious to him how spontaneous currents in the brain could be linked to movement controlled by local cortical regions. The neurophysiologists who followed Caton in commenting on the phenomenon shared this viewpoint. Although they did some tentative work with the ‘spontaneous current,’ it always took a back seat to the real problem of brain localization. Vladimir PravdichNeminsky, a physiologist at Kiev, recorded the ‘spontaneous current’ in 1912 with an Einthoven string galvanometer connected to the exposed cortex of a dog. Although he gave the phenomenon a name (the ‘electrocerebrogram’), it served as a mere baseline that could be transformed by applying various stimuli. The rhythm itself remained of little interest. Hearts and Minds Hans Berger was unlike earlier investigators who studied the brain’s electrical activity in that he recorded Pravdich-Neminsky’s ‘electrocerebrogram’ in human beings and was also deeply influenced by the rhetoric of holism. His psychiatric training precluded an emphasis on electrophysiology, and he was not interested in extending the theory of brain localization. He instead depicted the ‘electroencephalogram’ (as he renamed Pravdich-Neminsky’s phenomenon) as a tool for investigating how psychological, rather than motor, phenomena were produced by the brain as a whole. Moving away from the experimentally derived reflex paradigm of his predecessors (discussed in chapter 2), Berger anticipated that the EEG could measure the brain’s consumption of energy and thereby quantify thought, and that it might also be useful in diagnosing mental illness. Berger had little luck in the first area but achieved some limited success in the second, particularly in the case of epilepsy. His introduction of the EEG into the field of human research helped to change the orientation of electroencephalography away from brain localization and towards the meaning of alpha as a clinical phenomenon. Berger pursued psychophysiological problems much as Mosso had four decades earlier: he strove to find a way of correlating psychological activity to physiological measurement. Consequently, he emphasized the importance of the precise characterization of states over the experimental measurement of induced reactions, which had been the dominant paradigm of the laboratory-based psychophysics and physiological psychology advocated by Wundt. But Berger, who graduated with an MD from the University of Jena and became assistant to Otto Binswanger at

266

The Sleep of Others

the university’s psychiatric clinic in 1897, was more of a clinical investigator than an experimentalist. His early research had been dedicated to uncovering the changes in brain circulation brought on by various drugs, and, to this end, he used a plethysmographic assemblage identical to the one Mosso had described in his book on fear. A precise analysis of such physiological changes might, Berger hoped, allow him to correlate the dissipation of energy in the brain during mental processes. These efforts ended in failure, as did his related studies of brain temperature, which he had begun in 1907. It was around this time, Berger claimed, that he decided to study the brain’s electrical activity.26 Berger’s first efforts along these lines studied nervous transmission by electrically stimulating the cortex and then measuring the delay until a motor reaction appeared.27 Uninterested in developing localizationist theory, Berger simply used existing maps of the motor cortex to determine which regions were suitable for his experiments. After the First World War, he slowly began to abandon this research in favour of the detection, measurement, and recording of the spontaneous current that seemed to originate from the cortex. As the chair of psychiatry at Jena, Berger had virtually unlimited access to a large group of patients with skull injuries incurred during the war, and, by 1924, he had entirely set aside his stimulation experiments in favour of studying alpha: ‘What mattered to me now,’ he wrote in his first report on the human EEG, ‘was the investigation of the current oscillations present at all times that can be recorded from the surface of the cerebral cortex.’28 Shortly after recording the first human EEG in 1924, which he detected in a seventeen-year-old patient named Zedel (who, like Mosso’s star patient in Fear, had suffered from a cranial fissure), Berger celebrated the prospect of finally finding a investigative path that suited his interests: ‘Cortical currents [he wrote in his diary] (circulation, temperature, electrical processes!) and the hope so beautifully expressed by Mosso, which I experience time and again when I apply precise measuring instruments to the brain. A type of work which agrees well with me and my whole – psychophysiological – attitude.’29 But his enthusiasm seems to have been muted by the concern that his traces might actually be artifacts. Berger waited five years to publish his results, at which point he went to great lengths to demonstrate that the spontaneous currents were not due to blood flow, muscle currents, eye movements, or interference from the electrocardiogram. In effect, Berger was calibrating his discovery, setting the spontaneous currents of

Sleep Finds a Groove 267

the brain up against better-established phenomena that were coming to play an ever-larger role in the scientific medicine of the early twentieth century. In a sense, Berger’s hopes rode on the earlier success of the electrocardiogram, which had been established as a diagnostic tool at the dawn of the First World War, following Einthoven’s invention of a string galvanometer. Einthoven’s galvanometer, which used a silvered quartz string rather than magnetized mirrors or columns of mercury to detect voltage fluctuations, was the first instrument to display the complicated waveform of the ECG.30 By around 1906, stripped-down, inexpensive versions of Einthoven’s device were available on the German market, an important factor in making the ECG a viable tool for clinical diagnosis. Not surprisingly, almost all of Berger’s observations of the EEG were made on instruments that had originally been developed for the ECG. In 1926 he acquired a Siemens double-coil galvanometer, which he used in most of his published research. Such instruments provided Berger with an electrical rhythm that was well known and readily displayed, as well as being one to which the EEG could be calibrated, thus creating a stable physiological backdrop against which he could present his new discovery. His efforts to show that the EEG was not influenced by the ECG introduced the idea that the brain, like the heart, kept its own electrical rhythm. Indeed, he visualized the phenomenon in precisely these terms by arranging his graphical evidence in conventional anatomic order, with the EEG (representing the brain) on top of the composite image, followed by the ECG in the middle and a time code below (a trace made by a fork that oscillated at ten hertz). The EEG was for the brain what the ECG was for the heart. Berger also borrowed his nomenclature from the ECG. Disliking the combination of Greek and Latin elements in Pravdich-Neminsky’s ‘electrocerebrogram,’ he proposed instead ‘in analogy to the name “electrocardiogram,” the name “electroencephalogram” for the curve which here for the first time was demonstrated by me in man.’31 Berger identified two distinct rhythms in the EEG. The first, dubbed ‘alpha,’ had a frequency of about ten per second, and ranged from seven to fifteen microvolts. The second ‘beta’ rhythm was much faster (about twenty per second) and had a lower voltage (around two or three microvolts).32 It was the former rhythm, however, that captured Berger’s imagination. Contrary to his expectations, alpha waves disappeared whenever his subjects responded to a stimulus, be it the sudden noise of a cap gun being fired or the presentation of a simple mathemat-

268

The Sleep of Others

ical problem. In such situations, only the minuscule beta waves remained. Echoing Mosso, who had argued that even the slightest stimulus brought blood rushing to the brain, Berger had anticipated observing a similar response in the EEG. Stimulus would increase the current, not annihilate an entire class of waves.33 To explain the phenomenon, Berger turned to the concept of inhibition, which was then being promulgated by, among others, Ivan Pavlov (discussed in chapter 4). The waxing and waning of the alpha rhythm in relaxed wakefulness, as well as its extinction with mental effort, was, Berger argued, due to a ‘narrowing of consciousness’ as the excitation caused by a stimulus provoked the radiation of inhibition over the surface of the cerebral cortex.34 Alpha was not, in itself, a sign of conscious activity. It was more like an obligatory stage that, when consciousness performed on it, suddenly slipped from view. Berger reworked this interpretation into his analysis of two states of unconsciousness: sleep and epilepsy. Both featured the ‘flattening’ of the alpha rhythm, which Berger described as a form of ‘disinhibition.’ As discussed in chapter 4, inhibition was a term that variously described a state of control, order, and self-restraint. ‘Disinhibition,’ then, described a state in which any active force that enabled the focusing of attention and the maintaining of orderly consciousness had been attenuated or removed. For Berger, the confused, florid images of dreams were evidence that attention, which he assumed was a product of an inhibitory force, was largely annihilated in sleep. Sleep and epilepsy were equally described as ‘a disconnection of the cerebral cortex’ resulting from ‘an intense excitation’ that was directly conducted to the subcortical centre of the thalamus, where Economo had located a regulatory centre that normally blocked incoming stimuli in sleep.35 For Berger, the only substantive difference between sleep and epilepsy was that the latter included an inhibition of respiration and a disturbance of cerebral circulation.36 From his new perspective of alpha, these states could now be more accurately characterized by their transformation of the EEG, which signified a break in the continuity of the brain’s normal inhibitory work. Berger’s research was holistic in the sense that it was derived primarily from clinical observations rather than laboratory-based techniques of intervention. Once he had discovered alpha rhythms, his interests lay more in using them as technologically mediated signs of normal and pathological states than in uncovering the precise origins or physiological mechanisms of the phenomenon. Like all disciples of the graphical

Sleep Finds a Groove 269

method examined in chapter 2, Berger used graphical instruments to accomplish this goal, which required subjects to be capable of reporting on their own mental states. He then correlated these reports (the analogue of symptoms) to brain states as a whole, rather than pursuing the classical emphasis on localizing muscular responses in one or another area of the cortex. He did, of course, frame his claims in terms of particular brain centres. But these were regulatory centres, whose activity could be characterized only when the organism was investigated as an integrated entity; they were an anatomical region entirely different from the sensory or motor points on the cortex that had been the stuff of localization theory for the last sixty years. The rhythms of the EEG, which seemed to inscribe the brain’s activity as a whole, created a new language through which such systems could be described. It has long been customary to claim that Berger’s work on EEG was ignored (although a recent cultural history of EEG has challenged this view).37 If this is accurate, it might have been because his techniques were caught between a clinical holism that emphasized patient-centred narratives and a reductionist metaphysics that framed disease exclusively in terms of mechanically generated signs. For all that, one would have expected Berger’s work to be attractive to German clinicians in particular. The technique upon which Berger relied the most, electrocardiology, was well entrenched in German medicine by the dawn of the First World War. Bad Nauheim, for example, boasted a centre for the treatment of heart patients that drew students from around the world.38 There was also a concentration in Germany of technical expertise in the construction of medical instruments – all of Berger’s instruments came from the German manufacturer Siemens. But perhaps electrocardiography was almost too well installed to suit the tastes of some of Berger’s fellow psychiatrists, one of whom later recalled that ‘medicine as it was practised in university clinics then [around 1914] awakened in me and in many other young physicians the greatest doubt. Rounds were becoming increasingly impersonal; the wards, the dominance of laboratory tests and [electrocardiogram] curves revolted me. There was never a one-on-one conversation between doctor and patient.’39 The author of this statement, Viktor von Weizsäcker, was only one of many psychiatrists who began to turn towards psychotherapy in the 1910s and 20s as a means of encouraging communication with their patients. Berger published in psychiatric journals almost without exception. For those who agreed with Weizsäcker, Berger’s discoveries might well have appeared as little more thean the continuation of the reduc-

270

The Sleep of Others

tionist trend of relying on the technological mediation of symptoms for diagnosis over the patient interview. Nor, perhaps, did Berger’s work agree well with a genetic approach or the racial psychiatry that later blossomed under the Nazis. Indeed, Berger was opposed to Naziism, and was eventually forced to retire because his work received no support under the new regime. In despair over his isolation and his failed career, he killed himself in 1941.40 Neurophysiology as a Technological Enterprise, 1900–35 Within the international scientific community, it was neurophysiologists, not psychiatrists, who first responded to Berger’s discovery in any systematic way. In the United States, the EEG offered neurophysiologists the possibility of extending their technologically driven research into the field of neuropsychiatry, which was undergoing a considerable transformation (discussed in chapter 5). Funding organizations like the Rockefeller Foundation underwrote the shift in American medicine towards an increasing reliance on diagnostic technologies, and the EEG soon came to symbolize how such approaches might transform psychiatry. The mandate of the Josiah Macy, Jr Foundation, on the other hand, was to bring a knowledge of ‘the whole patient’ to bear on the practice of modern medicine.41 The officers of the Macy Foundation placed particular importance on understanding the effects of the emotions upon body function and the process of healing. They supported early EEG work, because they hoped it would supply yet another clinically useful physiological index of emotion. As we have seen in the case of Mosso, there was no inherent contradiction between investigating emotional states and deploying graphical technologies. Holistic rhetoric frequently railed against the technological ‘reduction’ of the patient’s illness to a series of mechanically derived signs, but technological progress and holistic medical research could operate in tandem. In the United States, these two approaches were allied through EEG research. Neurophysiologists began adapting EEG to their work simultaneously in England and in United States. Edgar Adrian (1889–1977) was the first person outside Germany to pay attention to Berger’s work. Adrian, who shared the Nobel Prize in Physiology or Medicine with Charles Sherrington in 1932, was probably the most famous neurophysiologist of his day. He had worked under Keith Lucas at Cambridge and eventually made his career there. Lucas (1879–1916) was part of the first generation of British neurophysiologists to reject Sherrington’s macro-level

Sleep Finds a Groove 271

experimental approach in favour of an assemblage radically reduced in scale.42 Whereas Sherrington had studied spinal reflexes by stimulating the skin of a decerebrate animal and then measuring the subsequent muscular response, Lucas isolated the smallest possible nerve-muscle unit or ‘preparation,’ stimulated the nerve, and then measured the subsequent muscular movement. Between 1904 and 1908, Lucas conducted a series of experiments that suggested an ‘all-or-none’ principle applied both to muscle fibre and to the motor unit (the group of muscle fibres stimulated by a single axon of a motor nerve). ‘All-or-none’ referred to the idea that the nervous system operated in a binary mode: nerves either fired, or they did not. Functional nervous impulses came in discrete units, not as a continuum. Lucas’s hypothesis that all nervous response (motor or sensory) depended only on the number of nerve impulses per unit of time that were transmitted represented an ‘engineering approach’ to the problem of nerve action. Whereas Sherrington’s work had drawn its authority in part from the delicate surgical procedures that produced experimental animals that were functional despite their effective lack of a cortex, Lucas’s work built systems up from the basic components of cells and tissues and inferred functional integration from the results. This shift also reflected contemporary electronics engineering’s embrace of the ‘all or none’ transmission of electrical signals. Indeed, the ‘all or none’ was often described as the ‘cable theory’ of nervous transmission, in an allusion to the conduction of electrical signals in a telegraphic cable.43 With the development of vacuum-tube amplification by the early 1920s, the investigation of the nervous impulse at the micro-level was all but consolidated as the primary experimental technique of neurophysiological research. Adrian’s award-winning summary of his teacher’s research, which he published three years after starting work with Lucas, reflected the central role the study of the nervous impulse then played in neurophysiology.44 The outbreak of war interrupted Adrian’s career just as he was beginning his medical studies in London. Adrian was ambivalent about the potential of extending his research into the clinic, but it seemed a reasonable way to contribute to the war effort, and it improved his job prospects.45 After the war ended, Adrian was pleased to leave medicine for good. A paper on the electrical testing of muscles that he managed to publish in 1919, was, he complained to his collaborator, Alexander Forbes, based on only a few cases since he ‘didn’t get many peripheral nerve injuries at Aldershot [the British Army base where the Cambridge

272

The Sleep of Others

Military Hospital was located].’ He looked forward to returning to the laboratory. ‘I shall have to look around for some ideas,’ he told Forbes, ‘at present my mind is a complete blank as regards physiology proper.’46 Alexander Forbes (1882–1965) was a crucial audience for Adrian’s concerns. He had been a student of Walter Cannon’s at Harvard and worked with Sherrington at Liverpool during the academic year 1911– 12. On a side trip to Cambridge, Forbes had also managed to work with Adrian in Lucas’s laboratory for three weeks. Adrian and Forbes quickly developed a close relationship, and communicated regularly thereafter. Together, they represented a central conduit through which neurophysiological research would soon be transformed on two continents. The ‘physiology proper’ to which Adrian referred in his 1919 letter to Forbes was laboratory-based work on nervous conduction at the micro-level, far removed from the clinical problems of neurology. Sherrington had taken ‘all-or-none’ as a conceptual framework that could help guide the study of nervous integration and would generate appropriate answers as to how the nervous system produced the sensation of pain. It was widely taught, complained Sherrington to his former student, that pain was the excessive stimulation of ordinary sensory nerves. But the all-or-none hypothesis might lead to the discovery that specific nerves were dedicated to communicating the sensation of pain and were activated only when stimulation reached a certain threshold.47 Adrian and Forbes’s ‘engineering approach,’ however, tended to frame the all-or-none hypothesis in terms of its instrumental solutions. In 1919, for example, Adrian’s most pressing difficulty was his inability to represent graphically the activity of a single nerve fibre. The mirror galvanometer was extremely sensitive but slow. The capillary electrometer and the string galvanometer, on the other hand, had quick response times but were not sensitive enough. Progress in the field would come first from technological developments, not through the appropriate application of neurophysiological theory to clinical problems. Forbes and Adrian quickly recognized the vacuum tube as the key. ‘The valve idea for magnifying the electric response sounds an excellent idea,’ wrote Adrian to Forbes early in 1919. ‘If you don’t make it work,’ he joked, ‘we shall have to breed a new kind of frog with a large electric response.’48 In his Nobel Lecture, delivered some thirteen years later, Adrian continued to depict the past (as well as the future) of neurophysiology almost exclusively in terms of these sorts of technological developments: ‘The signals which they [the nerves] transmit can only be detected as changes of electrical potential, and these changes are very

Sleep Finds a Groove 273

small and of very brief duration. It is little wonder therefore that progress in this branch of physiology has always been governed by the progress of physical technique and that the advent of the triode valve amplifier has opened up new lines in this, as in so many other fields of research.’49 Lucas had apparently proposed that vacuum-tube amplification be used for neurophysiological research, but, following Lucas’s death in a flying accident in 1916, Adrian took considerable time in following up on his master’s suggestion.50 It was Forbes who first brought tube amplification to Cambridge, on Adrian’s invitation, in 1921. In the ten years since they had first met, Forbes had established his reputation as a talented physiologist in his own right. Like Lucas, Forbes had been raised in a family deeply involved in the communications revolution brought on by the introduction of the telephone and wireless telegraphy. Lucas’s father was a self-taught engineer involved in the construction and maintenance of submarine telegraph cables, while Forbes’s father was the president of Bell Telephone Company (his maternal grandfather was the American poet and philosopher Ralph Waldo Emerson). Forbes cut his teeth on physiological research under Cannon at Harvard and received his MD in 1910. Knowing his student’s penchant for physiological research over a medical career, Cannon advised him to go to Liverpool to study with Sherrington. Forbes returned from England in 1912 and began using the string galvanometer to combine the work of Sherrington and Lucas by recording the motor impulse in a spinal reflex, rather than measuring the movement of the muscle. Although Forbes’s early recordings were not of a particularly high calibre, his work was important because it effectively replaced muscular response with graphical inscriptions as the medium through which the reflex arc could be observed.51 Reverberations from Forbes’s early work continued throughout the 1930s and beyond and coincided with the tenure of his former colleague at Harvard, Alan Gregg, as director of the Medical Sciences Division at the Rockefeller Foundation (1930–50). In many ways, Gregg would perpetuate the technological vision of biomedical progress that had produced his and Forbes’s first joint publications in 1915. The U.S. entry into the war in 1917 proved equally influential to Forbes’s research. Forbes, whose devotion to the Allied cause was clearly influenced by the letters he had received from his British colleagues, was an outspoken proponent of America’s entry into the European conflict.52 He was also deeply involved in wartime research, beginning with his work on wireless apparatus for airplanes in the Harvard Physics

274

The Sleep of Others

Department in 1916. It was this project that introduced him to the three-electrode vacuum tube – the triode or ‘audion’ – which was also being developed for government and military purposes by Harold Arnold, who worked in Robert Millikan’s laboratory at the University of Chicago as well as with Bell Laboratories. The ‘thermionic valve’ was initially developed by John Ambrose Fleming in England, in an effort to create a wireless system that featured a visible, rather than audible, display mechanism. Fleming’s ‘valve’ was able to convert the oscillating signal of a wireless set to a direct current capable of driving a mirror galvanometer. In the United States, Lee de Forest created a similar device, dubbed the ‘Audion,’ or triode, which included a third plate – a grid inserted between the anode and the cathode that allowed for the control of current by changing the voltage between the grid and the cathode. A small change in the grid current would produce a much larger change in the anode current, thus making electronic amplification possible. The triode was quickly deployed as a telephone repeater and was also used in the first transcontinental wireless communication from New York to San Francisco in 1915. These developments marked the dawn of amateur radio, and those who could afford the equipment began receiving and sending their own wireless broadcasts. With the everincreasing participation of the United States in the European war, however, military applications began to take precedence. The triode’s ability to tune in to specified frequencies had potential applications for navigation, and in mid-1916 Forbes began working for the navy, developing ‘radio compasses’ that used rotating loop antenna to determine the direction of a broadcasted signal. The use of several, well-separated broadcast signals made it possible for sailors to determine their precise location at sea. By the time he was demobilized in early 1919, Forbes had become convinced that the audion could be a key component in the growth of neurophysiological knowledge. He immediately contacted Arnold, who had left the University of Chicago to work at Bell Laboratories, about the possibilities of using the triode to amplify action currents. Within a year, Forbes had co-authored a groundbreaking paper describing the application and potential use of vacuum-tube amplification in neurophysiological research.53 The paper coincided with the return of amateur broadcasting in the United States in 1919. By 1922, there were 253 broadcasting stations in the country, and sales of triodes, as ‘receiving tubes’ marketed through the Radio Corporation of America (RCA, which distributed the wares of Westinghouse and General

Sleep Finds a Groove 275

Electric), were going through the roof: in two years, tube sales had effectively increased by a factor of ten, for a total of 11.35 million units in 1924.54 The audion (and its imitators) was rapidly becoming an ubiquitous feature of American life. The records that accompanied Forbes’s 1920 paper illustrated the new degree of magnification that could be obtained with the triode. The action currents of frog nerves that had once been almost completely illegible at one millimetre in height were now visible as a clear spike of twenty-five millimetres or more.55 Adrian was clearly impressed, and, seeing an opportunity to chart out new regions in physiological research, invited Forbes to return to work with him in Cambridge.56 Adrian, who seems to have been unsure as to whether or not such things could be obtained at Cambridge, reminded Forbes to bring his own valves with him: ‘Do please bring over some valves; I believe they can be obtained over here but at any rate you know the habits of those you have used ... We are just getting a new C.S.I. [Cambridge Scientific Instruments] string galvanometer & the lab should be particularly good for valve amplification as our 100 V DC mains are supplied by accumulators ... However, I haven’t yet thought of any epoch making research. I should like to find out more about the reactions of sensory nerves e.g. the optic or a dorsal root of the chord, but they are all so horribly small!’57 The tide had clearly turned. Whereas Forbes had left Britain with a remarkable new instrument in 1912, he was now returning with evidence of American technological ingenuity in hand. And as Adrian’s inability to think of any ‘epoch making research’ clearly indicates, the technological capacities of Forbes’s amplifier were beginning to shape the course of neurophysiological research. Research programs were being devised to suit the instruments available, rather than the other way around. EEG Abandoned: Adrian and Matthews It was in this technologically charged era of progress in neurophysiology that the EEG was first taken up outside Berger’s clinic in Jena. Although Adrian has traditionally been given credit for promoting the use of EEG as an investigative tool, this process unfolded simultaneously on both sides of the Atlantic.58 Forbes’s story is the more crucial one for illustrating the history of sleep, but Adrian’s approach is also useful, in that it illustrates the direction the EEG might have taken outside an overtly

276

The Sleep of Others

clinical context. After winning the Nobel Prize in 1932, Adrian began searching for a new direction once again. He abandoned the finegrained detail of nervous transmission for a new challenge that was more in keeping with the holistic tenor of the times, and began research on the electrical activity of the cortex in 1933. He had come across spontaneous electrical activity in his study of non-mammalian brains in 1931 but had dismissed them as artifacts.59 Two years later, Adrian and a colleague, Brian Matthews, turned to studying the reactions of an entire layer of nerve cells in the cerebral cortex of rabbits. The existing literature, they found, focused almost exclusively on the responses of local topographical regions of cells and was largely unhelpful. But in the process, they came across Berger’s unusual series of papers. Within a few months, Adrian and Matthews had successfully recorded the alpha rhythm and had arranged for a live demonstration at an upcoming meeting of the British Physiological Society in Cambridge in 1934. They published two papers on the EEG that year, and Adrian planned a tour of the most important centres of neurophysiology in North America to propagate the technique. Backed by his formidable international reputation, the EEG would quickly become a mainstay of neurophysiological research. The import of the discovery seemed obvious. What, precisely, it signified, was not so clear. Adrian and Matthews gave Berger full credit for discovering the EEG, which they proposed to rename the ‘Berger rhythm,’ but were hesitant about his interpretation of it. It seemed unlikely, they argued, that the alpha rhythm represented the activity of the entire cerebral cortex in ‘relaxed wakefulness,’ as Berger suggested. The term was vaguely clinical and seemed to lack the spatial and metric precision required by modern neurophysiology (there is no indication that they knew anything of Jacobson’s work, discussed in chapter 5). Besides, the rhythm seemed strongest near the visual centre of the occipital cortex, and it could be induced, Adrian and Matthews found, by setting the ‘flicker rate’ of a light stimulus at ten hertz.60 In place of Berger’s vague phenomenology, Adrian and Matthews proposed the idea of ‘pattern vision.’ Alpha’s presence indicated that the organism had ceased to identify spatial patterns, a process governed by the occipital lobe. Their eponymous renaming of the EEG was in fact a rearguard attempt to defuse Berger’s holistic view that the entire cortex was the source of the rhythm. They also rejected Berger’s arguments about the EEG being a sign of generalized inhibition. Dispensing with Pavlovian theory, Adrian and Matthews argued that the rhythm represented the ‘negative rather than the positive side of cere-

Sleep Finds a Groove 277

bral activity, [since] it shows what happens in an area of cortex which has nothing to do.’61 Within two years, Adrian had entirely abandoned his work on the EEG. When taken in the context of his efforts as an experimentalist, this wholescale rejection is perhaps not surprising. An absence of activity was hardly an appropriate object for laboratory-based research (as the question of sleep had shown more generally). Perhaps more important, Adrian’s research had been primarily shaped by problems in nervous transmission. He was not interested in psychophysiology, where a noninvasive technology like the EEG was most at home, and he was inexperienced at handling human subjects. It was his student William Grey Walter (1910–77), who pursued electroencephalography with passion and focus in England for the next twenty years.62 Unlike Adrian, Walter followed a clinical trajectory that, in many ways, bore the ‘psychobiological’ imprimatur of the Rockefeller Foundation (see chapter 5), which sponsored Walter’s research from 1939 to 1955.63 At the outset of his work in 1935, Walter already had something Adrian did not: an almost unlimited supply of human subjects, which he encountered upon taking up residency at the Maudsley Hopsital that year. Henceforth, Walter was consumed, as were many research-based clinicans, by the prospects held out by EEG for differential diagnosis and psychological and character assessment. The Conceptual and Clinical Origins of EEG in America An approach similar to Walter’s was taking shape in the northeastern United States just as Adrian and Matthews were debating the significance of the Berger rhythm. There, philanthropic agencies were lined up to support home-grown developments in diagnostic technology, as well as research into the role of the emotions in medical practice. Both of these ideals were deeply embodied in the EEG, and neurophysiological research in this region was considerably influenced by Forbes’s work, whose research program was broader than Adrian’s. Unlike his mentor, Walter Cannon, Forbes pursued a form of research that was itself driven by the developments in communications technology then transforming many facets of American life.64 Cannon’s concept of ‘homeostasis,’ a term he coined in 1926, had forged a powerful analogy between the self-regulation of the physiological body and that of the political body.65 Forbes’s analogy was more limited and invoked not control so much as communication: the nervous system, he suggested, functioned in a man-

278

The Sleep of Others

ner not so different from the technological advances that were then restructuring social communication, such as the public-address system, the radio, and the telephone. This was a parallel drawn less from the conceptual sphere as it was from the practical. After all, it was instruments derived from these very same technologies that Forbes and his students were using to investigate the nervous system in the first place. The ability of such instruments to create multiple kinds of inscriptions was at least as important as the problem of what physiological phenomena these inscriptions actually represented. When taken together and compared against other graphical signs (then accumulating in hospitalbased medicine), novel signs like the EEG could have meaning without pointing directly to physiological structures or processes. Its meaning was directly linked to the phenomena of organization, which neurophysiologists, like experimental biologists in general, were rapidly beginning to see as the most fundamental aspect of life itself. Upon his return to Boston from Cambridge in the middle of 1921, Forbes began preparing a massive review of the current state of neurophysiological research in the light of several recent advances made by Lucas, Adrian, Lapicque, and himself.66 The focus of his article reflected his own attempts at reconciling the vast differences that separated the phenomena of spinal reflexes from the work done on nerve-muscle preparations. In comparison to results gained from nerve-muscle experiments, for example, the reflex arc exhibited slower rates of conduction, fatigue upon continued stimulation, after-discharge (continued response after stimulation has ceased), and a greater variability of threshold. But these differences, Forbes speculated, must be chimerical; the two approaches would eventually dovetail and could even now be ‘generalized’ so as to form a grounding for psychological theory. At least, this was the ideal, since it followed the pattern laid down by the success of experimental physics in reducing chemical elements to their constituent subatomic parts. Yet this reductionist approach would, Forbes insisted, leave subjective experience and its analysis untouched: ‘Such a generalization need not affect any philosophical position we may take as to the relation between cerebral activity, as viewed from the objective standpoint, and consciousness as known to us through our own subjective experience. It should not seek to rob conscious life of any of its subjective properties. It would merely, if substantiated, effect a simplification of the physical aspect of the activity of the nervous system.’67 Forbes had sent Sherrington a draft of the paper sometime during

Sleep Finds a Groove 279

the latter half of 1921. Sherrington’s response was that Forbes’s microstudy of nervous tissue, which used electrophysiological methods to reduce the physical basis of mind to its simplest elements, was in fact an extension of his own work on spinal reflexes. Whether or not this was actually the case is irrelevant. The point here is that Sherrington acknowledged (as did Adrian) the novelty of what Forbes was attempting to do: create a neurophysiological assemblage that used electronic amplification to do away entirely with visible muscular movement as an indicator of nervous activity. Such a proposal seemed to open up, thought Sherrington, a sort of technologically demonstrable monism that found the potential of mind – the ‘germ of psychosis,’ as he called it – in all life. Or, at least, all life possessing a nervous system. He presented this perspective to Forbes in a whimsical diagrammatic form, as though spinal centres and nerve-muscle preparations were components taken from a yet-undiscovered circuit drawing. His drawings suggested that the ‘more trenchant’ claim was the idea that mind could be reduced to the simple interaction of nerve and muscle, without the integrating force of the spinal reflex that had so preoccupied his own research (fig. 23a). Sherrington went on to speculate about William James’s response to such a claim: ‘Would his eyebrows have lifted?’ he wondered, only to suggest that James would have located the real genius of mind in a vital force, here pictured as an amoeba (fig. 23b). Forbes’s reaction was, as always, imbued with sobriety: ‘What you say about the “germ of psychosis” is most interesting. I think I am in the amoeba camp. I discussed the question the other day with L.J. Henderson, and he said it depended upon what you mean by “germ”; if potentiality, then ameba; if beginning actuality, then perhaps integration; i.e. the entire organism minus all its component parts, thus stressing the principle of organization as against the units organized. He also said that in one sense he would be inclined to place the “germ of psychosis” in C, H, O and N.’69 Forbes’s reduction of life and mind to a ‘principle of organization’ was a quintessentially American position, and it would soon come to dominate, among other things, American neurophysiologists’ thinking about the nervous system.70 Encouraged by Rockefeller support of technologies that could bridge the gap between the laboratory and the clinic, and spurred on by the holistic approaches financed by the Macy Foundation, neurophysiologists found political and social systems of self-regulating order and organization mirrored in nature.

280

The Sleep of Others

23a An excerpt from Sherrington’s letter responding to Forbes’s ambivalence regarding the reduction of subjective experience to nervous conduction.

An Early Career in EEG: Hallowell Davis The technological reorientation of neurophysiology promoted by Forbes at the micro-level was pursued by his students in other quarters, in other ways. While he did participate in some EEG research, Forbes was not as committed to it as a transformative investigative device as was his student Hallowell Davis (1896–1992). Davis, working alongside his wife, Pauline, saw the EEG as a means of linking the most recent advances in neurophysiology with problems of psychiatric diagnosis, a topic of little interest to Forbes. The Davises were not alone in thinking that such a tool had the potential to reshape psychiatric practice: Alfred Lee Loomis, a millionaire financier and amateur scientist with a private laboratory in Tuxedo Park, New York, shared their belief that the device had viable (and perhaps profitable) clinical applications. Together, this group attempted a number of novel applications of EEG and discovered a curious fact about sleep’s rhythmic organization in the process. Hallowell Davis’s expansive vision for the EEG derived in part from his experiences as an undergraduate at Harvard. He was deeply influenced by Lawrence J. Henderson, a physiologist whose work in the biochemistry of fatigue eventually led him to create a field he described as

Sleep Finds a Groove 281

23b Continuation of the excerpt from Sherrington’s letter. (Courtesy Harvard Medical Library in the Francis A. Countway Library of Medicine, AFP 22, 733)

‘clinical sociology,’ which studied social organization by physiological methods. Davis graduated with a BA in chemistry in 1918, ten years before Henderson established his fatigue laboratory at Soldier’s Field in Boston. But the two maintained a close relationship, and Davis collaborated with Henderson and his students on several occasions over the following decades. Rather than pursue a PhD, however, Davis chose to enter Harvard Medical School upon graduation. He completed his MD in 1922, and Forbes immediately offered him a position in his laboratory. At Forbes’s urging, Davis soon began to study with Adrian in Cambridge, recognizing that his original field of study, in the physiology of

282

The Sleep of Others

blood chemistry, was then saturated with researchers while tube-driven electrophysiology was only just getting started.71 His work with Adrian was cut short, however, when his fiancée, Pauline, contracted typhus while doing relief work in Constantinople. In 1923, just two years after Forbes had taught Davis how to build a vacuum-tube amplifier, Cannon offered him a permanent position in the department.72 Most of the research Davis conducted during his long tenure at Harvard (1923–46) was devoted to the physiology of hearing. This field was obviously ripe for change with the development of vacuum-tube amplification, since the signals from individual sensory nerves could be recorded with a degree of precision that was inconceivable only a few years earlier. Davis immersed himself in this study and endeared himself in the process to Forbes, who suffered from partial deafness and was continually asking Davis for new types of hearing aids.73 Electroencephalography presented a similar opportunity to Davis to test the limits of the new electrophysiological technologies. Late in 1933, and well before Adrian and Matthews presented their experiments on the ‘Berger rhythm,’ Davis’s first PhD student, Bill Derbyshire, came across Berger’s work.74 Sceptical, Davis immediately proposed that Derbyshire and a medical student, Howard Simpson, test Berger’s results. It seemed unlikely to Davis that the activity of millions of neurons could be so perfectly synchronized as to produce such slow, regular potentials. And there were problems with the experiments right from the start. Although Forbes’s laboratory featured some of the most sophisticated electrophysiological equipment then available, they seemed unable to detect the presence of alpha in either Derbyshire or Simpson. Finally, they were able to detect a distinct rhythm of ten hertz that was taken from Davis’s own scalp. Davis immediately solicited his colleagues to volunteer for the experiment, and he found that they could be almost evenly divided between those who showed alpha and those who did not. Davis’s emphasis on individual differences was no accident. Unlike Adrian, who followed the micro-anatomical strategies of Lucas, Davis’s study of auditory physiology had made him familiar with psychophysiological investigative techniques and associated psychological approaches. Davis knew many psychologists, was married to a practising psychiatrist, and had even debated the Harvard experimental psychologist E.G. Boring on the nature of Helmholtz’s ‘mechanical acoustic analyser.’75 Adrian and Matthews also observed individual differences in ‘showing alpha’ but considered this a purely technical difficulty.

Sleep Finds a Groove 283

Davis, however, quickly turned it into a focal point for his EEG research, and it guided many of the technical developments that followed. Photographic records, for example, had been the mainstay of most electrophysiological research to that point. While this method of recording was useful for creating clear and precise images of electrical events in a nerve, they were laborious and expensive. Adrian and Matthews prized their precision but also developed an ink-writer, so they could generate results more quickly, and with less expense.76 Davis, on the other hand, found photographic records less helpful for psychophysiological experiments, in which subject responses were calibrated to physiological parameters over substantial periods of time. More important, he wanted to observe behaviour and brain events ‘on-line,’ in synchronicity.77 To this end, he set his technician, E. Lovett Garceau, to the task of developing an ink-writing oscillograph. Once again, communications technology played an important role: within a few months, Garceau had modified a Western Union ‘undulator,’ originally designed to record signals from transatlantic submarine cables on a 5/8s wide strip of paper, to inscribe brain waves. The Davises shared their interest in EEG with several colleagues at Harvard, including Erna and Frederic Gibbs and William Lennox. But while the Gibbses and Lennox soon made their reputations by successfully using the EEG to diagnose temporal lobe epilepsy, such welcome and valuable discoveries were not as readily forthcoming for the Davises.78 Hallowell’s enthusiasm for the EEG was somewhat tempered, at least when he spoke of it with Cannon. In his routine correspondence with the departmental director, Davis did not mention the EEG at all until 31 July 1934, more than half a year after he his first successful recordings. He then noted, almost in passing, that Frederick A. Gibbs, ‘who is working with us at present and plans to work with us next year on cerebral action currents,’ wanted to renew his position as a research fellow in neuropathology, then funded by the Macy Foundation.79 The fact that he described alpha as ‘cerebral action currents’ despite being unable to link them to any ‘action’ (muscular or otherwise) suggests a certain confusion about the phenomena, or at least the desire to maintain that they were of a kind with the action currents in nerve-muscle preparations that were then the stock-in-trade of neurophysiological research. Davis went on to describe his work in auditory physiology, saying nothing about the recent developments in creating viable ink-and-paper recordings. Indeed, the only mention he made of Garceau was to address Cannon’s perennial complaints about the technician, who apparently was in the habit of

284

The Sleep of Others

keeping a dog with him at all times and who tended to use ‘his radio for sounding broadcast through the first floor the music that he pleases to listen to.’80 The situation came to a head a year later, when Cannon instructed Davis that the dog had to be ‘kept away from the laboratory and the radio no longer used except for strictly experimental purposes.’ Davis, who had long been concerned with Garceau’s low salary, passed on the message to his technician. Garceau left within a week and quickly established himself as a successful instrument manufacturer.81 Davis’s communication with Forbes, however, indicates that he was well aware of the potential utility of the EEG. His first mention of the EEG came in a letter dated 9 June 1934 and sent to Forbes in Naples, Italy, where he was then vacationing. Davis had obviously just attended – or at least heard about – the American Neurological Association meeting that had taken place in Atlantic City on 4 June where Adrian announced his study of the ‘Berger rhythm’ to an American audience. Davis’s comments suggest that he was already considering using the EEG as a tool for investigating individual difference, and was under no delusions as to their distinctiveness from traditional action currents: ‘The dope from Adrian and Mathews [sic] that the slow cortical waves of about 10 per second are actually composites is surprising. I shall be extremely interested to see their evidence. The big waves often seem to have such an individuality, often a personality, that I find [it] difficult to accept Adrian’s dope offhand; yet from the point of view of explanation it is very very comforting.’82 Adrian and Matthews’s view that the alpha rhythm was simply the synchronization of millions of neurons with nothing to do was a ‘comforting’ physiological explanation. Yet Davis, like Berger, clearly thought that the EEG presented an entirely new kind of evidence that he hoped to link to ‘individuality’ and ‘personality.’ Adrian’s stature in the field, however, made for an obligatory point of passage, and Davis set to work characterizing alpha. At the end of July 1934, Davis wrote Forbes (who was still in Italy) that everyone in his lab was ‘about ready to climb a tree’ because of ongoing difficulties in his auditory experiments. The one bright spot was that Davis had managed to follow Adrian in abolishing alpha rhythms by visual stimulation: ‘On the other hand we used our heads to good advantage the other night and got our own cortical potentials meeting Adrian’s description. This can be done without putting the needle into the scalp. The currents are abolished by low visual stimulation.’83 Davis began conducting his own public demonstrations of the EEG shortly after Adrian returned to Cambridge. Visualizing the electrical

Sleep Finds a Groove 285

activity of the brain evoked a great deal of popular interest in the 1930s, not unlike the excitement the discovery of X-rays had created at the turn of the century. Such demonstrations resonated with the drama of stage performance and helped bring other researchers to the field. Davis’s first public demonstration at Harvard, for example, attracted the attention of the Gibbses, who had been working with epileptic patients under Lennox at Boston City Hospital.84 The Gibbses brought an epileptic patient into Davis’s laboratory in December 1934. By the end of the year, they had successfully recorded the classic three-per-second ‘spike and wave’ pattern that characterized an attack of petit mal epilepsy. Their report was one of the first publications on brainwaves in North America.85 The discovery eventually transformed how epilepsy was diagnosed, since EEG patterns came to supplant reliance on either patient histories or clinical observation of a seizure. Equally important, the discovery of so-called ‘larval’ and undetected forms of petit mal epilepsy helped determine the direction of future EEG research, for proponents suggested that mental lapses, inattentiveness, and fits of moodiness or frustration might all be forms of ‘epileptoid’ behaviour detectable by EEG. This search for hidden cases of ‘cerebral dysrhythmia,’ which, by some estimates, affected some ten million Americans, animated Forbes’s and the Davises’ wartime research, as they attempted to use EEG to screen those aviation cadets thought to be plagued by this sinister pathology.86 The explosion of interest in EEG and Davis’s enthusiastic demonstrations began to resonate in the obscure corners of unorthodox medicine, a trend that continues to the present day. In October 1934 Cannon (who must have seemed nearly humourless to Davis by this point) complained to Davis of having received ‘some queer mysticism which is the outcome of newspaper publicity.’87 Cannon was likely referring to a group calling itself ‘The Eastern Electronic Research Association’ (EERA), who wanted to have Davis and Derbyshire demonstrate the EEG at their winter meeting in New York in January 1935.88 The EERA’s express mandate was to encourage ‘the development and application in practice of methods employing the detection and identification of human radiations as an aid to diagnosis and the use of short radio waves as a means of treatment.’ Cannon, whose work with X-rays had left him with radiation burns that caused him considerable chronic pain, would seek out treatment through electrophysiological monitoring a few years later.89 But the EERA seemed to have more metaphysical aims in mind, and Cannon and Davis shared a common suspicion of the group. They fired off a curt rejection a few days after receiving the invitation.

286

The Sleep of Others

This sort of engagement with the margins of scientific research did not, however, prevent Davis from continuing his public demonstrations of EEG. Aside from his native interest in the device, he clearly saw it as a weapon in the public-relations battle with anti-vivisectionists, against whom Cannon (and A.J. Carlson) had proven to be tireless adversaries.90 The EEG became front-page news in mid-1935, when the New York Times ran a story about one of Davis’s demonstrations under the headline ‘Electricity in the Brain Records a Picture of Action of Thought.’91 The reporter, whom Davis had invited to play the role of experimental subject, described how he had been attached to an electroencephalograph and asked to relax until the EEG showed the alpha rhythm. Davis demonstrated how light and noise blocked alpha and concluded with the request that the subject perform a mathematical operation. After giving his answer, the subject’s EEG returned to alpha, only to change again into a ‘thought-wave’ as the reporter described the phenomenon. Davis apparently whispered to his colleagues that ‘“He [the subject] is now checking the answer to see if he was right,”’ which, the reporter testified, was exactly what he had been doing. In the wake of the article’s exaggerated claims about EEG as a mind-reading device, Davis assured Cannon that he had been very careful to explain what the demonstration ‘did and did not mean.’ Regardless, the potential benefits of such publicity, thought Davis, outweighed other considerations: We got what I consider a particularly good write-up ... Since returning home we have been besieged by reporters, photographers, movie men and representatives of radio broadcasting companies. We have drawn the line on the movies, but did allow our picture to be taken, with the apparatus as we had it in Detroit ... I have been somewhat emboldened in this policy of giving material to the newspapers, with the question of antivivisection very much in mind, by the remarks made by Dr. Whipple at the Federation [of American Societies for Experimental Biology] banquet. He strongly urged the utmost coöperation with the newspapers ... I hope that you do not feel that it cheapens our work or seems like undue seeking of notoriety.92

Tuxedo Park: The First Sleep Laboratory Hopes for the EEG resonated in the popular and scientific imagination. For the first time, it seemed possible to identify and correlate objectively, in real time, kinds of brain activity to mental activity. Thought had suddenly gained an automated and public expression that could be dis-

Sleep Finds a Groove 287

played with minor modifications to existing technologies. Little wonder that entrepreneurial amateurs took it up with a passion. The most important of these was surely Alfred Lee Loomis, who met Hallowell Davis for the first time late in 1934. Loomis, who had made a small fortune as a Wall Street lawyer before the stock market crash of 1929, enjoyed a reputation among scientists, businessmen, and politicians as one of the most powerful and trustworthy patrons of science. In the early 1920s, he had created his own laboratory in his house in Tuxedo Park, New York, where he began to develop a number of instruments in conjunction with an old friend, the experimental physicist Robert W. Wood. Loomis, who was trained in the physical sciences and seemed to know practically everyone (his cousin was the future secretary of war, Henry L. Stimson), was the perfect patron for Wood, who had complained bitterly to the Rockerfeller Foundation about his lack of funding for new instruments at Johns Hopkins University.93 All of the scientific work at Tuxedo Park revolved around the development and deployment of large and expensive precision instruments. The best known of these was the microwave radar set he built using a klystron tube, which ultimately led, in the summer of 1940, to Loomis’s being enrolled in forming the Radiation Laboratory (RadLab) at the Massachusetts Institute of Technology as part of the American war effort.94 During the 1930s, Tuxedo Park was also filled with, among other things, a 40c spectrograph, a ‘microscope-centrifuge,’ three Shortt pendulum clocks (there were only five others in the world, all in major astronomical observatories), and a private telephone line that carried a calibration signal from the quartz crystal oscillators developed by Bell Laboratories. Loomis’s passion for timekeeping coincided with his growing interest in biology in 1934, when he and E. Newton Harvey, a physiologist at Princeton, built a kymograph with an eight-foot-long drum (fig. 24). The instrument, reputed to be the largest in the world, generated inscriptions on a single piece of paper, thus enabling an experimenter to compare visually a large amount of information in a single glance, instead of going through hundreds of feet of paper tape. Loomis and Harvey had built the kymograph in order to observe slow changes in physiological activity over long periods of time. Its horizontal drum featured a red and a green pen that travelled once every minute around its forty-four-inch diameter, inscribing heart rate, blood pressure, and the like for several hours at a time.95 In keeping with the technological developments that were then taking place in American medicine, Loomis was particularly interested

288

The Sleep of Others

24 Loomis’s Tuxedo Park laboratory was filled with large and expensive instruments. This kymographic drum enabled Loomis and his co-investigators to record periods of sleep for up to eight hours at a time. (Alfred Lee Loomis, E. Newton Harvey, and Garret Hobart, ‘Electrical Potentials of the Human Brain,’ Journal of Experimental Psychology 19 [1936]: 249–79, 253, courtesy Henry Loomis)

in improving the electrocardiograph, which was still based upon Einthoven’s string galvanometer. He began experimenting with the placement of the electrodes and discovered that, when he placed one electrode on a subject’s head, the baseline was unsteady. Loomis suspected that this rhythmic activity was not artifact but was actually recording electrical activity in the brain. In the autumn of 1934, he turned to an old acquaintance, Alexander Forbes, for advice regarding this strange phenomenon. Naturally, Forbes introduced him to Davis, who had only just begun to study the EEG. Sleep was one of the first objects to take shape around the EEG in Tuxedo Park. Loomis’s sharp entrepreneurial eye combined with the popular and scientific interest in sleep that had grown out of the epi-

Sleep Finds a Groove 289

sodes of encephalitis lethargica, Pavlovian theory, and the use of sleep in psychiatric therapy (discussed in chapter 4). Although both Davis and Loomis began to study the uses of the EEG in sleep during 1935, there were marked differences between their approaches. When the work that Davis and his colleagues published at Harvard that year is compared with what was coming out of Tuxedo Park, it is clear that sleep was a marginal issue for the former group, who focused on the potential uses of EEG in neuropsychiatric diagnosis.96 The research at Tuxedo Park, on the other hand, focused almost exclusively on sleep. Why? The question is an important one, since our aim is to uncover how sleep became a scientific object. The EEG is a crucial part of this story, but EEG research took place in several different contexts simultaneously. Only Tuxedo Park generated knowledge of sleep’s temporal organization – the EEG of ‘sleep stages’ – that provided a foundation upon which twentieth-century sleep research could be built. What was it about Tuxedo Park that made it so amenable to this sort of research? Ease of access to investigative resources offers one possible answer. Epileptics were readily available for study at Harvard Medical School, while Loomis could only solicit friends and family to serve as experimental subjects in his private laboratory. Physical location certainly counted for something. Despite concurrent efforts to make biological labs look more natural, the Massacheusetts General Hospital in Boston was a surely a more artificial setting for sleep than the bedrooms in Loomis’s Tuxedo Park home.97 Professional orientation also played a role. On the one hand, Loomis lacked the medical training that the Harvard group enjoyed, so his interest in using EEG to parse out epileptics from other neuropsychiatric illnesses was limited at best. His background in physics and engineering, however, combined with his passion for precision instrumentation, loomed large in his choice of research topics, and his manner of pursuing them. Loomis’s fascination with sleep as a null state of the brain converged, as did most Tuxedo Park research projects, upon the use of instruments. Shortly after meeting Davis, Loomis, Harvey, and Hobart (a technician) quickly converted their giant kymographic drum to record brainwaves for a full eight hours. Recording brain waves during sleep, of course, was not a novel project in 1935. But no one had yet taken up the task of creating a comprehensive picture of EEG in sleep. Instead, EEG researchers had been content to compare samples of EEG taken throughout the night, implicitly assuming that there was, at most, a quantitative difference between these measurements. With their greatly enhanced recording capabilities, Loomis and his co-workers were sud-

290

The Sleep of Others

denly able to frame sleep as an object possessing a distinctive temporal structure. Their depiction of sleep as having ‘stages’ that could be identified only by continuous EEG recording broke sleep away from its dependence upon subjective consciousness. The sleeping brain was engaging in activities that could be observed by instruments, and of which the sleeping mind knew nothing. Loomis, then, definitively aligned sleep with mechanically generated signs, much as he had done with physical entities like the radiation emitted from a klystron tube. At Tuxedo Park, sleep achieved objective existence for the first time. Loomis was well aware of the unique properties of his EEG assemblage, and he routinely pointed this out in his publications, noting, for example, that his kymographic drum produced ‘a sheet of paper 8 feet long and 44 inches wide’ and was ‘a great improvement over the use of ticker tape which would require a ribbon of paper over half a mile long.’98 Experiments and research issues were soon orchestrated around the instrument’s capacities. Questions of EEG and brain localization, for example, were entirely ignored in favour of perfecting the technical aspects of their laboratory for observing sleep. The lab featured a shielded, furnished sleeping room separated from the control room by some sixty-six feet, and an amplifier that could be tuned to a desired EEG frequency and then connected to an ‘integrator,’ which could automatically detect and record fluctuations in rhythmic physiological activity and calibrate it to other recorded material (fig. 25). Microphones and mirrors were deployed to monitor behaviour without awakening or interrupting the subject and to help to determine which waveforms were the product of brain activity and which were artefacts of facial movements or other muscular contractions. There was even a phonograph to play a recording of Ravel’s Bolero as a standardized background stimulus to assist in the determination of normal alpha patterns for each subject. Mechanizing the production of data at Tuxedo Park was not simply another chapter in the automation of scientific labour; it was a virtue to be pursued in the name of capturing sleep in its natural habitat. ‘Perhaps the most important point to emphasize in regard to our technique,’ mused Loomis, ‘is the automatic control of all recording at a distance from the subject, who remains undisturbed throughout the experiment.’99 Staging Sleep: The First EEG Conference Loomis’s efforts at automating the observation of sleep through EEG helped to create sleep as an independent object, much as photographs

Sleep Finds a Groove 291

25 The sleep laboratory at Tuxedo Park. The young boy is probably Hallowell and Pauline Davis’s son. (Life Magazine, 18 October 1937, 56–8)

292

The Sleep of Others

had appeared to separate images from their creators nearly a century earlier. And, just as the advent of photography liberated the production of certain kinds of images from the artist who drew and painted, so, too, did EEG help to separate sleep from the sleeper. But the EEG resonated at a disciplinary level as well. While the Chicago neuropsychiatry project floundered and eventually imploded (see chapter 5), the EEG became a focal point upon which disciplines as diverse as physics, physiology, psychiatry, and neurology began to converge. Loomis’s laboratory at Tuxedo Park provided the perfect backdrop for such a convergence, located, as it was, outside formal academic and university circles even as it was set deep in the heart of the industrial and political power represented by the northeastern American elite. On 10 November 1935, Loomis held a conference on ‘The Electrical Potentials of the Brain’ that featured six presentations. Hallowell Davis offered a literature review, Herbert Jasper described the occipital origins of EEG, Loomis described his unique kymograph and gave a laboratory tour, and Harvey described the cortical potentials he and Loomis had recorded during sleep. Another paper, whose author was not listed in the official program of the day’s events, described EEG in animals, while a final paper was simply listed as ‘to be announced.’100 Although no attendance record was kept, a list of those who had confirmed attendance a week earlier numbered fifty participants.101 The novelty of EEG at the time can hardly be overestimated: only five papers had yet been published on the EEG in English. For many participants, it was likely the first time they had seen the phenomenon demonstrated. The potential importance of the device was underscored by the presence of Max Mason and Warren Weaver from the Rockefeller Foundation, and the president of the New York Neurological Institute, Floyd B. Odlum. For his part, Davis described the conference in unequivocal terms as ‘the most exciting scientific conference’ he had ever attended.102 The guests, most of whom came from the northeastern United States, represented an eclectic range of disciplines. Physiologists composed the largest single group, accounting for about a third of the total number of participants. Among this group were found many of the discipline’s current and future leaders, including Alexander Forbes, Joseph Erlanger, Herbert Gasser, Ralph W. Gerard, and John F. Fulton. But a growing group of self-identified medical or biophysicists made up nearly a fifth of the assembled group and included Detlev W. Bronk (future president of Johns Hopkins University), Kenneth S. Cole, and the president of MIT, Karl T. Compton. The EEG’s ability to bring psychiatry and neurol-

Sleep Finds a Groove 293

ogy together is evidenced by the fact that psychiatrists and neurologists were about equally represented and together made up roughly a quarter of the participants. Among these clinicians were a number of individuals who would soon help to forge such interdisciplinary connections, such as Stanley Cobb (Harvard), David Slight (Chicago), and Helen Flanders Dunbar (Columbia). In effect, Loomis played the same role in the development of the EEG as he would in the formation of the Rad Lab at MIT four years later. His deep personal and business connections with high-level administrators of universities and philanthropic organizations enabled him to bring EEG to the attention of scientific power brokers like Mason (who was now president of the Rockefeller Foundation), Weaver, and Herbert Gasser, who had just left Cornell to become director of the Rockefeller Institute for Medical Research. By enlisting individuals from all sides of the scientific enterprise around a technological system capable of representing brain activity in novel ways, Loomis engineered the EEG’s entry into American biomedicine in a way that was likely even more effective than Adrian’s public demonstrations a year earlier. The staging of sleep had begun. The idea that sleep came in cycles of distinctive stages developed between 1934 and 1937. The first few articles to appear in Science and the Journal of Experimental Psychology simply announced the discovery of a number of new kinds of brainwave, since Loomis’s patient kymograph revealed that the world of EEG was populated by more than just alpha and beta. There were ‘trains’ of regular ten-per-second waves, ‘spindles,’ which were short bursts of activity in a low-voltage background, slower, ‘saw-toothed’ waves with a high voltage, and periods of random activity that seemed to have no discernible pattern at all. In interpreting these waves, Loomis and his colleagues initially adopted fairly traditional neurophysiological and psychological approaches. In a 1936 paper they abandoned the naming of the waveforms altogether because ‘a logical system of nomenclature’ had to wait until the local origins of the waves were determined.103 Lacking a traditional neuroanatomical anchor, Loomis abandoned the colloquial names of ‘alpha’ and ‘beta’ and simply numbered all the different forms ‘I’ to ‘VI’ (see fig. 26). Some of the waveforms appeared during wakefulness as well, and Loomis’s tentative use of the same numerals to label his experimental subjects as a ‘Type II’ or ‘Type IV,’ depending upon which waveform seemed to dominate their EEG, indicated an interest in using the new device to study individual differences. Subjects were also classified according to the promi-

294

The Sleep of Others

26 The six different waveforms identified by the Tuxedo Park investigators and organized into sleep stages the following year. (Alfred Lee Loomis, E. Newton Harvey, and Garret Hobart, ‘Electrical Potentials of the Human Brain,’ Journal of Experimental Psychology 19 (1936): 249–79, 255, courtesy Henry Loomis)

nence of Type II (alpha) waves in wakefulness or the frequency of the appearance of Type III waves (‘spindles’) in sleep. One of the more unusual observations coming out of Tuxedo Park was that the appearance of Type II waves was not limited to those subjects who were in a state of relaxed wakefulness with their eyes closed. In most subjects, the same waveforms appeared in sleep, and the very same forms could even be provoked by mild stimulation that disturbed, but did not awaken, the sleeping subjects. Even that old psychological saw, hypnosis, returned to the investigative stage. In this instance, Loomis’s EEG simply confirmed what had long been maintained: sleep was vastly different from the hypnotic state. David Slight, who would soon replace Grinker at Chicago, brought a patient who had been hypnotized several times to Loomis’s laboratory, where Loomis demonstrated that Type II waves were absent when the hypnotized subject opened his eyes, just as they were in normal wakefulness. If, on the other hand, it was suggested to the patient that he were blind, the Type II waves would continue when he opened his eyes.104

Sleep Finds a Groove 295

Hypnotism, then, did have measurable physiological effects; but the term ‘hypnotic sleep’ was a misnomer, since the two states were identifiably different from the new vista offered by EEG. It was not until the middle of 1937 that Loomis and his colleagues began to organize their different waveforms into distinctive patterns. After assessing eighty-four records from thirty different subjects, Loomis and his collaborators concluded that there were definitive ‘states of sleep’ that suddenly changed from one to another throughout the night, and that the appearance of a series of these states could be correlated with movements in sleep, the application of external stimuli, and even dreams.105 The authors quickly moved on to an enthusiastic discussion of how their work could be applied to the psychology of individual differences, suggesting that subjects might be divided into two groups, ‘the alpha and the non-alpha,’ and that the ‘extreme types of record’ in each case might be linked to personality or mental and emotional characteristics.106 But such attempts had thus far been elusive. It was at this point that the authors turned their attention from the particular to the universal and began to focus resolutely on sleep. Abandoning their reluctance to use colloquial names such as ‘alpha’ or ‘spindles’ to describe the waveforms, they now linked these forms as five distinctive rhythms that appeared regularly appeared throughout the night in a sleeping subject’s EEG: A – Alpha. Alpha rhythm appearing in trains of various lengths. The eyes may be slowly rolling, under closed eyelids, as indicated by an electrode over the eyebrow. B – Low voltage. A quite straight record, with no alpha rhythm and only low voltage changes of potential. Rolling of the eyes may occur. C – Spindles. Line slightly irregular with 14 per second spindles of 20–40 microvolts every few seconds. D – Spindles plus random. The spindles continue together with large random potentials 0.5 to 3 per second. The random voltages may be as high as 300 microvolts. E – Random. The spindles become inconspicuous, but the large random potentials persist and come from all parts of the cortex.107

The Tuxedo Park researchers were no closer to the ideal of linking particular waveforms to distinctive cortical regions than they had been two years earlier. But this no longer seemed to matter. As they began to take sleep as their object of study, they began to situate different waveforms

296

The Sleep of Others

not by space but by time. Sleep’s structure (its ‘architecture,’ as sleep researchers now say) was now definitively temporal. Honouring tradition, sleep, ideally, still had to share its new stage with dreaming. Frustratingly, Loomis and his colleagues had observed a link between dreaming and a sleep rhythm only twice. In one instance, a subject’s record went from a ‘stage D’ to a ‘stage B’ pattern without any observed bodily movement. Eight minutes later, he moved, woke up, and reported a dream that seemed to be associated with the recent movement. In the second case, the subject’s record jumped from a stage D to a stage B pattern after an experimenter turned the knob on a laboratory door, awakening the subject who proceeded to report a dream that seemed as though it had been provoked by the noise. The investigators arrived at the conclusion that dreams were likely to be associated not with any particular pattern of EEG waves but rather with the state of sleep. Like so many other physiologists and psychologists before them, the Tuxedo Park investigators seemed to prefer to picture dreams as the product of internal or external stimuli, rather than a phenomenon that possessed a distinctive rhythm of its own. Following Loomis’s 1935 EEG conference, other investigators across the United States began to take up the electroencephalographic study of sleep. The first of several papers on EEG coming out of Carlson’s Department of Physiology at the University of Chicago appeared in 1937 and was co-authored by Ralph Waldo Gerard (1900–74), who would later become the first director of the influential Neuropsychiatric Institute of the University of Illinois. The paper confirmed that there were a number of distinctive EEG rhythms that regularly appeared during sleep, and it then proceeded to correlate these rhythms to depth of sleep.108 Following the pattern set by the Tuxedo Park investigators, Gerard and his colleagues established a uniform level of noise stimulus by foregoing Ravel’s Bolero for a, 1000-hertz signal amplified through a loudspeaker in the sleep room. They quietly whispered ‘Are you awake?’ to their sleeping subjects and counted the number of seconds that elapsed before the subject responded as a measure of sleep depth. On this basis, they argued that the first two hours of a sleep period were spent in a deep sleep, which then became progressively lighter until the fifth hour, at which time it either stayed constant or cycled again before awakening. More important, their measurements of sleep depth could be positively correlated to brainwave patterns. Deep sleep was associated with a prominence of slow, high-voltage delta waves (two to three per second), while lighter sleep featured the faster, low-voltage alpha and

Sleep Finds a Groove 297

beta waves. The next paper coming out of Chicago clearly showed the influence of Kleitman’s passion for identifying the endogenous rhythms of sleep that existed apart from any relationship to external stimuli.109 The Tuxedo Park group had argued that a sleeper’s EEG continually shifted from one pattern to another through the night. Movement or external stimuli tended to shift the record from a ‘deeper state’ (D or E) to a ‘lighter state’ (A or B). But, to that point, no normal pattern of stages appearing throughout an unbroken night’s sleep had been documented. By 1939, the Chicago group was arguing that uninterrupted sleep generated a normal cycle of six EEG stages, which, even if they were briefly interrupted by movement or stimulus, would inevitably return to a predictable cycle. This 1939 paper also featured a brief investigation of dreaming that followed the same logic of linking experience, such as depth of sleep, to endogenous brain activity. Instead of attempting to identify dreams with a particular stimulus, as the Tuxedo Park group had done, the Chicago investigators simply woke up their subjects during different stages of sleep and asked them if they had been dreaming. But, for all their difference in method, their conclusion was practically the same as that proffered by Loomis’s group. Dreaming, they observed, was ‘present most of the time’ during the night and ‘a subject abruptly awakened, almost at any time during the night, can recall having dreamed.’ They did, however, note that the ability to recall a dream did seem to be positively correlated with the presence of alpha waves in sleep: ‘The longer the immediately preceding period with no alpha waves, the less is the [dream] recall, and when this period is about a minute (especially if delta waves are present), there is no trace of a dream’s having been in progress.’110 This experiment, while it did not depict the presence of dreaming in the stark binary terms that Kleitman and Eugene Aserinsky later described as ‘REM’ and ‘non-REM sleep,’ nonetheless seemed to offer, in embryonic form, the notion that dreaming itself had a rhythm. Neither Gerard nor Kleitman, however, took up this possibility in their research over the following decade. Gerard, who was keenly interested in the EEG, tended to use it as an example of how all organized systems, be they biological (he called them ‘animorgs’) or social or political (‘inanimorgs’), tended towards a greater degree of organization through greater integration and specialization.111 The very concept of freedom in such systems was redefined as more and more functions became automated. The electroencephalographic evidence that ‘nerve cells can continue to beat electrically in a constant environment within

298

The Sleep of Others

the body’ and that ‘this beat can be modified by all sorts of external circumstances’ demonstrated, for Gerard, the immense amount of organizational work required to maintain a stable state in which the smallest amount of freedom (as consciousness) could appear.112 It was in this sense, which Gerard felt was utterly opposed to the ‘terrorism, falsifications, irrationalism and martial orientation’ of the Nazi state, that all systems nonetheless tended towards ever-increasing levels of social control. ‘As history unfolds,’ Gerard paradoxically concluded, ‘I am confident that man will find himself more subject but less slave.’113 Kleitman, in contrast, had no truck with such high-flung rhetoric, regardless of the worthiness of the cause. His empirical caution and disinclination to speculate (which was surely derived in part from Carlson’s leadership) effectively prohibited him from drawing grand conclusions about the greater significance of the EEG; it also seemed to prevent him from actively engaging the problem of dreaming, even when given a viable laboratory tool for this purpose. He continued to be preoccupied with linking sleep and motility and conducted little EEG research until the early 1950s. Multiple Meanings Despite their pervasive dedication to EEG, neither Hallowell nor Pauline Davis ended up focusing their efforts upon the study of sleep. In contrast to Forbes’s work, which had, in part, allowed the instrument to frame its own research problems, the Davises clearly attempted to fit EEG into existing problems in diagnosis. This is not particularly surprising, if we keep in mind the fact that Loomis’s work was self-funded and drew upon his long-standing fascination with instruments, while the Macy Foundation bankrolled the Davises’ work at Tuxedo Park for three summers from 1937 to 1939. As already noted, the official madate of the Macy Foundation entailed a commitment to biomedical holism. The foundation had been formed in 1930, and it emphasized the need ‘to develop more and more in medicine, in its research, education and ministry of healing, the spirit which sees the center of all its efforts in the patient as an individuality.’114 This would be accomplished through research that created ‘not only a clear conception of the patient as a total organism but also a concern for the patient as an individual personality often in need of sympathetic insight which the family physician of old was able to offer.’115 In the Davises’ hands, the EEG was to provide a technological link between

Sleep Finds a Groove 299

the progress of modern American biomedicine and the development of a more holistic medical practice. A report Hallowell sent to Dr Frank Fremont-Smith, head of the foundation’s medical office who later went on to organize a groundbreaking series of conferences on cybernetics in the 1940s, indicates the strong clinical focus of the Davises’ research: ‘The purposes of our survey have been to test our previous conclusions as to the individuality and the stability of the EEG pattern; to describe and characterize EEG patterns for purposes of classification and comparison; to seek for possible correlations between features of the EEG pattern and other characteristics of the subjects, whether mental, physical, or physiological; and to determine the range of variation of normal patterns, in order to establish a background against which abnormalities of clinical significance may be recognized.’116 Sleep was only a marginal aspect of the Davises’ work at Tuxedo Park. In 1938, for example, they published, in conjunction with Loomis and his co-workers, a study of EEG changes provoked by stimuli presented in sleep, with a particular focus on how EEG patterns at sleep onset could be correlated to the subjective experience of the ‘floating’ feeling many reported feeling just before falling asleep.117 It was rather the problem of individual differences and how it could be linked to Hallowell’s blossoming interest in psychoanalysis that took centre stage. Psychoanalysis, as we have already seen in our discussion of the formation of the neuropsychiatry unit at Chicago (chapter 5), was a controversial facet of this push for holism in neuropsychiatry; the official line of the Macy Foundation, for example, seems to have been that holism was worth funding but psychoanalysis was not.118 Its proponents during the 1930s included several notable neurophysiologists who initially took it upon themselves to ground psychoanalytic theory in the anatomy and physiology of the brain. Hallowell Davis’s attempt to apply electroencephalography to the question of personality was a case in point. Already in 1935, he had teamed up with Leon J. Saul, a psychoanalyst, to use the EEG to study patients at Franz Alexander’s Chicago Psychoanalytic Institute. Davis, who had initially been turned down by the Macy Foundation for work on the project, was undeterred, since he was confident that Forbes’s old colleague, Alan Gregg, could bring the Rockefeller Foundation around to support the work.119 Their plan was to use EEG to classify neurotic patients and to detect and measure physiological and emotional changes invoked by psychotherapy. The Rockefeller Foundation was already supporting Alexander’s efforts to create a new field of psychosomatic medicine that would incorporate

300

The Sleep of Others

psychotherapy and orthodox medical practice in order to treat simultaneously the emotional and psychological effects of disease, and Davis felt that his work added an experimental, laboratory-based aspect to this same tradition. Davis’s first public discussion of this work took place in June 1937, during the American Neurological Association’s convention. At the ‘Neurological Supper Club,’ a gathering of elite members of the association, Davis outlined how he was trying to bring EEG to bear on the practice of psychoanalysis. His work met with considerable scepticism and protest, although its depth was evident to Davis only after Cannon passed on many negative comments he had overheard later that year. Cannon did not identify his source or sources, which must have made the problem seem all the more threatening to Davis, who was obliged to guess at who was attacking him. ‘I am not surprised,’ wrote Davis, ‘that such a comment as you quote should come back to you, as two or three of the members, notably Dr Myerson and [the] Bronson brothers, I could feel bristling with hostility (and perhaps disgust?) from the moment I mentioned psychoanalysis.’120 Concluding this letter to his departmental head, Davis noted that ‘I think that it is quite possible that there may be further reverberations from my venture into dangerous territory, and I should like you to have on record these comments on the situation.’ While Davis readily agreed, in his communication with Cannon, that ‘psychoanalytic description’ suffered from ‘shortcomings,’ he clearly hoped that the EEG would serve as a bridge between his own field and Saul’s, just as electroencephalography had started to bring together neuropsychiatry and physiological research through the Gibbs’s studies of epilepsy. Refuting an anonymous critic, Davis insisted that ‘of course he has not seen any correlation between psychoanalysis and electroencephalograms. I believe that our own group is the only group which has had that opportunity.’ But, though Davis and Saul had decided to be ‘much more cautious’ in their decision to publish their findings, Davis remained deeply devoted to the project: ‘In fact,’ he wrote, ‘I believe so strongly in the ultimate validity and importance of this field of inquiry that I am planning to be psychoanalyzed myself in order to obtain firsthand insight into the field and not be forced to rely on second-hand opinions.’121 While one can only speculate about the results of Davis’s personal encounter with analysis, it seems that very little came of his hopes to use EEG to ground psychoanalytic practice. The Macy Foundation, in the end, did come through with funding for the project, and, in

Sleep Finds a Groove 301

his progress report, Davis announced that the EEGs of patients at Alexander’s institute did not differ substantially from their control groups, and that ‘in spite of considerable changes in the emotional state of many of the individuals’ treated by psychoanalysis, these were effectively undetectable by EEG. The only major difference seemed to fall on the well-established fault line created by EEG itself – namely, that ‘persons who tend to be passive and dependent show many alpha waves, while those who are very active and driving do not.’122 Why would an experimental physiologist, particularly one who was a member of the elite group of ‘axonologists’123 that studied the mechanisms of nervous transmission, bother with psychoanalysis? Davis’s hopes for the EEG were merely one manifestation of the exceptional situation brought on by the convergence of a reformist drive in the care of the mentally ill, the appropriation of psychoanalysis by neuropsychiatrists, the reification of technology as a sign of biomedical progress, and the holistic rhetoric that permeated the scientific atmosphere of the 1930s. Asylum reform had been under way in the United States since the early twentieth century. Psychiatry had long been based on the care of the mentally ill housed in huge asylums, but even as the numbers of patients in such state-run or charity institutions were dramatically expanding, psychiatrists began to see the future of their profession in private practice and scientific, hospital-based medicine – something that the stigma of asylum medicine seemed to preclude. Psychiatrists thus began to model themselves upon the neurologists, who had forged a lucrative practice with a scientific basis, in an urban setting, treating the neuroses of the wealthier ranks of American society. Outpatient psychiatric clinics that treated acute cases of mental illness rather than warehousing chronic cases soon began to spring up across America. In the wake of A Mind That Found Itself (1908), a thinly veiled plea for reform by a former psychiatry patient, Clifford W. Beers, psychiatrists began to forge alliances with various movements (particularly the National Committee for Mental Hygiene, which was funded by the Rockefeller Foundation until it became radicalized in the late 1930s) to relocate psychiatry within scientific medicine. This transformation affected professional accreditation and the emergence of medical subspecialties as well. Whereas the term ‘neuropsychiatry’ had once been a purely political accommodation to bring together these fiercely opposed groups to care for the wounded during the First World War, by 1934, the American Board of Psychiatry and Neurology had been established to certify both groups in concert.124 This same

302

The Sleep of Others

period witnessed the appearance of a series of organic therapies (discussed in chapter 4) whose effectiveness would, it was hoped, clear the asylums of their ever-growing populations. Psychosomatic medicine was also receiving widespread philanthropic support.125 The 1930s was a time when an astonishing array of treatments and ideologies could, and did, pass for progress in the medical treatment of the mentally ill. Psychoanalysis was part and parcel of this transformation, and so it did not prove difficult for a talented, well-connected, and dedicated investigator like Hallowell Davis to move from any one of these subfields to another. Indeed, the Davises’ wartime EEG research reflected exactly the same pragmatic, eclectic approach Hallowell had shown in Tuxedo Park. When Loomis abruptly closed his laboratory in the summer of 1939 to work secretly on microwave radar, the Davises did not have to wait long to find an another avenue for EEG research. With the massive increase in American military spending that began early in 1940 and culminated in the draft that September came a dire need to evaluate inductees efficiently. By the summer of that year, the Davises and Forbes were using the EEG at the Pensacola Naval Base in Florida to assess the emotional, physical, and mental attributes of cadets and thus to determine their fitness as pilots. The project lasted for three years, ultimately ending in failure, as investigators concluded that EEG could not reliably predict whether or not a cadet would succeed as a military pilot. The fact that the very definition of an ideal pilot changed over the course of the war certainly did not help the Davises’ efforts, but the salient point is that they and Forbes extended the Macy Foundation’s project of biomedical holism into the field of psychophysiological testing of military personnel. Like the patients whose ideal medical treatment should include sympathetic insight from their physicians in addition to a battery of diagnostic tests, the trainees at Pensacola were to be evaluated in terms of their mental, physical, and emotional performance before they ever stepped into the training planes. In both instances, the analysis and treatment of the ‘whole’ person would, with the help of the EEG, produce an efficient and scientific result. The EEG was born of a climate of medical and scientific reform that emphasized technological innovation in biomedical research. The fact that Alfred Lee Loomis, one of the greatest scientific entrepreneurs in the United States, took such an active interest in the EEG testifies to the widespread potential that this instrument seemed to have in the 1930s. Like Mosso’s ergograph, the EEG relied on technological systems already entrenched in diverse fields, and it was thereby able to bring

Sleep Finds a Groove 303

together investigators from neurology, psychology, and physiology. But what made the EEG different was its ability to generate an entirely new kind of information. The ergograph was created with the explicit aim of understanding fatigue by mechanically isolating and recording bodily movements that were already part of subjective experience and independent observation. The EEG, on the other hand, was an accidental creation that presented investigators with an entirely new kind of physiological activity. Its central phenomenon, the alpha rhythm, appeared, for the most part, where experience was missing. The EEG made something out of nothing. Little surprise, then, to find that it took more than a decade for investigators to conceptualize the ways in which brainwaves could have meaning. In retrospect, sleep might seem an obvious target for a device that creates information about the brain and the mind in the absence of experience. But, although sleep was frequently included among the diverse projects of early EEG researchers, it was, at most, a marginal problem that was largely irrelevant for the dominant medical and psychological problems of the day. Even Kleitman, the foremost sleep researcher in the 1930s, paid little attention to the EEG. Before sleep could be sustainable as an investigative object of laboratory-based science, it needed a battery of phenomena built up around it. The discovery of sleep stages provided one component of this battery. The return of scientific interest in dreaming brought on by psychoanalysis provided another. The Invisibility of Rapid Eye Movements In retrospect, the research conducted at Tuxedo Park in the late 1930s seemed to offer ample opportunities for the discovery of rapid eye movements, just as Nicholas Vaschide’s all-night observations of sleepers had done at the turn of the century. Why did REM remain invisible, despite such an intensive study of sleep? Hallowell Davis himself has suggested that the group at Tuxedo Park was dedicated to investigating the EEG phenomena that were unique to sleep (slow, high-voltage waves and the singular bursts of high-voltage activity known as ‘K complexes’), rather than the low-voltage waves that resembled alpha and that later turned out to be associated with REM: Alfred [Loomis] was greatly interested in rhythms, particularly the sleep spindles (14 Hz) and the delta waves and the K complexes as sleep counterparts of the alpha rhythm, and he used them to define his stages of the

304

The Sleep of Others

onset of sleep. He gave very little attention to the low-voltage state of sleep later in the night that is now known as REM sleep. We let a big one get away that time, although the sheets from the chronograph, laid out on the pingpong table in the game room, revealed at a glance the periods of slow-wave sleep and the periods of low-voltage sleep. We simply set the latter aside as uninteresting, – a return to light sleep. It was more fun to work with floating [‘lapses’ of consciousness before sleep onset] and K complexes, which could be obtained very reliably on the drum and try to make something happen in the late low-voltage stage ... We saw REM but didn’t identify it as a separate state. We worked where the action was.126

Putting aside the conundrum of claiming to have seen something without having identified it in any way, it is clear that the ‘action’ for the Tuxedo Park group, as far as sleep was concerned, was in using EEG to distinguish sleep from wakefulness in an objective, visually demonstrable way. Dreaming fit uncomfortably into this project, since it relied heavily on subjective reports which the EEG seemed to be on the verge of replacing (as least in its most successful applications, such as the diagnosis of epilepsy). The Tuxedo Park group did attempt to correlate dreaming to particular stages of sleep, but the value of such an attempt was unclear. This was not ‘where the action was’ and was unworthy of sustained study. Perhaps it was even a little bit dangerous. Mapping previously uncharted territory in sleep made good sense. It could be accomplished with little conflict. But to suggest that there was a form of knowledge to be had about dreaming that was almost completely outside its definition as a private experience inevitably ran up against many conceptions of dreaming, not the least of which was that proffered by Descartes: dreaming was the mind’s activity in sleep, and, like thoughts, dreams must be continuous. If they were not, the mind would necessarily come in and out of existence, a potentially difficult prospect for a culture that had so recently exchanged its notion of ‘soul’ for that of ‘mind.’ Psychoanalysis translated poorly, or not at all, into early EEG research. Davis’s courageous attempts to use EEG to uncover hidden emotional or physiological results of analytic therapy ignored dreams, and for good reason: the effects of psychoanalysis on dreams were already well documented through phenomena such as transference and countertransference. The situation hardly cried out for physiological evidence that such effects were taking place. Of course, those who were predisposed to reject psychoanalysis in the first place tended to look at dreaming as beyond the pale of scientific investigation. Rapid eye movements

Sleep Finds a Groove 305

remained invisible to Edmund Jacobson for precisely this reason. It would have been impossible for him to reconcile progressive relaxation with the idea that dreaming was a rhythmic brain activity. Jacobson thought that dreaming disturbed, rather than protected, sleep, so the idea that dreaming’s function might have observable physiological signs was inconceivable. Progressive relaxation was a method of restoring health, and part of this restoration involved the minimization or even the annihilation of dreaming. Even though Jacobson had the technical ability to detect REM periods, such a discovery could have no place in his biomedical and psychological schemas. For his part, Kleitman tended to depict sleep as an extension of relaxation, just as Jacobson did. Dreaming was little more than the intrusion of mental activity into the body’s nocturnal quiescence. Besides, after Piéron, dreaming was a problem best left to philosophers and their psychological fellow-travellers. His method of enforced insomnia defined the appropriate limits of sleep as a physiological, laboratory-based problem. Kleitman’s extension of this method to incorporate human subjects and, most notably, himself did not include references to dreaming. Sleep was a kind of human performance, the periodicity of which could, Kleitman hoped, be manipulated with the aim of increasing the health and efficient productivity of populations, be they naval watchmen or infants in their cribs. Graphical methods dominated Kleitman’s efforts, but only to the extent that they retained their late-nineteenth-century characteristic of being read as a visual record of physical movement. Kleitman’s insubstantial engagement with electroencephaolography is curious in retrospect, given the EEG’s current importance in sleep research and medicine. But the incongruity vanishes when we consider both Kleitman’s own focus on sleep motility and the EEG’s initial reception as the ‘mind’s inscription.’ The fact that the EEG cycled through different waveforms during a night of sleep was just another fact for the first doyen of sleep research, an accretion to his already well-stocked cabinet of curiosities. Dreaming’s place was not here, on public display. Although early Chicago EEG researchers acknowledged that dream recall might be associated with distinctive brainwave patterns in sleep, there was no connection to be had between this approach and the private, closed world of analyst and analysand, where ‘the vast psychiatric literature dealing with the dynamic properties of dreams’ dominated.127 It was not until EEG became an established and foundational practice that new concepts of sleep could take hold of investigators’ imaginations. Until this point, Kleitman continued to depict sleep as a relative

306

The Sleep of Others

absence of motion and consciousness. Although EEG laboratories did exist, most notably for a brief time in Tuxedo Park, electroencephalography had yet to become a way of life for investigators. New Subjects, Old Methods: The 1940s The trajectory of Davis’s investigations reveals the early hopes for the EEG in terms of its ability to classify human subjects, both for medical and for military purposes. The stability of an individual’s brainwaves was thought to reveal a hidden trait that could be correlated to an existing category, such as ‘emotionally unstable,’ ‘nervous,’ ‘calm,’ ‘dominant,’ ‘aggressive,’ and the like. Davis’s work with the EEG did not differ a great deal from Jacobson’s experiments with relaxation: they both used electrographic traces as a sort of sophisticated graphology (or perhaps even physiognomy) thought to betray the individual differences that held the key to understanding physiological and psychological performance. Sleep was only a marginal feature of such a research program. After all, what relevance could it have for the psychology of personality? Davis and Loomis had demonstrated that the EEG of sleep, like alpha, was surprisingly constant. But Loomis’s passion for detecting and measuring rhythms was set within Davis’s ambitions for the EEG as a tool for the study of individual difference – a study that relied on how people behaved, not how they retired from the world in sleep. Jacobson, and even Kleitman, adopted a similar perspective, depicting sleep exclusively as a passive state. Its relationship to normal psychology existed only insofar as it could be modified by habit, through conditioning that was either self-imposed or a product of childhood training. During the 1940s and 1950s, Kleitman expanded his research into three main areas: shift work, biological rhythms, and sleep hygiene for infants. The first was the realization of his initial suggestion that the study of sleep would be relevant to the physiology of work.128 In the late 1940s, for example, Kleitman spent several weeks on the submarine U.S.S. Dogfish, conducting mental and physiological tests on sailors to determine their performance under a variety of sleep schedules.129 He found that the sailors’ test scores tended to decrease along with their body temperatures. One of his recommendations was that the traditional system of split watches (four hours on, eight hours off) should be changed to correspond more closely to the diurnal temperature rhythm. Submarine warfare, of course, had played a major role in the Second World War, but the strategic importance of this particular class

Sleep Finds a Groove 307

of ships continued into the Cold War as the possibility of submarines carrying nuclear weapons became a reality. Kleitman’s study was part of a larger movement in human-factors research that had rapidly emerged as a key facet in military engineering.130 The need to create soldiers and machines in concert had become readily apparent in aviation during the Second World War and had helped bring the field of aviation medicine to the forefront of war research. Just as airplanes provided a unique environment in which to study, for example, respiratory physiology and the effects of g-force on reaction times, so, too, did submarines provide a sort of underwater laboratory in which the limits of sleep/wake schedules could be analysed. The study of shift work was merely an applied aspect of another of Kleitman’s long-standing projects: the study of biological rhythms. The formal origins of the study of recurrent biological changes can be traced back to the International Society for the Study of Biological Rhythms, a group of mostly medically oriented researchers from Sweden, Germany, and the Netherlands, which held its first conference at Ronneby, Sweden, in August 1937.131 Kleitman’s 1949 review of the subject carefully wrote his own research into this history, noting that his own Sleep and Wakefulness had no less than five chapters devoted to the topic of periodicity.132 Rhythm had long been an important dimension of sleep research: Piéron, in his effort to avoid the teleology embraced by Claparède’s concept of sleep as an ‘active defence,’ relied on the phenomenon of rhythm in organic and inorganic matter to fit sleep within the context of organic memory (chapter 3). Sleep was a habit, not a defence. In 1949 Kleitman offered a similar definition of rhythm, through a Pavlovian analogy: ‘A rhythm may be likened to a conditioned response, which is also individually acquired and depends on an extrinsic reinforcement for its establishment, yet will persist for a shorter or longer period of time in the absence of such reinforcement.’133 Kleitman’s work continued to characterize sleep as a rhythm throughout the 1940s. By the end of the decade, however, he began to make inroads into another field of applied research: sleep and child development. The study of infants’ sleep was a natural merger of Kleitman’s evolutionary theory of sleep, the post-war ‘baby boom,’ and the rise of child psychology in the United States. As discussed in chapter 5, Kleitman had proposed a developmental account of sleep in 1939, arguing that sleep was a passive, ‘default’ state punctuated by episodes of wakefulness that were linked to the physiological demands of nutrition and excretion. The diurnal rhythm of sleep, marked by a consolidated period of somno-

308

The Sleep of Others

lence lasting about seven to nine hours, was not a pre-determined physiological response but was instilled in the infant through ‘family and community routines.’ Kleitman had not, however, published anything in the field of infant or child sleep until the late 1940s, and he relied upon the work of others in this regard. Although it seems likely that he had paid systematic attention to the sleep of his two daughters, Esther and Hortense, it was not until 1949 that Kleitman began to formally incorporate the study of infant sleep into his research. That year, the New York Times reported that scientists at the University of Chicago had just received a $10,000 grant ‘for a study of the sleeping habits of babies.’ The money came from Swift, a meat-packing company based in Chicago, and the project was clearly geared towards the corporation’s interests. ‘The scientists will seek to determine,’ the article declared, ‘whether a 25 per cent increase in protein content of the infants’ diet will induce a more restful slumber. The babies will have specially prepared meats. An apparatus attached to the crib will record every movement made by the child, indicating the soundness of the sleep ... [the scientists] added that distraught parents might receive some welcome news this fall.’134 The researchers in question went unnamed, but there is little question that Kleitman was the central investigator: the focus on product testing for the food industry and the peculiar nature of the recording apparatus all bore his signature. Kleitman’s research came in the wake of a flurry of popular publications on childrearing, the most notable of which were authored by Frances L. Ilg and Arnold Gesell, both from the Yale University Child Development Clinic, and, of course, Dr Benjamin Spock.135 Kleitman’s first popular article on sleep hygiene for infants appeared in the early 1950s, but he had already published or was in the midst of writing a number of academic articles on the subject.136 None of them, however, was based on EEG. Although Kleitman was clearly well aware of sleep stages by this point, he obviously did not consider EEG to offer any sort of dramatically new perspective on the study of sleep. Instead, he continued to deploy his motility recorders, by now nearly three decades old, to study sleep in new subjects. In practice, if not in theory, Kleitman continued to define sleep in terms of movement. From Observation to Fact: Eye Movements, Dreaming, and the Sleep Lab It was in this context of a stable, well-defined, but rather staid research program that Eugene Aserinsky began working in Kleitman’s laboratory

Sleep Finds a Groove 309

in 1950. Like any other field of historical research, autobiographical accounts of discovery in the history of science can never be considered authoritative. But Aserinsky’s is particularly problematic, since it is marked by his strong sense of injustice at having been denied the full share of the credit for the discovery of rapid eye movements in 1953. His two published accounts need to be read in the context in which they appeared. The first is found in a 1977 sociological study that took sleep research as a case study in how scientists assess risk in making choices about research.137 The French authors invited Aserinsky to write a preface to the book, and he took the opportunity to establish his singular claim to having discovered REM. He developed this claim further in an article published two decades later, in an early volume of a journal explicitly devoted to the history of the neurosciences.138 While the former preface contained scant detail regarding the discovery, the latter contained rather too much. The editors of the journal apparently asked Aserinsky to retract several of his negative statements regarding Kleitman (who was still alive at the time). Aserinsky, clearly seeing this as his first real opportunity to make his case, refused. Much to the chagrin of a former student of Kleitman’s who had since established herself as one of the leading historians of the field, the journal’s editors ultimately accepted Aserinsky’s article, unchanged.139 For a historical study taking a ‘long view’ of the unfolding of sleep as a scientific object, this controversy poses few problems. If the task of the historian is not to dispense honours to individuals but to understand the myriad contexts in which an object takes shape, the question of ‘who discovered what,’ is a poorly formulated one. The real question, at least in this case, is how something came to be. Regardless of the veracity of Aserinsky’s personal characterization of Kleitman, we have seen that Kleitman’s efforts would probably never have produced a discovery like REM. And we have already discussed how eye movements in sleep were observed, then summarily dismissed, by Jacobson. Something needed to have changed between 1938 and 1953. Aserinsky’s own account affords some important clues as to what those changes actually involved. Aserinsky (1921–98) had an eclectic educational background: he had studied social science and Spanish and been both a pre-medical and a dental student at several different institutions, yet he had no degree when he enrolled as a graduate student at Chicago.140 He had hoped to study organ physiology, but, since cellular physiologists had come to dominate the Physiology Department at Chicago by 1950, Kleitman became his supervisor by default. Not surprisingly, Aserinsky began by

310

The Sleep of Others

taking up the problem of infant sleep, in particular, determining whether or not a specific rhythm (the rate of blinking) could be used as a criterion of sleep onset in infants. After some initial attempts, Aserinsky complained that the project was misconceived, for it was proving difficult to determine what counted as a ‘blink’ for infants in their cribs. He proposed a new project in which he would apply a binary system, comparing periods of eye movement to periods of quiescence. Kleitman agreed, and Aserinsky eventually determined that his infant subjects’ sleep was populated by periods of approximately twenty minutes’ duration in which their eyes did not move at all. Aserinsky’s observation revealed nothing about sleep onset, but since it added yet another sleep rhythm to Kleitman’s growing collection, Kleitman suggested that Aserinsky start working towards a PhD immediately, without finishing his master’s degree. Aserinsky accepted, yet, according to his autobiographical account, he recognized the risks that his choice entailed: if he failed to find something noteworthy, he could well end up with no degree at all. It was in the process of starting his doctoral research that Aserinsky adopted two novel approaches to his work. On the one hand, Kleitman encouraged his student to begin to conduct similar research on periods of eye movement in adult subjects. This was clearly an effort to link Aserinsky’s curious observations with Kleitman’s evolutionary hypothesis of sleep. Did the periods of eye quiescence in infants change as sleep patterns developed and became consolidated in the adult organism? Askerinsky’s own account of the situation, however, suggests that his supervisor was largely unconcerned with how eye movements might be linked with EEG. Kleitman’s interest in the project was not, for example, profound enough to warrant the purchase of any new equipment for this task, and Aserinsky was stuck with an archaic Offner Dynograph that he discovered ‘stored in the bowels of Abbott Hall.’141 Given Kleitman’s limited interest in and exposure to EEG research in the first place, it is perhaps not so surprising that he would not have had a more sophisticated device in use in his laboratory. In fact, Aserinsky’s personal account of REM’s discovery even claimed that Kleitman had no sleep laboratory at all at this point! But this statement is a typically ahistorical one. It assumes as a given what was really an accomplishment that followed the discovery of REM – namely, the centrality of EEG to laboratory-based sleep research. In any event, the Offner Dynograph was a pen-and-ink device, so it was at least more appropriate to cost-effective recording than the ones

Sleep Finds a Groove 311

using bromide film that had preceded it. But the machine had never been used for recording eye movements, and Aserinsky soon discovered that these were difficult to distinguish not only from skin and EEG voltage fluctuations but also from the sixty-hertz interference coming from the electrical wiring of the laboratory. In desperation, Aserinsky approached Frederic Gibbs, whose experience in recording EEG was almost unrivalled by the early 1950s. Even Gibbs, the great champion of EEG in diagnosing epilepsy, recommended to Aserinsky that he abandon the electrical recording of eye movements entirely, because the problem of distinguishing signal from noise was endemic and essentially unresolved in the electrooculogram (EOG). In its place, Gibbs advocated Kleitman’s long-standing preference, mechanical recording. Aserinsky, it seems, was not willing to give up so easily. One suspects that perhaps this student, now in his early thirties, was more seriously infected with the passion and hype that surrounded electroencephalography during the 1940s and 1950s than his supervisor, who was nearly twice his age. But the old guard of Chicago physiology was not, for all that, a rigidly homogenous group. It would not have been particularly difficult for Aserinsky, without ever leaving the city, to find someone who had long expertise in electrophysiological recording and who placed vast stores of confidence in how such instruments could literally reshape the course of physiological research. Edmund Jacobson had already established himself as an authority on EOG through his electrophysiological studies of the late 1920s and 1930s, and it was well known that he had also made some tentative suggestions about the relationship between dreaming and eye movements in sleep. You Can Sleep Well was, after all, a popular book, and its author was a well-established, if unconventional, physiological researcher and clinician. So it is hardly surprising to find that Aserinsky conferred with Jacobson when he was confronted with difficulties in recording eye potentials.142 Whether or not Aserinsky took from Jacobson the idea that eye movements could be linked to dreaming, or whether (as Aserinsky claimed) it came purely from his own desperate attempts to discover something – anything – that would earn him the PhD is, in a sense, to miss the point that Aserinsky was working in an entirely different context from Jacobson. Electroencephalography had assumed a much greater significance in both brain research and in psychiatry, and the latter had been indelibly stamped by psychoanalytic preoccupations which included interest in dreaming. During the 1930s, Jacobson could afford to be cavalier in his dismissal of dreams as investigative objects worthy of electrophysiologi-

312

The Sleep of Others

cal study. He was, after all, an independently wealthy, highly successful clinician with a well-established program of research. The EEG was then still a novel device with footholds only in select neurological diagnoses. Dreaming might have had some significance for the psychoanalytic therapy of the day, but the subject was almost entirely unrelated to experimental biology or physiology. It was of only incidental interest to even the most dedicated sleep researchers and attempts to link psychoanalysis with the laboratory-based life sciences were, as Hallowell Davis’s experiments testified, ill-advised. Laboratory-based researchers were better off to ignore dreaming in the 1930s, just as experimental psychologists had given up on hypnosis three decades earlier. Aserinsky, on the other hand, found himself in a tenuous financial, educational, and professional situation, and was afforded few luxuries. But more important, psychoanalysis during the immediate post-war period was on the brain-research agenda in a way that it had not been a decade and a half earlier. When Aserinsky saw a chance to make what he thought to be an important statement about psychoanalysis via the EEG, he took it. Following his consultation with Jacobson, he soon began his all-night sleep recordings of adults. He recognized that he needed to discover something comparable to the twenty-minute period of ocular quiescence he had recently found in infants. Lacking any apparatus similar to Loomis’s giant kymograph, Aserinsky was obliged instead to watch his archaic machine devour the requisite half-mile of paper tape, in the hope that he might notice some transformation of the electrical record as it was happening. What he found were periods not of eye quiescence but of eye movement. After having visited Jacobson and conducting a thorough literature review on the subject of eye movements, Aserinsky was well prepared to relate this phenomenon to dreaming. Psychoanalysis and the Triumph of Investigational Practice Aserinsky’s desperate search for ‘something’ in his darkened sleep laboratory was shaped by two emerging practices. The first was the EEG’s ability to create novelty out of nothing. This had already been well noted in the discovery of sleep stages, which presented a brain at work where investigators had formerly thought there was only the absence of consciousness. It was natural, especially for someone whose study of sleep had begun only after this discovery, to anticipate that additional applications of the method of all-night recording could produce additional phenomena, if only one looked hard enough. But what should

Sleep Finds a Groove 313

one look for? A growing tendency among scientists of mind and brain was to extend their authority into a controversial domain by applying the techniques and technologies of brain research to the study of psychoanalytic concepts. Such boundary work, in which both the authoritative scope of particular fields and the so-called norms of the scientific enterprise are redefined, found a new home in the EEG laboratory of the early 1950s.143 These efforts to bring the evidence of the physiological laboratory to bear on psychoanalysis were a small part of the exceptional growth that took place in the sciences of mind and brain during the 1950s.144 Although the chemical theory of nervous transmission had been proposed some five decades earlier, it was only in 1953, when the renowned neurophysiologist and future Nobel laureate John C. Eccles finally recanted his belief in a purely electrical theory of nerve-impulse conduction, that it could be said that the theory became established fact. This new emphasis on neurochemistry was paralleled by a return of the localizationist hypothesis. The Horsley-Clarke stereotaxic instrument (the ‘stereotack’), revived in the late 1920s by S.W. Ranson at his Institute of Neurology at Northwestern University in Evanston, just north of Chicago, gave neurosurgeons a precise and standardized means of identifying local centres of brain activity below the level of the cortex. Regulatory centres became the main target of such investigations. A crucial development here was the 1949 announcement by Giuseppe Moruzzi (1910–86) and Horace Magoun (1907–91) that the reticular formation was the centre responsible for blocking afferent stimuli during sleep.145 In itself, the idea of a subcortical regulatory centre for sleep was hardly new. Based largely on neuroanatomical studies and post-mortem evidence, Economo had proposed such a region in the 1920s. A decade later, the Belgian neurophysiologist Frédéric Gaston Bremer (1892– 1988) demonstrated such a region through two neurophysiological ‘preparations’ in which he transected a cat’s brain either at the lower medulla (encéphale isolé) or at the midbrain, just behind the origin of the oculomotor nerves (cerveau isolé).146 The cuts effectively isolated highly specific areas of the cat’s brain from its body. As Kleitman noted a few years later, many investigations had been made ‘on the behavior of an animal without a brain. Bremer studied the behavior of a brain without an animal.’147 But this was not quite true: Bremer had observed the extremely passive and lethargic behaviour of his cats following such operations, and postulated that the isolated regions interrupted sensory stimuli, thereby causing sleep. In Chicago two decades later, Magoun

314

The Sleep of Others

and Bremer’s student, Moruzzi, coupled the stereotack with the EEG to offer another interpretation of these classic ‘preparations.’ They electrically stimulated the isolated brain regions and observed EEG patterns now devoid of any ‘noise’ produced by the animal’s motor reactions. More important, they observed a regular alternation between the fast, low-voltage EEG of wakefulness and the slow waves typical of sleep in encéphale isolé, which disappeard in cerveau isolé. They concluded that the reticular formation was a region of the brain associated with ‘cortical tone’ that lay outside traditional sensory pathways and that was responsible for regulating sleep in the absence of any sensory stimulation whatsoever. Their results were published in the first volume of what rapidly emerged as the most important EEG journal in the world, EEG and Clinical Neurophysiology.148 The ability to visualize the brain apart from behaviour led several leading American neuropsychiatrists to argue that neurophysiological methods might eventually provide a biological basis for psychoanalytic ideas. Another early EEG researcher, Stanley Cobb, for example, argued in 1949 that Freud’s categories of ‘Super Ego,’ ‘Ego,’ and ‘Id’ could be directly mapped on to brain areas recently identified by neurologists and neurophysiologists (fig. 27).149 Cobb was at Harvard at the time, but Freud’s influence was just as strong in Chicago, where Magoun and Moruzzi were working. Magoun, an experimental neurologist at the Illinois Neuropsychiatric Institute as well as at Northwestern, agreed that psychoanalytic theory could, just like evolutionary concepts, be readily incorporated into current brain research.150 While writing histories of their own field decades later, investigators like Magoun effectively purged their accounts of the past by failing to give any forceful role or even mention to such speculations.151 In his own retrospective account, Aserinsky at least gives such ideas a negative influence. Following his observations of periods of eye movement associated with dreaming, he asked Dr Nathaniel Apter, who headed the Psychiatry Department at Chicago, whether or not infants, in whom he had not yet observed similar periods of eye motility, dreamed. His response, according to Aserinsky, ‘was in the affirmative and predicated entirely on Freudian concepts.’ The anecdote served to establish Aserinsky’s own heroic selfaccount: ‘As an experimentalist,’ he mused, ‘I considered his reliance on Freud as a sort of religious faith, and therefore his answer was of no value in helping me understand why I had apparently not seen REM in infants.’152 As the example of Hallowell Davis indicates, there was nothing about

Sleep Finds a Groove 315

27 Stanley Cobb was one of a number of neuropsychiatrists who argued for a synthesis of psychological and physiological approaches to mental activity. (Stanley Cobb, ‘Human Nature and the Understanding of Disease,’ in Nathaniel W. Faxon, ed., The Hospital in Contemporary Life [Cambridge, Mass.: Harvard University Press, 1949], 108–36, 113)

being an experimentalist that somehow prohibited the pursuit of developing psychoanalytic theory. Davis’s work showed how it was possible to be part of the inner circle of neurophysiology and yet still maintain an active research program inspired by psychoanalysis. Dreams were an integral and highly visible part of psychoanalytic therapy, making them an active and contentious arena of investigation in a way they had not been fifty years earlier. During the early 1950s, a Jungian psychologist, Calvin Hall, published the first major systematic survey of dreaming since the American psychologist Mary Calkins and her student, Grace Andrews, had published their survey of psychology students’ dreams in 1900.153 The scope of Hall’s project was a curious mismatch with his

316

The Sleep of Others

conclusions. He had collected over ten thousand dreams by questionnaire and, following an analysis of their themes, plots, settings, and characters, concluded that dreams were far more mundane and much more like waking life than Freudians had led people to believe. Even for psychoanalysts, dreams were becoming domesticated as standardized objects of investigation, quite unlike the rare and exotic creatures of earlier ages. The extensive currency of psychoanalysis in the United States during this period, coupled with a new and highly rationalized view of the dream, made a correlation between dreaming and physiological activity a potential object for any physiological researcher, regardless of whether or not they approved of psychoanalysis. It is not surprising, then, that once Aserinsky had determined that adult sleep was punctuated by several periods of eye movements, he decided to place the pursuit of an association between these periods and dreaming ‘very high on the agenda for further exploration.’154 Kleitman, Aserinsky later recalled, remained quite sceptical about the existence of such a relationship. This might seem odd, given that Kleitman had coauthored an article almost fifteen years earlier that considered the possibility of a link between dreaming and brainwave patterns. But the two men’s investigational contexts were literally generations apart. Kleitman’s interest lay fixed in mental performance, and the ability to recall dreams was beyond the pale of such a research program. For him, dreams were no more legitimate objects of physiological experiment than they had been for Jacobson. Indeed, according to Aserinsky’s memoir, Kleitman himself performed rather poorly as Aserinsky’s experimental subject. The second night Kleitman spent in Aserinsky’s makeshift sleep laboratory, he exhibited three different episodes of REM but acknowledged that he had been dreaming only once. This was an improvement over his first night, during which time Aserinsky had been able to record no REM periods at all. But it was hardly definitive. Despite his resistance to the idea, Kleitman recognized that the project had some merit. So he encouraged Aserinsky to present his research at the upcoming meeting of the Federation of American Societies for Experimental Biology (FASEB), which was to be held in Chicago in 1953. Before the meeting, however, Kleitman insisted that his daughter, Esther, be used as an experimental subject. Aserinsky later claimed that he ‘could not fathom’ why Kleitman ‘would subject his daughter to an experience that was less than pleasant,’ but this perhaps indicates only changes in experimental mores and Aserinsky’s lack of awareness of earlier traditions.155 The use of family members as subjects

Sleep Finds a Groove 317

28 The first publication of graphically recorded rapid eye movements. Aserinsky and Kleitman calibrated vertical (RV) and horizontal (RH) eye movements to body motility and to EEG (RF). (Reprinted with permission from Eugene Aserinsky and Nathaniel Kleitman, ‘Regularly Occurring Periods of Eye Motility, and Concomitant Phenomena, during Sleep,’ Science 118 [4 September 1953]: 273–4, copyright 1953, AAAS)

for sleep research was hardly unusual – earlier investigators had typically used whatever subjects were available. Hallowell Davis used his children, his colleagues, and Loomis’s servants as subjects at Tuxedo Park. Kleitman had observed his daughters’ sleep on numerous occasions, and Esther, who was almost thirty by this time, had conducted numerous experiments with her father and would later published one paper with him and help to prepare the second edition of Sleep and Wakefulness.156 No doubt, Nathaniel thought that his daughter might serve as the most reliable witness then available to confirm Aserinsky’s suspicions. These initial trials must have been successful, because Aserinsky made the first public announcement of REM at the FASEB meeting in 1953.157 In their Science paper that appeared a few months later, Kleitman and Aserinsky used the EEG to show not only that rhythmic eye potentials were not artifacts of bodily movement or of EEG but that these potentials were clearly related to a low-voltage brain wave pattern (figure 28).158 In

318

The Sleep of Others

case the EEG provided insufficient evidence that the subjects were actually sleeping, Aserinsky filmed his subjects and indicated that he had seen the sleeping subjects’ eyes move beneath their lids directly, without the aid of any instruments. Most important, the authors made it clear that they felt eye movements could now be definitively related to periods of dreaming. Although only ten of the twenty experimental subjects had been questioned as to whether or not they had been dreaming, of the twenty-seven times those ten subjects had been interrogated during ocular activity, dreams ‘usually involving visual imagery’ had been reported twenty times. The remaining seven times were described as ‘failure of recall’ or else ‘the feeling of having dreamed.’ In contrast, when subjects had been awakened and questioned during periods of eye quiescence, twenty-three interrogations produced nineteen incidences of ‘complete failure of recall,’ with only four reports of either having dreamed or the feeling of having dreamed. Eye movements in sleep had received cursory attention by physiologists and psychologists long before 1953, but Aserinsky and Kleitman were the first to document and comment on the regularity of such movements.159 Earlier, eye movements had been seen as nothing more than irregular episodes of night life. And so the simple fact, discovered by Aserinsky and Kleitman, that there were regular periods of REM during sleep would have been an important contribution to sleep physiology. But the discovery went beyond this, for Aserinsky’s and Kleitman’s decision to relate REM to dreaming made it part of the expansionary program of neurophysiology in the post-war period. By calibrating REM against the sleep stages of the EEG, Aserinsky and Kleitman had turned dreaming into a physiological event suited to the new era of autoregulation: dreaming was a rhythm, rather than a response. All that said, this concept was still undeveloped in 1953, and it is tempting to cite biographical reasons for this. Aserinsky’s relationship with Kleitman had been a rocky one at best, and as soon as he completed his dissertation in mid-1953, he moved on, taking a one-year research position at the School of Fisheries at the University of Washington in Seattle and then obtaining a position at Jefferson Medical College in Philadelphia, where he remained until 1976. But REM did not remain undeveloped simply because Askerinsky had left the field of research or because of Kleitman’s lack of interest in the topic. If the citation record is any indication, the research itself simply did not fit in the popular or in the scientific imagination. It attracted very little scientific interest – Aserinsky and Kleitman’s two 1953 papers received a

Sleep Finds a Groove 319

29 Citations of Aserinsky and Kleitman, ‘Regularly Occurring Periods of Eye Motility’ (SCI and SSCI combined). This paper in Science was cited less than twenty times in the decade following the 1953 announcement of their discovery of REM.

mere handful of citations until the mid-1960s (see fig. 29). In contrast, James Watson and Francis Crick’s revelation of the structure of DNA (published that same year in the journal Nature) had been cited more than sixty times by the end of 1954. Press coverage was also minimal. According to Aserinsky, Kleitman asked him to hide behind a door during a press interview to offer corrections to his supervisor later, if need be.160 This event, if it actually transpired as Aserinsky claimed, generated little more than ill will between him and his supervisor. The New York Times relegated Kleitman’s statement that ... ‘“it will be possible, for the first time, to obtain objectively data on incidence, frequency and duration of dream episodes and to relate a dream pattern to other characteristics of personality or living habits”’ to a minor column on the back pages.161 The immediate future of REM lay in the movement to link psychiatric practice with neurophysiology, something that engaged neither Aserin-

320

The Sleep of Others

30 Michel Jouvet and William Dement in San Franciso, California, 1964. (Courtesy Michel Jouvet)

sky nor Kleitman, both of whom lacked medical degrees. In December 1952, a young medical student named William Dement (b. 1928) joined Aserinsky in Kleitman’s laboratory (figure 30). Dement, who had a keen interest in psychiatry, had heard Kleitman lecture on sleep, and asked to work with him. Kleitman immediately threw him in with Aserinsky, who was then just beginning to rework his research towards eye motility in adults. Dement’s first paper on REM appeared in 1955, just as he was finishing his MD and well after Aserinsky had left the laboratory. Within five years, he had established three major directions in which experimental work on REM would later develop: the comparative study of REM in normal and abnormal subjects, the attempt to link eye movements to dream content, and the examination of the effects of REM deprivation. In Dement’s hands, REM became the bridge that finally united Kleitman’s sleep laboratory with neuropsychiatry at Chicago. Aserinsky suggested on several occasions that the real purpose of Dement’s first paper on REM was to validate the original work he and Kleitman had published in 1953, and perhaps even to satisfy his supervisor that the results ‘were not a fluke.’162 But this is not entirely accurate. Dement’s purpose was rather to extend REM into psychiatric practice by

Sleep Finds a Groove 321

using it to compare the physiology and psychology of normal and schizophrenic subjects. This clinical application of REM, replete with psychoanalytic overtones, gave it an interdisciplinary scope that had been completely lacking in Aserinsky’s research. Instead of going to Nathaniel Apter for psychoanalytic speculation, Dement asked him for experimental subjects. The schizophrenics Apter allowed Dement to ‘use’ turned out to have the same REM periods as normal subjects. The difference was in their description of their dreams, which, Dement claimed, featured ‘isolated, inanimate objects, apparently hanging in space, with no overt action whatsoever.’163 Their inability to offer proper dream narratives after being awakened during a REM period was, for Dement, an indication of their mental pathology, which he illustrated by citing their impoverished accounts of ‘hats’ and ‘ripped coats.’ But this sort of investigation, thought Dement, offered more than just another diagnostic tool: such research, he hoped, might shed light on the very nature of schizophrenia, because the awakening of patients during REM sleep could evoke a kind of dream material that was inaccessible in regular psychoanalysis. After receiving his MD, Dement continued to push on the boundaries of psychiatry and physiology. He stayed in Kleitman’s laboratory, earning his PhD in 1957, and then interned for a year, becoming a resident fellow in psychiatry in 1958. In the process, he published another two papers, this time in the Journal of Experimental Psychology. Here, Dement began to develop what he later called the ‘scanning hypothesis,’ arguing that eye movements in sleep were actually following the activity retrospectively described in a dream report.164 Dement’s ‘scanning hypothesis’ was a clear attempt to drive REM beyond its role as a physiological index of dreaming and relate it to the actual content of dreams themselves in normal subjects. The hypothesis was highly contentious, because it tended to rely on isolated incidences of high correspondence between dream reports and eye movements awash in a sea of dream narratives, bearing little or no relationship to the recorded data. In sharp contrast to the work of Hall on statistical analysis of dream reports, Dement followed the spirit of nineteenthcentury physiological psychologists by reasoning about the nature of dreams on the basis of sparse but carefully generated examples. Three records showing prominent vertical movement of the eyes readily corresponded to dreams of a hoist going up and down a cliff, climbing a series of ladders, and shooting baskets; records with mostly horizontal movement linked easily to a dream in which two people were throwing tomatoes at each other. These attempts to develop REM as a form of

322

The Sleep of Others

dream recording evoked some quite pointed responses. In a famous essay, Norman Malcolm, a philosopher of mind and disciple of Wittgenstein, argued that physiologists like Dement had completely misunderstood the normal usage of the verb ‘to dream.’165 Such a term, argued Malcolm, could never be uttered in the present tense (‘I’m dreaming’) because sleep was, by definition, a state of unconsciousness. The meaning of the verb could be inferred only from the meaning assigned to the act of describing or reporting a dream. Rapid eye movements certainly reflected some sort of physiological activity, but it did not represent what was ordinarily meant by dreaming.166 David Foulkes, a psychologist who began working in Kleitman’s laboratory shortly after Dement, similarly argued that dream reports could be elicited from non-REM and REM periods alike. The nature of the reports differed, but they were equally incidences of ‘mentation’ during sleep. In other words, laboratory research was not so much documenting incidence of dreaming as it was changing the very meaning of the term.167 These critiques, however, did nothing to prevent a growing number of psychologists and neuropsychiatrists from using REM to develop their own laboratory-based programs of dream research. The EEG was a familiar clinical tool by this point, and it took little technical effort or skill to add two more channels to a physiograph in order to measure eye movements. Because the detection of eye movements relied on the application of a well-established technology, REM sprang up everywhere there was an interest in the systematic study of dreaming. Many of the most outspoken critics were obliged to experiment with REM in order to counter Dement’s ideas. Calvin Hall, for example, founded an ‘Institute of Dream Research’ in Florida in 1962, received several grants from the National Institute of Mental Health, and began to study REM dreams. In 1966 he published the first of a series of works attacking the very idea that dreams recorded in sleep laboratories could be considered representative of the normal dreams of home.168 The former, Hall argued, were unusual and anomalous, because they often featured conflicts in which the sleep laboratory predominated as a symbol. But even though Hall was defending his survey-based method of content analysis by attacking the status of REM as an investigative tool, the fact remained that he was really doing little more than extending the reach of REM, just as Dement had initially hoped: ‘It seems reasonable to conclude [wrote Dement] that an objective measurement of dreaming may be accomplished by recording REM’s during sleep. This stands in marked contrast to the forgetting, distortion, and other factors that are involved

Sleep Finds a Groove 323

in the reliance on the subjective recall of dreams. It thus becomes possible to objectively study the effect on dreaming of environmental changes, psychological stress, drug administration, and a variety of other factors and influences.’169 Laboratory investigations of REM were starting to invert the significance of dreams. Whereas Freud had once proposed that dreams were the surest point of access to the unconscious mind, Dement was now arguing that dreams, like temperature or blood pressure, could signify internal psychophysiological changes. REM had become one of a battery of indicators of health that relied on the graphical method of visualization to depict physiological self-regulation. In his next major article, Dement pushed REM further into the mainstream of physiological investigational practice.170 Here, he presented his theory of ‘dream deprivation,’ in which he argued that a certain amount of dreaming – equivalent to the sum total of all REM periods in a single night – was required for psychophysiological health. When deprived of this ‘dream time,’ subjects’ brains would automatically and consistently attempt to increase their REM sleep on subsequent nights. If the deprivation continued, the subjects would begin to suffer adverse psychological effects. In its essence, Dement’s experiment was little more than Piéron’s and Kleitman’s method of ‘experimental insomnia’ adapted to the use of EEG and the subsequent discoveries of sleep stages and REM. The method employed a classic technique of physiological experiment, in which the normal function of a organic phenomenon was examined by eliminating it and observing the results. In terms of REM sleep, this was a relatively simple but extremely time-consuming strategy. Each of Dement’s subjects was obliged to endure five consecutive nights of ‘dream deprivation,’ in which they were awoken every time they entered a REM state, determined by characteristic eye traces and an active, lowvoltage EEG pattern. They were then allowed a number of ‘recovery nights,’ in which their sleep was monitored in the laboratory but not disturbed. An identical routine was then employed targeting nonREM sleep. In all, the experimental trial involved between twenty and thirty nights for each of the eight subjects, including several nights of obtaining baseline values for each subject. Dement found that subjects experienced a dramatic increase in their ‘dream attempts’ after REM deprivation. That is to say, time spent in REM increased dramatically in the nights immediately following the periods of deprivation. The average amount of time spent in REM, which Dement described as ‘dream time,’ had been eighty minutes out of the total six hours of sleep time,

324

The Sleep of Others

or 19.5 per cent. In the recovery night that followed the first series of deprivation experiments, REM sleep rose to 29 per cent of the total sleep time. A similar disruption of sleep patterns was not observed when subjects were deprived of non-REM sleep. Equally significant were the psychological changes Dement observed. Anxiety, irritability, and difficulty in concentrating typically appeared after the periods of ‘dream deprivation.’ Non-cooperation also became a notable factor; Dement noted that one subject ‘quit the study in an apparent panic,’ and two others were able to endure only four nights of interrupted sleep, rather than the five that had been adopted as experimental protocol. While conceding that none of these changes was ‘catastrophic,’ he concluded that ‘it is quite possible that if the dream suppression were carried on long enough, a serious disruption of the personality would result.’171 A new vision of the functional significance of dreaming was emerging – one that treated the duration of dreaming as a measurable component of physiological and psychological health. Dement had taken the tools, concepts, and practices formerly limited to that last of the nonnaturals, sleep, and used them to turn Freud’s claim that dreams protected sleep into a claim that dreams themselves needed protection. Despite Dement’s explicit psychoanalytic slant, his conclusions were a far cry from Freud’s insistence that whatever the subject reported as a dream had to be taken as such by the analyst, regardless of what may have been altered or forgotten in the process of recollection. For Dement, censorship and repression were entirely displaced by the immediacy of the EEG record. The same technique dispensed with Freud’s argument that a comparative study of the manifest dream report and the latent dream wish could reveal the function of dreaming. The rhythmic nature of REM periods, coupled with the phenomena of ‘REM rebound’ following periods of deprivation, seemed to demonstrate that dreaming served a function much more direct than Freud had thought. It was not the case that dreams protected sleep; it was rather the other way around, in that sleep’s rigid temporal architecture, invisible before the EEG, protected dreams. Whereas Aserinsky and Kleitman’s original discovery intimated that the discovery of REM could completely replace psychoanalytic investigation, Dement’s proposed a novel interpretation of the field. The appeal of the latter was immediate and profound and ultimately led to the creation of the sleep clinic as we know it today.

7 Begin the Begin

On its own, the discovery that dreams could be calibrated to eye movements in sleep was not enough to recreate sleep for the laboratory. The concept of ‘dream deprivation’ helped link REM to psychoanalytic concepts, but this alliance between the laboratory and the couch was tenuous at best. As some investigators quickly realized, the reinvention of sleep as a scientific object required a particular kind of social organization to promote such an end.

‘I find that I am dreaming with increasing clarity (lucidity?) but they remain mysterious. I cannot connect them to [my] “emotional” domain (either latent or immediate). Further, try as I will, I cannot make connections to my “memory” domain (either remote or immediate). And, most remarkably, few of the participants are familiar. That latter fact has always been the most puzzling aspect of dreams that no “theory” seems to deal with effectively.’1 So wrote Wilse B. Webb (b. 1920), a psychologist at the University of Florida, Gainesville, in a letter around 1990. The tone of the letter is a testament to the enormous changes the study of sleep and dreaming had undergone. A century earlier, psychologists like Ladd would have readily incorporated such introspective observations into the very core of their experimental studies. A few decades later, a psychoanalytic psychiatrist would almost certainly have taken his own inability to make sense of his dreams as grist for self-analysis and professional development. But Webb seems resigned to the private nature of his comments, sharing them only with his correspondent, David Foulkes (b. 1935), a fellow psychologist and sleep researcher at the University of Wyoming. This is not quite true, of course. The letter was placed by Webb into

326

The Sleep of Others

the archives of the Associated Professional Sleep Societies (APSS), which he himself had helped to establish at the University of Chicago. It is perhaps not too much to suggest that Webb, a historian in his own right, was somehow acknowledging that concerns about one’s own dreams were objects better suited to the historian than the experimental psychologist. They could be rendered public only when locked into the specific chronology and context that makes up an individual’s life. If this interpretation is right, it is little more than an extension of what Webb had been doing since the early 1960s, when he helped create the Association for the Psychophysiological Study of Sleep. This organization, which no longer exists, was a predecessor of the current APSS and the first group to be signified by that abbreviation. The transformation of the organization, from a group dedicated to studying sleep’s psychophysiology to a society of professionals whose main interest was in the clinical aspects of sleep, encapsulates sleep’s shift from the personal to the public. The APSS, in both its guises, effectively divorced the experience of sleep from its orthodox investigation. Under its rubric, those who studied sleep did so as an organized group, not as individual researchers working in isolation. And the thing they studied – sleep – became a property of populations whose sleep was defined not as private experience but as a problem of public health. Dreams’ Decline The nature of this transformation almost certainly took many of its participants by surprise. The discovery of REM was not immediately taken up by a mythical horde of crypto-sleep researchers, eagerly awaiting the discovery that would finally unite their field. Such a group did not yet exist, and would not for some time. Still, important threads of this fabric could already be found. Neurological research was just beginning to delve into deep brain structures and their regulatory role in various brain states, thereby seeming to sidestep a naive reduction of all consciousness and behaviour to the activities of precisely defined cortical regions. The influence of psychoanalytic concepts within the technocratic, eclectic, and pragmatic values of American psychiatry made ample room for projects that defined dreaming according to laboratorybased parameters. The EEG grounded REM in the problem-based, interdisciplinary aims established in American biomedical science established in the 1920s and 1930s. From this latter perspective, REM seemed like an extension of the biomedical technology that had already solved

Begin the Begin

327

several chronic neurological puzzles, epilepsy among them. Perhaps dreams were next. Perhaps the limits to how deeply science could penetrate subjectivity were not as severe as once thought. Perhaps Descartes’s supposition, whereby dreams appeared as the most rudimentary evidence that a realm of pure subjectivity really existed, could be demonstrated false. Over the decades that followed REM’s discovery and elaboration, however, the law of diminishing returns seemed to apply to this field no less than it did to any other. Even as the circle of self-identified ‘sleep and dream researchers’ expanded rapidly during the 1960s and 1970s, significant new discoveries about dreaming were not forthcoming. Dement’s vision of an absolute correlation between eye movements and dream images proved controversial and difficult to demonstrate, even though this notion continued to resonate throughout the field and in the popular press. Likewise, his idea that the ‘REM rebound’ effect represented a ‘pressure for dreaming’ seemed incomplete, since it was unable to explain the ontogeny of REM sleep and could not account for the fact that depressed patients seemed to benefit from treatment with tricyclic drugs, despite the REM suppression that such drugs produced. Changes in the nature and direction of the research were perhaps more radical than the discoveries themselves. Although REM’s origins lay in physiology, laboratory-based investigations of sleep and dreaming drew closer to psychiatry and neurology throughout the 1960s. As we have seen, physiologists initially paid little heed to REM. Aserinsky, in fact, was obliged to defend his doctoral work in front of a panel of cellular physiologists, most of whom took little interest in the gross functioning of the brain as registered by the EEG. Like Kleitman, Aserinsky worked in considerable isolation and had a hard time motivating fellow physiologists to participate in his research.2 As an MD, however, Dement had access to subject populations far beyond anything Aserinsky could possibly muster. Just as important was Dement’s use of schizophrenic patients as subjects and his setting of REM within a diagnostic context, which made the problem of reliable testimony about dreaming irrelevant. Whether or not his subjects’ dreams actually corresponded to what they reported was relatively unimportant. What counted were the differences between what schizophrenics and normal subjects said when they were woken up in the middle of a REM period. Although REM did not turn out to be a particularly useful diagnostic marker for madness, Dement’s application of Aserinsky’s investigative procedure to problems in psychiatry and psychology nonetheless foreshadowed the future

328

The Sleep of Others

direction of sleep and dream research in the United States. The longstanding association between dreaming and madness was not to be abolished overnight, as it were. Instead, Dement’s ‘dream deprivation’ hypothesis traded on the Freudian notion that dreams had a function in maintaining a veneer of sanity laid overtop the irrational unconscious. This idea eventually resonated with psychiatrists and psychologists alike, both of whom had substantial access to potential subjects in the form of patients and students. Naming the Field Not surprisingly, this clinical direction was only barely perceived by early REM researchers. After all, sleep had precious little in the way of distinctive pathologies. Encephalitis lethargica had practically disappeared after the 1930s. Narcolepsy was still considered an exotic condition that was frequently misdiagnosed or attributed to interpsychical conflicts that were beyond the pale of experimental investigation. Hysteria and its analogue, hypnotism, were long gone, and with them the source of earlier clinical interest in sleep. Epilepsy and coma had been definitively separated from sleep through electroencphalography. Insomnia was common enough, but its routine appearance as a clinical symptom merely betrayed the existence of a range of mental disorders, such as depression, anxiety, or mania, none of which had any particular affinity with sleep. The essence of sleep, as the historian of medicine Georges Canguilhem once observed, was ‘to let life go without calling it to account.’3 Sleeping implied an ability to sink beneath the threshold of concern about one’s health that grounded a patient’s symptoms. So sleep could never itself become the foundation of pathological knowledge; its disturbance would only indicate that some unrelated pathological process was at work. The trajectory of sleep research, if we examine it from the perspective of the history of the APSS, appears somewhat differently than what Canguilhem might have predicted. Rather than beginning with pathological symptoms generated from the consciousness of the individual patients, and then moving towards an experimental science of physiology, sleep research inverted this path. It began, in the early 1960s, with an interest in physiological mechanisms and experimental methods, and only later moved into the clinic, once sleep had been established as an active, dynamic process that could be visualized and captured without recourse to subjective symptoms. Early REM investigators saw themselves as carv-

Begin the Begin

329

ing the tiny niche of sleep and dream research out of the larger field of psychophysiological research, which was itself just beginning to take institutional shape in the United States. The first edition of the ‘Psychophysiology Newsletter,’ for example, appeared in 1955, the same year Dement published his first paper on REM. Six years later, the first meeting of what would soon be known as the Association for the Psychophysiological Study of Sleep took place, just a year after the creation of the Society for Psychophysiological Research (SPR). The term ‘psychophysiological’ is rarely used by sleep researchers today, and the word has been dropped from the American version of their formal organization altogether. But how did researchers use the term ‘psychophysiological’ in the 1950s and 1960s? Psychophysiology has generally been ignored by historians of medicine who have successfully analysed the process of ‘molecularization’ and the transformation or wholesale reinvention of biomedical fields such as immunology, genetics, and microbiology.4 A parallel development in physiology – the creation of ‘neuroscience’ out of cognitive psychology, behavioural science, neurology, and neurophysiology – has recently gained more sustained attention. But psychophysiology is nowhere to be found in these accounts, perhaps because it adopted the automated aspects of the new biomedicine while resisting its ‘miniaturization.’ Rather than examining ever-smaller entities, psychophysiology focused on the use of ‘field surveillance’ to make the behaviour of entire organisms subject to scientific investigation. Developing systems of recording ‘the covert proceedings of the organism relevant to a psychic state or process ... with minimal disturbance to the natural functions involved’ was a key goal for psychophysiology and effectively differentiated it from other modes of psychological or physiological investigation.5 According to the first president of the SPR, Chester W. Darrow (1893–1967) of Chicago’s Institute for Juvenile Research, recording would one day enable psychophysiology’s reach to extend to the furthest depths of human activity: As we turn and view the exponential future progress of science in general and psychophysiology in particular we quail at the attempt to portray the possible psychophysiological achievements of tomorrow, the 21st century. The technology of recording and analyzing data will by then have reached such an ideal of efficiency and reliability that standardization will be inevitable. The investigator will be able to buy his data-gathering, computation and analysis ready-made, in a box, on a rack, or maybe in a beautiful table

330

The Sleep of Others

model. By then the registration of organic changes will have circumvented the ball-and-chain limitations of laboratory instrumentation. No longer shall we have to immobilize the subject on a cot in a sound-proof room. He will be completely ambulatory. Adorned with pellet-size teletransducers he will face the realities of life on the stock exchange, in the boudoir, or on the battle field. The psychophysiologist’s problem will be, at a safe distance, to trail him with portable recording equipment, to track him adequately from the central control tower of the laboratory, or to transcribe later the portable tape concealed in the subject’s left hand coat pocket. The subject’s major problem will be to differentiate his real thoughts from the ones he is supposed to have, and to conceal his richer inner life from the all-seeing eye of the electropornograph.6

Darrow’s prognosis was obviously tongue-in-cheek, but the hope of using recording to overcome the artifice of the laboratory was taken quite seriously. It also had some deep historical roots in the work of physiologists like Marey and Mosso, as discussed in chapter 2. The term ‘psychophysiological,’ however, had a more specific meaning for the small group of investigators who gathered together at the Billings Hospital in Chicago on the weekend of 25–26 March 1961. The meeting had been convened largely through the efforts of Kleitman’s former student, Allan Rechtschaffen (b. 1927), and William Dement, who had by then relocated to Mount Sinai Hospital in New York City. Although some participants’ retrospective accounts of this meeting claim it to have been dedicated to the study of sleep, it is clear that dreaming dominated the official agenda, which was formally described as a ‘Meeting of Researchers in Field of EEG and Dreams.’7 In a letter inviting a colleague to participate in the upcoming gathering, Rechtschaffen explicitly called it ‘a meeting of dream researchers’ who were working in ‘the area of EEG and dreaming.’8 Indeed, the only session devoted to sleep at the 1961 meeting concerned the standardization of scoring EEG stages. The remaining three sessions dealt with the relationship of the EEG stages to the ‘verbal report,’ dream theory, and ongoing research. But things soon changed. As early as December 1961, Rechtschaffen had begun circulating a letter to participants in the March meeting, describing the new group as ‘The Association for the Psychophysiological Study of Sleep.’9 The numbers of participants was growing at exactly the same time that the association was defining the ‘psychophysiology of sleep’ as its main object. By the time a formal structure emerged in 1964, the APSS had gone from a casual group of about

Begin the Begin

331

twenty-five to nearly two hundred participants.10 But what had happened to dreaming in the interim? For many members of the burgeoning society, the term ‘psychophysiological’ provided a bulwark against some aspects of psychoanalytic theory. The majority of the participants in the 1961 meeting were either psychiatrists or psychologists, and many of them conceptualized REM in terms of the Freudian theory that they had gleaned from their formal education or, in Dement’s case, through their own readings of Interpretation of Dreams.11 Their interpretation of Freudian theory was, in some sense, unorthodox: it tended towards an extreme hypostasis, treating dreams as static objects capable of mechanical verification, rather than as dynamic products of a therapeutic encounter (as discussed in chapter 3). The use of EEG to localize the process of dreaming and evoke dream narratives in a laboratory seemed to render dreams into reliable, standardized signs of mental illness. Dement and Charles Fisher (1908–88), a senior psychiatrist at Mount Sinai, began moving in this direction early on. At the 1961 meeting, they presented a report on a psychotic twentyfive-year old veteran who suffered from a series of sleeping spells and fugue states, which eventually developed into paranoia.12 Fisher and Dement rejected the previous diagnosis of narcolepsy and decided the patient suffered from a psychosis. The patient, who received Stelazine (an anti-psychotic drug) and psychotherapy, readily agreed to sleep in the laboratory ‘where he always felt comfortable and protected,’ according to his psychiatrists. Fisher and Dement acknowledged that this manifestation of transference allowed them to take eighteen nights’ worth of EEG recordings, which led them to the conclusion that their patient was dreaming too much: he spent 50 per cent more time in REM sleep than did normal subjects his age. Other psychiatrists offered less heterodox interpretations. Frederick Snyder, Walter Hamburger, Harry Trosman, and William Offenkrantz, for example, all presented papers on the psychoanalytic interpretation of dreaming at the 1961 meeting. Hamburger, from the University of Rochester, argued that the fact that REM first emerged in Chicago indicated the continuous line that could be drawn from the work by Alexander at the Chicago Psychoanalytic Institute, which dated back to 1932.13 But several REM researchers took issue with this appropriation of their work and described the psychoanalytic presentations as ‘incomprehensible.’ Calling their own work ‘psychophysiology,’ a group of prominent investigators soon emerged who contrasted what they depicted as the vagaries of psychoanalytic language with the technical

332

The Sleep of Others

precision of their instruments and their procedures for scoring EEG records. Led by Rechtschaffen and his students, the ‘Chicago School’ dismissed the psychoanalytic reification of the therapeutic context of dream investigation as readily as they had rejected philosophical concerns about how their new experimental context was changing the definition of dreaming.14 In their eyes, neither Freud nor Wittgenstein could undermine the authority of the graphical record. By invoking sleep, rather than dreaming, the APSS’s very name carved out a new space that distinguished laboratory-based research from any Freudian or behaviourist antecedents. The psychophysiological nature of their investigations allied their work with the study of organic systems over that of mere behaviour. This new form of ‘basic research’ – the term ‘neuroscience’ had not yet been invented – would tolerate criticism of the laboratory context for the study of dreaming and its analysis of ‘verbal behaviour’ in dream narration, but only barely. Its emphasis was on defining the parameters of normal sleep through an intercalibration of physiological measures in sleep and wakefulness, and an intracalibration of the various sleep stages, whose function remained mysterious. With normal sleep thus defined and the relationship between REM and dreaming deemed to have been satisfactorily resolved, the new leaders of the APSS now turned towards an analysis of sleep’s mechanisms and its functions. Dreaming could be contained within the laboratory. A ‘place’ for dreaming had been created: conceptually, in its night-time recurrence as part of the sleep cycle; physically, as an object extracted from the body of the sleeping laboratory subject via the EEG.15 Dreaming was still in some sense the ‘prize,’ but the APSS determined that a thorough study of sleep was the only possible vehicle by which it could be won. The Biological and Clinical Contours of the Early APSS This psychophysiological mode of investigation blossomed during the 1960s and early 1970s. Most of the APSS’s members were engaged in what they felt to be ‘basic research’ on sleep and dreaming. In other words, they tended to study the mechanisms and functions of sleep and dreaming through laboratory-based experimentation with both animal and human subjects. Despite the fact that many APSS members were trained clinicians, the questions they asked rarely focused on pathology or disease. Instead, their problems involved a diverse combination of experimental and natural-historical approaches: Which regions of the

Begin the Begin

333

brain were responsible for turning REM and slow-wave sleep ‘on’ or ‘off?’ Which hormones or neurotransmitters were involved? Did all animals display REM sleep, or was this phenomenon restricted to mammals and birds? Why did infants spend so much time in REM sleep? But, despite this diversity, APSS members were united in their goal of creating a new physiological language through which dreaming and sleep could be conceived of as active processes. Of course, in most investigators’ eyes, it was not only humans that were doing the dreaming. The use of animal subjects, however, was initially an insignificant feature of early APSS meetings. There were no animal studies presented at the inaugural meeting, and only one, Webb’s description of a sleep-deprivation study in rats using a ‘wakefulness enforcer wheel,’ appeared in the 1962 meeting.16 At this same gathering, Kleitman, now retired and living in Santa Monica, presented a purely theoretical paper entitled ‘Phylogenesis and Ontogenesis of Dreaming – a Speculation.’17 It was only at the 1963 conference that an entire session was devoted to neurophysiological research. This was a landmark conference for a number of reasons. It was the first time this group of investigators had met together outside Chicago (it took place in downstate New York), and it was also by far the largest gathering yet, with almost two hundred in attendance. This rapid growth demanded a formal administrative structure and a process of fee gathering, for which the position of secretary-treasurer had been created late in 1962. It also brought a substantial international dimension to the meetings, and this wider scope served to detach the APSS from clinical styles of investigation even further. Studies of the sleep of finches, crocodiles, turtles, dolphins, and any number of other animals provided REM, and thereby dreaming, with a natural history. The phylogenetic aspect of this work consisted mainly in taking the established practices of EEG recording of sleep and applying them to whatever animals could be acquired. But this research was also allied with a classic style of neurophysiological investigation that identified circumscribed brain regions with behavioural phenomena. This latter style of research was somewhat different from earlier studies in brain localization, in that it studied regulatory systems rather than sensory or motor function. Earlier localization studies had relied on pathology (as did von Economo’s speculation that sleep was regulated in the midbrain, because such structures were destroyed in encephalitis lethargica), or they had involved ‘reflex’ experiments that identified regions of the cerebral cortex and associated them with observed behaviour

334

The Sleep of Others

through electrical stimulation. By the 1950s, microelectrode recording was beginning to play a more central role in determining the regulatory functions of the deep, subcortical structures of the brain by studying spontaneous, rather than evoked, electrical activity in single nerve cells.18 Spontaneous electrical activity had been thought to be a cortical property and was thus typically investigated in human beings, through the EEG. By the early 1950s, this tradition was beginning to change as the use of implanted microelectrodes merged with the long-standing tradition of studying behaviour in decorticate animals. Cats and dogs, for example, had long been established as appropriate subjects for studying the neurophysiological mechanisms of sleep behaviour. As discussed in chapter 4, Pavlov had established the use of dogs as subjects for both chronic and acute experiments by 1900; both Kleitman and the Belgian neurophysiologist Frédéric Bremer followed in this tradition. The discovery of REM in non-human mammals added a modern neurophysiological cachet to the ancient idea that animals dreamed. The unique place of the domesticated pet in human society surely had some impact on this development. Unlike cows, gorillas, or rabbits, cats and dogs were readily available and thoroughly domesticated sources of experimental material, whose primary function in human culture was not to work or to serve as food but (I would argue) to reflect human selfimage. The fact that pets were such close companions opened up the possibility that they, like their human masters, could dream. And, since no animal’s sleep had been subject to as much casual observation as that of cats and dogs, making an inference about their mental activity from their overt behaviour did not involve an unacceptable leap of faith. Their ‘otherness,’ however, made their sacrifice for experimental uses acceptable. As the technique of microelectrode recording spread throughout the 1950s, some investigators began to use this technique to study sleep as a process of regulation within different regions of the brain. In 1958, while Dement was using electroencephalography to document the presence of REM periods in cats, Michel Jouvet and a colleague, François Michel, implanted microelectrodes in a cat’s brain with the aim of studying the role of the neocortex in learning. Their investigation soon took a new turn when they found that their cats continued to sleep even after the total ablation of the neocortex and its corresponding inhibition of the slow-wave activity in the thalamus deemed to be essential to sleep.19 More important, the hypertonia that affected the decorticate cats’ mus-

Begin the Begin

335

cles appeared to cease with a rhythmic regularity. Jouvet was thoroughly familiar with Pavlov’s inhibitory theory of sleep, having completed his MD thesis in 1956 on the role of the neocortex and the reticular formation in conditioned reflexes. Thus, he quickly set to work, attempting to understand the nature of the process that inhibited muscular hypertonia during sleep. He found that these periods were associated with ‘spindle’ activity in the reticular formation, an active EEG in the cortex, and the rapid movements of the cat’s eyes beneath its closed lids. The fact that Jouvet arrived at this particular stage of sleep through animal, rather than human, research had a considerable impact on his later analysis of this phenomenon. He readily associated this ‘paradoxical sleep,’ so named because it combined an active EEG with muscular paralysis and a raised arousal threshold, with a dream state. But his exclusive use of animal subjects, and his focus on the inhibitory phenomena of the EMG, a sort of absence, rather than the overt activity of eye movements, led him to biological, rather than psychophysiological, theories of this unique sleep state. It was as though the inhibition of muscular tone in animals worked as a surrogate for the ability to awaken the subject and elicit a verbal account of a dream. Following perhaps more closely the tradition in sleep research first laid down by Piéron, which thoroughly delineated physiological from psychological investigations of sleep, Jouvet initially attempted to explain the paradox of an active cortex and an inhibited musculature in sleep in terms of its biological function, rather than framing ‘dream states’ within sleep stages as a question about the mental function of dreaming. With the appearance of animal surrogates for human subjects, the distance from the individualized, intimate phenomena of dreaming and the generalized, objectified phenomena of REM sleep became even greater. After the 1963 meeting, animal research began to figure prominently in the APSS annual gatherings and publications. This tendency was paralleled by a decline in the initial hope, held by several psychiatrists, that the sleep EEG would allow them to overcome the phenomenon of repression by ‘capturing’ dreams just as the patient’s mind was creating them. This approach assumed that dreams, as discursive objects, were also therapeutic tools. But, as the 1960s wore on, fewer and fewer American psychiatrists held to this idea as an article of faith. For that growing cadre of clinician-scientists whose first commitment was to basic research, REM posed questions about dreaming that were effectively unrelated to the mundane work of psychiatric practice. From a structural perspective, however, REM research continued to

336

The Sleep of Others

be closely allied to medicine. The majority of investigators in this small field were affiliated with a local hospital service or medical school. Psychiatric clinics continued to provide subjects in the form of patients, and medical students offered themselves up for similar purposes. Typically, sleep laboratories were located in the psychiatric wings of hospital wards, a dramatic shift from Kleitman’s ill-fated attempt to insinuate his research within the University of Chicago’s hospital system during the 1930s. Perhaps most important, the majority of the funding for sleep research during the 1960s came from the National Institutes of Health, whose own laboratories in Bethesda supported at least four investigators working on sleep.20 A pattern of financial support that clearly separated physiological research into basic biological mechanisms from studies directly linked to disease began to emerge in the United States during the 1960s, as large scale, statistically based clinical experiment began to dominate therapeutic investigations.21 This was well in keeping with the tenor of medical and psychiatric practice in the United States during that decade. The rhetoric of the time was permeated with hostility to what seemed to constitute established authority. Medical practitioners faced serious challenges to their paternalistic tendencies from within and from without. Women and minority groups demanded greater access to medical education, while at the same time asserting their right to reclaim control over their own bodies, which had been ‘medicalized’ by such recent technological developments as the birth control pill and the tranquilizer. Psychiatry, in particular, was demonized, as the various poles of the ‘anti-psychiatry’ movement declared mental illness to be little more than a ‘myth’ and called for a deinstitutionalization of psychiatric patients, the liberation of the unconscious, and a radical re-evaluation of Freud, who had invested so much interpretive authority and emotional power in the role of the analyst. The power of this critique, sometimes spearheaded by psychiatrists and psychoanalysts themselves, was further amplified in popular culture through films and novels offering fictional and factual accounts of the serious abuses suffered by psychiatric inmates. These critical perspectives helped transform both the organization and the practice of psychiatry. The mentally ill, finally accorded rights, emerged as ‘people,’ not ‘patients.’ They consumed the psychiatric services of open clinics instead of being institutionalized in asylums, many of which were demolished as relics of a barbaric past. In the ideal therapeutic encounter, they were no longer passive objects of medical interven-

Begin the Begin

337

tions but actively chose their treatment, whatever it might be. With the progressive dissipation of the boundary between insanity and normality and the development of ‘community psychiatry,’ psychiatrists were obliged to abandon their tradition of assuming their patients to be deviant, delinquent, or diseased and then prescribing a treatment. Instead, they needed to make their diagnostic criteria explicit and their procedures consistent, as they were held to be in medicine at large. Part of this process involved discovering useful, objective, and standardized diagnostic markers, and Dement’s work with schizophrenics reflected the same trend. Although he used mentally ill subjects in his experiments, he did not do so in order to further refine existing disease classifications or test new treatments. Instead, he used these patients to uncover the basic structures of certain diagnostic phenomena – in this instance, dreams. Such developments ultimately helped catapult sleep research from the basements of physiology departments to an integral part of the wards of hundreds of hospitals and clinics. But, in the process, the psychiatric conceptualization of dreams as cryptic signs of madness disappeared, as the sleep clinic began to turn to diseases that were firmly entrenched on the boundary between the physical and the mental. The analysis of movement as a state-dependent phenomenon turned out to be one important means of distinguishing the work of the burgeoning sleep clinic from traditional neuropsychiatry. But it did not start out this way. Jouvet’s reinterpretation of REM as ‘paradoxical sleep’ and his corresponding emphasis on the muscular atonia that accompanied such periods provided an important counterpoint to the American schools of sleep research that developed in Chicago, New York, and Los Angeles during the 1960s. Whereas many, and perhaps most, American researchers felt that REM’s future lay in neuropsychiatry applications, Jouvet’s group in Lyon focused instead on functional interpretations of paradoxical sleep. For the Lyon group, ‘paradoxical sleep’ was not a quasi-return to consciousness marked by reports of dream images. It was rather an intensely deep period of sleep marked by cortical activation and a form of muscular paralysis.22 These phenomena – not the presence or absence of dreams – were what required explanation. Certainly, the differential use of human subjects influenced each approach; after all, the dreams of the cats favoured by Jouvet could only be inferred. Paradoxical sleep, argued Jouvet, might actually be a mechanism whereby genetic ‘programs’ could maintain an individual’s psychological individuality in the face of the variable environmental influences upon neuronal activity. In other words, paradoxi-

338

The Sleep of Others

cal sleep was ‘iterative’ and involved repeating and reinforcing patterned activity in the cortex and in the subcortical and spinal motor neurons. In the latter case, muscular atonia would serve to inhibit the acting out of stereotyped movements in sleep, while the ‘motor programs’ ran. Just as the graphical method of Marey and Mosso helped them frame psychological problems in physiological terms, so, too, did Jouvet use the EEG and EMG to de-emphasize, rather than calibrate, dreams as evidentiary phenomena. The Case of Narcolepsy Despite the differences between what might generally be dubbed the American and French traditions, these two strains of research came together in the early 1970s around the problem of narcolepsy. As discussed in chapter 2, narcolepsy had been described as an independent disease entity during the 1880s. Gélineau’s claims for the specificity of narcolepsy had relied primarily upon the nature of the attack, which featured an intrusion of sleep, sometime complete with dream recall, upon wakefulness. In contrast, epilepsy and hysteria were not just temporally misplaced phenomena: their seizure-like nature and associated amnesia made them entirely unlike normal experience. Other neurologists, however, soon began to place greater emphasis on the cataplectic aspect of narcoleptic attacks than on their inopportune timing. In 1902 a German clinician had made a forceful argument to this effect, and his emphasis on the patient’s complete loss of muscle tone as a prominent sign narrowed the diagnosis of narcolepsy considerably, making genuine cases of the disease even more rare than they had been under Gélineau’s description.23 The Viennese neurologist Emil Redlich (1866– 1930), who, like Economo, had been a student of Jauregg, became one of the foremost authorities on the disease during the 1910s and 1920s, precisely because he had observed so many patients, the number of whom came to a grand total of nineteen over a sixteen-year period. Redlich’s analysis of this affectiver Tonusverlust (loss of tone in emotion) described how his patients became overwhelmed by emotion and literally collapsed during their attacks.24 If not accompanied by a loss of consciousness, argued Redlich, these crises engendered such an extreme feeling of helplessness as to be almost a form of waking paralysis. Like Economo, Redlich identified the cause of the disease in a brain lesion, located in the ‘sleep centre’ in the diencephalon. Other investigators, however, were less willing to identify a single

Begin the Begin

339

region as the only centre responsible for sleep. A London neurologist, W.J. Adie (1886–1939), was equally smitten with the use of cataplectic attacks to diagnose true narcolepsy.25 But, following Pavlov, Adie argued that sleep was a generalized cortical inhibition that could have either cortical or subcortical origins. As Pavlov’s excitable dogs that fell asleep in their stands demonstrated, monotonous stimuli, like brain lesions, could cause sleep attacks. In Adie’s interpretation, narcolepsy was not characterized by a pathological form of sleep. Rather, it was the incursion of the state of sleep that was normal for persons of a certain constitution, or predisposition, the origins of which, Adie admitted, were completely unknown. This ‘constitutional’ perspective on narcolepsy received its most systematic expression in a 1934 paper published by Max Levin (1901–74), a neuropsychiatrist and clinical director of the Harrisburg State Hospital in Pennsylvania. Levin is best known for his description of the syndrome of periodic somnolence accompanied by a rapacious appetite and schizophreniform symptoms that now bears his name (Klein-Levin Syndrome). Two years before this analysis, however, narcolepsy commanded Levin’s attention: he estimated that, during the 1920s, he had seen ten times as many narcoleptic patients as he had the decade previous.26 He blamed the increase in large part on the rapid and widespread popularity of the automobile, which, he argued, required exponentially more acts of discrimination and, subsequently, increased inhibition of positive reflexes than ever before. The inhibition naturally radiated across the cortex, frequently causing sleep at precisely the moment when extreme vigilance was demanded. The human cortex, concluded Levin, was simply unable to develop at the same rate of progress as did ‘life-threatening machinery,’ which put those with a constitutional tendency to inhibition at risk. At the same time, an opposing tendency was emerging in which clinicians drew upon recent theories of sleep to expand the diagnosis of narcolepsy as a mental disease with somatic manifestations, a perspective that eventually came to dominate American approaches to the condition. More than two decades after the New York neurologist Charles L. Dana published his incidental observations on ‘morbid drowsiness and somnolence,’ Carl D. Camp offered the first complete description of narcolepsy in the American literature.27 Camp, unlike his predecessor and fellow Philadelphian Weir-Mitchell, was obliged to come to terms with the inroads that psychology had made on territory traditionally held by physicians. So he was well familiar with some of the more recent functionalist theories of sleep, and his 1907 presentation of a narcolep-

340

The Sleep of Others

tic patient featured clinical observations followed by pages of theoretical analysis. Camp was no psychoanalyst, but he did comment on his patient’s dreams, which were frequent, vivid, and generally unpleasant: the forty-one-year-old labourer dreamed, for example, of being strangled by a woman or imprisoned in ‘an asylum for the self-polluted.’ But in Camp’s opinion, such dreams did not point towards any grand internal conflict but simply reflected his patient’s unhappy state. Combining the ideas of Pierre Janet on neuroses with those of Claparède on sleep, Camp argued that his patient’s natural and functional impulse to sleep had degenerated into a pathological desire which ruled the mind like an ‘imperative idea.’ The fact that, no matter how short the sleep attacks, the patient always awoke feeling refreshed and strong, was evidence that the patient’s sleep was a perversion of a normal function and not the product of an organic need. Caffeine, which should have worked had the normal or pathological desire to sleep been due to organic causes, utterly failed. Camp’s remedy was thus, like those of Janet, primarily psychological: static sparks applied to the head, along with a good dose of suggestion, generated considerable benefit, and the patient’s number of daily attacks dropped off considerably. This interpretation of narcolepsy as the somatic manifestation of a psychological disorder permeated American neurological research into the disease well into the 1930s. As the field of clinical psychology expanded, neurologists generally rejected the notion that cataplexy accompanied by sleep attacks constituted genuine narcolepsy. Instead, they maintained a more generous definition that encompassed the disease’s unusual repertoire of symptoms, and set this against its increasing morbidity. Neurological literature filled with accounts of patients suddenly falling asleep while playing cards, making business deals, or even while having sex. Safety was a central concern, since doctors reported train brakemen falling asleep on the job and automobile enthusiasts overwhelmed by sleep before they could pull off the road. Conversely, some patients were able to continue their efforts despite being incapacitated by their attacks. One telegraph operator, for instance, seemed able to continue her ‘automatic’ behaviour of transmitting messages in her sleep; another victim, a typist, was able to produce coherent but irrelevant copy in the midst of an attack.28 What Gélineau had described in 1880 as ‘a rare or at least little-known neurosis’ could be found, forty years later, almost anywhere. The twentieth century was, of course, no stranger to new diseases and even new epidemics. As classic nineteenth-century epidemics, such as cholera,

Begin the Begin

341

typhoid, diphtheria, and tuberculosis, began to wane, new ones seemed to be taking their place. The influenza pandemic of 1918 was the most dramatic example of this new breed of epidemic, but neurological infections, too, were on the rise. Some of them, like poliomyelitis or cerebral meningitis, were familiar, but not in their epidemic form. Others, like encephalitis lethargica, were unprecedented. Though there was no etiological analysis of narcolepsy as a ‘filterable virus disease,’ the ‘narcoleptic syndrome,’ like that of Parkinsonianism, was seen as a possible sequelea of encephalitis lethargica. Sleep had also emerged as a health issue that was, as both Jacobson’s and Levin’s work demonstrates, closely linked to the peculiar form of twentieth-century urban life. Sleep’s successful cultivation seemed to some to be threatened by the business mores and technological contraptions of modern civilization, and this naturally resulted in a sharp increase in sleep disorders whose manifestations were both widespread and varied. New therapeutic possibilities also increased interest in narcolepsy as a biomedical problem. Ephedrine sulphate, introduced to Western medicine from the Chinese pharmacopoeia in the 1920s, largely through the work of Ko Kuei Chen (1898–1988), seemed to provide some relief from narcolepsy’s symptoms.29 Chen, who returned to his native China after graduating with a degree in pharmacy from the University of Wisconsin, worked with an American pharmacologist, Carl F. Schmidt (1893–1965) at the Peking Medical College, where they isolated ephedrine from Ma Huang, a traditional remedy derived from a species of evergreen shrub. Chen and Schmidt demonstrated ephedrine’s usefulness in treating asthma, and described its power as a stimulant and vasoconstrictor. In 1925 Chen returned to the United States and convinced the Eli Lilly company to manufacture the drug on a large scale. Within two years, the American Medical Association (AMA) had accepted ephedrine as a standard drug, and Eli Lilly was receiving more orders for the product than it could fill. Ephedrine remained one of its best-selling products until China’s communist revolution in 1949 cut off its access to the herbal source of the drug and obliged it to switch to pseudoephedrine, which could be derived from plants found outside China. By the early 1930s, ephedrine was being used as an experimental treatment for narcoleptics. Randomized, controlled clinical trials of drugs were still decades away, but some clinical researchers were nonetheless committed to working with as large a patient population as possible. This tendency was particularly pronounced at the Mayo Clinic in Minnesota. The Mayo, in alliance with the University of Minnesota, was

342

The Sleep of Others

then one of two or three medical schools in the United States offering specialized graduate medical education, in contrast to the more traditional process of specialist certification through state boards run by the AMA.30 With its wide referral network and graduate program in neurology, the Mayo was a perfect place to study systematically the effectiveness of a new drug on a large patient pool suffering from a disease that seemed to be rapidly emerging from its former obscurity. Here, narcolepsy could take on the mantle of a fully fledged research object, rather than a clinical curiosity. Two Mayo researchers, H.A. Cave and Luman E. Daniels, published the results of their study of ephedrine in narcoleptics in 1931 and 1934, respectively.31 Daniels’s study was based on the largest number of cases: he had treated or interviewed 47 patients personally and had access to records of another 100 cases. Like many of his colleagues, Daniels explicitly eschewed a classificatory system that carved up nervous disease into ‘organic’ or ‘functional’ categories.32 The debate over such categories, Daniels argued, was utterly sterile. What was required instead was a rich phenomenology of nervous disease in which anatomical seats of illness might be identified but which nonetheless maintained a close and careful analysis of the all symptoms, be they psychological or somatic. To this end, he offered a detailed description of narcolepsy’s symptoms, including the premonitory appearance of night terrors and horrifying dreams, hallucinations, spontaneous feelings of helplessness or paralysis, insomnia, excessive weight gain, disrupted sexual development or a deteriorating libido, excessive urination, and drooping eyelids (ptosis). The recurrent appearance of sleep attacks, however, remained the pathognomonic sign of the disease. Cataplexy, Daniels explained, had to be excluded as a necessary symptom of true narcolepsy for eminently practical reasons: such a stringent condition would have left him with practically no patients to study. He nonetheless took a considerable interest in describing and accounting for cataplectic seizures in the disease. Following the work of German neurologist Curt Rosenthal (1892– 1937) and the provocative clinical phenomenology of the Parisian neuropsychiatrist Jean Lhermitte (1877–1959), Daniels argued that these episodes of muscular atonia were actually a ‘state of partial sleep’ and that they were simply one more manifestation of the same process that provoked sleep attacks.33 The nature of this process, however, remained mysterious. Daniels concluded that narcolepsy was more or less unrelated to heredity, alcoholism, personality, or physical type and that it was only somewhat

Begin the Begin

343

correlated to headaches, influenza, endocrine function, or head injury. A study of related diseases did little to clarify matters. Encephalitics, it seemed, could develop ‘symptomatic narcolepsy,’ which, Daniels admitted, was for all intents and purposes indistinguishable from ‘true’ narcolepsy. The resemblances between narcolepsy and epilepsy, however, were merely superficial, since victims of the latter could not be revived out of their seizures, and, unlike narcoleptics, epileptics often responded well to bromides. In therapeutic terms, ephedrine sulphate seemed to provide some relief: Daniels found that sixty-eight of eightyfour patients treated with the new drug experienced some benefit, although he admitted that the treatment, which had been as short as a few days in many cases, would likely be less promising over a longer period. But the existence of any benefit at all indicated to Daniels that the sleep attack did not represent a need for sleep; rather, it was likely caused by a disruption in a sleep-regulating brain centre. This centre might be located where von Economo and Redlich had suggested: somewhere in the posterior floor of the third ventricle. But this neuroanatomic speculation did not prevent Daniels from also following a more psychological line of reasoning and characterizing the symptoms of narcolepsy as ‘little more than gross exaggerations of normal weaknesses of the flesh.’34 In other words, narcolepsy existed on a continuum with normal feelings of sleepiness and emotional stress, and, as such, it was a constitutional disease that in many ways marked a failure of the will that would be difficult, if not impossible, to trace back to any brain lesion identified in a post-mortem. The Flesh Transformed Clinical applications of electroencephalography would eventually transform narcolepsy from a disease characterized by the exaggerated defects of the will that Daniels had invoked into a disorder of sleep regulation. But this change was slow in coming. By the early 1940s, neurologists and psychiatrists had already begun to use the EEG to differentiate narcolepsy from epilepsy.35 Indeed, the extensive use of electroencephalography in the study and diagnosis of epilepsy had even generated speculation about a widespread, yet hidden, disorder dubbed ‘cerebral dysrhythmia,’ which some held responsible for symptoms as seemingly minor as bursts of emotional distress, inattention, or temporary memory lapses.36 In contrast, the expansion of electroencephalography during the 1940s served to delimit narcolepsy by confirming the attitude most

344

The Sleep of Others

clinicians already held – namely, that narcoleptic seizures bore little resemblance to those of epilepsy. It was not this technological development but the rise of psychoanalysis that first began to transform the diagnosis of narcolepsy in the immediate post-war period. Before the Second World War, clinicians had invested little significance in narcoleptics’ dreams. At most, the horrific dreams reported by many patients served only to reflect the somatic phenomenon of cataplexy and paralysis. The development of psychoanalysis in the United States helped reverse this causal order. Noting that cures, or at least a diminution in the frequency and intensity of the sleep attacks, could be had through psychotherapy, many analysts suggested that the symptoms of narcolepsy were in fact psychogenic.37 In a sense, this was not a new claim: Gélineau himself had advocated psychotherapy, albeit not of a psychoanalytic variety, naturally, as a viable treatment for the disease. But the psychoanalysts’ claim that narcoleptic symptoms could be interpreted as symbols for underlying mental conflicts was new. Among these clinicians, Daniels’s attempt to study the disease and its treatment through large patient populations was abandoned in favour of an intensive analysis of a single patient’s dream images for signs of pathological repression. The urgency of accounting for the increasing incidence of the disease seemed to be supplanted by its symbolic value in demonstrating the vast causal force of the unconscious mind. This symbolic approach started to merge with the new EEG-based dream research during the early 1960s. Not surprisingly, the shift happened first in Chicago, where narcolepsy became the first disease to be made over by the new practices of the sleep laboratory. In 1960 Gerald Vogel, a psychiatrist at the University of Chicago, presented his report of a forty-two-year-old Black man who had suffered from sleep attacks since the age of thirteen.38 First diagnosed with narcolepsy in 1951, he began taking amphetamines, which eliminated the sleep attacks entirely. He took up drawing, which impressed his friends and colleagues at the shipping and receiving company where he worked. But his wage as a shipping clerk was poor, and his new side occupation as a commercial artist brought little extra income, because he lacked the confidence to demand the fees that his talent was worth. Frustrated, he sought psychotherapy. Today, one might expect that the key to this patient’s neurosis might lie in the exploitative and discriminatory social conditions in which he found himself: for an intelligent, talented, but poorly educated Black man aspiring to improve his economic and social status, the United States during the late 1950s was a breeding ground for frustration. But

Begin the Begin

345

Vogel identified the source of his patient’s anger elsewhere: the patient’s mother had rejected him as a child and turned his care over to his grandmother. The patient’s subsequent fear of abandonment, argued Vogel, manifested itself in destructive fantasies towards women, which appeared in his art, and in his dreams. Vogel suspected that his patient’s sleep attacks, like his art, were actually a defence against inappropriate impulses. The sleep attacks would shunt the hostile feelings out of the stream of consciousness and redirect them towards the more appropriate medium of dream imagery. If this were true, Vogel hypothesized, his patient would likely begin to dream after falling asleep much sooner than would normally have been the case; that is to say, he would have a greatly reduced period of ‘dream onset,’ a concept that had only recently been delineated following the discovery of REM. To defend his idea, Vogel borrowed data from Rechtschaffen and Edward Wolpert that demonstrated that normal subjects began dreaming one-half to three-quarters of an hour after first falling asleep. Vogel’s patient, who spent less than a full night sleeping in a laboratory, inevitably began to dream within four to eight minutes after falling asleep. The patient’s dreams, the reports of which Vogel tape-recorded, generally featured the experimental context itself; more particularly, they contained images which the patient later associated with his fear that the treatment would fail, and that his doctor, like his mother, would ultimately abandon him. Harry Trosman, another psychoanalytic psychiatrist at Chicago, had worked with Vogel on this project. So had Rechtschaffen, who helped Vogel record and interpret the EEG. But this alliance between psychoanalysis and the psychophysiological study of sleep, important as it was for getting REM research off the ground, was relatively short-lived. Over the next six years, Rechtschaffen and Dement published several studies demonstrating that Vogel’s suspicion was correct: narcoleptics did indeed experience early onset of REM periods.39 But, despite the persistent psychoanalytic interests of Dement and some of his collaborators, psychoanalytic theory played virtually no role in these publications. That is to say, while the functional accounts of sleep and dreaming might have continued to influence Dement, he rejected the symbolic interpretation of symptoms to arrive at these conclusions. Indeed, his publications on the topic acknowledged Vogel’s 1960 paper in passing, if at all. Instead, Dement, Rechtschaffen, and their colleagues forged a diagnostic system that relied equally on the narcoleptic’s symptoms, sleep stages as signified by the EEG, and differential diagnosis. By 1966,

346

The Sleep of Others

this system was strong enough that the etiological theory provided by Vogel’s dream analysis could be ignored. Electroencephalography ultimately tempered early sleep researchers’ attitude towards psychoanalytic theory. But so, too, did the use of large populations for the study of narcolepsy, which was in sharp contrast to the highly individualized traditions in psychoanalysis. All aspects of clinical research, from drug testing to risk-factor analysis, had been affected by this turn to population-based investigations.40 Sleep research proved to be no exception, but it worked on a far smaller scale than multicentre, nation-wide studies designed to evaluate the relationship between coronary heart disease and what are now known as ‘risk factors.’ Working out of Stanford University, Dement and his collaborator, Stephen Mitchell, took an out advertisement in the San Francisco Chronicle in 1963 soliciting the participation of people who suffered from overwhelming sleep attacks.41 Within a year, they had received over one hundred responses and had opened a clinic in which they conducted all-night polygraphic observations of their patients. Their ‘sleep clinic,’ such as it was, had a short life. After a year of evaluating and treating desperate patients, Dement realized that his National Institute of Mental Health grants were not enough to run such a large practice; meanwhile, insurance companies refused to pay for the patients’ extensive drug regimens. But this exposure to such a large pathological population had enabled Dement and Rechtshaffen to return cataplexy to the diagnostic heart of narcolepsy, and to reject definitively the older assessment of narcolepsy as a pathological exaggeration of normal sleepiness, which could not, they argued, account for the ‘auxiliary symptoms’ of cataplexy, hypnagogic hallucinations, and sleep paralysis.42 It was these latter symptoms, the authors claimed, that truly defined narcolepsy as an independent disease. Earlier neurologists, of course, had worked similar ground. Some, like Daniels, had suggested that the ‘constitutional’ element of narcolepsy expressed itself in different ways in different individuals. Some narcoleptics collapsed when they burst out laughing; others did not. Others, like Redlich, had insisted that true narcolepsy must be accompanied by cataplexy. But this debate had been set within the confines of the dominant epistemology of the day, when pitched battles were fought over the extent to which nervous diseases could be identified with specific regions of the brain. The rise of psychoanalysis and the discovery of REM changed the nature of these debates by promoting the notion that a laboratory-based study of dreaming was both possible and desirable. Although the phe-

Begin the Begin

347

nomenon of dreams appearing immediately upon falling asleep – hypnagogic hallucinations – had been commented on by observers since the mid-nineteenth century, it was not until its regularity in narcoleptics could be demonstrated in the sleep laboratory that it emerged as a pathological sign. Hypnagogic hallucinations first appeared as a sign of disease precisely because early investigators like Vogel wanted to use REM to legitimate the symbolic interpretation of dream images. The importance of this particular sign carried over into later studies of the disease, even though its interpretation was radically transformed to signify a pathological failure of the ‘system’ that normally controlled REM. The fact that the sleep-regulating function of the ‘system’ in question – the reticular activating formation – was largely speculative at this time was besides the point. It was the ability to create normal sleep phenomena with the EEG that counted, because such creations allowed the deviant sleep of narcolepsy to be redefined according to objectively generated signs. Routinized laboratory-based practice replaced the vagaries of talk therapy; waveforms supplanted free association. The emphasis upon cataplexy and sleep paralysis further redefined narcolepsy as a somatic disorder. Dividing their patients into those with and without cataplexy, Dement and Rechtschaffen proceeded to demonstrate that almost all patients in the former group had a REM latency of less than two minutes following sleep onset, while most of those in the latter had much longer, or even normal, REM latencies. The link that had been established by Dement and Jouvet’s earlier work between REM (or paradoxical sleep) and muscular atonia effectively naturalized this new classification of a population of narcoleptic patients. Muscular atonia was a normal feature of REM periods; the experience of sleep paralysis commonly reported by narcoleptics must therefore be evidence of ‘dissociated’ REM periods that appeared too early, outside the normal temporal architecture of sleep. This ‘dissociation of REM sleep,’ as defined by the polygraphic recordings of the sleep laboratory, soon emerged as the new conceptual foundation of the narcoleptic syndrome and provided the basis of Dement’s revival of a sleep clinic of Stanford in 1970. Graphical recording became the very basis of the clinic’s existence. Hundreds of patients, many of them desperate, frustrated, and undiagnosed, responded to newspaper and TV advertisements soliciting people who suffered from sleep attacks. When they arrived at the clinic, they were treated to an arsenal of recording procedures that ran the gamut from brief nap recordings to full-blown twenty-four-hour polygraphic surveillance. But

348

The Sleep of Others

these patients were able to appear and be treated only in the wake of some dramatic institutional changes in American biomedical research and health-care delivery. An emphasis on population-based clinical research made the formation of such patient groups necessary; the creation of Medicare and Medicaid in 1965 made therapeutic funding for such ‘experimental’ diagnosis and treatment possible, a development that eventually spread to private insurers as well. With these transformations in policy, narcolepsy became the ‘wedge’ through which sleep research could be conducted and funded. It also helped to distinguish sleep research and diagnosis through the sleep laboratory as a distinctive form of medical practice that was related, but not reducible to, neurology or psychiatry. In this regard, the APSS recognized early on the importance of standardized criteria for scoring EEG, EOG, and EMG sleep recordings and created a manual to this effect in 1968.43 Sleep research had recreated narcolepsy as its proprietary disease by emphasizing the ideal of ‘total surveillance’ that had been so recently expressed by psychophysiologists. Although an interest in interpreting the content of dreams had helped to instigate this practice, it was soon abandoned. While sleep had gone public in a new and technologically mediated way, dreams were again consigned to the impenetrable realm of private consciousness.

8 Insomnia Returns

By the early 1970s, sleep researchers had successfully recreated narcolepsy as a model disease that could align the work of their sleep laboratory with clinical observation. The progress of laboratory-based studies of sleep was no longer determined by the problem of dreaming. But narcolepsy was a relatively rare disease, so sleep research’s foray into the domain of neurology was still limited. It was the return of insomnia as a biomedical problem that would ultimately reorient sleep research towards the much larger domain of public health. Changes in American drug legislation and a new depiction of drug addiction as ‘dependence’ changed both the image of the insomniac and the way the condition was investigated. The diagnosis and treatment of insomnia came to rely upon laboratory-based studies of sleep. This development, in turn, generated a set of diagnostic criteria that created a new category of disease, sleep disorders, and a new medical specialty that treated them: sleep medicine. Severed from dreaming, standardized in the clinic, and routinized in the laboratory, sleep shed its impenetrable, private nature and emerged to become a public concern by the end of the twentieth century.

The Multiple Contexts of a Lonely Experience Insomnia has proven a crucial component of the medical knowledge of sleep at various times during the nineteenth and twentieth centuries. By the late nineteenth century, ‘nervous’ or ‘neurasthenic’ Americans began to include sleeplessness as an important symptom of their distress. The 1920s and 1930s saw the evolution of the full-blown insomniac, whose routine struggle for adequate sleep provided psychiatrists and neurologists with a typology, and patients with an identity. Psychoanalysis had identified sleep as something so necessary for mental

350

The Sleep of Others

health that dreams had evolved to protect sleep. Psychiatrists, following Pavlov, used sleep therapy, accompanied by the recent invention of a powerful new class of hypnotic drugs, the barbiturates, as a means of restoring sanity to their disturbed patients. But, if insomnia was a routine concern for clinicians, its status in terms of medical research was virtually non-existent. Sleep, as I have argued in past chapters, was a thoroughly private phenomenon until around the beginning of the twentieth century. By the same token, insomnia was sleep’s mirror image. Its experience was personal, as was knowledge and diagnosis of its condition. The physician’s role was simply to facilitate the treatment of what the patient already knew to be the problem. This situation began to shift during the 1960s. Sleep research expanded, as more and more sleep laboratories began to appear within psychiatric clinics across the United States. Those that had once been used to study dreaming began to ‘retool,’ shifting the focus of their attention towards conditions formerly deemed marginal to neurology and psychiatry. In those disciplines, the question of insomnia provided a veritable vacuum. It was too mild a condition to warrant psychiatric institutionalization. It was too psychogenic for neurologists to take seriously. And it was far too routine a problem for harried general practitioners to become particularly concerned. Insomnia was so mundane as to be normal. What made it different was the fact that it was self-diagnosed, and self-medicated, either through over-the-counter drugs or through prescriptions patients requested from their physicians. Debates over the role of hypnotic drugs coincided with this expansion of the sleep laboratory into clinical medicine in three important ways. First, the enormous popularity of the so-called ‘minor tranquilizers’ from the mid-1950s on had helped to contribute to the mundane nature of insomnia in clinical practice. A simple prescription would cure it, as it would ‘those edgy feelings’ of anxiety or depression. But, in comparison to the older barbiturates, the cost of these new medications was high. The extraordinary profits of drug companies attracted the attention of anti-trust reformers, who questioned both the cost and the effectiveness of this new class of drugs. Sleep researchers thus entered the fray and began evaluating these drugs through their new, laboratorybased technologies and concepts. Secondly, the image of drug addiction was undergoing a radical transformation. In a landmark 1962 decision, the United States Supreme Court declared addiction to be a disease, not a crime. Addiction was, once again, part of public health, and it came to be replaced with the more expansive concept of ‘drug dependency.’

Insomnia Returns

351

Unlike addiction, dependency incorporated the excessive use of drugs not typically considered to be narcotics, such as the sleeping pills and tranquilizers. These drugs, sleep researchers argued, fundamentally altered normal sleep and, as a result, affected daytime performance at work or at home. The sleep laboratory here appeared as a useful vehicle for setting precise limits to the otherwise vague concept of dependency. Finally, an old communion between sleep and death was revived during the 1970s, as the federal government called upon sleep researchers to investigate the safety of barbiturates, which were frequently a part of suicide and overdose deaths. The subsequent investigation by the National Academy of Sciences’ Institute of Medicine (IM) gave sleep research a clinical forum and public profile it had not enjoyed since the discovery of REM. Fear and Madness Insomnia’s relationship to madness had been a tentative one throughout the first half of the twentieth century. As discussed in chapter 3, psychoanalysis itself tended to pare insomnia down to its psychogenic roots, just as it had done with neurasthenia.1 But insomnia was not solely, or even generally, in the hands of psychoanalysts. It was distinguished from madness by several important factors (examined in chapter 4). First, the investigations of Claparède, Pavlov, Piéron, and Kleitman all defined sleep as a bodily performance that had evolved, in both the species and in the individual. In any case of its disruption, this function could naturally be restored by creating the conditions appropriate to re-enact the performance, as Jacobson’s ‘progressive relaxation’ was meant to demonstrate. Memories and the repression of unconscious desires had little role to play here, as Freud freely acknowledged in his discussion of the ‘actual’ neuroses.2 Secondly, the restorative value of sleep, though poorly understood, encouraged a sleep therapeutics that circulated around the use of hypnotic drugs. The rapid expansion of such therapies into the asylum during the 1920s and 1930s served to reinforce the characterization of insomnia as an organic condition. In spite of its psychological components, insomnia, like pain, remained thoroughly grounded in the personal and medical care of the body. The interplay of insomnia and madness, then, operated simultaneously at a social and an individual level. In 1929 Robert Kingman (1876–1940), a physician from Brooklyn, New York, observed that ‘the dangers of insomnia have been so widely and generally exaggerated that

352

The Sleep of Others

the average person becomes little short of panic stricken when sleeplessness attacks him.’3 Neurasthenics and those suffering from ‘cerebral hyperanemia’ would have made up a sizable portion of his private practice; many of his patients would undoubtedly have been familiar with such diagnoses before they even set foot in his office. Certainly, they were aware, as were Jacobson’s patients, of the dangerous links between insomnia, mania, and the harried nature of life in a modern American city. For Kingman, this situation presented a unique problem. Insomnia had become such a well-publicized phenomenon, he argued, that the fear of insomnia had itself become a cause of insomnia. Drawing upon Pierre Janet’s notion of an idée fixe, Kingman presented a picture of the insomniac as someone who suffered from a sort of circular madness, in which modern media had so exaggerated the problem of insomnia that, for many people, the normal rhythms of sleep and wakefulness had become distorted by excessive emotion. Recognizing the physiological need for sleep, people experienced an overstated desire for it, which was counterbalanced by a pathological fear of continued wakefulness. This battle between two ‘complexes’ ultimately left its victims lying in bed, hopelessly awake. Sleep came only when the insomniac resigned himself or herself to fate, declaring ‘‘‘What’s the use of trying any longer; there’s no time for a decent sleep now anyway.”’ At this point (which Kingman thought usually occurred around five o’clock in the morning), sleep would finally come. Optimistically, Kingman looked forward to the time when insomnia would be framed as a serious research problem, subject to ‘the principles of the new school of behavioristic psychology in the shape of psychoanalysis.’ But this was far in the future. In the interim, he trotted out the rather meagre fare of sleep hygiene, adapted to the modern conditions of urban existence. People should not sleep much past daybreak, he insisted, nor should they remain in bed longer than eight or nine hours. Sleep was a state initiated and dominated by habit and emotion; the ‘age of the efficiency expert,’ however, implied that people could and should bring their desires in line with the science of work physiology. People did not and should not require as much sleep as was the case in the days before artificial lighting, industrialization, and the demands for a highly productive workplace. The ‘root of the trouble’ for most insomniacs, Kingman argued, ‘is in their minds and their emotions, not in their muscles or their elimination.’ Clearly, the neat Freudian separation between soma and psyche that helped to signal an end to

Insomnia Returns

353

neurasthenia by the early twentieth century had passed over insomnia. Body, mind, and culture were inextricably linked in sleep and its disturbances, and memory, which had played a crucial role in understanding dreams, served little or no function in insomnia, where the flexibility of behaviour implied that insomniacs could be bent to the ever-changing conditions of labour and culture. The emotionally deranged insomniac described by Kingman had his early-twentieth-century counterpart in the self-medicated user of sleep remedies. Like other disturbances of the Galenic non-naturals, problems with sleep could always be treated with various medications, and the increasing restrictions placed upon such remedies contributed far more to insomnia’s rise as a medical condition than any analysis of emotional complexes. Before 1900, opium, morphine, alcohol, belladonna, hyoscine, Indian hemp, chloral, and bromide were all indicated as treatments for sleeplessness. Combinations of these were produced as proprietary drugs, which were (by definition) protected by patent and generally available without a doctor’s prescription. This situation began to change by the turn of the century, when concerns about excessive opiate use merged with the expansion of psychiatry and the increased authority of scientific medicine. In the process, the image of the junkie as a criminal psychopath emerged, and with it, legislation, institutionalization, and expert intervention designed to define and control the addict’s place in society.4 The habitual use of opiates and alcohol had, of course, been a subject of medical interest at least as far back as the early nineteenth century, when the state began to assume a more active role in policing its populations’ health. Some sanitarians linked the opiate habit with poverty, degeneracy, and moral failure in working-class mothers, while insurance companies worried that long-term use led to premature illness and death, and developed annuity tables to suit. Medical practitioners, some of whom were themselves habitual users of opiates, frequently invoked professional ethics in their debates over how to keep their patients from becoming addicted to the drugs they prescribed. Drug legislation typically focused on consumption, not production. In the United States, the passage of the Pure Food and Drug Act in 1906 merely prohibited pharmaceutical firms from making false claims about the ingredients in their trade remedies. The Harrison Act (1914), on the other hand, made it illegal to sell opiates without the doctor’s original prescription, thus legally placing the maintenance of addicts under medical control. The AMA had successfully lobbied for such reforms in an effort to gain control over the unrestricted sale of patent medica-

354

The Sleep of Others

tions, the main ingredient of which was frequently opiates or alcohol.5 But the medical control of addiction did not last much past the 1920s. Consensus regarding the physiological damage caused by addiction was elusive, and withdrawal symptoms were not enough to convince the profession that addiction was a disease rather than a vice. At the same time, a new psychiatric model emerged in which the addict was construed as a (typically young, Black, male, and urban) psychopathic deviant who sought out drugs for pleasure – a model that fit well with the Treasury Department’s rigorous prosecution of addicts for possession under the Harrison Act. By the end of the 1920s, the majority of federal convicts were in jail because they had violated the Harrison Act. Once a disease, addiction had now become a crime. These developments may seem a long way off from the question of insomnia, but they are an intrinsic part of its development as a disease. The widespread use of barbiturates in medical and psychiatric practice discussed in chapter 4 was generally unaffected by the new American drug legislation aimed against opiates, cocaine, and (during Prohibition) alcohol. Barbiturates did not seem to produce dramatic withdrawal symptoms when their use was suddenly discontinued, and they were not associated with the shady underworld of crime, prostitution, racial pollution, and deviancy that was generally perceived to be an integral aspect of narcotic use.6 On the contrary, these new drugs were hailed as the linchpin of a modern, scientific psychiatry that had finally overcome the barbaric practices of restraining the mad in favour of proper medical treatment.7 For its part, psychoanalysis had made acceptable the notion that nervousness was on a continuum with madness, so insomniacs could indulge in the same drugs that were then putting asylum patients to sleep. Besides, there was no legislation to stop them: before the Second World War, barbiturates and bromides were readily available without a prescription.8 This is not to say that concerns about sedative use did not exist. Admittedly, many neurologists and some general practitioners likely shared Kingman’s belief that insomniacs were delusional. So the act of seeking out drugs for relief, rather than pleasure, was a logical extension of the neurosis that caused the disease in the first place. Others were more explicit about the dangers of sleeping pills; in the process, they were also more critical of the medical establishment. Unlike Kingman, Jacobson was outspoken on this point. The insomniac protagonist of You Can Sleep Well encounters the perils of sleep through sedation on his pilgrim’s progress towards ‘progressive relaxation.’ Armed with the

Insomnia Returns

355

recommendation of a stranger and a prescription borrowed from an old friend, Jacobson’s imaginary patient tries both bromides and barbiturates. Initially, the results of each are pleasing, but, soon enough, he finds that he is taking more and more of the drugs yet sleeping worse than ever. His lack of energy causes his salesmanship to slip, and financial disaster looms for the patient and his family. The kindly physician from down the street diagnoses him with ‘bromide psychosis’ – a common enough problem, the doctor notes, given how many people foolishly take such drugs without a physician’s supervision. ‘I should call it a vice,’ the doctor continues,‘except that there is no pleasure connected with it. Here you are, developing a ‘sedative habit,’ until you are so dependent on your daily dose that you tremble at the very thought of going without it ... do you consider it natural and efficient to live under the influence of sedatives in order to avoid fears of lying awake? Mental hygiene slipping, I should call it ... do you want to be dependent upon artificial depressants? Is it natural and desirable to live in a world clouded and discoloured from the after-effects of drugs? I would advise you to give them up altogether. If you cannot readily do so at once, you had better diminish the dose a little each day, until you regain a natural, free life.’9

Jacobson’s fictional patient was no addict. Indeed, the image of the addict in the late 1930s, when You Can Sleep Well was first published, was the very opposite of that of Jacobson’s patient, who was nothing but a thinly disguised veil for Jacobson’s middle-class reader. The addict was Other. The insomniac or sedative user was the Self, or, as the subtitle to Jacobson’s book had it, ‘the Average Person.’ Sedatives did not threaten the health, the morality, or the economic vitality of the nation so much as they undermined the individual’s freedom of will and strength of character. Carefully attending to one’s body through the practices of progressive relaxation, Jacobson advised, would restore the will that had been damaged by the careless use of sleeping pills. The Changing Face of Addiction The message was timely. You Can Sleep Well was published in 1938, the same year that saw the passage of the Food, Drug, and Cosmetic Act in the United States. This act, like its 1906 predecessor, was intended to protect the consumer from unsafe drugs. Although the bill had been

356

The Sleep of Others

proposed back in 1933, its final passage was in response to the ‘Elixir Sulphanilamide tragedy’ of 1937, in which a pharmaceutical firm, the S.E. Massengill Company, created and sold a liquid version of their popular anti-streptococci drug using diethylene glycol as a solvent. The mixture was extremely toxic, and nearly one hundred people, many of them children, died before the drug was pulled from the shelves.10 Not surprisingly, the 1938 act emphasized the need for manufacturer’s labels to describe the relative safety of the drugs in question. Regulations for enforcing the act by the Federal Food and Drug Administration (FDA) were written later that year, following industry proposals. They stipulated that manufacturers had to list ingredients in a standardized manner, as well as provide instructions for proper use. Drugs labelled ‘Caution: To be used only by or on the prescription of a physician [or dentist or veterinarian],’ however, were exempt from these conditions. It was assumed by the two FDA officials who accepted the industry proposal, Drs Theodore Klumpp and J.J. Durrett, that the physician writing the prescription would explain the drug’s proper use and attendant dangers to the patient. The law was explicitly designed to accommodate and encourage the culture of self-medication by ensuring the safety of the products available. But an opposing trend emerged between 1940 and 1960. The class of prescription-only drugs quickly grew, as drug companies used the ‘prescription-only’ status to market drugs to consumers while simultaneously reducing their liability.11 For its part, the FDA declared certain drugs, like the barbiturates or the sulphanilamide drugs, to be too dangerous to be self-administered, but too useful to be banned outright.12 In either situation, the increase in the proportion of prescription-only drugs relied upon and enhanced the professional authority of the physician over all other medical and paramedical practitioners. In 1949 retail druggists, who had recently begun to bear the brunt of FDA prosecutions for illegally refilling prescriptions, asked Congress to restore this right. The FDA responded by arguing that it was the federal government, and not industry, that should determine whether or not a drug should enjoy prescription-only status. Drug manufacturers, invoking fears of ‘socialized medicine’ at congressional hearings, successfully prevented the FDA’s proposal from becoming part of the 1951 legislation that formalized the distinction between over-the-counter and prescription medicines. The latter category included habit-forming drugs, as well as those that were considered dangerous or toxic without expert supervision, and it applied to all new medicines. Thus, a new and large

Insomnia Returns

357

class of pharmaceuticals had come into existence: drugs that were recognized as addictive but that nonetheless had medical value and were not primarily defined as part of the hedonistic, deviant, or criminal subculture of addiction. The distinction between ‘iatrogenic’ and ‘recreational’ forms of addiction started to unravel by the early 1960s. A 1962 Supreme Court decision ruled that a California law making it illegal to be an addict was unconstitutional. The defendant in the case (Robinson v. California) had been arrested and convicted by a municipal court simply because he had needle marks on his arms and had admitted to using narcotics in the past. The Supreme Court overturned the decision, arguing that addiction was properly defined as a disease, not a crime. A White House Conference and Presidential Commission on Narcotic and Drug Abuse was organized that same year. It, too, pulled back from a punitive model of addiction, depicting addicts as deserving victims rather than criminals and supporting the prosecution of drug traffickers instead. After considerable debate as to whether barbiturates were addictive or not, the first federal law against depressants and stimulants was finally passed in 1965.13 The extension of the concept of addiction to ‘dependence’ had thus given legal status to the ‘sedative habit.’ It was in this context that post-REM sleep research made its first forays into the domain of public health. Out of the Asylum: American Psychiatry and Psychoactive Drugs The ‘therapeutic revolution’ of the mid-twentieth century, signalled by the discovery and widespread application of antibiotics, was also accompanied by the creation of an entirely new field, that of psychopharmacology. Of course, drugs that affected sensation, perception, and cognition were at least as old as alcohol and opium. Bromides, barbiturates, and amphetamines had equally forceful effects, and, when they had been introduced around 1900, psychiatrists had hailed them as ‘revolutionary’ in much the same way that meprobamate (Miltown), chlorpromazine (Thorazine), or the benzodiazepines (such as Valium) would be celebrated during the 1950s.14 But the latter class of drugs was said to lack the ‘totalizing effects’ of the earlier ‘sedatives’ and ‘hypnotics.’ Experimental animals and psychotic patients alike appeared calm and even indifferent when they were given these new drugs. In Europe, researchers had initially described the drugs’ effects using existing terms: their ‘sedative’ effects on patients were due to the drugs’ ‘neuro-

358

The Sleep of Others

leptic’ properties. In other words, they cured psychosis by inhibiting muscle tone until a ‘crisis’ had passed. In the United States, however, the term ‘tranquillizer’ soon caught on after Frank Berger, Miltown’s creator, suggested the word as a useful way for people to distinguish between his new drug and the older sedatives. Tranquillizers rapidly infiltrated popular consciousness. As the historian of psychopharmacology, David Healy, observes, the success of these drugs was as much due to the new context in which they appeared as it was to their chemotherapeutic activity. American psychiatry entered its second great transformation following the end of the Second World War. Fifty years earlier, psychiatrists had successfully remade their field as a medical specialty. But, even though their new status as medical professionals meant that psychiatrists were no longer mere custodians of the mentally ill, psychiatry itself continued to be associated with asylum practice and its attendant stigmas. By the 1940s, psychiatrists were beginning to break with the asylum. Psychoanalysis, which provided the bulk of intellectual and educational leadership of the field, had already started to encourage work in group and community settings. The Second World War opened up new problems and opportunities for physicians with an interest in mental illness: the large numbers of soldiers traumatized by their wartime experiences needed treatment quickly, without recourse to institutionalization. Typically, they were treated by physicians with little or no experience in the asylum, but with at least a limited exposure to psychiatric practice. The large population of soldiers confronting similar kinds of emotional distress and the demands of war meant that group therapy appeared as the most viable mode of treatment. After the war ended, these physicians were anxious to establish themselves as psychiatrists in private practice, rather than compete for what few institutional positions were available in a saturated job market. This move away from the asylum had its counterpart in the 1963 legislation that promoted the creation of Community Mental Health Centers.15 Such legislative efforts at reform were accompanied by the more radical ‘anti-psychiatry’ movement, which was simultaneously an internal critique of psychiatry’s power over its patients, a drive for professional reorientation and reorganization, an attempt to expand the concept of human rights to include psychiatric patients, and a general cultural transformation in attitudes towards what constituted mental illness.16 Public statements or social movements opposed in some way to psychiatric practice had certainly preceded the 1950s and 1960s. The first use of the term ‘anti-psychiatry’ appears to date back to 1908, fol-

Insomnia Returns

359

lowing the sensational trial of a German tavern keeper who had ‘liberated’ an inebriate Scottish priest confined in a German religious asylum. Bernhard Bayer, a German psychiatrist, described the subsequent popular sentiment against his profession as ‘anti-psychiatrische.’17 But the anti-psychiatry movement of mid-century was not tied to a single nation, issue, or event; it was a diffuse, widespread campaign to change the very idea of madness. Along with legislative reforms and therapeutic developments, the anti-psychiatry movement served to create a climate in which mental illness and mental health were conceptualized as lying on a continuum. Former asylum inmates were prescribed medication and psychotherapy, then ‘freed’ to walk the streets. Nervous or depressed patients became less liable to seek an organic diagnosis from their neurologist and instead began to visit their neuropsychiatrist, looking for a prescription. The diverse forms of sleep therapy and the use of sedatives to treat psychoses were institutional treatments, in that they required a medical staff large enough to conduct and monitor the therapeutic regimens and corresponding restrictions on the patients’ mobility. As asylums and the support for them began to crumble, so, too, did these therapies. ‘Tranquillization’ quite literally enabled the elimination of asylum beds, since former inmates could now be treated yet sleep elsewhere. The attendant difficulties associated with the health and welfare of such patients outside the asylum, unfortunately, usually fell below the threshold of biomedical or political interest. As far as the neuroses were concerned, however, tranquillization proved to be a lucrative enterprise. Relentless medical advertising claimed that these drugs could relieve ‘nervousness,’ ‘anxiety,’ ‘depression,’ or ‘emotional and muscular tension,’ and almost always mentioned that sedation or drowsiness – properties once thought curative – would not occur.18 The manufacturers of one product, the suggestively named ‘Suavitil,’ described their drug as an ‘antiruminant’ that promoted sleep not by its sedative or hypnotic effects but by ‘reducing repetitive thinking.’ Most minor tranquillizers were deemed capable of performing a similar double-duty of calming during the daytime and promoting sleep at night.19 Following the spectacular success of chlordiazepoxide (Librium) in 1960 and diazepam (Valium) in 1963, a number of similarly structured sleeping aids, including nitrazepam (Mogadon) and flurazepam (Dalmane), were launched. These sleep remedies were commercially successful but somewhat less so than their ‘anxiolytic,’ or ‘anti-anxiety,’ counterparts.20 In any event, the prescribing practices of physicians were clearly changing. Prescrip-

360

The Sleep of Others

tions for sedatives and hypnotics began to drop around 1964, as the new drugs were launched that claimed to treat the specific symptoms of the mental disturbance, be it anxiety, depression, or sleeplessness. Sleep was losing its anodyne status and was being remade as a specific site of mental and organic disorder. Sedatives, Insomnia, and the Sleep Laboratory The introduction of tranquillizers in the 1950s and 1960s did not so much eliminate the use of sleeping pills as bring to light a particular conception of insomnia as a manifestation of anxiety. This concept clearly predated the tranquillizers: Kingman had argued that the insomniacs were primarily anxious about the amount of sleep they required, and that a cure would be forthcoming once they recognized this simple fact. Jacobson described anxiety in terms of ‘tension,’ a clear analogy to the sensation of muscular strain or effort. His patients would recover from their insomnia once they gained control over their muscular sensations. In contrast, the tranquillizers supposedly affected neither the mind nor the body but the brain. All cognitive and motor functions were supposedly left unchanged, leaving the brain to continue its normal regulatory work, which naturally included sleeping. The concept of ‘drug dependence’ that emerged out of the tranquillizer era followed a similar pattern. The physical effects of withdrawal, once depicted as the primary stigmata of addiction, became less important than the spectre of the routine use of an unnatural substance to the point of reliance or ‘dependency.’ Consider, for example, the following bit of testimony gathered during a series of congressional hearings regarding the abuse of benzodiazepines. Having given birth to seven children in eight and a half years, Mrs Hinton, a housewife from Plainfield, New Jersey, felt emotionally exhausted. She saw a psychiatrist, who prescribed her Valium. She took it every day but soon found that she required more than the prescribed amount. So she visited several other psychiatrists, and got prescriptions from each. Ultimately, she claimed to be unable to do anything without the drug: ‘I went through different stages with it,’ she testified, ‘but I had to have it. I had it by my night table and in the morning before I even got up, I took the Valium, because I felt that I could not even wash my face unless I had the Valium.’21 Mrs Hinton’s testimony was recorded in 1979, but the image of dependency her story depicted had been part of popular knowledge and psychiatric concern for at least a decade.22

Insomnia Returns

361

Tranquillizers were the regulatory power of the brain made external. During this same period, sleep had also emerged as a regulatory force, rather than a simple elimination or reduction of consciousness – a theory of sleep that older hypnotics and sedatives had once exploited. The EEG, which had been the prime mover behind this new view of sleep, soon brought insomnia, drug dependency, and the benzodiazepenes under its domain in the sleep laboratory. Federal legislation dovetailed with the progress of clinical experimentation in bringing insomnia into the sleep laboratory. Demonstrating the efficacy of a drug was intimately tied up with safety requirements since the passage of the 1938 Food, Drug, and Cosmetic Act.23 After all, the risks involved in taking a drug could be outweighed by its potential benefits in treating one disease, in which it seemed effective, but not in another, in which it had little or no effect. But it was not until 1962 that federal legislation (popularly known as the Kefauver-Harris Amendment) was passed that insisted that drug companies demonstrate this efficacy through ‘adequate and well-controlled investigations’ using ‘appropriate statistical methodologies.’24 And it was only in 1970 that the FDA formally laid out standards for such studies, which involved the use of a contemporaneous control group assigned at random that permitted quantitative evaluation.25 With this precipitous increase in legislation and enforcement during the 1960s and 1970s, the FDA became engaged in controlling the traffic in barbiturates (through its Bureau of Drug Abuse Control) while, simultaneously, traffic in narcotics and marijuana remained under the administration of the Treasury Department’s Bureau of Narcotics.26 The FDA’s bureau created standards, supervised industry studies of drug safety and efficacy, and performed its own studies of popular medications, particularly those sold over the counter. The expansion of both the concept of addiction as dependence and the authority of the FDA to police the efficacy of all drugs turned self-sedation into a new political and medical object. Habitual barbiturate use became the focus of an entirely new concern – death by overdose – while the efficacy of over-the-counter sedatives quickly became the target of FDA intervention. The concern about barbiturate use was international in scope. As early as the mid-1950s, the World Health Organization had suggested increased control of barbiturate sales, although it stopped short of calling the drugs addictive, describing them instead as ‘habit-forming.’27 The United Nations repeatedly passed resolutions with a similar sentiment throughout the 1960s, and, by the middle of the decade, the enor-

362

The Sleep of Others

mous increase in sales of barbiturates and tranquillizers was receiving regular attention in the lay press.28 The medical press also began to take notice. A lead article in the British Medical Journal in 1968, for example, noted that prescriptions for sleeping pills accounted for 10 per cent of all prescriptions in Britain, and that sales of sedatives and tranquillizers had increased by 535 per cent in the United States between 1958 and 1965.29 Reports of young Londoners injecting hypnotics intended for oral use intensified the campaign against barbiturates.30 Medical concern regarding barbiturate use was based, in part, upon recent laboratory-based sleep research. Insomnia, as we have already observed, had not been an attractive problem for medical researchers before the mid-1960s. The potential significance of dreaming or its deprivation had preoccupied the psychiatrists, psychologists, and neurologists who made up the core of the emerging field of sleep research. A 1967 survey of sleep researchers found that only one of 218 respondents named insomnia as one of their ‘areas of specialization.’31 A similar lack of interest in insomnia characterized the APSS annual meetings. The term was rarely mentioned in titles or abstracts before 1969, at which point Anthony Kales, Joyce D. Kales, and Allan Jacobson, working out of the Psychiatry Department of the University of California at Los Angeles (UCLA), described the studies of insomnia treatments conducted at their sleep clinic at Cedars-Sinai Hospital. These studies dated back several years and were part of the growing interest in using the sleep laboratory to assess over-the-counter and prescription remedies for sleeplessness. The first major study came from the Department of Psychiatry at the University of Edinburgh, where Ian Oswald and R.G. Priest reported that patients trying to give up their sleeping-pill habit needed, on the average, five weeks before their normal sleep, as determined by the EEG, was restored.32 Three years later, Oswald produced a major review of a number of popular sedatives, in which he documented their adverse effects on normal sleep and looked forward to a time when a natural ‘hypnotoxin’ could be synthesized.33 Similar studies were emerging from across the United States, particularly from UCLA and the University of Florida College of Medicine in Gainsville.34 Before 1965, a handful of investigators had used the sleep laboratory to reinforce the existing theory that insomnia was a neurosis, not a physiological disorder. Patients at a clinic in Paris, for example, were treated to EEG studies that demonstrated that their complaints about lack of sleep were often fictional.35 Others offered a more nuanced view. Work

Insomnia Returns

363

conducted at Rechtschaffen’s laboratory in Chicago suggested that the difference between self-described ‘poor sleepers’ and ‘good sleepers’ was a matter of perception, not reality. Poor sleepers reported an average of fifty-nine minutes to fall asleep, but the average time it took them to fall asleep in the laboratory was fifteen minutes, just eight minutes longer than the ‘sleep latency’ of good sleepers.36 Personality tests also indicated that poor sleepers also tended to be more anxious and depressed than good sleepers. But these subjects (they were students, not patients) were not simply the mentally-ill-in-waiting: polysomnography (PSG) indicated that they tended to spend less time in REM sleep and more time in Stage II sleep than the so-called good sleepers, suggesting that insomniacs might have a distinctive form of sleep that distinguished them from traditional categories of psychiatric illness on the one hand, and from conventions of the complaining hypochondriac, on the other.37 Despite these differences, a common rationale underwrote all these studies – the conviction that the sleep laboratory was the most appropriate site for the study of insomnia. This unprecedented development was fuelled by concerns over drug dependency. Neither Kleitman nor Edmund Jacobson, both of whom had written on insomnia and conducted EEG studies of sleep, had put these two together in the late 1930s. Nor had this approach taken shape during the early years of the APSS, when researchers had focused almost exclusively on dreaming and basic neurophysiology. By the late 1960s, students of insomnia acknowledged that, more than a decade after the discovery of REM, EEG-based research had failed to establish the normal function of sleep.38 Nevertheless, they insisted that the measurement of sleep through EEG was the only appropriate means of delineating normal sleep, even without recourse to the problem of function. By the early 1970s, the importance of this practice had increased considerably. This was not in any way due to the success that sleep researchers had in using the EEG to diagnose insomniacs; rather, it was the result of their ability to establish a quantitative and graphical means of measuring the effects of sedatives on normal sleep. During the late 1960s, a series of conferences organized by Anthony Kales and held at the Brain Research Institute (BRI) at UCLA indicated the possibilities of such a direction. The use of the EEG to diagnose insomniacs was not at issue here. Indeed, as insomniacs remained the province of the general practitioner and the office psychiatrist, subjects were easily found by placing ads in local newspapers. Insomnia was readily acknowledged as a condition that was

364

The Sleep of Others

self-diagnosed, but it had become a significant medical concern by virtue of the fact that prescriptions were needed to obtain many sleep remedies.39 So these studies focused not on the diagnosis but on the efficacy of the drugs used to treat it. The sleep laboratory became the bastion of reason in dealing with the unreasonableness of insomniacs and the physicians who treated them. In his preface to the published papers from the 1968 symposium, for example, Kales invoked the irrationality of Americans spending ‘millions of dollars annually for various drugs, sleeping aids, and other remedies in a frantic search for the “good night’s sleep,”’ and he set this trend against the carefully controlled studies from the sleep laboratory, which generally indicated that such remedies were of extremely limited value.40 The new mission of the sleep researchers was to get this information out to the clinicians who were prescribing the drugs. Insomnia was still of little interest for sleep researchers, even if they were clinicians. But as a drug user and potential addict, the figure of the insomniac held court at the BRI sessions. In the 1968 conference, for example, five papers were presented that evaluated the effects various drugs had on sleep, while only one attempted to characterize insomnia by psychophysiological measurement. The fact that such conferences were skewed towards the assessment of drugs is perhaps not surprising, given that the pharmaceutical firm Hoffmann-LaRoche was a major sponsor of the meetings. This trend continued throughout the 1970s, as sleep researchers began to delineate the problem of insomnia along two major lines: the ongoing evaluation of the efficacy and dangers of sleeping pills, and the importance of educating clinicians regarding the nature of sleep and its manipulation by drugs. The first of these approaches quickly merged with FDA activities as the agency became an active participant in the approval process for new drugs. Not only could the FDA stipulate the terms under which drugs could be tested, the agency could also conduct its own reviews, which it did through the Drug Efficacy Study. The study, which was composed of a panel of experts in various medical specialties, operated under the auspices of the National Research Council of the National Academy of Sciences and ran from 1966 to 1969. These experts would not conduct systematic trials themselves, but would rather determine the fate of a drug by reviewing existing data and by drawing upon their own clinical experience. In 1972, following a failed legal attempt by a pharmaceutical firm to prevent the FDA from denying approval to one of its new drugs, the agency began to apply retroactively the terms of its review

Insomnia Returns

365

boards to all drugs introduced since 1938. The 1972 review included a comprehensive study of over-the-counter medications and was frontpage news when it was first announced.41 Its primary target was the socalled ‘mood drugs,’ which included sedatives and stimulants. In fact, the agency had already started to move against such drugs: in Brooklyn, two years earlier, agents had seized forty cases of a popular sedative, arguing that the drug had not yet been approved. The manufacturer claimed that such approval was not needed, because the drug was merely a combination of ingredients already accepted by the agency.42 These activities were undertaken in the full glare of public opinion. Part of the transformation of the addict into a deserving victim involved acknowledging that people could be coerced into taking drugs, and the participation of the drug companies in such coercion was extensively explored in the Nelson hearings on competitive problems in the drug industry between 1967 and 1979, which included over 17,000 pages of testimony and exhibits.43 The drug firms’ television advertisements were routinely blamed for the perceived epidemic of illegal drug use among America’s youth, since they tended to imply that the most routine forms of pain, discomfort, and unhappiness could be cured instantly and effortlessly. The most maligned of these ads, for example, depicted worn-out housewives transformed into charming hostesses with the help of a pill. And, if such behaviour was acceptable to parents, their children were that much more likely to use drugs to solve their own problems (or so the argument went).44 This response against sedative overuse was part of the general disillusionment with medical science and authority, which has been characterized by one historian as the ‘myth’ of the overmedication of America.45 But this myth was equally a part of post-war medical reform, particularly in mental health, where some psychiatrists paradoxically embraced ‘anti-psychiatry’ in an attempt to continue the reorientation of their field away from an asylum-based practice and towards the treatment of ‘community nervousness.’46 Sleep researchers equally drew upon the rhetoric surrounding overmedication to further their own efforts to expand the domain of the sleep laboratory and to educate physicians about sleep and its disorders. Insomnia became the fulcrum that provided leverage for both reforms. In a 1971 review of the insomnia literature before and after the discovery of REM, two psychiatrists at the University of Florida at Gainsville noted that, while psychoanalytic approaches had generated a considerable number of theories about how anxiety, fear, masochism, and other psychopathological factors could produce insomnia, consensus

366

The Sleep of Others

remained elusive.47 The authors in no way rejected the usefulness of psychoanalysis in diagnosing and treating insomnia, but they did argue that the sleep laboratory held the key to future research. In contrast to psychoanalysis, EEG investigations did not begin with claims about the underlying causes of insomnia but rather offered new descriptions of the condition in terms of how it deviated from normal sleep. The longterm promise of such an approach was that such deviations, once correctly identified, could be corrected by the synthesis of a naturally occurring ‘hypnotoxin.’ In the meantime, the problem of addiction figured front and centre. Echoing a sentiment expressed by the Kaleses a year earlier, the University of Florida authors recommended that the FDA immediately begin a campaign of ‘utilizing EEG sleep study reports in the evaluation of hypnotic drugs,’ and that drug companies be required to ‘include quantitative and qualitative sleep pattern changes in advertising literature distributed to the medical profession.’48 This new phenomenology of sleep disorders would be enhanced by the extension of psychophysiological ‘telemetry systems,’ which would enable EEG monitoring of subjects in their routine environments. Computer scoring of such records would make the process more efficient, and longitudinal EEG studies would help identify those members of the population ‘at risk’ for insomnia. An increased financial commitment to sleep research on the part of the federal government was also necessary, since this would enable a comprehensive study of all the relevant physiological, psychological, social, and environmental factors that played into the disease. ‘Hopefully,’ the authors concluded, ‘the sleep EEG will someday be a vital part of the perennial routine checkup and will play an important role in preventive medicine.’49 The transformation of drug addiction into drug dependence, increasing regulation of sedatives, and the ambitions of sleep researchers were starting to turn sleep research towards the clinic. Narcolepsy continued to command interest as a model sleep pathology, but its relevance was limited because it was so rare. Insomnia, a widespread phenomenon with intimate connections to drug abuse, brought sleep research firmly into the domain of public health for the first time, despite the tendency among clinicians to depict insomnia as a symptom rather than a disease unto itself. As a consequence of these developments, the sleep laboratory began to take on new roles during the 1970s. On the one hand, the laboratory became a testing ground for assessing the effects of drugs on the sleep patterns of patients. This research program, if it could be

Insomnia Returns

367

called such, quickly expanded to incorporate the study of not just sedatives but all psychoactive drugs. On the other hand, the laboratory emerged as a diagnostic tool. Sleep researchers insisted that an integral part of educating physicians about sleep involved establishing a procedure of ‘differential diagnosis’ in sleep disorders. Evidence from the sleep laboratory had demonstrated that sleep was an active state. On that basis, sleep researchers argued, sleep must have its own specific pathologies, which, as it turned out, could be properly diagnosed only through the use of the EEG. Neither the general physician nor the traditional psychiatrist, however, was capable of properly diagnosing sleep pathologies on the basis of patient histories and physical exams alone. These diagnoses had to be confirmed in the sleep laboratory. Such arguments effectively created the sleep clinic. Translating Neuroscience for the Clinic: The Brain Research Institute By 1980, there were at least thirty-four sleep clinics across the United States, most of which were less than five years old.50 Fifteen of these were run by psychiatrists and nine by psychologists, which was a testimony to the deep significance sleep had had for those fields for the better part of the twentieth century. Neurologists headed eight of the clinics, while only two were run by internists. Many of these clinics had emerged from sleep laboratories. That is to say, the sites had originally used human or animal subjects to investigate the normal parameters of sleep, as had Kleitman’s laboratory at the University of Chicago, but many were now also being used in medical diagnosis and treatment. Though Jacobson’s Laboratory for Clinical Physiology had offered a model of such a transition, it was stillborn, since Jacobson operated in an ill-defined space between academic medicine, psychology, and mainstream medical practice. Jacobson successfully combined research and clinical practice in his laboratory, but his definition of insomnia as a problem of ‘will power’ could never reach beyond the confines of individual sleep hygiene. Sleep had not yet been defined as a social issue, a problem of the other. The creation of neuroscience as an investigative field helped insomnia emerge as a public-health problem by offering up addiction as a potential platform upon which sleep research could move into the clinic. The phenomena of addiction had long been a useful device through which physiological research could grasp and reshape pathol-

368

The Sleep of Others

ogy and thereby contribute to both medical practice and public health. For example, one of the first investigative procedures adopted by Kleitman to study conditioning in the 1920s was to make dogs addicted to morphine and then measure changes in their levels of salivation as this reflex waxed and waned. Kleitman’s dogs, of course, were not models of drug addicts, for the latter were considered by most investigators of the day to be psychopaths and criminals, whose behaviour could hardly be replicated by experimental animals. Kleitman’s dogs were actually models of conditioned reflexes, the cyclical nature of which he then adapted to his analysis of sleep and wakefulness, as discussed in chapter 5. It was this contribution to the concept of conditioned reflexes, and not his work on sleep, that Kleitman had identified in a 1975 interview as his most important contribution to physiological research.51 When such conditioning was invoked to reformulate drug addiction as drug dependency during the 1960s, sleep emerged as a public-health issue, with the full weight of institutional and instrumental forces behind it. Unlike earlier experimental psychologists, post-war investigators turned their attention away from evoking and measuring bodily movements or glandular secretions, and instead united the EEG with stereotactic surgery to study the transmission of information between the sensory and motor nerves, regulatory structures in the deep brain, and the cerebral cortex. The synthesis of this research ultimately became an integral feature of ‘neuroscience,’ a term used to distinguish the experimental study of the normal flow of information in the nervous system from the neurologists’ analysis of brain anatomy and pathology. This distinction between the clinical orientation of neurology and experimental neuroscience was not, however, particularly clear to actors in the field during the 1970s. In an exchange of letters in 1978, for example, Horace Magoun, whose work in neuroanatomy was discussed in chapter 6, approached Francis Schiller and Webb Haymaker to help him write a book on ‘the founders of Neuroscience.’ Schiller asked, somewhat caustically, how the ‘neurosciences’ differed in any way from the ‘neurology’ featured in his and Haymaker’s previous book, which had included discussions of brain anatomy, physiology, pathology, surgery, and chemistry.52 Magoun took the charge seriously, responding that ‘neurology’ merely named a clinical specialty and was thus too narrow to accurately reflect the ‘aggregate,’ interdisciplinary, and scientific nature of ‘neuroscience.’ The latter, he claimed, ‘has come to connote, together with the neural sciences, the related behavioral and communi-

Insomnia Returns

369

cative sciences’ that provided the bedrock of a comprehensive study of the brain as a biological, medical, physiological, psychological, and sociocultural object.53 In the United States, no place exemplified neuroscientific research better than Magoun’s Brain Research Institute at UCLA. Although it emphasized basic research over clinical investigation, the BRI was, curiously enough, also home to one of the earliest sleep clinics to study insomnia and its relationship to hypnotic drugs. In addition, the BRI was the site of the Brain Information Service (BIS), an organization that produced a series of key publications in sleep research during the 1960s and 1970s, before the field had an official journal of its own. An analysis of how sleep was treated at the BRI is thus a useful way of understanding how the transformation of insomnia as a disease influenced the way sleep research was conducted at the institutional level during the 1970s. Magoun and Moruzzi’s 1949 discovery of the ability of the ‘reticular formation’ to regulate periods of sleep and wakefulness, discussed in chapter 6, had helped focus brain research on regulatory systems in the following decade. At the so-called ‘Laurentian Conference,’ held in Sainte-Marguerite, Quebec, in August 1953, participants struggled to understand how this rather ill-defined area of the brainstem could constitute a ‘system’ that was somehow responsible for conscious awareness but that clearly operated outside traditional sensory channels involving the cerebral cortex.54 The reticular formation was equally prominent in a 1957 conference held in Detroit, as well as a conference held in Moscow the following year, which led to the creation of the International Brain Research Organization (IBRO).55 This concept of ‘system,’ which was also transforming immunology at roughly the same time, offered a new view of the brain that was part way between the views espoused by morphological and behavioural research.56 The reticular formation emerged as a ‘non-specific’ arousal system, in that its location was not discrete but rather a diffuse, ‘net-like’ crossing of the posterior and anterior nerve fibres from the spinal cord. Nor did its activity correspond directly to any one of the special senses, thus rendering ineffective traditional psychophysiological experimentation, which relied on the observation and measurement of patterned behaviour following stimulation. The reticular formation appeared to be the crucial system that made just such overt behaviour possible by engineering wakefulness. It could be studied only electronically, by a process that treated the brain as part of a circuit that typically began with the electrical stimula-

370

The Sleep of Others

tion of a region deep in the brain and culminated with an EEG analysis of the cortex’s response. The microphysiology of brain science seemed to be rapidly reducing the study of the regulatory systems responsible for, among other things, sleep and wakefulness (see fig. 31). For many neurologists, psychologists, and even physiologists, the direction of this research represented something of a threat. A focus on regulatory systems implied a rejection, or at least a deferral, of analysing the messy complexities of behaviour in favour of EEG studies whose only relationship to human or clinical phenomena was through the medium of a waveform. One prominent critic, the British neurologist Francis M.R. Walshe (1885–1973), suggested in a 1955 letter to Magoun that the hosts of the Laurentian Conference had assigned far too much importance to the reticular formation. The result was that they had ended up ignoring the cortex, long established as the engine of behaviour among brain scientists: ‘No analogy is perfect, of course, but a rough one is that your reticular system provides the traffic lights that coordinate and make possible the traffic in the cortex and the spinal cord. However, the traffic lights are not the traffic, and this is the distinction that [Herbert] Jasper and [Wilder] Penfield seem in my judgement to have failed to appreciate. For them the reticular system is both traffic and lights, and the cortex is an intermittent spectator of the show.’57 Magoun responded by agreeing that others shared Walshe’s objections but that a rapprochement would ultimately be found between those interested in the cortex and those studying the brain stem. Walshe was unmoved, and noted that, in his attempt to provide a comprehensive review of the literature on the reticular formation, he found the writing on this particular topic to be filled with ‘redundancies’ and ‘obscurities.’ It was upon precisely this hope that clinical and basic research could work together in harmony that Magoun founded the Brain Research Institute at UCLA.58 Electroencephalographic research played a major role in the BRI, with special programs devoted to the mathematical analysis of EEG data by computer and the creation of telemetric techniques to monitor EEG and stimulate midbrain structures with implanted electrodes taking up large portions of the institute’s annual budget during the 1960s. When funding for the BRI peaked at just over $10 million in 1968, such research clearly held promise, but it was not all-consuming. About two-thirds of the research budget came from the National Institute of Neurological Diseases and Blindness, a division of the National Institutes of Health, while the remaining third came from the U.S. Air Force, which supported the institute’s Space Biology

Insomnia Returns

371

31 By the early 1970s, sleep research was beginning to orient itself more towards sleep pathologies than to the question of dreaming, with a subsequent increase in interest in non-REM sleep. This 1972 image depicts the neurophysiological mechanisms thought responsible for non-REM sleep (or ‘synchronized sleep’ [SS]). The negative signs depict active inhibition of the midbrain tegmental reticular formation (crosses) by areas in the basal forebrain (to the left) and the pons (to the right), resulting in a reduction and syncronization of the impulses between the cortex, ‘unspecific’ nuclei in the thalamus (the eggshaped figure in the centre), and the reticular formation. (Roger Broughton and Frederick Snyder, ‘Sleep and Clinical Pathological States,’ in Michael H. Chase, ed., The Sleeping Brain: Perspectives in the Brain Sciences [Los Angeles: Brain Information Service, 1972], 1: 365)

372

The Sleep of Others

Program. State funds, which were primarily focused on clinical care, amounted to no more than 5 per cent of the overall budget.59 Initially, the BRI was also supposed to be closely affiliated with the Neuropsychiatric Institute at the UCLA Medical School, but this association had become strained after both institutions had relocated to the same complex in 1961.60 Magoun, an anatomist by training, seemed to have little personal interest in clinical problems, since he tended to view patients as experimental subjects who might shed light on brain structure and function. The Clinical Neurophysiology Laboratory represented the most radical expression of laboratory–clinic relations within the BRI. There, researchers ran a series of trials in which they implanted microelectrodes, via stereotaxic surgery, in the deep-brain structures of patients with epilepsy and Parkinson’s. The electrodes often served triple-duty, since they were used to monitor seizure activity and drug effects, evoke responses by electrical stimulation, and destroy small regions of the deep brain in lieu of more invasive surgical procedures. This program continued throughout the 1960s but was eventually restricted to the study of drug-resistant epilepsy in 1971, after the introduction of L-dopa and a number of new neuroleptic drugs reoriented the treatment of the majority of such patients away from surgery and towards chemotherapy.61 The principal investigator of the program after 1971, Paul H. Crandall, later recalled how unusual it was that this research never encountered protests similar to those that evolved after 1970, when Vernon Mark and Frank Irvin published their controversial book, Violence and the Brain, which advocated brain surgery as a potential means of controlling some forms of criminal behaviour.62 Crandall credited the liberal atmosphere (for the investigators, if not for the patients) at the BRI for the continuation of such research, but he also noted that Vernon and Mark’s work was based on behaviour while the BRI’s was grounded in neurophysiology. Sleep, however, was still considered to be largely under the province of behaviour, and interest in it at the BRI was intimately linked to the experimental model involving the use of implanted electrodes in epileptic patients. One research program involved an electroencephalographic search for seizure activity during sleep in epileptic patients. The other, more substantial effort, led by Carmine Clemente, demonstrated that electrical stimulation of the basal forebrain could be used to establish a conditioned reflex to a constant tone that produced behavioural

Insomnia Returns

373

and EEG manifestations of normal sleep.63 The ‘dream research’ espoused in the APSS meeting of the early 1960s, however, was so far removed from the micropractices of the new neuroscience as to be nearly invisible. For investigators like Magoun, such research was still too clinical and too focused on mind, not brain, to be cultivated within his institute. Yet, by the late 1960s, this situation had begun to change. The initial psychiatric interest in using EEG to study dreaming started to migrate towards mainstream medicine and public health through insomnia research. In the process, sleep became a vehicle whereby EEG, the dominant imaging technique of cutting-edge neuroscience, could finally become translated into routine and widespread clinical practice. The career trajectory of one researcher, Anthony Kales, is particularly helpful in understanding this development. Kales (b. 1934) completed his MSc at UCLA in psychiatry in 1963. The topic of his thesis, ‘Reportable Mental Activity during Sleep,’ was a timely one.64 As already noted in chapter 6, the very concept of a ‘dream laboratory’ was coming under attack by a number of psychologists during the early 1960s. Deriving their arguments largely from experimental observations rather than from a perspective based on the philosophy of language, these psychologists criticized existing laboratory studies of dreaming along two main lines: the distinction between ‘dreaming’ and ‘non-dreaming’ could not be strictly mapped on to reports following awakenings from REM and non-REM sleep; and the laboratory context itself affected not only sleep patterns but dream reports as well.65 Foulkes, for example, began to study the correlations between stages of personality formation and reports elicited from subjects awoken in both REM and non-REM periods. By the mid-1960s, after he settled at the University of Wyoming, he quickly became the most outspoken proponent of studying the ‘sleep mentation’ that occurred at all hours of the night.66 Others took a more radical view. Hall’s content analysis of dreams extracted from laboratory subjects or recorded by subjects at home indicated that the nature of the former environment so distorted the psychological process of dreaming that its data could not be extrapolated beyond the laboratory walls.67 But such criticisms were the exceptions, not the rule. The majority of the new generation of REM researchers generally rejected the charge that, by applying the common parlance of ‘dreaming’ and ‘sleeping’ to the highly uncommon and interventionist context of the laboratory, they had actually changed the very meaning of these terms. REM researchers shrugged off this criticism by appealing to probabili-

374

The Sleep of Others

ties: it was more likely that awakened subjects would report having dreaming during a REM period than during any other.68 Kales’s thesis reviewed this literature and described a laboratorybased content analysis of dreams in which he acted as both subject and experimenter, a not uncommon practice for that particular era of dream research. Kales relied on the presence of absence or narrative structure, which he described as ‘progression’ or ‘elaboration,’ to distinguish ‘dreaming’ from ‘thinking’ in the reports he obtained from himself and his one other subject. He agreed with the critics, but only to the extent that some form of mentation, be it dreaming or thinking, did indeed appear to be present throughout sleep. It was also clear that the laboratory context did indeed influence both the manifest content of dreams (Kales did not attempt to uncover the latent content) and the patterns of sleep. Subjects appeared to require several nights to adjust to the laboratory conditions, and, even then, they continued to dream about the experimenter, the apparatus, or the sleep laboratory itself. But, Kales insisted, reports of dreaming, rather than merely thinking, were far more frequent in REM awakenings than in non-REM ones. Kales’s observations were neither novel nor controversial. But they did reflect the state of the art in 1963, when EEG analysis in the sleep laboratory was still primarily aimed at understanding the relationship between REM and dreaming. What is particularly noteworthy about Kales’s career is that he did not pursue this direction of research over the following few years. Rather, he began to apply his technical experience in the sleep laboratory to the clinical problem of insomnia. This shift began to take place in the mid-1960s, shortly after the Kefauver-Harris Amendments had been adopted and just before the FDA had set down the new guidelines for testing drug safety and efficacy. The hybrid nature of Kales’s new research appeared in a 1967 paper, in which he combined dream-recall studies with REM deprivation and an analysis of drug effects on sleep.69 The study of sedatives quickly took priority over his earlier interest in dreaming. The following year, he organized a symposium on sleep at UCLA, in which clinical questions about sleep pathologies and hypnotic use and withdrawal were as well represented as physiological studies of sleep and dreaming.70 That the changes in drug legislation had had some effect on this shift towards the clinic is borne out by the fact that pharmaceutical firms had begun to sponsor Kales’s research. In comparison to the monies provided by the National Institute of Health and the U.S. Air Force, the rel-

Insomnia Returns

375

ative amount of drug sponsorship was quite small. The total funding for projects run by BRI members for 1968–9, for example, was a record $10.37 million, but the contribution from pharmaceutical firms accounted for a mere $50,000 (or just under .5 per cent).71 The funding came from a host of companies: Sandoz, Geigy, Pfizer, HoffmanLaRoche, Arnar-Stone, Abbott, and Lakeside. But the fact that the funding was so minimal and spread between so many companies should not mask the reality that 80 per cent of these funds were going towards Kales’s study of the ‘effects of sedatives on sleep patterns,’ a trend that continued until Kales left UCLA to join the Department of Psychiatry at the Hershey Medical Center at Penn State in 1971. Changing drug legislation had brought the pharmaceutical industry into the sleep laboratory, and it was now beginning to turn at least some investigators’ attention away from dream research and towards insomnia. Upon his arrival at Penn State, Kales set out to make this new relationship between the sleep laboratory and the sleep clinic more concrete. In a 1974 paper he and his wife, Joyce D. Kales, proposed a new normative vision of laboratory-clinic relations in sleep research.72 The study of sleep, they argued, had arrived at a crossroads. Sleep laboratories that had once elicited data about normal sleep were now turning towards pathology. Consequently, sleep clinics, which placed laboratory research in an outpatient setting, had begun to appear across the country. But were these new sleep clinics the physical site of an emerging medical specialty? Or were they more akin to a traditional biomedical laboratory, which served as a place for analysis and research, the results of which could then be communicated to the practising physician? Dement and his associate at the Stanford Sleep Clinic, Christian Guilleminault, had assumed the former position at the APSS meeting in San Diego in 1973. Anthony and Joyce Kales had taken up the latter argument. Using the problem of insomnia as their model, the Kaleses insisted that the primary task of the new sleep-disorders clinics was to screen populations and diagnose patients for such disorders by a variety of methods, including psychological tests, psychiatric interviews, and allnight EEG studies (see fig. 32). Epidemiological information and individual diagnosis would be communicated to the physician, who, as a general practitioner, internist, psychiatrist, or neurologist, was already established as the primary actor in the diagnosis and treatment of sleep pathologies, among which insomnia, of course, was the most common. Another major role of the laboratories housed within these clinics

376

The Sleep of Others

would involve the evaluation of new and existing drugs that affected sleep, with particular attention to their withdrawal effects or potentially addictive qualities. This work would be conducted for pharmaceutical firms in their efforts to meet FDA regulations (see fig. 33). Finally, a limited amount of laboratory time would go towards the psychophysiological study of normal sleep that had first launched the sleep laboratory. In the Kaleses’ scheme, the ‘sleep specialist’ did not, and should not, exist. This position was not the logical consequence of surveying the entire field of sleep research; rather, it was the product of their appropriation of insomnia as a model disease, which, they argued, was typically the product of a psychopathological problem that the patient refused to recognize.73 This underlying psychopathology belonged to the province of the psychiatrist or neurologist, not of a new specialist, the foundation of whose expertise lay primarily in the scoring of sleep stages. Kales spoke from considerable authority on this last point, since it had been he and Rechtschaffen who had co-edited the first manual for standardizing techniques, signs, and terminology in sleep research back in 1967.74 Pharmaceutical firms and physicians alike, the Kaleses complained, encouraged the insomniac’s delusion. On the one hand, physicians, in their ignorance, would prescribe ineffective or even dangerous drugs for their insomniac patients, without bothering to counsel these patients to get at the psychological root of their problem. On the other hand, companies would market their prescription sedatives as specific remedies for insomnia, rather than as the symptomatic treatments they actually were. The Kaleses were particularly acerbic in their contention that drug companies misrepresented laboratory data regarding drug effects on sleep stages and REM suppression to hawk their wares.75 Advertisements that cited evidence correlating REM deprivation to psychosis ignored the inconclusive results of such studies; in fact, the Kaleses pointed out, there had been little progress in determining the function of any specific sleep stage.76 What laboratory evaluation could provide, they argued, was an understanding of the global effects of drugs on individual patients’ sleep over the long-term, information that the one or two nights of polygraphic studies cited in most drug advertisements inevitably failed to provide. Again, Anthony Kales was in a strong position to make such points, for he had, quite literally, written the book on this topic. In 1971 he and Louis Lasagna (1923–2003) had crafted an experimental protocol for evaluating hypnotics for the FDA. Lasagna later became an outspoken critic of the FDA’s increasingly

Insomnia Returns

377

32 The education of physicians regarding sleep disorders first became a major concern of the sleep-research field during the 1970s. This image depicts the physician, not the sleep-disorders specialist, as the target of information about sleep pathologies. (Anthony Kales, Edward O. Bixler, and Joyce D. Kales, ‘Role of the Sleep Research and Treatment Facility: Diagnosis, Treatment and Education,’ Advances in Sleep Research 1 [1974]: 391–415, 392)

33 This depiction of the interactions between patients, physicians, sleep laboratories, the FDA, and pharmaceutical firms was largely modelled on the publichealth debates surrounding insomnia during the 1970s. (Anthony Kales, Edward O. Bixler, and Joyce D. Kales, ‘Role of the Sleep Research and Treatment Facility: Diagnosis, Treatment and Education,’ Advances in Sleep Research 1 [1974]: 391–415, 412)

378

The Sleep of Others

strict and inconsistent application of its own rules, which, Lasagna argued, deterred pharmaceutical firms from developing new drugs.77 The redirection of Kales’s research interest from a laboratory-based dream science to the diagnosis and treatment of insomnia was an important part of a larger process – the diffusion of the sleep laboratory beyond the narrow goals of neuropsychiatric diagnosis. In contrast to the largely internal significance of Dement and Guilleminault’s unravelling of narcolepsy, the application of the sleep laboratory to the wellknown, but seemingly intractable, problem of insomnia resonated widely. The American public proved eager to hear about the corrupt practices of drug companies, the compromised (or even unethical) position of the medical practitioner, and the widespread hypocrisy of many citizens, who lashed out at drug addicts and dope smokers while husbands boozed it up after work and wives popped pills to soothe their frazzled nerves. What the Kaleses were suggesting was not ‘anti-psychiatry.’ After all, they were proposing to give the psychiatrist, among other practitioners, more power to diagnose psychopathologies, not less. Yet the reforms proposed by the Kaleses did have some of the trappings one might expect from a ‘fellow-traveller.’ The idea of a biomedical technocracy was decidedly out of vogue during the mid-1970s, and their criticism of the idea that there needed to be a ‘sleep specialist’ played into such misgivings. The ‘whole patient’ they argued, required caring, understanding, and proper counselling, not more drugs. Most important, patients were not simply objects of medical knowledge. They were also consumers, and, as such, they had the right to expect their government to protect them from predatory business interests. The primacy placed on physician education and drug regulation in the Kales’s vision of the sleep clinic had considerably more in common with congressional anti-trust hearings of the 1960s and 1970s than with any talk of a free and unfettered medical marketplace. The Institute of Medicine Report Protecting consumers in the medical marketplace had been a recognized role of government since the ‘trust-busting’ days of the Progressive era. The new drug laws of the 1960s, as I have already pointed out, had originated in congressional inquiries into price-fixing in the drug industry. Before news of the thalidomide disaster spread, legislators were less interested in the safety and efficacy of the drug industry’s products than they were in ensuring that the pharmaceutical market

Insomnia Returns

379

remained competitive. Estes Kefauver himself had started his series of congressional hearings with the intention of changing patent legislation to restrict the formation of pharmaceutical monopolies; safety and efficacy simply proved to be the back door through which parts of his proposed bill were passed into law. By the 1970s, however, questions of safety, not competition, began to dominate the political management of the medical marketplace. In addition to its established role of regulating the medical profession and its affiliated industries, the American government took on the additional task of protecting consumers from themselves. This process had already been instigated with the spread of the concept of iatrogenic ‘dependence’ and the decriminalization of addiction in the 1960s. As a result, governments and researchers alike began to examine the dynamics of drug prescription and consumption anew. In the case of sedatives and hypnotics, the upshot of these new investigations was an increase in calls for the creation of a ‘rational’ basis for prescribing such medications by expanding the field of sleep research. This investigation began, however, with the irrational. In August 1977 the recently elected U.S. president, Jimmy Carter, established a commission to study the use of barbiturates in medical practice through the National Institute of Drug Abuse (NIDA). The Carter administration was considering a ban on barbiturates, not so much because these drugs were addictive as because they were so frequently cited in deaths due to accidental overdose and suicide. Alternative sleeping pills, in the form of the benzodiazepines, were now available, and it seemed reasonable to assume that the substitution of these safer drugs for the barbiturates could help prevent citizens from taking their own lives. That same year, the NIDA published a study that noted that barbiturates were involved in almost 5,000 deaths each year. This attracted the attention of the Institute of Medicine of the National Academy of Sciences, whose steering committee and advisory panel consisted of a number of prominent sleep researchers as well as several psychiatrists and pharmacologists. The IM took up the issue, with a mandate to examine barbiturate use and its relationship to drug overdose and suicide, and produced a report two years later.78 But this report proved far more expansive than a mere drug survey. Following, as it did, several decades of publicity that had enshrined tranquillizer use as a symbol of how the nation’s medical practice had been corrupted by industry and compromised by ignorance, the report called for further, internally generated medical reform.79 The report’s conclusions were somewhat surprising. Rather than ban-

380

The Sleep of Others

ning barbiturates as sedatives and replacing them with the benzodiazepenes, the authors urged the federal government and the medical profession to take insomnia seriously and support sleep research in universities and education in sleep disorders in medical schools. This appeal was based on their conclusion that, contrary to current belief among many medical practitioners, benzodiazapenes were not safer than barbiturates in any straightforward way. True, it was much more difficult to ingest a deadly dose of flurazepam than it was to kill oneself with barbiturates, but most drug-related deaths also involved alcohol, and, in such situations, both kinds of sleeping pills could be lethal. The epidemiology of suicide by drugs was no more comforting. Between 1954 and 1968, barbiturate overdose accounted for about anywhere between 60 and 80 per cent of drug-related suicides in the United States Between 1970 and 1976, this figure dropped to 30 per cent, which reflected the rapid decline in barbiturate prescriptions following its rescheduling in 1973.80 Flurazepam, which began to outsell barbiturates in 1974, had clearly replaced the older drugs as the sleeping pills of choice by the end of the decade. Yet the overall rate of drug suicides had declined only slightly, as people turned to cocktails of anti-depressants and alcohol that ultimately ended their lives. Things were little improved on the addiction front. If the definition of addiction was restricted solely to the appearance of physical symptoms upon withdrawal, then the benzodiazepenes were clearly superior to the barbiturates.81 Yet the ground easily shifted to ‘dependence,’ and, on this count, flurazepam fared poorly. Like the rest of the benzodiazepines, it was suspected of helping to create a situation in which patients felt helpless without their drugs, even after the condition that had prompted their initial prescription had apparently been successfully treated. Safety issues were further complicated by the fact that benzodiazepine metabolites remained in the patient’s system for up to eight days – far longer than was the case with barbiturates. This prompted concerns regarding the risk of impaired performance, especially in the operation of heavy machinery or cars, long after the drug regimen had ended. So why had the medical profession not responded accordingly and proceeded to prescribe sleeping pills with additional caution, rather than simply substituting the newer drugs for barbiturates? The difficulty lay in the disconnect between insomnia as a problem of medical practice and insomnia as an object of medical theory and research. While the latter took nothing for granted, the former, argued some critics, simply approached insomnia as a familiar part of medicine’s ‘therapeutic

Insomnia Returns

381

ritual,’ which was just then being subjected to historical, sociological, and psychological analysis.82 A Boston psychiatrist, Ernest Hartmann (b. 1934), addressed this irrational side of medicine in his popular 1978 book on insomnia, The Sleeping Pill. Hartmann’s message, that ‘insomnia is not an illness for which a sleeping pill is the cure,’ was grounded in his claim that the prescription and consumption of such drugs was in fact an unrecognized and illegitimate substitution for a genuine therapeutic encounter.83 Like some of his sociologically and historically oriented counterparts, Hartmann suggested that the incursion of biomedical technologies into all aspects of medical practice had corrupted the genuine human concern and respect between doctor and patient that had long characterized medicine and that constituted an integral part of its therapeutic effects.84 Insomniac patients, Hartmann insisted, denied that they suffered real psychological problems, while physicians, in their desire to reassure patients of their sanity and to meet expectations of a cure, responded with a prescription. Hartmann’s historical and ethical position was buttressed by a number of laboratory studies that suggested that a placebo was often as effective as a hypnotic in improving sleep. Much of this work was his own, based upon a sophisticated experimental protocol which recorded the EEG effects of drugs and placebo in a subject for up to eighteen months, thereby allowing placebo effects to be compared directly with hypnotics in the same subject. This process, he admitted, was far too expensive to be used for clinical trials of hypnotics, a procedure with which Hartmann was intimately familiar, having served as a panel member for the FDA from 1972 to 1976. But his intensive exploration of the psychodynamic context of insomnia complaints and drug prescription and consumption did suggest the possibility of insomnia research moving towards an integration of the psychological, sociological, and neurophysiological sciences. Such an investigative practice, which would presumably ground the diagnosis and treatment of insomnia in the rational analysis of the disease’s irrational aspects, would require a new scheme of funding, one that originated with government funding agencies rather than drug companies. And this was precisely what the IM report aimed at when it called for ‘new, more sophisticated multi-disciplinary research to be initiated to test the efficacy of various pharmacological, psychotherapeutic, behavioral, and psychosocial treatment approaches to various types of insomnia.’85 The report concluded with a plea that a new authoritative body be created under the auspices of the U.S. Public Health Service

382

The Sleep of Others

(the same organization that oversaw the work of the National Institutes of Health) which would work with the specialty societies, medical associations, and academic health centres to coordinate sleep research and the medical management of sleep disorders, while also increasing public awareness of the multifaceted problem of insomnia. Although such a recommendation might have made for good science policy, it made poor press. Rather than focus on the call for increased funding for sleep research, it was the report’s ambivalence regarding the relative benefits and risks of hypnotics that received the lion’s share of coverage by the media. A front-page article in the New York Times, for example, observed that too many Americans were taking too many sleeping pills over too long a period of time.86 The Journal of the American Medical Association (JAMA) offered a bit of sober self-reflection in a news article, noting that ‘a new wave of criticism of the drug-prescribing practices of American physicians is taking shape’ in the wake of the IM’s report, and observing that advertisements for flurazepam inevitably failed to indicate the very small number of patients who had been tested to demonstrate that the drug was effective over a twenty-eight-day period.87 Readers of the JAMA were, of course, well familiar with such ads, usually for Hoffmann-LaRoche’s Dalmane, which they would have seen in virtually every issue since the early 1970s. A summary in Science, on the other hand, pressed much harder on the drug companies by suggesting that flurazepam was addictive, although less so than the barbiturates, and that drug companies simply ignored or misrepresented these risks.88 Science actually reproduced a Dalmane ad that featured a male head at the horizon of a surreal landscape dominated by EEG tracings, with the taglines ‘Architect of Sleep’ and ‘Sleep Laboratory Proof’ emblazoned on the bottom corner. The author of the article had clearly taken a cue from the IM report’s criticism of such advertising, which, it claimed, misrepresented insomnia as a unitary disorder that could be cured by a specific remedy, whose effectiveness over a precisly defined period of time had been clearly demonstrated. Nowhere, the report observed, did these ads indicate that such studies (two, in fact) had involved a carefully selected group of only five chronic insomniacs. ‘The emphasis on the sleep laboratory technology,’ the report concluded, ‘distracts attention from the absence of data on subjective relief and daytime measures of drug-induced improvement or impairment.’89 The public response to the IM report captured the attention of both the report’s authors and the drug companies. The former, fearing that physicians would abruptly take their patients off hypnotics,

Insomnia Returns

383

claimed to have ‘explicitly eschewed the terms “drug dependence” and “addiction”’ because so little was known about these terms, and had instead framed the problem as ‘nightly reliance on drugs for sleep.’90 A scientist from Roche laboratories, on the other hand, argued that the Science article confused ‘tolerance,’ or the loss of effect at therapeutic doses, with ‘addiction,’ defined as physical dependence, which was rarely found with benzodiazepines. As to the misrepresentation of laboratory evidence, the Roche scientist insisted that the relevant data dealt with ‘insomniac’ patients whose condition and drug response were objectively measurable and that the increased precision of measurement eliminated the need for a large-population study. The author of the Science article responded to the first criticism by noting that the IM report had indeed used the term ‘dependence’ in its description of benzodiazepines. To the second, he argued that the distinction between insomniac and non-insomniac patients who took flurazepam was spurious, because the overwhelming majority of people who took the drug – or any drug – for insomnia did not suffer from the classic version of the disorder that could be objectively demonstrated in the sleep laboratory. Regardless of which side of the debate investigators found themselves on, it was becoming increasingly clear that public and institutional concerns about addiction were providing a platform upon which insomnia, and with it, sleep research, could emerge as a public-health issue. The IM report even functioned as an internal critique of sorts, since it levelled precisely the same criticism at the authors of the very studies that the pharmaceutical firms had cited in their ads. Sleep researchers, the report suggested, had themselves fallen victim to the routinized generation of visual and graphical data that were in some ways detached from the experience of sleeping, or of suffering from its lack. ‘There must be clear definitions of the specific problems and of the goals of the treatment,’ the report stated. ‘Above all these goals must be related to the actual problems. Many sleep laboratory studies have omitted significant descriptions of the subjective complaints of the individuals whose responses to hypnotic drugs were being measured. In fact, in a good number of studies the subjective effects upon the individual have gone unreported in the midst of a flurry of EEG tracings.’91 The charge was not particularly new – Moruzzi, the co-discoverer of the ascending reticular formation’s role in regulating sleep and wakefulness, had argued a decade earlier that the neurological sciences were beginning to ignore the behavioural aspects of sleep in favour of a preoccupation with its mechanisms.92

384

The Sleep of Others

The sleep laboratory, like many bastions of biomedical confidence in post-war science, had come under attack from within over the course of the 1960s and 1970s. Carried to the fore by concerns about addiction, insomnia’s dual nature as a psychopathological suspect with a somatic name represented both a challenge and an opportunity to sleep research. It offered a lucrative and expansive segue from the laboratory into the clinic, while at the same time it obliged sleep researchers to reconsider their reliance upon the EEG, which had initially proven so successful in establishing their field as an objective, laboratory-based solution to age-old questions about dreaming. Symptoms and Signs: Creating Sleep Disorders Medicine Among many sleep researchers, the prospects of creating a medical specialty out of their field seemed almost at hand. An editorial appearing in the first volume of Sleep, the official journal of the APSS, presaged the Institute of Medicine report by suggesting that 1979 would be ‘an even more promising year for sleep as a full-fledged triple-threat academic discipline.’93 The editors noted that all hope was centred upon sleep’s clinical prospects, since some forty million Americans were afflicted with sleep problems yet few medical schools devoted any time at all to instruction in sleep disorders. The U.S. Public Health Service had already launched an initiative in furthering sleep research, which would blossom following the release of the IM’s report. Consensus in the medical field, which was almost entirely lacking in the study and treatment of insomnia, was to be a major goal following the IM report. One article expressed little more than cautious optimism that even a definition of insomnia – not as lack of sleep but as the complaint of sleep loss – could be determined in the absence of ‘the classification of psychopathology in some enduring manner’ and ‘the forging of a consistent attitude to the discrepancy between subjective and objective estimates of sleep loss.’94 Such ambivalence and uncertainty was by no means limited to the field of sleep disorders; it had come to characterize American psychiatry by the end of the 1970s. The profession, which had long prided itself on its eclecticism and pragmatism, was in the process of a major purge of the psychoanalytic concepts that had dominated psychiatric theory and practice since the 1950s. Diagnosis was the most prominent battleground of this conflict, and standardization and codification became the order of the day, as the third edition of the Diagnostic and Statistical Manual of Mental Disorders, more com-

Insomnia Returns

385

monly known as the DSM-III, was almost completely revised to espouse biological and psychopharmacological concepts of mental illness in place of psychodynamic approaches. The IM’s report had tacitly paid tribute to this transformation by acknowledging that, in contrast to ‘the opiates, the antidepressants, or the neuroleptics (antipsychotic drugs), for which there is broad scientific consensus about some of the neurochemical mechanisms associated with their psychological effects,’ the ‘precise mechanism’ by which hypnotics worked remained unknown.95 Unlike other their psychiatric counterparts, who relied upon concepts like the ‘dopamine receptor theory’ to tie together the investigation, diagnosis, and treatment of schizophrenia, sleep researchers lacked a ‘platform’ whereby they could link similar elements together in a integrated system dedicated to the medical problems of sleep.96 Sleep researchers readily acknowledged these problems. Whereas their field had once been filled with psychiatrists and psychologists optimistically applying the hybridized styles of experimental psychology and ‘dry’ electrophysiology to the problems of sleep and dreaming, research was now becoming infused with the tools and techniques of ‘wet’ neurochemistry and its applications for newer classes of drugs. Daniel F. Kripke, a psychiatrist and sleep researcher working out of the San Diego Veterans Affairs Hospital who had served on the IM panel, lamented the fact that many investigators had abandoned the ‘aged’ sleeping pills ‘for romances with the younger antipsychotics, antidepressants, lithium salts, and so forth.’97 He compared this to the tendency during the 1950s and 1960s to reject ‘old-fashioned and modest technologies of solar heating and windmills in favour of nuclear energy,’ and concluded that ‘youth had triumphed over sober policy.’ What little integration sleep researchers did enjoy originated with the EEG, a poorly understood but widespread technology then nearly half a century old. Consensus, in this instance, had to be built up from the clinical end of the field, rather than the ‘engineering’ side of basic research, which was then encountering some resistance in terms of its inability to incorporate the subjective nature of insomnia (as well as of dreaming and sleeping in general). Standardization, as always, was key. Efforts to this effect had already been made ten years earlier by Rechtschaffen and Kales, in their manual of EEG scoring for sleep research, which had been universally adopted. But the process of standardizing the diagnosis of sleep disorders began in earnest only during the early 1970s. One of the first efforts to create a nosological schema for sleep disorders came from Roger Broughton, a neurologist at the University of Ottawa. Broughton

386

The Sleep of Others

(b. 1936) had studied at Marseilles from 1962 to 1964 under the French clinician Henri Gastaut (1915–95), who specialized in clinical neurophysiology and epilepsy. Gastaut assigned Broughton to the problem of determining whether the somnambulism and bed-wetting experienced by some epileptic children in his clinic was an essential part of their epilepsy, or simply coincidental.98 Broughton, following a series of polygraphic studies of these patients, concluded that these symptoms were unrelated to epilepsy and rather amounted to disorders unique to sleep. Upon his return to Canada, Broughton continued to use the EEG to map out the domain of sleep-specific disorders, including nocturnal enuresis and night terrors in children and ‘incubus attacks’ in adults.99 In keeping with the tenor of the times, his use of the EEG did not preclude a psychoanalytic account of the aetiology of these disorders. For example, he accounted for the fact that children suffering from night terrors tended to awaken with bloodcurdling screams during non-REM sleep by suggesting that their primordial urges had broken through into consciousness without the beneficial work of distortion and condensation that produced normal dreams. For Broughton, as for Dement, electroencephalography functioned as a means of inscribing the activities of the sleeping mind. It did not necessarily reduce these activities to brain functions. Broughton drew upon his experience with epilepsy when he proposed a classification of sleep disorders as part of a workshop on the ‘nosology and nomenclature of the sleep disorders’ organized for the APSS annual meeting in 1972.100 Sleep disorders, he argued, were similar to epileptic seizures and should likewise be classified according to their symptoms, because this was how patients presented to clinicians and because the exact causal mechanisms remained unknown. Broughton thus proposed five classes of sleep disorders. In his scheme, the two cardinal symptoms of genuine narcolepsy, sleep attacks and cataplexy, fell into two different categories. The former was a specimen of ‘excessive sleep,’ while the latter was paradigmatic of sleep disorders involving motor phenomena, which also included sleep paralysis, snoring, and restless-legs syndrome. ‘Inadequate sleep’ covered all variations of insomnia, while sleep disorders involving psychosensory phenomena, such as sleep terrors, provided a counterpoint to the category encompassing motor phenomena. Finally, sleep disorders involving autonomic functions described a fifth category, which included enuresis, periodic respiration, and painful erections during sleep. Broughton’s emphasis on symptoms rather than pathophysiology or etiological theory was characteristic of the transitional situation in

Insomnia Returns

387

which sleep researchers found themselves during the 1970s. On the one hand, their emphasis on laboratory-based observation indicated their desire to model their burgeoning medical specialty after conventional biomedicine, which reached a diagnosis according to a perceived ‘fit’ between clinical symptoms and signs generated by biomedical assays. But the fact that the object in their laboratories was not merely a pathological sample but a complete human being made it difficult to put such a reductionistic model into practice. The penchant for aspects of psychodynamic or psychoanalytic theory thus persisted in sleep research in the face of widescale attempts to render psychiatry ‘biological.’ Sleep labs, after all, were different creatures than those that received blood, urine, or tissue samples, or even those that undertook the labour to generate images, such as X-rays, directly from the patient’s body. As several historians and sociologists have repeatedly pointed out, such laboratories have the luxury of treating their samples as pure commodities and can easily exercise control over the labour process that converts such samples into diagnoses in an efficient and rational manner reminiscent of that which takes place on any factory floor.101 In the sleep laboratory, on the other hand, the patient’s very activity of sleeping, rather than any product of that activity, is the phenomenon being converted into grist for the diagnostic mill. Recording nocturnal behaviour is extraordinarily time-consuming and expensive, and there is less room for an economy of scale. Specialized physical sites that are both soundand light-proof need to be built. Mutually convenient times need to be established for the recording process. One or more well-trained EEG technicians need to be present all night to ensure that the recording procedure does not fail. Patients need to be convinced that the process is comfortable, convenient, and safe – issues that are of little interest to blood or sputum samples. Given such conditions, it is unsurprising that diagnosing a patient in a sleep laboratory was an unlikely option during the 1970s. In order for sleep medicine to expand, it could hardly rely on the widespread access to such facilities among physicians. Thus, its proponents continued to rely on the patients’ symptoms in their nosological systems, a trend that, in some senses, went against the reductionist grain of twentieth-century biomedical tradition. The First Sleep-Disorders Manual The initial drive towards standardization nonetheless culminated in 1979, when the APSS and the Association of Sleep Disorders Centers

388

The Sleep of Others

(ASDC) published the first edition of the Diagnostic Classification of Sleep and Arousal Disorders (DCSAD). Like the Institute of Medicine’s report of the same year, the DCSAD was devised against the backdrop of physicians’ so-called irrational treatment of sleep problems in the midst of the appearance of standards in the diagnosis of narcolepsy and the spread of sleep clinics across North America. The content of the DCSAD was a perfect mirror of its composition, in that it appealed to the expertise of individual practitioners rather than to any existing consensus in the field. In February 1976 the ASDC created a committee that consisted largely of psychiatrists who directed sleep clinics, with the aim of creating diagnostic standards for all sleep disorders.102 The sleep laboratory had been one origin of this drive for standardization, and the ASDC had recently applied the specifications of Rechtschaffen and Kales’s 1968 manual to the clinical polysomnography that inhabited the new sleep clinics spreading across North America. The DCSAD, however, was equally motivated by the movement to reform psychiatric diagnosis, which culminated in the DSM-III and the ninth revision of the World Health Organization’s International Classification of Diseases (ICD-9). As such, the ASDC’s standards were designed more as a set of research parameters proposed by experts in the field than as reflections of classifications already in existence. The document was generative, not representative. Like the Institute of Medicine’s report, the DCSAD emphasized the irrational way in which sleep disorders were currently diagnosed and treated. Current standards of scientific medicine, it argued, had been completely inverted. When confronted with complaints about sleep, physicians rejected the conventional hierarchy of sign over symptom and simply took patients at their word. They then proceeded to prescribe drugs for sleep disorders without ever verifying the accuracy of the patients’ reported symptoms, and often without following up on the results of the treatment. The prescribing of hypnotics to treat insomnia, was, of course, the most salient case in point. Yet so little was known about normal sleep that classifying sleep disorders according to pharmacological results seemed an unlikely starting point for a useful nosology. Pathophysiological signs, which could be visualized through PSG, seemed more appropriate for this task. But, although normal sleep architecture had been reasonably well established, the significance and function of the various sleep stages remained unknown. They were purely descriptive at this point, and, as already noted, much criticism had already been levelled against pharmaceutical firms for misrepre-

Insomnia Returns

389

senting this fact in their advertisements. Availability of PSG was also a problem, since most clinicians outside major urban centres were unable to refer their patients to a sleep laboratory for diagnosis.103 The solution advocated by the DCSAD was to organize sleep disorders according to symptoms. These symptoms, however, had to be finely pared in order to produce the desired differential diagnoses. In particular, the DCSAD created two major categories of sleep disorder: disorders of initiating and maintaining sleep (DIMS) and disorders of excessive somnolence (DOES). Insomnia was the model of the former, while narcolepsy provided the model of the latter. Most clinicians were well able to distinguish the two disorders from each other. Insomnia looked much like the neuroses, even when it appeared in its ‘idiopathic’ or ‘psychophysiological’ form, and resisted identification with emotional distress or a deviant personality. The patient’s mental life dominated insomnia’s symptomatology. Narcolepsy, on the other hand, continued to appear as a sort of epilepsy, defined, as it was, by seizure-like sleep attacks and motor dysfunction. But, while narcolepsy maintained its integrity within the DCSAD, insomnia quickly fractured into ‘the insomnias.’ Most diagnoses falling under this category (the DIMS) were associated with psychiatric disorders. In general, the inability to fall asleep was linked with ‘symptom and personality disorders’ – that is, anxiety and its many manifestations. Difficulties in maintaining sleep, on the other hand, were associated with the ‘affective disorders,’ the most significant of which was major depression. The archetypical insomniac, whose most persistent worry concerned his or her inability to fall asleep, reappeared under the guise of ‘psychophysiological insomnia.’ This was considered the most common category of insomniac and was also the one held responsible for the majority of sleeping-pill prescriptions.104 ‘Transient’ or ‘situational’ insomnia followed an emotionally charged ‘precipitating event,’ such as a sudden death in the family or a divorce. If the insomnia lasted longer than three weeks, however, the insomnia was reclassified as ‘persistent.’ Causation was reascribed to ‘two separate but reciprocally amplifying sources of interference with sleep: somatized tension-anxiety and conditioned association.’105 Patients ‘discharged’ their anxiety into physiological channels, while they also developed a conditioned response to the context of sleeping, either in the form of external stimuli, such as their bedroom, or internal ones, such as the attempt to try to fall asleep. Typical patients would report sleeping better in a strange place, such as in a motel while on vacation, than at home. Practitioners were warned that the disorder

390

The Sleep of Others

‘was difficult to clarify diagnostically,’ in part because ‘the picture is often additionally clouded by chronic use of hypnotics.’106 In many ways, the DCSAD was a diagnostic culmination of several notable controversies and clinical investigations that had surrounded sleep for much of the twentieth century. The three-week cut-off point that separated transient and persistent psychophysiological insomnia approximated the length of time that the most popular hypnotic of the time – flurazepam – was supposed to maintain its effectiveness. This implied what the Institute of Medicine report had stated outright – that hypnotics were overprescribed and were therapeutically useful only in a minority of insomnia cases. The fact that the sleep laboratory qualified as an ‘unfamiliar place’ in which such patients tended to sleep better further diminished its status in the diagnosis of psychophysiological insomnia, and alluded to the debate regarding the utility of relying on ‘laboratory dreams’ in psychological research. Instead of being encouraged to rely on technologically generated signs, physicians were advised to pay more attention to their patients’ complaints, which might reveal useful clues about the contextual nature of their insomnia and lead towards appropriate classification and therapy. The fact that persistent insomnia was described as an ‘interaction’ between the tendencies to somatize anxiety and to associate feelings and objects with sleeplessness paid tribute to the pre-REM sleep medicine proposed by Jacobson and Pavlov. In such situations, drugs were of limited usefulness and could be supplanted by behavioural therapy. By emphasizing symptoms over signs, the DCSAD indicated that the practice of sleep medicine was to remain closely allied more to psychiatry than to internal medicine or even neurology. Although early interest in dreaming and REM-related pathology seemed to be in the process of being supplanted by work done in narcolepsy and insomnia during the 1970s, sleep medicine continued to be tied to psychiatric practice through the patient interview. Listening to patients, a practice advocated by both the IM report and the DCSAD, took priority over routinely sending patients to spend a night or two in a sleep laboratory for diagnosis. Yet the use of the sleep laboratory for diagnosis was nonetheless growing. Indeed, insomnia seemed to be furnishing a prime opportunity to put the sleep lab to use. One 1975 poll indicated that, of a sample of 574 patients diagnosed by 73 clinicians, 447 (roughly 78 per cent) suffered from some form of insomnia, two-thirds of whom had insomnia complicated or caused by drug or alcohol abuse.107 In contrast, only about 10 per cent of the total number of diagnoses were for narcolepsy.

Insomnia Returns

391

Patients spent an average of 3.11 nights sleeping in the laboratories, at a mean cost per night of about $220 (the equivalent of about $830 in 2006 dollars).108 The study concluded that the use of the sleep laboratory for diagnosis had become part of accepted clinical practice and that health-insurance agencies should therefore pay the costs of the procedure. Most agencies then considered diagnosis in the sleep laboratory ‘experimental,’ however, and laboratory costs were taken out of research or hospital funds or charged directly to the patient. The poll, which was sent out only to APSS members, was perhaps less a measure of accepted clinical practice than it was an indication of the clinical direction that the sleep-research field was taking during the 1970s. Most of the respondents had a sleep lab, and of the thirty-eight sleep labs represented, twenty-three were located in hospitals, medical schools, or medical-sciences buildings. Certainly this presence of sleep laboratories within hospitals, coupled with the sleep researchers’ drive for public representation in the debates over insomnia, was an important factor in the roughly twofold increase of sleep disorders diagnosed by all clinicians in the United States between 1975–6 and 1979.109 Debates over the use of PSG in the diagnosis of insomnia began to appear in the late 1970s and early 1980s. Several psychiatrists, including Anthony Kales, decried the overuse of sleep laboratories in diagnosing the disorder, which, he argued, tended to detract from the comprehensive approach needed to manage the problem.110 One potential solution to this situation was to encourage the complete physical and operational integration of the sleep clinic within general hospital practice, rather than to have free-standing sleep centres that were built around a sleep laboratory and that operated on a referral basis.111 Variants of this debate persisted for decades, as clinicians argued over whether to refine existing nosologies according to the theoretical aetiology of the disorder or to deploy purely empirical or operational definitions instead. The very idea that a typical ‘insomnia patient’ could be reliably modelled around a set of either organic or psychological symptoms began to fall apart.112 Indeed, the diagnosis had considerable difficulty fitting anywhere, even as attempts to rationalize it encouraged the spread of the sleep lab. The complex analysis championed by the DCSAD emphasized the individuality of the disorder and thus had considerable appeal for practitioners of psychosomatic medicine, who frequently treated insomnia with biofeedback (harkening back to the work of Edmund Jacobson) or by various cognitive-behavioural therapies. But such a highly individualistic disorder was unlikely ever to link up to

392

The Sleep of Others

rational drug treatment fashioned on the notion that specific medications should be found to target specific causes of disease. While the initial reforms of the diagnosis shared a common origin with the more general reforms sweeping through American psychiatry during the 1970s, the following decades saw a greater divergence, as the latter became grounded in the pharmacological treatment of depression and psychosis derived from analysis of neurotransmitter behaviour supposedly specific to each mental illness. Polysomnography and a New Classificatory Regime While the routine use of clinical PSG in insomnia fell under attack by the late 1980s, epidemiological studies of the disease were criticized for the poverty of their methodology, which failed to meet the stringent statistical requirements of ‘evidence based medicine,’ or EBM.113 Nevertheless, PSG diagnosis received its official endorsement in the diagnosis of all sleep disorders with the 1990 publication of the International Classification of Sleep Disorders (ICSD), which replaced the DCSAD. Unlike the DCSAD, which asked a panel of experts to exercise their best clinical judgment to arrive at a consensus regarding each diagnostic entry, the ICSD sought refuge in the faceless bureaucracy of questionnaires, which polled sleep clinicians’ response to the DCSAD. Whereas the earlier nosology had classified diseases according to symptoms and explicitly rejected the ‘unitary’ approach as less a classificatory system than a mere list, the ICSD explicitly endorsed the latter approach. The great advantage here, the ICSD compilers argued, was that each diagnosis would have a single entry, thus making accurate epidemiological and clinical research possible. Testifying to the recent alliance between the sleep laboratory and the hospital, the authors of the ICSD rejected a ‘systemoriented approach’ ordinarily grounded upon theories generated by basic research, in favour of a ‘multidisciplinary approach’ that ‘would allow a synthesis of physiology, pathophysiology and symptomatology.’114 The ICSD authors reached back to the founding father of their field for justification of this ‘clinical turn.’ In the process, they quite literally rewrote history, portraying Kleitman’s Sleep and Wakefulness – a compendium of physiological research written by an experimental physiologist – as a work of clinical medicine: ‘The distinction into intrinsic and extrinsic sleep disorders divides the major causes of insomnia and excessive sleepiness into those that are induced primarily by factors within the body (intrinsic) and those produced primarily by factors out-

Insomnia Returns

393

side of the body (extrinsic). This grouping of the sleep disorders initially had been proposed by Nathaniel Kleitman in his extensive monologue [sic] on the sleep disorders that was published in 1939.’115 The misrepresentation (Kleitman had actually dedicated one of eight sections of his book to what he called ‘Spontaneous Changes in the Sleep-Wakefulness Cycle’) was doubly odd, since the APSS had funded the reprinting of Sleep and Wakefulness in 1987, just a few years before the appearance of the ICSD. In any event, this shift of sleep research towards the clinic was indicative of the direction of the American medical enterprise as a whole, which had been moving away from basic research in the medical sciences since the late 1960s. The new post-war clinical research, however, was not built up out of the experience of individual clinicians so much as it was the product of a thoroughgoing standardization: in the nomenclature of diseases; in the determination of correlated ‘risk factors’ over the discovery of mechanical causes; in the manufacture and use of bioassays; in the application of diagnostic procedures; in the education of physicians, nurses, and technicians; in the design of hospitals; and in the construction of medical instrumentation. At the end of this litany of standardized and codified data lay the new rationality of scientific medicine that was grounded in tendencies and probabilities rather than causes. In sleep medicine, such an approach effectively dispensed with symptoms. Pathophysiological signs generated by either PSG or the Multiple Sleep Latency Test (MSLT), both of which had been standardized, underwrote the organizational structure of the ICSD. Patient complaints of being unable to sleep or feeling sleepy all the time were no longer reflected in sleep medicine’s official nosological scheme, since insomnia and hypersomnia (which included narcolepsy) were collapsed into the ‘dyssomnias.’ This group of disorders, which were a pathological alteration of normal sleep patterns, contrasted with the ‘parasomnias,’ such as nightmares or sleep paralysis, in which pathological events intruded into the domain of normal sleep, the function of which had yet to be enunciated. REM sleep, for example, could be almost completely abolished with certain psychopharmaceuticals, with no untoward effects. And there was always a plentiful supply of short and long sleepers to serve as counterpoints to the traditional eight-hour rule. But none of this mattered much, because the temporal grasp of the sleep laboratory was most effective when it was allowed to squeeze data from deviations, either in the overall durations of parts of sleep architecture or from selected instants of the same.

394

The Sleep of Others

Past idiosyncrasies of the insomniac’s symptoms – the denial of psychopathology, the fear of insomnia, the disconnect between feeling sleepless at home but sleeping well in a strange environment – were relegated to the margins of the ICSD. The curious interaction between an insomniac’s perceptions and his or her sleep behaviour lapsed as a research program in sleep medicine. Addiction, which had once brought insomnia and sleep research to the front pages of national newspapers, lost its status as a central concern for psychiatric practice. Indeed, addiction to psychopharmaceuticals was hardly even acknowledged, let alone investigated.116 Insomnia had brought sleep medicine into the domain of public health. But, even when it was distanced from psychiatric disorders, as it had been by the creation of ‘psychophysiological insomnia’ in the DCSAD, the reliance on symptoms over signs in its diagnosis implied that insomnia might still collapse into psychiatry. It could never be a platform upon which sleep medicine could achieve recognition as an independent field or medical specialty. By the close of the century, the problem of insomnia had become almost irrelevant to sleep medicine. ‘Whatever happened to insomnia?’ mused a 1991 editorial in the American Journal of Psychiatry.117 The prevalence of insomnia complaints among Americans continued to be high, and hypnotics were still some of the best-selling drugs around. Yet a review of twenty years of publications on primary insomnia could find only twenty-seven studies that met current scientific standards.118 Admittedly, empirical standards had changed since the early 1970s, with more emphasis now placed on the use of control groups and randomization in medical research. Many of the earlier insomnia studies relied upon a clinician’s best judgment for diagnosing insomniac subjects rather than the application of codified rules following the appearance of the DCSAD in 1979. The use of control groups was also problematic and expensive when it came to integrating PSG into clinical research, since research protocols were time-consuming and subjects’ continuing participation depended upon ever-increasing amounts of monetary compensation. Self-experiment, a mainstay of sleep research from the mid-nineteenth century until the 1960s, was no longer a valid investigative strategy. The arrival of the sleep of others as the central object of sleep science and medicine had come close to eliminating the subjective facets of sleep’s most prevalent disorder.

9 Breathe

The transformation of sleep apnea from an obscure curiousity to the central disease in the new field of sleep medicine began in North America during the 1980s. Without the expansion of the sleep laboratory fuelled by the creation of insomnia as a public-health issue, this transformation might never have happened. While insomnia remained in ontological limbo, stuck part way between psychological disorder and somatic complaint, sleep apnea benefited enormously from its thoroughly organic nature. Its status as a public-health problem hidden even from those who suffered from it reflected the objectivity enshrined in the practice of the post-REM sleep laboratory.

The Rise of Sleep Apnea and the Consolidation of Sleep as an Active State Insomnia research was not simply killed off by the intransigence of the disease, the recriminalization of addiction, or ambivalence about the role of PSG in its diagnosis. Positive factors were also at work. The attention of practising sleep researchers was rapidly turning towards an entirely new disorder, sleep apnea; in the process, a new and different kind of sleep researcher, unconcerned with the classic problems of mind and body that had marked neuropsychiatry for the last century, now began to dominate the field. Unlike insomnia, sleep apnea was unlikely to be considered a psychiatric disorder, since its primary symptoms featured none of the classic aspects of mental illness – delusions, psychotic episodes, personality alterations, and mood imbalances. On the contrary, sleep apnea was characterized, at least initially, by the more pedestrian phenomena of extreme sleepiness and obesity, neither

396

The Sleep of Others

of which provided much in the way of material that could be reworked into a viable object of the enduring but unpredictable romance between sleep, dreaming, and madness. More important, the utility of PSG in the diagnosis of sleep apnea was unquestionable. More often than not, sufferers were unaware that they repeatedly stopped breathing for long periods while sleeping. Their quotidian fatigue, sleepiness, difficulty in concentrating, memory failure, and loud snoring were usually ascribed to character, not pathology. Unlike insomnia, which was inevitably selfdiagnosed, sleep apnea required the intervention and interpretation of the expert sleep clinician.1 While the link between insomnia and public health had been framed in terms of ‘community psychiatry,’ sleep apnea allowed sleep medicine to break away from psychiatry and move into the mainstream. Less than two years after the ICSD classification appeared, an editorial in the New England Journal of Medicine argued that the high prevalence of sleep apnea should draw attention to ‘the staggering impact of sleep disturbances on the health and welfare of our society, an impact that rivals that of smoking.’2 The fact that the editorial had been written by a respiratory physiologist indicated how far sleep research had come out from under its domination by neuropsychiatry and psychology. In a sense, sleep research had returned to its late-nineteenth-century origins in the investigation of the fatigue through the graphical methods of Marey and Mosso. Dreaming, which had been enthroned as the most important (and intractable) phenomenon of sleep since the early twentieth century, had finally been overturned (see fig. 34). The clinical and somatic reorientation of their field was already apparent to most sleep researchers by the late 1970s. Even as some of the leaders within the APSS tried to reassure their membership that clinical problems would not overtake their field, others suggested that such a direction was both inevitable and desirable. As the editors of Sleep put it in 1978: ‘We appear to have crossed a sort of threshold in terms of public and professional awareness of the existence of specific and serious sleep disorders, the best example being the primary and secondary sleep apnea syndromes.’3 Dement was even more straightforward about the prospects of sleep medicine. In his introduction to a monograph that introduced sleep apnea to an American audience that same year, he echoed earlier speculations regarding the routine use of EEG surveillance in psychophysiological research and insomnia diagnosis. For the first time, he suggested that such work could actually reveal the existence of new pathologies in normal subjects. In speaking of a group of

Breathe 397

34 Sleep research, 1961–98. A glance at the relative proportion of papers published or presented in sleep research gives an idea of how the field changed from 1960 to 2000. Studies of dreaming and sleep physiology began to give way to clinical studies during the 1970s. By the mid-1980s, sleep apnea had been recreated as the core disorder of sleep medicine, with the result that clinical studies of sleep constituted the majority of all papers published on sleep medicine.

Italian researchers, he noted how he was ‘particularly impressed by their original observations on normal subjects who snore heavily. The fact that such people can have periodic apneas and hemodynamic abnormalities has very significant implications for clinical sleep research. We can now foresee the day when the periodic medical checkup will include a careful review of the related symptoms and when there can be no clean bill of health from a thorough physical examination backed by all the technical armentarium of scanners, chemical analyses if the tests were done only in the waking state.’4 This concept of a ‘state-dependent pathology’ proved crucial to sleep apnea’s trajectory as a diagnostic category. The notion that sleep was a state rather than a mere diminution of wakefulness had been an important consequence of sleep’s relocation from the bedroom to the labora-

398

The Sleep of Others

tory. The idea that such a state was an ‘active’ one, although first proposed by psychoanalysts in the early 1940s, had been consolidated through the coordination of the REM research of the 1950s and 1960s and the study of how the ascending reticular formation discovered by Magoun and Moruzzi was actively inhibited to produce sleep.5 The sleep laboratory was thus primarily responsible for creating sleep as a state in which physiological phenomena occurred which could not be predicted or explained as a product or consequence of wakefulness. And, if sleep had its own unique physiology, it could surely have its own pathologies. Of course, both narcolepsy and insomnia long pre-dated the arrival of the sleep laboratory and had historically been incorporated into patients’ complaints and clinical observation in the absence of any concept of sleep as an active state. The sleep laboratory ultimately served only to refine these diagnoses, not reinvent them. In the case of sleep apnea, the situation was rather different. Before the creation of laboratory-based sleep research, patients’ complaints of sleepiness, excessive snoring, or of waking with the sensation of choking, if they mattered at all to the clinician, were not causally linked to events that took place during sleep. This changed when sleep emerged as an independent state, a process that began with the discovery of sleep stages and expanded following the discovery of REM. With this new conception of sleep, the door was open for state-dependent disorders, like sleep apnea, to appear. Sleep apnea was the apotheosis of the twentieth-century practice of studying the sleep of others in the laboratory that had supplanted the study of sleep through introspection. Origins Sleep apnea was forged out of the phenomena of narcolepsy and insomnia just as the latter were becoming codified during the 1970s. Attacks of irresistible sleep were common to both narcolepsy and sleep apnea. But cataplexy, the most important of the ‘pathological manifestations of REM sleep’ that were accepted in 1975 as an essential feature of narcolepsy, was absent in sleep apnea.6 Only a few years earlier, researchers at the Stanford Sleep Clinic had discovered that a small number of insomniacs also appeared to suffer from periodic breathing in their sleep, and sleep apnea soon became the most significant of the ‘organic diseases’ that could cause insomnia.7 Thereafter, sleep apnea enjoyed a rise to prominence among the sleep disorders that was nothing short of meteoric. This effect was par-

Breathe 399

ticularly pronounced in North America. The first paper to discuss breathing difficulties during sleep at the annual APSS meetings was in 1973. Of all papers on sleep disorders presented at these meetings between 1973 and 1977, around 30 per cent discussed insomnia, 20 per cent discussed narcolepsy, and about 15 per cent were on sleep apnea.8 This situation completely changed during the 1980s as sleep apnea came to dominate the agenda of sleep medicine. Between 1978 and 1982, the number of papers devoted to insomnia fell to around 15 per cent of the total, those considering narcolepsy rose to just over 30 per cent, and sleep apnea accounted for 40 per cent. From 1983 to 1987, insomnia and narcolepsy each accounted for about 14 per cent of the total number of papers on sleep disorders, while sleep apnea now made up more than half of the total. Breathing difficulties in sleep had been virtually invisible to clinical observation before the late 1970s. Comprehensive reviews of the pathological states of sleep presented by a neurologist and a psychiatrist at the First International Congress of the APSS held in Bruges in 1971 – the same conference during which Frédéric Bremer announced that sleep research had finally become ‘a discipline unto itself’ – made no mention whatsoever of sleep apnea (although the published ‘discussion’ following one paper did focus on the condition, as noted below).9 Even when clinicians looked hard for such patients, they were difficult to find. In 1974 Dement and Guilleminault reported diagnosing only 20 patients with sleep apnea from a pool of more than 250 patients who had been referred to their clinic for complaints of sleep disturbances.10 Six of these patients had initially complained of insomnia, leading the investigators to suggest optimistically that a large number of sleep-apnea patients might be found through a differentiation of insomnia. A year later, Dement reported diagnosing 36 patients with upper-airway sleep apnea from an unselected group of 200 patients complaining of excessive daytime sleepiness.11 Narcolepsy was by far the more common diagnosis, accounting for more than two-thirds of the remaining patients. In 1972 sleep-disordered breathing was considered to be a rare syndrome: one investigator was able to locate a mere sixty cases reported in an exhaustive survey of the literature.12 Ten years later, one American study reported that nearly one thousand patients had been diagnosed with the disorder in a two-and-a-half-year period.13 The Wisconsin Sleep Cohort Study, the most reputable of all epidemiological studies of the prevalence of sleep apnea among the general population, suggested in 1993 that 2 per cent of women and 4 per cent of men in the middle-aged workforce met the

400

The Sleep of Others

minimal requirements for sleep-apnea syndrome, a figure that far exceeded the projected morbidity of narcolepsy.14 Fat How could such a widespread condition have been almost entirely missed by clinicians before the early 1970s? In part, the early standards set for the sleep laboratory were to blame. Rechtshaffen and Kales’s 1968 manual established the protocols for sleep research, and they utilized only EEG, EOG, and EMG. On this count, respiration was not an integral part of sleep research. Yet such an explanation can account for only why sleep apnea went unnoticed in the laboratory, not the clinic. If sleep apnea was such a serious disorder, it is reasonable to assume that patients must have complained of symptoms and clinicians must have somehow responded to these complaints. The history of such symptoms, however, is rather complex. Medical knowledge groups symptoms in different ways at different times, and so the attempt to project retrospectively our current classificatory systems onto the past is problematic. Whereas clinicians today would be loath to offer a definitive diagnosis based only on a few symptoms gathered by a non-physician or someone with unorthodox medical training, clinicians (and a few historians) frequently do just that when it comes to interpreting the dynamics of disease in the past. But past descriptions of disease, be they from medical practitioners or lay people, often tell us much more about the thought processes of the people creating the descriptions than about what is actually being described. The history of sleep apnea, then, is the story of how a chronic feeling of irresistible sleepiness turned into a disorder of breathing in sleep. Sleepiness, both as a sensation and as a symptom, has always been with us. But what did commentators say about it during the early nineteenth century, when sleep was just beginning to change from its Galenic status as a non-natural to a localized phenomenon of the brain? Cabanis, as we observed in chapter 2, argued that sleep was related to blood flow in the brain. His argument was based largely on an analogy: dreaming was a kind of delusion, akin to episodes of madness, and so the brains of sleepers, which could not be seen, probably resembled the brains of madmen, which seemed to be suffused with blood upon dissection. But Cabanis, an idéologue who wanted to give a physiological account of mind, said little about sleepiness as a clinical symptom. Macnish’s Philosophy of Sleep (also discussed in chapter 2) is a more helpful guide, for a

Breathe 401

variety of reasons. Sleep was little more than a curiosity for Cabanis, but Macnish elevated it to the interface between life and death, imagination and reason, identity and anonymity. A writer of Gothic tales, some of which may have been published under the pseudonym ‘James Hogg,’ as well as numerous clinical anecdotes, Macnish had a dualistic attitude to sleep that typified the outlook of the Romantic authors he often cited. As one commentator has noted, sleep had a plurality of meanings at this time. It was ‘the last word in sloth, a word for stillness, silence and seclusion, debility and imbecility; but out of these sweet nothings came forth strength, and it was also a word for sensation and imagination.’15 Hygiene, natural history, psychology, and medicine intersected in Macnish’s study of sleep. All that was missing – from our current perspective – was experiment, a process that Macnish seemed to reserve for his own identity as an author. There is yet another reason why we should pay attention to what Macnish said about such mundane things as daytime sleepiness: it is he, and not Charles Dickens, who is likely the source of the symptoms that made up the so-called ‘Pickwickian syndrome,’ which some sleep researchers have depicted as the historical precursor to sleep apnea.16 The syndrome was endured not by the main character of Charles Dickens’s The Posthumous Papers of the Pickwick Club (1836–7) but by Joe, the fat, redfaced servant-boy whose capacity for sleep was rivalled only by his appetite. Dickens combined Joe’s exceptional somnolence together with a pronounced stupidity to advance the plot of his story.17 At one point, for example, Joe attempts to deliver a message. But, in his habitual waking stupor, he enters a sitting room without knocking, only to witness a gentleman with his arms around a young lady. The couple does not want the affair to be revealed to her father, so they bribe Joe with food and money to keep the secret. Joe, unfortunately, is barely capable of understanding the secret, never mind keeping it. He smirks and winks to his new confidants throughout the following meal, to the astonishment both of the father (who thinks the boy is drunk) and of Pickwick, who fears that Joe is ‘a little deranged.’ Both men demand an explanation, and the secret quickly spills out of Joe’s mouth like so much undigested food. Dickens apparently claimed to have based Joe’s character on that of a bully from his childhood, but this seems disingenuous, since Joe is dim-witted, not malicious or tyrannical. Macnish is a much more likely source, given Dickens’s and Macnish’s shared interests in mesmerism, physiognomy, and phrenology. Dickens was a keen follower of the latter:

402

The Sleep of Others

phrenological language had already permeated Dickens’s earlier work, Sketches by Boz, and Dickens’s old friend and physician, John Elliotson, was both a mesmerist and a defender of the phrenological faith (he had been elected president of the London Phrenological Society in 1824). We have already seen how phrenology influenced Macnish’s revisions of Philosophy of Sleep in 1834. Macnish’s Anatomy of Drunkenness clearly influenced Dickens while he composed Bleak House in the early 1850s, for they both contain an account of the spontaneous combustion of a drunkard.18 Dickens kept a copy of this book, as well as an edition of Philosophy of Sleep, at his library at Gad’s Hill. Even if this relationship between Macnish and Dickens is purely coincidental, the fact remains that their accounts of sleepiness are remarkably similar. More so than sleeptalkers, sleepwalkers, or fabulously long sleepers, chronically sleepy people were a distinct kind of person for Macnish. He portrayed sleepiness not as an accidental affliction but as an integral part of temperament or character. Macnish complained that such individuals were ‘dull’ and ‘heavy-headed,’ ‘with passions as inert as a Dutch fog,’ for no better reason than because that was simply the sort of people they were: ‘The cause of this constitutional disposition to doze upon every occasion, seems to be a certain want of activity in the brain, the result of which is, that the individual is singularly void of fire, energy, and passion. He is of phlegmatic temperament, generally a great eater, and very destitute of imagination. Such are the general characteristics of those who are predisposed to drowsiness: the cases where such a state coexists with intellectual energy are few in number.’19 Like Dickens’s Joe, Macnish’s chronic sleepers were incapable of selfcontrol, since ‘the soporific tendency springs from some natural defect, which no medicinal means can overcome.’ Obesity for both was merely an outward sign of a more general structural failure, which, according to physiognomy and phrenology, equally incorporated character, temperament, and intellect. Macnish’s drowsy patients (if he indeed ever had any) could no more cure themselves than could Joe, whose incapacity revealed itself by his failure to master that most elementary example of self-restraint, keeping a secret. The decline of phrenology and physiognomy notwithstanding, the association between fat and drowsiness persisted well into the twentieth century, even as their unkindly relationship to dullness of intellect began to fade. In 1889 Richard Caton, whose electrophysiological contributions were discussed in chapter 6, presented what he described as a case of narcolepsy to the Clinical Society of London.20 His patient, a

Breathe 403

poulterer from Liverpool, had been in fine health until 1881, when he rapidly began to put on weight. He soon began to fall asleep at the theatre, while riding the bus, on walks, and even when he served customers. The patient stood about 5’ 5’’ and weighed over 220 pounds when Caton first saw him. His nervous system was fine, and his sleep, which Caton had the opportunity to observe when the patient fell asleep upon examination, seemed sound. But it was marked by excessive salivation and punctuated by ‘dyspnoeic attacks’ during which the patient’s attempts to breathe were foiled by what Caton thought was an obstruction of the glottis. Caton diagnosed narcolepsy (then a novel diagnosis) and hypothesized that ‘convulsant and salivating poisons’ were the cause. Remarking that the patient’s difficulties had subsided during a brief stint doing fieldwork in Wales, Caton advised him to lose weight and then gave him naphthalin, iodoform, and charcoal ‘to produce intestinal antisepsis, and to destroy and eliminate poisons generated in the intestine and in the blood.’ The patient improved within a few days, and his ability to remain awake increased as he lost more weight. In his discussion of the paper, the Society’s president, Christopher Heath, noted the similarities between Caton’s patient and the fat boy from Dickens’s Pickwick Papers.21 Caton was not looking at a patient with sleep apnea – such a diagnosis did not exist at that time. For Caton, his patient’s difficulties in breathing in no sense caused his sleepiness; after all, the patient did not complain of this symptom, since he never woke up during these attacks. Like the salivation, Caton considered the attacks to be the result of a poison that diminished the ability of the patient to remain conscious during the day. And neither of these symptoms was as significant as the obesity, which he attempted to treat first. Nor did Caton suggest, as does most of the current work on sleep apnea, that his patient’s problems might in any way be widespread; on the contrary, he admitted that his patient was an anomaly, and he appealed to his Society fellows for help on precisely this basis. Respiration during sleep was no more significant to the most astute of Caton’s contemporaries. The famed clinician William Osler, in early editions of his Principles and Practice of Medicine, made but a single comment linking sleep and obesity as a ‘constitutional disease,’ namely, that ‘an extraordinary tendency in excessively fat young persons is an uncontrollable tendency to sleep – like the fat boy in Pickwick. I have seen one instance of it. Caton has reported a case. Sainton (Narcolepsie et obésité, Rev. Neurologique, 1901) regards it as auto-toxic in origin.’22 In

404

The Sleep of Others

earlier editions of Osler’s monumental work (for example, the fourth edition, from 1902), he did not mention seeing a case personally, nor did he allude to Dickens’s writings. In later editions, the condensed entry fell under ‘diseases of metabolism,’ in which Osler noted that ‘it is quite possible that this narcolepsy is also a manifestation of disturbed internal secretions.’23 In regard to sleep disturbances, Osler was far more interested in undermining Hammond’s popular, but ill-founded, concept of cerebral hyperaemia than to bother with the exotic curiosities of narcolepsy. In his section on ‘diseases of the nervous system,’ he emphasized that ‘older writers,’ who laid such great stress on ‘apoplectic serosa’ or the ‘so-called cerebral hyperaemia,’ were mistaken in ascribing fainting, drowsiness, giddiness, insomnia, hallucinations or delirium to a ‘congestion’ of the brain. These symptoms were due to changes in the quality of the blood which affected cerebral nutrition, and not to any hydraulic modifications.24 Other writers of the time described altered respiration during sleep as either a symptom of cerebral haemorrhage25 or as a sign of the onset of the nervous degeneration called ‘locomotor ataxia.’26 They said nothing about obesity. In some instances, writers simply ignored somnolence associated with obesity altogether, perhaps because they considered it to be a true, but trivial detail that overweight people often snored or felt sleepy all the time.27 Such commonplace truths contrasted sharply with the anomalies and aberrations that distinguished neurology from other medical specialties. Breathing During the 1950s, a new group of researchers began to take an interest correlating breathing patterns to obesity.28 These were not neurologists but cardiologists and internists whose efforts were directed to understanding how shallow breathing or ‘alveolar hypoventilation’ contributed to heart failure, which had only recently appeared as the most important chronic disease affecting Americans. Changes in the bureaucratic structure of U.S. medical funding accounted for most of this research. As we have already noted, various organizations lobbied Congress during the 1940s to restructure the National Institutes of Health around research into groups of clinical diseases, rather than around the basic biomedical sciences.29 By 1949, the National Heart Institute, the National Institute of Dental Research, and the National Institute of Mental Health had joined the National Cancer Institute under the aus-

Breathe 405

pices of the new NIH. Funding dramatically increased as well, skyrocketing from around $4 million in 1947 to over $100 million ten years later. These researchers were primarily engaged with symptoms they could correlate to some measurable sign, and they were particularly interested in the relative oxygen and carbon dioxide levels in the lungs and blood, the precise analysis of which had been a major project of wartime aviation medicine at the NIH.30 Hypoventilation and polycythemia (an increase in the number of red blood cells), they argued, could be accounted for by an increase in weight. Obesity diminished lung capacity, and, some suggested, made the respiratory centre in the brain less sensitive to increased carbon dioxide in the lungs (hypercapnia), which would ordinarily increase respiratory effort. Organs and tissues would thus receive less oxygen (hypoxemia), and the condition could potentially end with right-sided heart failure. Weight loss was the primary therapy, and one clinician emphasized the ‘constitutional’ nature of the problem by turning to literature and dubbing it the ‘Pickwickian Syndrome.’31 But it was the heart failure that could develop out of the ‘vicious cycle’ between shallow breathing and increased obesity that remained at the focal point of this research. Somnolence was a mere epiphenomenon that served to mark the patient for the clinician. The EEG, like cranial X-rays, served only to rule out neurological disease or trauma. Without the physical space and the unique practices of the sleep laboratory, somnolence was merely a guide to a literary past, not a future direction for biomedical research. The Difference a Sleep Laboratory Made The work of a small group of European neurologists began to change the concept of Pickwickian syndrome during the mid-1960s. Less than a decade later, the idea that periodic breathing during sleep was exclusively due to a pathological modification of respiratory centres in the brain was in decline. An obstructed upper airway now seemed responsible. Admittedly, it appears counterintuitive that neurologists would be responsible for a shift in causal accounts going from a brain mechanism to an airway obstruction. Indeed, most neurologists and psychiatrists interested in sleep during this period were preoccupied with psychopathology and dreaming. Respiratory changes in sleep seemed insignificant when compared to the explanatory potential that insomnia, narcolepsy, or night terrors had for neurological knowledge. But twentieth-century neurological research was largely dominated by tech-

406

The Sleep of Others

nique and technology. The vicissitudes of practice and the imperative to generate phenomena could overwrite disciplinary coherence. Neurologists studied the Pickwickian syndrome because they already had the instruments and the spaces to produce novel results, not because they intuitively recognized the disorder’s significance for public health. In fact, most neurologists felt that the confluence of sleep research and public health was to be found in psychosomatic medicine. Sleep appeared to them as an adaptive behaviour. Knowledge of its mechanisms and function would enable clinicians to develop appropriate combinations of psychological and physiological therapies to control the most notable sleep pathologies, insomnia and narcolepsy. This was, at least, the prevalent attitude at a symposium on the ‘Physiological, Pharmacological and Clinical Aspects of Sleep,’ organized by the Brain Research Institute of the University of Zurich in September 1964. The preface to the published proceedings indicated that there was ‘an increasing part played by psychological factors in the genesis of disease,’ and that this was ‘ the price we pay for the rapid advance in our civilization and standards of living.’ Sleep disturbances were either ‘a warning sign the organism is in danger’ or a ‘main symptom of psychosomatic disease.’32 Alongside the experts who presented papers on insomnia, hypnotics, and deprivation experiments at the Zurich symposium were two neurologists from the University of Freiburg, Richard Jung and Wolfgang Kuhlo, who reported on their work with the Pickwickian syndrome.33 Clearly imitating work already done with insomniacs a few years earlier, Jung and Kuhlo followed their review of major examples of abnormal sleep with a description of what they claimed to be the first all-night recordings of Pickwickians. Internists, they argued, had been fixated on obesity as the dominant symptom. All-night PSG, in contrast, generated a different sort of physiognomy, one that made a laboratory-based distinction between narcolepsy and the Pickwickian syndrome possible. In general terms, cataplexy was unique to the former, while apnoeic periods were specific to the latter. Both symptoms, they argued, were probably caused by specific brain pathologies. Pickwickianism was not, they insisted, a qualitative change from the normal. Obesity was largely irrelevant to the disease: their patients lost weight but continued to suffer periodic breathing in their sleep. Noting that other investigators had observed a decreased CO2 sensitivity in sleep, Jung and Kuhlo described Pickwickianism as ‘a caricature of the normal,’ in which they compared the sudden (and admittedly pathological) interruption of breathing

Breathe 407

during sleep to the sudden transition from waking to sleep in normal persons.34 This paper effectively took Pickwickianism out of the hands of internists and set it in the neurological domain by making sleepiness, rather than obesity, the pathognomonic sign of the disorder. Henceforth, signs gathered in the sleep laboratory, rather than gasometric data gathered from the wakeful patient, would generate the appropriate diagnosis. The direct observation and recording of apnoeic episodes took epistemological precedence over the measure of their effects. A few neurologists already equipped with similar investigative systems began to take notice. Henri Gastaut, for example, began to study the Pickwickian syndrome at his Neurobiological Research Unit at Marseilles shortly after hearing of Jung and Kuhlo’s research.35 Gastaut specialized in epilepsy, and so he was very familiar with the EEG as a research tool and even had a sleep laboratory for studying nocturnal seizure activity. Jung and Kuhlo’s new approach to Pickwickianism fit perfectly within such a system. A few weeks later, Gastaut had found a patient – an obese and somnolent male carpenter in his mid-fifties – and enlisted a neurologist and a pulmonary physiologist to perform the recordings.36 Gastaut’s patient had been predisposed to doze off at inopportune times even before he started putting on weight. While a soldier in the Second World War, he had fallen asleep after taking refuge in a tree in the midst of combat and was quickly captured by the enemy after the battle ended. Following the war, he rapidly began putting on weight, and his condition worsened. He would fall asleep while on the phone or driving a car but would awaken ten to fifteen times a night, rarely getting more than a half-hour of sleep at a time. Respiratory and blood gasometry tests conducted during wakefulness revealed no pathology. The patient would continue to doze off during the day, even though his oxygen saturation and CO2 levels were normal. During the night, however, these tests were coupled with standard PSG to generate new differences among apnoeic episodes. By measuring airflow at the mouth and comparing it with a diaphragmatic EMG and an abdominal actogram, Gastaut and his colleagues demonstrated that most apnoeas were caused by an obstruction in the upper airway and not by a disordered central regulatory mechanism that resulted in a decreased CO2 sensitivity, as Jung and Kuhlo had suggested. The hypoventilation and right-sided heart failure observed by earlier investigators were merely effects of the obesity, argued Gastaut – they played no essential role in the Pickwickian syndrome.

408

The Sleep of Others

If neither hypoxemia nor hypercapnia caused the daytime sleepiness, what was to blame? Gastaut suggested two hypotheses, both of which fixed any future study of Pickwickianism in the sleep laboratory. On the one hand, the disorder could be ‘some sort of narcolepsy,’ in which the brain-stem centre that regulated sleep and wakefulness failed. His patient did complain of irresistible sleep attacks during the day, which seemed to support an affinity, if not an identity, with narcolepsy. But there was no evidence of cataplexy, and recordings of the daytime attacks revealed that they did not take place during REM periods, which went against the definition of narcolepsy as ‘dissociated REM sleep.’ Another hypothesis, which Gastaut favoured, held that the patient was sleepy during the day because of continual interruptions of his sleep at night resulting from the collapse of his upper airway. The work of the Marseilles investigators opened up the study of Pickwickianism to neurologists and neuroscientists in two ways. On the one hand, their reliance on the sleep EEG to differentiate the new syndrome from narcolepsy indicated that Pickwickianism could be diagnosed with well-established neurological instruments and procedures. On the other hand, their hesitant conclusion that their patient was an anomalous case, and that the majority of Pickwickians suffered from a primary disturbance in the neurological regulation of sleep-wake schedules related, in some unknown way, to their obesity, held out the prospect that such could yet be discovered among sufferers of the disorder. In either case, Pickwickianism was a clinical research problem that was practically ready-made for neurologists interested in sleep. By the time of the 1971 APSS conference in Bruges, Pickwickianism had attracted enough attention to become an interesting sideshow to the main stage dominated by narcolepsy and insomnia. Although Broughton and Snyder’s presentation on the clinical phenomena of sleep had ignored the issue, for example, it quickly came to the fore in the discussion that followed.37 At issue was the supposed ‘central’ or ‘peripheral’ origin of the disorder. Many investigators now acknowledged that the obstruction of the upper airway was the immediate cause of the respiratory difficulties in sleep. In Bologna, Italy, several neurologists working with Elio Lugaresi had demonstrated that tracheostomy, combined with weight loss, would often reduce or entirely eliminate the nocturnal apnoeic episodes, thus suggesting that it was ‘peripheral’ obstructions that were responsible. Yet this procedure seemed equally effective in patients whose periodic breathing had a central origin (which had been indicated by their recorded failures to make efforts to breathe during

Breathe 409

sleep). It was also well known that sedatives could induce such episodes in those patients who had been ‘cured’ by tracheostomy, again indicating that the problem was ‘central.’ It was difficult to argue clearly one side or another in such instances, and some were frankly agnostic. The Parisian clinician and EEG researcher Betty Schwartz, for example, suggested that Pickwickianism was probably not even a coherent syndrome. Periodic breathing could be found in asymptomatic, normal people who were not overweight, and it also developed in some narcoleptics. Some so-called Pickwickians would lose their periodic breathing for no apparent reason, or the problem would subside if they simply slept on their sides. Schwartz underlined the artificiality of the syndrome classification by noting that most sleep laboratories obliged their patients to sleep on their backs during recording, even though many complained they could sleep only on their sides (fig. 35). Broughton, who chaired the discussion, concluded that ‘probably what we call Pickwickian really represents a spectrum of patients who have in common only periodic respiration and excessive sleep.’38 In terms of the progress of sleep medicine, however, the question of whether or not Pickwickianism was an independent syndrome was in some sense irrelevant. The most active participants agreed that the question was resolvable only by PSG research. Pickwickianism had thus been rendered a novel problem for a laboratory-based sleep medicine. Here was a potential success amidst considerable clinical failure. Many of the participants at the 1971 Bruges session acknowledged that their field’s original ambitions to provide neuropsychiatry with key diagnostic and investigative tools and concepts had fallen on hard times of late. The sleep EEG had not emerged as a useful way of differentiating psychiatric disorders, and it had contributed nothing to psychiatric therapy. The function of the various sleep stages, including REM, remained a mystery, and deprivation studies had failed to shed any light on why people dreamed. The solution to this impasse, proposed by Kales and welcomed by many participants, was to integrate the sleep laboratory with a patientcentred psychiatric practice more effectively, by combining sleep-laboratory data with psychological tests and patient histories to provide better evaluation of the effects of all drugs over long periods of time. This long-standing preoccupation with psychopathology was a product of REM’s history. Interest in the sleep laboratory, particularly in the United States, had evolved around psychiatrists’ aspirations to redefine their field around their newfound abilities to record and measure dreaming. In contrast, those who had first taken an interest in Pickwick-

410

The Sleep of Others

35 These images of a ‘typical Pickwickian’ under observation in a sleep laboratory were used to illustrate how such patients, usually depicted as obese, would have difficulty breathing when they slept on their back, as was conventional practice with polysomnography of the day. (Roger Broughton and Frederick Snyder, ‘Sleep and Clinical Pathological States,’ in Michael H. Chase, ed., The Sleeping Brain: Perspectives in the Brain Sciences [Los Angeles: Brain Information Service, 1972], 404 and 405)

Breathe 411

ianism held few pretensions about solving the riddle of mind/body relations through the sleep laboratory. C.A. Tassinari, one of Gastaut’s colleagues at Marseilles, went so far as to suggest that the intimate connection between the sleep laboratory and psychiatry was about to dissolve: ‘The polygraphic study of sleep for abnormal movement disorders, epilepsy, narcolepsy, Pickwickian syndrome, etc., is giving us many useful results because the methods allow documentation of the symptoms and eventually go on to physiopathogenic hypotheses … In contrast, the study of sleep in psychiatric disorders seems not to be giving us many interpretable results. I would say there are not many reasons to think that sleep studies are relevant to psychiatric disorders.’39 Lugaresi agreed, and, with a little protest from some of the psychiatrists present, the session concluded. A major symposium in Rimini, Italy, the following year served to reinforce the various positions on the Pickwickian syndrome. The Marseilles school became more insistent on obstruction-caused nocturnal insomnia and daytime sleepiness, while the Bologna neurologists argued that nocturnal apnoeas and daytime sleepiness were independent effects with central causes. Although Lugaresi, who had organized the meeting, had invited a number of respiratory physiologists, the neurologists were in the clear majority, and none of them seemed to agree with the older claim that Pickwickianism was best understood as a breathing disorder and should be studied by blood gas analysis.40 The lack of consensus regarding the etiology, pathophysiology, and methodology, or even basic definitions of ‘central’ and ‘peripheral factors,’ prevented the unconditional acceptance of the notion that the Pickwickian syndrome was a sleep disorder. One Baltimore physician, for example, curiously suggested that taking barbiturates constituted a ‘peripheral factor,’ and that ‘psychological needs’ qualified as ‘central factors’ in the Pickwickian disorder.41 Nonetheless, all commentators agreed that Pickwickianism was exceedingly rare.42 This status did not change until two developments occurred: the dissociation of obesity from Pickwickianism, and the institutionalization of the sleep EEG as the standard instrument for the diagnosis of narcolepsy, already discussed in chapter 7. Narcolepsy, Patient Populations, and Risk Both of these developments took place during the mid-1970s. Dement, who was not present at Bruges, nevertheless sensed that Pickwickianism would soon be an important facet of sleep medicine. He successfully

412

The Sleep of Others

convinced Guilleminault to return to Stanford to help establish a sleep clinic there, and they quickly set to work.43 Their subsequent contributions to the debate over Pickwickianism had their greatest impact in how they consolidated the use of the laboratory in the investigative and clinical practices surrounding sleep disorders. Subjects at the Stanford clinic presented repeated apnoeic episodes during sleep that seemed linked to the presence of a sleep disorder. Like Gastaut’s group, Dement’s team presented an enormous array of nocturnal recordings taken from a relatively small number of subjects, and then drew a sharp contrast between this clear evidence of pathology and the non-pathological blood-gasometry data acquired while the patient was awake. Respirologists’s old claims that Pickwickianism was diagnosable by changes in the blood were thus quickly supplanted by a demand for allnight sleep recordings. This gesture anticipated a massive expansion of the sleep laboratory; so, too, did the choice of subjects. In fact, the original four subjects at Stanford were not Pickwickians at all. None was obese, nor did any complain of apnoeic episodes during the day or night. Rather, they were insomniacs and narcoleptics whose primary complaint was of extraordinary sleepiness during the day. Unlike Gastaut’s patients, who presented themselves when their obesity and sleepiness had become overwhelming, the Stanford subjects had responded to ads, placed in local newspapers and on TV, that called for sleepy subjects to participate in a study. The two investigations thus offered rather different epistemological trajectories. Gastaut’s research followed a traditional clinical model: the patient presented himself or herself for diagnosis and treatment, and became the object of a conventional medical process that translated subjective sensations of illness into neurological signs. The idealized message to fellow clinicians was specific, individualized, and incisive: ‘Your patient presents these symptoms. Appropriate analysis will confirm this diagnosis. The following therapies are indicated. These physiological and pathological processes may be involved.’ In contrast to this conventional medical process of reification, the Stanford investigators from the very beginning drew upon a biomedicalresearch model strongly influenced by the detection of ‘risk factors,’ in which laboratory-generated signs were used to redefine symptoms in the name of improving public health. Advertising asked subjects, drawn from the general population, to consider their experiences as potential objects of research, not as mere (or necessary) symptoms. The knowledge gleaned from this research was subsequently translated into the

Breathe 413

language of risk, as the Stanford investigators linked the possibility of heart failure to the incidence of apnoeic episodes in sleep. The following message (paraphrased in my own words) thus obtained: ‘A sample of individuals with the following experiences or symptoms are shown to suffer the following problems in their sleep. The subjects, who are drawn from a larger population of people with sleep disorders, are unaware that they suffer from these problems. These problems put people at reater risk of death and disease but can and should be revealed by institutionalized laboratory analysis on a large scale.’ An important secret about sleep lurked, not so much within regions of the brain, but among populations of citizens. This is not to say that the brain was unimportant; indeed, Guilleminault and Dement cogently argued that the apnoeas that united their group of subjects were not obstructive and must therefore have a central cause in the ascending reticular formation. But the cause was of less immediate consequence than the ability to differentiate a new group of patients – ‘sleep apneacs’ – from already existing populations of insomniacs and narcoleptics through laboratory diagnosis. Distinctive PSG traces replaced obesity as the primary signifier of disturbed breathing in sleep, just as they distinguished narcoleptics with ‘REM sleep onset’ from those patients who simply suffered from excessive daytime sleepiness. Such a development, with its subsequent expansion of the sleep laboratory, could not have taken place within an investigative system such as Gastaut’s, which restricted its scope to individual patients rather than to populations of un- or under-diagnosed subjects. The Birth of Sleep Apnea The sleep laboratory took on an enormous new significance as the physical locus for the study of Pickwickianism during the 1970s. But changing therapies were also responsible for the expansion of sleep medicine in this period. Tracheostomy was introduced in 1969 as a more effective solution to Pickwickianism than weight loss.44 A decade later, the fact that tracheostomy seemed to cure apnoeas with ‘central’ or ‘peripheral’ origins offered convincing evidence that upper-airway obstructions were the main cause of patients’ breathing problems.45 Obesity, along with daytime hypoventilation and hypersomnia, was interpreted by sleep researchers as either an effect or an epiphenomena that followed years of chronic breathing difficulties during sleep. Working backwards within the disorder’s natural history, researchers suggested that exces-

414

The Sleep of Others

sive snoring might indicate an increased risk of developing a breathing disorder in sleep. Surgical intervention had coupled with PSG to usurp the significance of obesity in Pickwickianism. The older ‘literary’ term was replaced by an entirely new, and much broader, disorder, variously known as ‘hypersomnia with periodic apneas’ (HPA), ‘sleep apnea syndrome’ (SAS), or simply ‘sleep apnea.’ Sleep apnea described many patients formerly classified under a variety of older disorders, including narcolepsy, insomnia, Pickwickianism, Ondine’s curse, and sudden infant death syndrome (SIDS). But, while the success of tracheostomy helped relocate interest in sleep apnea from the brain to the upper airway, the procedure was still fairly radical, and its use was generally restricted to a minority of the patients diagnosed with sleep apnea. Since most drugs proved ineffective, strict dietary regimens were the most common therapeutic approach. The majority of the patients, most of whom were overweight, preferred trying to lose weight to having a hole cut out of their throats. Physicians even presented their patients with the PSG evidence itself in an effort to convince them of the urgency of the situation.46 Alcohol was also demonstrated to be a complicating factor, and patients were encouraged to give up drinking. Most important, however, was the risk posed by the hypnotics, whose depressive effects on the central nervous system exacerbated existing apneas during sleep. This problem was especially pronounced among the elderly, who were also some of the most frequent users of hypnotic drugs. The ability to identify apnoeas in the sleep of insomniacs and narcoleptics thus proved useful in preventing drug therapies from becoming a complicating factor in those illness, even in the absence of an appropriate therapy for sleep apnea. It was not until 1981 that such a therapy appeared, thus launching a new era in which sleep research became dominated by a respiratory disorder rather than by a neurological disease or a psychopathological illness. The invention of continuous positive airway pressure (C-PAP) by Colin Sullivan, an Australian respirologist, effectively transformed the nature of sleep apnea by making it a treatable disorder built around PSG diagnosis and monitoring.47 Sullivan had been introduced to breathing disorders during sleep while working in Toronto, and developed C-PAP upon his return to Australia. The device, which consisted of little more than a blower motor connected to a face mask, worked on the principle that increased air pressure in the nasal cavity could prevent the upper airway from collapsing during sleep. The technique worked, and, within a few years, commercially manufactured machines became available.

Breathe 415

The application of C-PAP quickly turned the sleep laboratory into a lucrative opportunity for enterprising physicians. Because the field was not yet an officially recognized medical specialty, any doctor with an EEG could open a sleep laboratory. A single night of recording could cost at least several hundred dollars, and it was universally acknowledged that proper diagnosis could take place only after baselines were established for each individual patient. Some laboratories insisted upon several nights of recording to establish the optimum operating C-PAP pressure, thus forcing total costs for diagnosis and therapeutic evaluation into the thousands of dollars. Professional integrity and scientific authority were at stake for sleep researchers, whose work was rapidly becoming dominated by clinical interests. Their response was typical of any social group that aspires towards professional and specialized recognition: they attempted to create a system of internal self-regulation by enforcing standards among their own. Such attempts, of course, pre-dated both C-PAP and the vogue for sleep apnea. As discussed in chapter 8, the American Sleep Disorders Centers, for example, had been instrumental in creating the DCSAD in 1979. This was an attempt to make effective communication among sleep researchers possible, to legitimate sleep medicine as a potential medical specialty, and to render sleep disorders comprehensible to non-specialist physicians who were usually the first to see patients with sleep complaints. The publication of the DCSAD and the release of the Institute of Medicine’s report the same year served these purposes admirably. These two texts represented an attempt at professional consensus around the nature of insomnia, then universally recognized as the most common sleep complaint, and forced its recognition as a serious public-health problem insofar as its treatment involved the risk of drug dependence. This consensus reflected American sleep research’s roots in post-war psychiatric practice, in that it emphasized diagnosis through extensive patient interviews and psychological testing rather than laboratory analysis. Sleep apnea, however, was a quite different beast. Its pathophysiology was purely mechanical and unclouded by any of the mentalistic premises that still seemed to determine insomnia. Its diagnosis (and its therapeutics after 1981) proceeded by purely technological means, through the sleep laboratory. Aside from the numerous physicians who were beginning to establish their own sleep laboratories around 1980, respiratory physiologists were also starting to take an interest in sleep medicine as a potentially new arena in which to extend their expertise

416

The Sleep of Others

in assessing cardio-pulmonary performance. This ‘second generation’ of sleep researchers, like the first, entered the field as novices. They learned as they went along. This situation proved unacceptable to the first generation, who had aspired to establish their work as a legitimate research field and medical specialty. To control the situation, the Association of Sleep Disorders Centers, which traced its origins back to the mid-1970s, created a certification process that focused on the technology of diagnosis rather than its nosology. By 1982, the ASDC began soliciting all sleep centres to meet their accreditation standards, and a two-week course on sleep medicine had been organized at Stanford to provide novice polysomnographers with an introduction to standards in the field. This drive to standardize the technological dimensions of sleep medicine was paralleled by the rise of sleep apnea and the expansion of the sleep clinic. Dreams, insomnia, and narcolepsy had dominated the sleep-research agenda in 1978.48 But, around 1980, sleep apnea began to enjoy a precipitous increase in popularity relative to research in four other categories: neurophysiology, dreams, insomnia, and narcolepsy. Whereas the proportion of papers published on these five topics was roughly equal in 1979, by 1986, sleep apnea had come to account for half of such publications. Simultaneous with this trend was a veritable explosion in the number of sleep laboratories across the United States. In 1980 the number of sleep centres in the United States was estimated at thirty-four.49 Ten years later, a federal report suggested that there were probably more than a thousand sleep centres, one hundred and fifty of which had been formally accredited.50 During the 1980s, debate around sleep apnea oriented itself around PSG technology, just as many of the debates surrounding insomnia in the 1970s had taken drug dependence as their focal point. Initially, the diagnosis itself was an object of criticism. Some commentators observed that, while thousands had been diagnosed with sleep apnea in the United States, there seemed to be few patients in Britain with the disorder; one well-established British clinician, Ian Oswald, for example, reported seeing but one case of sleep apnea in his twenty-year career in sleep research. These researchers suggested that a higher prevalence of obesity in the United States might be to blame, thus implying that sleep apnea was not really a sleep disorder at all.51 Others argued that British physicians simply lacked the necessary number of sleep laboratories and the proper education to recognize the disease.52 As we have already seen, some proponents of the new diagnosis, seeking to demonstrate

Breathe 417

that sleep apnea had long existed but had been misunderstood in the absence of PSG, responded with historical analysis to this effect.53 The spectre of hundreds of thousands of undiagnosed patients proved indomitable. Estimates of the prevalence of sleep apnea varied wildly and were often restricted to specified populations: elderly hospital patients, Midwestern working men and women, and so on. A 1990 federal report suggested that anywhere between 1 and 10 per cent of the American population suffered from obstructive sleep apnea, roughly one hundred times the incidence of narcolepsy.54 Of course, narcoleptics inevitably approached a doctor complaining of the sleep attacks that characterized their disorder. Likewise, insomniacs were, by definition, people who complained of being unable to sleep. Many of those who had obstructive sleep apnea, on the other hand, might be clinically asymptomatic. Their condition put them at risk for developing hypersomnia or the sleep-apnea syndrome, which might lead to further cardio-pulmonary complications and even death. Having strictly delimited narcolepsy, and having failed to resolve the problem of insomnia, the sleep laboratory now reinvented itself as an epidemiological beacon, pointing towards the maintenance of the public’s health through the surveillance of the sleep of others. How effective was the sleep laboratory in accomplishing such a task? A U.S. government report indicated that, by the end of the 1980s, clinicians agreed that the sleep laboratory played an integral role in the diagnosis and therapy of sleep apnea.55 Sleep histories, along with physical and psychological exams, could suggest the presence of apnoeic periods in the night, but such findings could be confirmed only through PSG, a procedure that was also necessary to determine the appropriate level of C-PAP pressure. Outside the domain of sleep apnea, however, this consensus regarding the clinical role of the sleep laboratory remained uncertain. Narcolepsy could be definitively diagnosed on the basis of patient histories and clinical exams, particularly if the patient reported suffering from cataplexy and sleep paralysis as well as from sleep attacks. Some of the more exotic parasomnias, such as sleepwalking or ‘REM sleep behaviour disorder,’ in which patients lost the normal muscular atonia during REM periods and subsequently thrashed about in their sleep, required PSG. The fact that injurious or violent acts could take place during such episodes made a PSG record of these events crucial for litigation. A young Toronto man, for example, was acquitted of a murder charge in 1988 on just such a basis. Kenneth Parks rose early one morning, drove fourteen miles to his in-laws’ subur-

418

The Sleep of Others

ban home and, upon his arrival, beat up his father-in-law and stabbed his mother-in-law five times, killing her.56 He then drove to a police station, where he declared, ‘I think I have killed some people.’ But he claimed to remember nothing of his crime, and the prosecution could establish no motive. Unemployed and a chronic gambler, Parks also suffered from a severe insomnia. Yet it was not a conventional psychiatric assessment but his history of sleepwalking, evidenced by PSG records and expert testimony (including that of Roger Broughton), that helped him win his case.57 But diagnosing insomnia, the single most common sleep disorder, did not seem to involve recourse to the sleep laboratory at all. Insomnia’s origins in psychopathology could be revealed by a traditional psychiatric interview, as could psychophysiological or ‘primary’ insomnia. Interest in hypnotics and addiction had declined among psychiatrists and psychopharmacologists alike during the 1980s, and drug companies turned towards the production of anti-depressants and anti-psychotics, whose effects could be explained by reference to the neurotransmitter pathologies that supposedly caused these diseases. PSG, in contrast, was incapable of describing similar mechanisms at work in insomnia, since the disorder’s primary form had long been defined as ‘psychophysiological,’ – that is, irreducible to either mind or body. Without a steady flow of new hypnotic drugs, interest in applying the sleep laboratory to drug evaluation dried up. Dependency on hypnotics, be they barbiturates or benzodiazepenes, similarly failed to capture much public or scientific interest. The few times such concerns were invoked, they focused on the new drugs which, like Prozac, were used to treat major psychiatric disorders, not insomnia. With the diagnosis of narcolepsy solved and concerns about insomnia returning to the psychiatric domain, the ecology of sleep disorders shifted in favour of sleep apnea during the 1990s. The paths of public health and sleep research had once crossed through the disordered emotions, suicidal impulses, and addictive behaviour of the insomniac. They now met at the intersection of sleep apnea and hypertension. By the end of the decade, the definitive focus of sleep research was no longer on questions of mind and behaviour, but on the evaluation of physical health and risk. The subjective experience of sleep was no longer at the core of this field of research. It had been replaced by the sleep of others. The linchpin of this transformation was the technology of the sleep laboratory. The unrestricted spread of sleep laboratories had already

Breathe 419

prompted public-health officials to commission an investigation into the necessity of using such sites to diagnose sleep disorders.58 The subsequent report, published in 1991, noted the thirtyfold increase of sleep laboratories since 1980 and proceeded to outline a clear hierarchy in the disorders for which PSG was relevant to diagnosis and treatment. Sleep apnea, which purportedly affected between 1 and 10 per cent of Americans, topped the list. Narcolepsy, which affected .01 to .1 per cent of the U.S. population, came in a distant second, and was followed by the parasomnias. Insomnia, on the other hand, was still seen to be widespread but was depicted as a disorder best diagnosed and treated through the traditional psychiatric interview followed by treatment with cognitive therapy and drugs.59 These conclusions regarding the efficacy of the sleep laboratory, however, were based on retrospective accounts that simply reported on the use of PSG in diagnosis. Such accounts did not test the effectiveness of PSG diagnosis through controlled and randomized clinical studies. Project Sleep The creation of sleep as a substantive public-health issue was not accidental. In an attempt to capture the momentum following the Institute of Medicine’s 1979 report on sleeping pills, Dement and a number of other sleep clinicians began to reorient the APSS around such issues rather than around neuroscience research. The Carter administration responded by agreeing to fund ‘Project Sleep,’ a multidisciplinary investigation of the extent and impact of sleep disorders in American society. The project foundered almost immediately. When the Reagan administration came to power in 1981, all funding for Project Sleep was pulled, and sleep researchers were left wondering what prospects their field still had. The days of active government sponsorship of neuroscience seemed to be drawing to a close, and the sense of urgency surrounding insomnia and hypnotic overdoses had dissipated as the ‘war on drugs’ recriminalized addiction and focused attention on ‘recreational’ drugs instead of pharmaceuticals. Within a few years, however, a new model of public-health politics began to take shape in the United States that ultimately helped consolidate sleep research as a pre-eminently clinical affair. The appearance of acquired immune deficiency syndrome (AIDS) in the early 1980s brought with it not only a new kind of epidemic but an entirely new relationship between patients, clinicians, and researchers.60 AIDS patients,

420

The Sleep of Others

most of whom had already been stigmatized and marginalized as either homosexuals or drug addicts, began to group together in an effort to lobby government and industry to put the new epidemic at the top of the public-health research agenda. This ‘patient activism’ was supported by several prominent AIDS physicians and proved both successful and resilient at extending public-health education, securing research funding, and accelerating the pace of clinical drug trials. For the first time in the history of twentieth-century biomedicine, medical experts were obliged to surrender some of their authority to a group of patients and their supporters whose self-identity became dominated by the image of a single disease. Working out of his clinic at Stanford, Dement was close to the international epicentre of these developments in San Francisco. Impressed by the success of AIDS activists in drawing attention and funding towards their disease, he aspired to emulate their success for sleep disorders.61 He and several others convinced the APSS to hire a full-time lobbyist in Washington, D.C., to promote the cause of sleep disorders. By the end of 1988, they had made some limited progress. In November of that year, Congress passed legislation creating the National Center for Sleep Disorders Research (NCSDR) as part of the National Heart, Lung, and Blood Institute of the NIH. The NCSDR’s mandate was to assess the current state of knowledge of sleep disorders and make recommendations regarding the funding of future research. By the mid-1990s, the results of these investigations were published in two volumes as Wake up America: A National Sleep Alert. Project Sleep had been revived. The political lobbying and public-awareness campaign that generated Project Sleep had been in part inspired by the successes of AIDS activism. But, whereas the latter phenomenon was driven primarily by patients who identified themselves as members of marginalized populations and who focused on a single disease, the former was run by sleep clinicians and emphasized a spectrum of disorders seen in sleep clinics. Furthermore, in contrast to AIDS patients, whose idenity was closely bound up with their disease, people suffering from sleep disorders – with the possible exception of insomniacs – did not personally identify with their illness. Through campaigns directed at both patients and medical practitioners, Project Sleep aspired to provide just such an identity, both to patients and to a burgeoning clinical specialty. One often-cited statistic was that sleep disorders affected some forty million Americans yet 95 per cent of these victims remained undiagnosed.62 The economic consequences seemed dire. Direct costs of sleep

Breathe 421

disorders (medical and self-care) were estimated at $15.9 billion, while indirect costs were said to reach upwards of $150 billion through factors such as lost productivity, automobile accidents, and premature death.63 The authors of Wake up America painted a more illustrative picture for those policymakers who doubted the accuracy of such statistics: ‘On any given day,’ the authors argued, ‘a substantial number of Americans, perhaps the majority, are fundamentally handicapped on the job, in the classroom, or behind the wheel.’64 The report argued that sleep disorders were more common than schizophrenia, bipolar illness, anxiety, and addictive disorders combined. An estimated eighteen million Americans suffered from sleep apnea, in comparison to the twelve million victims of clinical depression.65 Yet sleep disorders remained a ‘silent epidemic.’ Government funding of the field had stagnated since the late 1960s, and even pharmaceutical firms had backed off in sponsoring research because of baroque FDA regulations governing the process of getting new drugs to market. A mere $44.6 million was being spent by the U.S. Public Health Service on sleep research, while $1.67 billion was going to cancer research, $1.58 billion to AIDS research, $672 million to the study of heart disease, and $147 million to the investigation of Alzheimer’s disease. Sleep disorders were equally invisible on a professional level. There were no appointed chairs in the field, and few research opportunities. Medical students received, at most, a few hours of training on the subject. Under-diagnosis was a perennial issue. Narcoleptics, the diagnosis of whose disorder was one of the most thoroughly standardized in the field, averaged fifteen years and five physicians before they were properly diagnosed.66 The effects of sleep apnea, despite its rapid proliferation as a recognizable disorder, continued to be ignored, even by its own victims. The report compared the cognitive deficits that accompanied such disorders to those resulting from a better-known phenomenon that had been widely publicized during the 1980s: drinking and driving. ‘Because individuals often do not recognize they are sleepy,’ observed the report’s authors, ‘they seldom guard against inappropriate sleep episodes. Much like the intoxicated driver, sleepy drivers do not realize they are incapable of adequate performance and may deny drowsiness and impairment.’67 Wake up America thus presented a much different scenario than that presented by either AIDS activists or the Institute of Medicine’s 1979 report on sleeping pills. AIDS and insomnia both traded on the individual subject’s self-identity as diseased. In the latter case, excessive selfmedication (potentially leading to suicide or overdose) was the con-

422

The Sleep of Others

cern; in the former, fears of active and systemic discrimination loomed large. In either instance, policymakers were faced with a ready-made population or group of populations requiring (or demanding) state intervention at various levels. The new brand of sleep-disorders activism, however, called for a rather different kind of intervention: the public – including many of the people who were said to suffer from the disease – had to be taught that sleep disorders were real diseases, and not merely a matter of individual temperament or character. Sleep disorders needed to be created, on both an institutional and an individual level. Hypertension, the Sleep Laboratory, and the Language of Risk As the authors of Wake up America astutely noted, public-awareness campaigns had effectively moulded a similar epidemic some twenty years earlier.68 Hypertension had been associated with heart attacks, strokes, congestive heart failure, and vascular disease since the 1950s, but it was not until the 1972 formation of the National High Blood Pressure Education Program that high blood pressure was systematically publicized as a serious health issue. By the time Wake up America was published, hypertension had successfully permeated public and medical consciousness, and low-fat diets, weight loss, routine blood-pressure monitoring, and drug regimens were standard concerns of middle-aged Americans and their doctors. The relationship between the epidemics of hypertension and sleep disorders was primarily an analogical one for the authors of Wake up America. While some of the working reports (volume 2) of the study focused on this issue, the executive summary emphasized the cognitive impairment and affective problems associated with sleep disorders. In keeping with the historical dominance of neuropsychiatry in sleep research, it was sleepiness and its detrimental effects that were portrayed as the central issue for sleep medicine, not heart disease or stroke. This situation quickly changed during the 1990s as respirologists and cardiologists came to dominate sleep research and its clinical practice. In contrast to neuropsychiatrists, these newcomers were much more engaged in the mechanics of organs, and much less interested in the vagaries of the mind. Consequently, their concerns were directed more towards the technological production of signs than the analysis of patient-reported symptoms. An epidemic that could be hidden even from its victims was hardly a novel concept to those clinicians and researchers who had built up hypertension on this same basis.

Breathe 423

Investigators in Bologna and Stanford had produced laboratory studies linking episodes of sleep apnea to elevated systemic and pulmonary arterial pressure in the early 1970s. But it was not until a decade later that these same sleep laboratories began to generate studies depicting the symptoms of sleep apnea – in particular, snoring – as ‘risk factors.’ Many of these studies showed that, like obesity or high blood pressure, snoring could be positively correlated to the incidence of cardio-pulmonary disease.69 Respirologists, long familiar with integrating risk factors into their approach to diseases such as asthma and emphysema, looked at these new results as a promising direction for sleep research. Many neurologists, however, were far more impressed with the rapid relief that C-PAP seemed to provide their sleep-apnea patients than with any prospective risk reduction. They were thrilled by the prospect of eliminating the cognitive and emotional symptoms that had originally brought their sleep-apnea patients in for consultation. After all, the general lack of viable therapies in neurology as compared to other specialties was so well known as to have been codified in the caustic aphorism ‘Diagnose, then adios.’ C-PAP must have appeared as a powerful tool in such a context. Nonetheless, the expansion of sleep apnea into the domain of public health during the 1990s brought with it economic concerns about the overextension of C-PAP and raised anew questions about the diagnostic capabilities of the sleep laboratory. Since its inception in the post-REM period, the sleep laboratory had served a variety of functions. It was, among other things, an investigative centre, a pedagogical tool, a social and architectural space, and a rhetorical weapon in disputes. Students, instruments, instructors, researchers, and patients interacted in the sleep laboratory to produce an investigative forum that inevitably attempted to replicate itself elsewhere, just as had happened throughout the history of modern medicine. But, by the mid-1990s, the anecdotal information and tacit knowledge that had been the engine of the sleep lab’s spread was no longer enough to justify its extension to populations at large. The transformation of sleep’s eclectic mix of psychological, neurological, and respiratory disorders to a public-health approach based on risk factors needed to be written in the language of epidemiology, not clinical observation. Critics of such an expansion first appeared in Britain, where the National Health Service (NHS), like many third-party insurers in the United States, were becoming wary of the over-diagnosis of this new disease at their own expense. In 1997 investigators at the University of

424

The Sleep of Others

Leeds published a survey in the British Medical Journal reviewing the evidence that linked sleep apnea with hypertension, as well as the those studies that evaluated C-PAP as a therapy.70 The evidence, they claimed, associating sleep apnea with cardio-pulmonary disease and stroke was weak, and randomized controlled clinical trials of C-PAP were practically non-existent. In the fifty-four studies analysed, subjects were usually either referrals to sleep laboratories, retirement-home residents, or respondents to ad campaigns by sleep researchers and clinicians. Largescale controlled trials, the authors argued, needed to be conducted before public-health services and insurers could make rational decisions about paying for sleep-apnea treatments. Other therapies for treating sleep apnea, including weight loss, also needed to be properly evaluated and promoted. The review’s authors challenged the notion that sleep apnea was an independent disease entity, as opposed to ‘a marker of or a symptom of obesity and aging,’ and concluded that ‘the relevance of sleep apnoea to public health has been exaggerated.’71 The BMJ article prompted a rapid response from sleep clinicians across Britain and North America. Several agreed that the link between C-PAP and the reduction of hypertension had not been demonstrated, but that this was beside the point, since C-PAP treated excessive daytime sleepiness (EDS), not hypertension.72 Epidemiological rigour was no substitute for clinical experience, they argued. No major studies had been undertaken to evaluate C-PAP, because its efficacy was obvious to those who prescribed it. Believing that C-PAP treated hypertension rather than EDS would be ‘a natural mistake,’ argued a group of Oxford clinicians, ‘for those with no experience of the disease and its management, who thus have no idea that it is mainly hypersomnolence which disturbs patients.’73 Others observed that C-PAP was a well-established treatment that had been endorsed by a number of institutions, including the Royal College of Physicians and the Medical Research Council; the Leeds epidemiologists, it was suggested, were carelessly jeopardizing proper patient care by giving the NHS an excuse to disallow payments for such treatments.74 They also cited studies demonstrating that weight loss was frequently unsuccessful in treating sleep apnea, and noted that clinical trials were of little use because the decision to use C-PAP was ultimately one based on the clinical judgment of individual cases, not the analysis of populations at large.75 The epidemiologists refused to allow the grounds of the debate (as they saw it) to shift. In a series of responses to their critics, they acknowledged that C-PAP was indeed effective for severe sleep-apnea patients,

Breathe 425

but that this in no way justified the application of the therapy to the 2–4 per cent of the middle-aged population who were said to suffer from low to moderate sleep apnea.76 The rationale for treating the latter group was based not only on C-PAP’s relief of daytime sleepiness caused by sleep apnea but also on its supposed role in lowering the risk of hypertension. Moreover, the very definition of sleep apnea as a disorder was unclear. As a complaint, sleepiness was reported by people with or without observable sleep apnea, and so the former was an unreliable diagnostic marker for the latter. An apnea-hypopnea index of greater than fifteen (meaning that respiration was interrupted fifteen times or more each hour in sleep) was no better, because population-based studies had demonstrated that, by such a criterion, between 10 and 20 per cent of middle-aged men had sleep apnea; yet this criteria itself was poorly correlated to EDS. Sleep apnea, critics maintained, was ‘a label in search of an illness.’ It described a phenomenon observable in the sleep laboratory but one that could not be reliably correlated with any particular constellation of symptoms. The implication of such arguments was that references to the efficacy of C-PAP in lowering hypertension were merely a pretence for the ‘technology creep’ that accompanied many new therapies and biomedical technologies. Although critics of the expansive diagnosis rarely made mention of history, they could well have. We have already seen in chapter 6 how the EEG was an ‘instrument in search of a diagnosis’ during the 1940s and 1950s, and that biomedical enthusiasm surrounding the device had effectively manufactured a novel and widespread pathological condition dubbed ‘cerebral dysrhythmia.’ Perhaps C-PAP was the 1990s therapeutic equivalent of the device that had launched the sleep laboratory some fifty years earlier. A few clinicians offered some limited support for the epidemiologists’ critique of how laboratory-generated signs were beginning to supplant sleepiness as the pathognomonic indicator for treatment by C-PAP. A randomized controlled trial of C-PAP, for instance, demonstrated that the treatment provided no benefits whatsoever to subjects with a very high apnea-hypopnea index (over thirty), but with no self-reported sleepiness.77 Subjective and objective tests of the quality or duration of patients’ sleep, their neurocognitive performance, and their blood pressure remained unaffected by the therapy. The implication was that the ‘total clinical picture’ could not be captured in the sleep laboratory and so the latter, perhaps, was not the reliable epistemological tool sleep clinicians had made it out to be.78 Others, however, continued to pursue

426

The Sleep of Others

the connection between the mechanics and measures of sleep apnea, hypertension, and C-PAP, with little regard for symptoms.79 One investigation at Toronto’s St Michael’s Hospital even demonstrated positive cardiovascular effects for C-PAP in heart-failure patients with sleep apnea, symptomatic or not.80 The investigators’ recommendation was to introduce routine PSG screening for all heart patients from the moment they entered the clinic, regardless of whether or not they complained of sleep problems. Sleep medicine, by this token, had progressed so far that its most dynamic and expansive research area could now effectively ignore the subjective experience of sleep altogether. By the early twenty-first century, subjective experience no longer provided the touchstone for scientific investigations into sleep. Sleep had become a laboratory-based object to be studied through the lens of recording technologies. Personal testimony regarding sleep’s onset, its duration, and its varying levels of unconsciousness had been almost entirely supplanted by the testimony of machines. The most idiosyncratic and personal aspect of sleep – the experiences of dreaming – had failed to be accounted for by the tools and techniques of the sleep laboratory, and dreaming subsequently began to recede from its central position at the very heart of sleep research. Many members of the first generation of REM researchers ultimately began to move away from the ideal of objectifying dreams in the laboratory. Instead, they turned to charting the mysteries of sleep through diverse applications of the EEG. Wilse Webb’s observations regarding the disconnect between his life-long study of sleep and his personal experience of dreaming, mentioned in chapter 7, indicate just how much REM research had recalibrated itself towards an objectification of sleep. This cognitive shift had been a long time in coming. As the number and importance of sleep laboratories grew, the real questions that animated sleep research came to be informed more and more by the experiences that investigators had with the tools and techniques of the sleep laboratory, rather than their own intimate relationship with sleep. They had created a new habitus, both for themselves and for their subjects. Psychiatric patients were one of the most accessible subject populations for the early dissemination of EEG during the 1940s and 1950s – a development that someone like Kleitman, whose own clinical designs had ended so miserably, could never have predicted. The success of the EEG in such situations was based upon its supposed ability to objectify species of madness, and early American dream researchers, inspired by their psychoanalytic meliorism, hoped to utilize REM in a similar fashion.

Breathe 427

The identification of REM periods at sleep onset among narcoleptics was certainly the most important development that emerged from this strain of research. Yet the clinical problems that had little or nothing to do with dreaming offered far more fruitful directions for sleep research than did the total surveillance espoused by psychophysiology during the 1960s. Narcolepsy’s symptoms were dramatic, but uncommon. Insomnia, on the other hand, was as widespread as its symptoms were mundane. Its status as a minor neurosis was transformed during the 1960s as a new image of addiction emerged. Hypnotics and sedatives became caught up in this conceptual transformation from addiction to dependency, and sleep research was drawn into the circle. Insomnia began to move from being nothing more than a symptom to an emergent independent disease, as described in the first systematic classificatory manual of sleep disorders, the DCSAD. Yet this migration of insomnia into public health was stillborn. Unlike major depression or even schizophrenia, psychiatrists never reached consensus regarding insomnia’s biological origins. The disorder continued to teeter between the somatic and the psychological. With the recriminalization of addiction during the 1980s, insomnia began to drop below the horizon of biomedical research. Insomniacs were inevitably self-diagnosed, and this process of self-diagnosis seemed to play into the physiological manifestations of the disorder. The ability of EEG rhythms to stabilize sleeplessness in the laboratory was largely limited to telling insomniacs that their fears about losing sleep were often misplaced. Paradoxically, this served to eliminate insomnia as a candidate for establishing sleep medicine as an independent medical specialty. Sleep apnea emerged as a much more robust problem for sleep clinicians. Its diagnosis and analysis was firmly grounded in laboratory-based practice and increasingly separated from subjective experience. Its symptoms invoked now-classic problems of fatigue, and its excesses were ultimately distinguishable from those of narcolepsy and insomnia only by PSG. With the invention of C-PAP, the extension of the sleep laboratory was assured, because this was the only place where therapeutic benefits could be both monitored and adjusted. As respirologists began to frame sleep apnea in terms of risk, sleep laboratories spread rapidly throughout North America. Larger and more diverse populations of patients came under the auspices of the sleep lab. Apneic indexes manufactured in the sleep laboratory supplanted excessive daytime sleepiness as the dominant feature of sleep apnea’s profile as a public-health issue. Sleep had finally arrived as an object unto itself. Experienced by no one individual, the

428

The Sleep of Others

kind of sleep (or its absence) that now fuelled the progress of the field was something that could be observed only in others, who slept in a laboratory and were continuously monitored by physiological recording devices. Sleep had finally been corralled by biomedical objectivity. It was not only the unconsciousness that lurked in the shadows of the night. Disease might also wait there, its silence broken only by the watchful presence of a mechanized monitor.

Epilogue

Why are there beings at all instead of nothing? In the summer of 1935, when Martin Heidegger presented this question to his audience at the University of Freiburg, he effectively ruled out the notion that it might, at least in part, be answered by historical research.1 History, after all, traded in the particulars of existence, while the foundational problem of metaphysics was that of the transcendental nature of Being. There was, of course, a history of Being. But Heidegger inevitably presented this history in a tragic mode. It began with the invention of reason in pre-Socratic philosophy and culminated in the forgetting of Being and the modern regime of the technical, the scientific, and the political (which seems to have captured Heidegger himself in the form of his controversial commitment to National Socialism).2 Certainly, something of the tragic mode persists in the story I have here presented about the transformation of sleep’s being as a scientific object. The domestication of the sleeping body by biomedical science now seems almost complete. Many mysteries continue to surround sleep and its various disorders, but, whereas ancient marvels like sleepwalking and sleeptalking were once offered up as vehicles for debating the nature of the mind and will, they now join novelties such as REM-sleepbehaviour disorder as curiosities, not so much of sleep as of the sleep laboratory.3 The old and the new appear as mysteries on the verge of a definitive solution, in part because they have been organized and redefined as objects through a common system of production, driven by a historically specific configuration of disciplinary, professional, economic, technological, intellectual, and political forces. This transformation of sleep from a facet of experience intimately

430

The Sleep of Others

allied to nothingness to a fully fledged scientific object seems, at first pass, to be a purely technical accomplishment. It also appears to have achieved its highest (that is to say, its most recent) expression in the rapid rise of sleep-disorders medicine, which, unlike its psychiatric predecessors, seems to have no truck with the problem of dreaming. But the first suggestion is untrue, while the second is misleading. The application of teleological thinking, first to dreaming and then to sleep, was an obligatory point of passage through which sleep had to pass before it could emerge as a scientific object. The notion of sleep as a function, championed by Claparède, vilified by Vaschide, and re-engineered by Piéron, made it possible for later investigators, like Kleitman, to begin to imagine what it might mean to study sleep in a laboratory. The fact that this formulation seems to have gained particular prominence in republican cultures that prided themselves on their secularism or their pragmatism is perhaps worth further exploration. In any event, Jacobson’s work offers a potential counter-example to easy generalizations, since he stood outside this ‘functionalist’ tradition of clearly separating subject and object but nonetheless pursued a purely technical path designed for clinical success. He thereby ignored the possibility of an independent existence for sleep and dreaming in the process. He also proposed a means of actively transforming experience through a practice that literally altered the being of sleep and dreaming. Ethnographic research will likely reveal the extent to which the spread of sleep laboratories and clinics has likewise changed experience; cross-cultural studies may illuminate both the variability and the inherent limitations of a laboratorybased knowledge of sleep. If the question of sleep’s function remains unresolved, it continues, a century later, to be an active topic of speculation and curiosity among neuroscientists and philosophers alike. Some of the most powerful arguments consider the integration of the polygraphic forms of sleep with the content of dreaming. If REM periods do indeed mark a unique kind of ‘mentation’ intimately linked to dreaming, they may be periods of brain activity designed to help eliminate useless or even dangerous cognitive patterns incapable of distinguishing neural ‘noise’ from meaningful signals.4 They may also be a self-sustaining way for the brain to generate, maintain, and enhance creative thought.5 And they could equally be an automated way for the brain to safely ‘rehearse’ the idiosyncracies of lived experience while motor responses are inhibited, thereby guarding and enhancing individual differences within a species.6 Tracing the natural history of the rhythmic patterns of REM and

Epilogue 431

sleep stages may even point towards their role in providing the very foundation of thought itself: such periodicity may have provided our evolutionarily challenged vertebrate ancestors with life’s first neural mechanisms capable of generating internal representations of an environment (that is to say, thought), be they accurate or illusory.7 The possibility also exists that sleep architecture and dream content are essentially unrelated.8 Dreams quite clearly serve a multitude of cultural, psychological, and social functions. Sleep’s physiological importance is equally obvious, even though we remain ignorant of how, exactly, its phenomenology relates to its various neural mechanisms and polygraphic structure. Brain- and mind-based explanations may trade in fundamentally different kinds of objects that no amount of neuroscientific or philosophical research can effectively hybridize. Functional questions place phenomena in the full glare of rigorous investigation and high expectation. When such questions are posed, we demand enlightenment. But, despite their centrality to mental and physical life, dreaming and sleeping have been decidedly marginal as objects of knowledge. This duality, at least as I have attempted to outline it here, has been expressed in various ways through different configurations of knowledge and practice. The story it involves has hardly amounted to a forgetting of being; it rather tells of its perpetual reinvention. Perhaps this expression of difference between the personal and the public, between experience and experiment, between brain and mind, between knowing and doing, may turn out to be an integral function of the objects that mark the boundary between knowledge and imagination. After all, we can always dream otherwise.

This page intentionally left blank

Notes

Introduction 1 Lorraine Daston, ‘The Coming into Being of Scientific Objects,’ in Lorraine Daston, ed., Biographies of Scientific Objects (Chicago: University of Chicago Press, 2000), 1–14; Ian Hacking, Historical Ontology (Cambridge, MA: Harvard University Press, 2002). 2 Georges Canguilhem, The Normal and the Pathological, trans. by Carolyn R. Fawcett in collaboration with Robert S. Cohen (New York: Zone Books, 1989), 209. 3 On sleep as a public phenomenon, see Simon J. Williams, ‘Sleep and Health: Sociological Reflections on the Dormant Society,’ Health 6 (2002): 173–200. 4 See, for example, Tiego Moreira, ‘Sleep, Health, and the Dynamics of Biomedicine,’ Social Science and Medicine 83 (2006): 54–63; and Matthew WolfMeyer, ‘Sleep, Signification, and the “Abstract Body” of Biomedicine,’ paper presented at the Economic and Social Research Council conference ‘Sleep, Medicine and Health,’ at the University of Warwick, 2 December 2005. 5 Roger A. Ekirch, ‘Sleep We Have Lost: Pre-Industrial Slumber in the British Isles,’ American Historical Review 106 (2001): 343–86. 6 For example, see Francis Schiller, ‘The Semantics of Sleep,’ Bulletin of the History of Medicine 56 (1982): 377–97. 7 For examples, see William Dement, ‘The History of Narcolepsy and Other Sleep Disorders,’ Journal of the History of the Neurosciences 2 (1993): 121–34, and S.L. Peng, ‘Reductionism and Encephalitis Lethargica, 1916–1939,’ New Jersey Medicine 90 (1993): 459–62. 8 For examples, see Norman Malcolm, Dreaming (London: Routledge and Kegan Paul, 1959); David Foulkes, The Psychology of Sleep (New York: Scribner’s, 1966), and ‘Dream Research: 1953–1993,’ Sleep 19 (1996): 609–24; Ian

434

9 10 11

12 13 14 15 16

Notes to pages 7–23

Hacking, Why Does Language Matter to Philosophy? (Cambridge: Cambridge University Press, 1975); Owen Flanagan, ‘Deconstructing Dreams: The pandrels of Sleep,’ Journal of Philosophy 92 (1995): 5–27. François Dagognet, Faces, surfaces, interfaces (Paris: J. Vrin, 1982). Jacques Attali, Histoires du temps (Paris: Fayard, 1982); Wilse B. Webb, ‘Sleep as a Biological Rhythm: A Historical Review,’ Sleep 17 (1994): 188–94. Jagdish Hattiangadi, ‘The Mind as an Object of Scientific Study,’ in Christina E. Erneling and David Martel Johnson, eds., The Mind as a Scientific Object: Between Brain and Culture (Oxford: Oxford University Press, 2005), 342–66. Roy M. Whitman, ‘A Decade of Dreams: A Review,’ International Journal of Psychoanalytic Psychotherapy 3 (1974): 217–18. Roger Broughton, personal interview (5 February 1999). Peggy Mason, personal interview (11 December 1998). Joe Kamiya, telephone interview (25 September 2002). Bourdieu’s clearest statement regarding his concept of ‘habitus’ can be found in Pierre Bourdieu, Science of Science and Reflexivity, trans. Richard Nice (Chicago: University of Chicago Press, 2004), 37–44.

1. The Persistence of Privacy 1 Daryn Lehoux, ‘Tropes, Facts, and Empiricism,’ Perspectives on Science 11 (2003): 326–45. 2 Lynn Thorndike, A History of Magic and Experimental Science, 8 vols. (New York: Columbia University Press, 1923–58); G.E.R. Lloyd, Magic, Reason and Experience: Studies in the Origin and Development of Greek Science (Cambridge: Cambridge University Press, 1979). 3 On Asclepius’ origins and the spread of his cult, see Emma J. Edelstein and Ludwig Edelstein, Asclepius: A Collection and Interpretation of the Testimonies, 2 vols. (Baltimore: Johns Hopkins University Press, 1945); C. Kerényi, Asklepios: Archtypal Image of the Physician’s Existence, trans. Ralph Manheim (New York: Bollingen Foundation, 1959); and Vivian Nutton, Ancient Medicine (New York: Routledge, 2004), 103–10. 4 See Edelstein and Edelstein, Asclepius, 2: 145–58 for a description of the rituals. See also Nutton, Ancient Medicine, 109. 5 Lynn R. LiDonnici, The Epidaurian Miracle Inscriptions: Text, Translation and Commentary (Atlanta: Scholars’ Press, 1995). 6 As quoted and translated in Charles A. Behr, P. Aelius Aristedes: The Complete Works, 2 vols. (Cambridge: Cambridge University Press, 1986), 2: 312. 7 Patricia Cox Miller, Dreams in Late Antiquity: Studies in the Imagination of a Culture (Princeton, NJ: Princeton University Press, 1994), 184–203; see also

Notes to pages 24–7

8

9 10 11 12 13 14

15 16

17

18 19 20

435

Howard C. Kee, ‘Self-Definition in the Asclepius Cult,’ in Ben F. Meyer and E.P. Sanders, eds., Jewish and Christian Self-Definition, vol. 3, Self-Definition in the Graeco-Roman World (Philadelphia: Fortress Press, 1982), 118–36. Nutton (Ancient Medicine, 110) notes that the Greek traveller Pausanias records similar rituals being practised at the oracle shrine of Trophonius at Lebadea in northern Greece, although it should be mentioned that this record comes from the second century A.D., well after the Asclepian cult was established in Graeco-Roman culture. Nutton, Ancient Medicine, 108–9. Ramsay MacMullen, Paganism in the Roman Empire (New Haven, CT: Yale University Press, 1981), 66. The nature of this intimacy is discussed in Miller, Dreams in Late Antiquity, 109–17. Nutton, Ancient Medicine, 109. Ibid. ‘Hippocratic tradition’ is a dynamic term that continues to evolve: see David Cantor, ed., Reinventing Hippocrates (Burlington, VT: Ashgate, 2002). In this instance, I mean it to include Galen and Galenic appropriations of the Hippocratic corpus – precisely the tradition that was transformed by the arrival of scientific medicine, according to Wesley D. Smith, The Hippocratic Tradition (Ithaca, NY, and London: Cornell University Press, 1979). I borrow the term from Ian Hacking, ‘Dreams in Place,’ Journal of Aesthetics and Art Criticism 59 (2001): 245–60. Ludwig Edelstein, ‘Greek Medicine in Its Relation to Religion and Magic,’ in Oswei Temkin and C. Lilian Temkin, eds., Ancient Medicine: Selected Papers of Ludwig Edelstein (Baltimore: Johns Hopkins University Press, 1967), 205–46. See also Oswei Temkin, Hippocrates in a World of Pagans and Christians (Baltimore: Johns Hopkins University Press, 1991), 79–85. The most cogent argument for the physician’s relatively lowly status in antiquity can be found in Vivian Nutton, ‘The Medical Meeting Place,’ in Ph.J. van der Eijk, H.F.J. Horstmanschoff, and P.H. Schrijvers, eds., Ancient Medicine in Its Socio-Cultural Context (Amsterdam and Atlanta: Rodopi, 1995), 3–26. Ibid., 17. Ibid., 4. For examples, see George Gordon Dawson, Healing: Pagan and Christian (New York: Macmillan, 1935); Roy Porter, The Greatest Benefit to Mankind: A Medical History of Humanity (New York and London: W.W. Norton, 1997), 53; and Jacques Jouanna, ‘The Birth of Western Medical Art,’ in Mirko Grmek, ed., Western Medical Thought from Antiquity to the Middle Ages, trans. Antony Shugaar (Cambridge, MA: Harvard University Press, 1998), 22–71.

436

Notes to pages 27–31

21 Nutton, Ancient Medicine, 64–6. 22 Ch. 87 of Hippocrates, Dreams, in Hippocrates, Nature of Man. Regimen in Health. Humours. Aphorisms. Regimen 1–3. Dreams. Heracleitus: On the Universe, trans. W.H.S. Jones (Loeb Classical Library, 1931). For a detailed analysis of this text, see R.G.A. Van Lieshout, Greeks on Dreams (Utrecht: HES Publishers, 1980), 98–103, 185–90; and Steven M. Oberhelman, ‘Dreams in GraecoRoman Medicine,’ in W. Haase and H. Temporini, eds., Aufstieg und Niedergang der Römischen Welt: Geschichte und Kultur Roms im Spiegel der Meueren Forschung, vol. 37 (Berlin and New York: Walter de Gruyter, 1993), 21–156. 23 Hippocrates, Dreams, ch. 89; see also Nutton, Ancient Medicine, 65–6. 24 Steven M. Oberhelman, ‘The Diagnostic Dream in Ancient Medical Theory and Practice,’ Bulletin of the History of Medicine 61 (1987): 47–60. 25 Ch. 71 of Hippocrates, Regimen 3, in Hippocrates, Nature of Man; see also Van Lieshout, Greeks on Dreams, 101. 26 Oberhelman, ‘The Diagnostic Dream,’ 48. 27 Aristotle, On Sleep, trans. J.I Beare in Jonathan Barnes, ed., The Complete Works of Aristotle, 2 vols. (Princeton, NJ: Princeton University Press, 1984), 453 b 26–7. See also Oberhelman, ‘Dreams in Graeco-Roman Medicine,’ 134–5. 28 Aristotle, On Sleep, 454 a 4–7. 29 Ibid., 454 a 15–454 b 8; Aristotle, On Plants, trans. E.S. Forster in Barnes, ed., The Complete Works of Aristotle, 816 b 29–37. 30 Aristotle, On Sleep, 455 a 3ff. 31 Ibid., 455 a 13–27. 32 Aristotle, On the Soul, trans. J.A. Smith in Barnes, ed., The Complete Works of Aristotle, 422 b 17–424 a 15; Aristotle, On Generation and Corruption, trans. H.H. Joachim in Barnes, ed., The Complete Works of Aristotle, 329 b 6ff. See also G.E.R. Lloyd, Aristotle: The Growth and Structure of His Thought (London: Cambridge University Press, 1968), 188. 33 Aristotle, On Sleep, 456 a 30ff. 34 Ibid., 458 a 25–32. 35 Ibid., 456 b 25–33, 457 b 7–10. 36 Ibid., 456 b 38–457 a 4. 37 Ibid., 456 b 27–32. 38 Ibid., 457 b 14–17. 39 Ibid., 457 b 20–458 a 6. 40 Ibid., 458 a 30–32. 41 Ibid., 455 b 13–28. 42 Ibid., 455 b 3–13, 457 a 7–14. 43 Aristotle, History of Animals, trans. d’A.W. Thompson in Barnes, ed., The Complete Works of Aristotle, 536 b 25ff.

Notes to pages 32–7 44 45 46 47 48

49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73

437

Ibid., 537 a 12ff. Ibid., 536 b 29–34. Ibid., 537 b 15. Lloyd, Aristotle, 69–70. Aristotle did refer to how the amount of movement of the blood in ‘sanguineous animals’ could affect the clarity of a dream (too many movements obscured the dream image). But this was simply an analogical extension of human experience; Aristotle did not, for example, attempt to demonstrate this point by relating it to observations of dogs barking in their sleep as a sign both of a dream’s clarity and of the lesser number of movements in the blood (Aristotle, On Dreams, trans. J.I. Beare, in Barnes, ed., The Complete Works of Aristotle, 461 a 24–30). Aristotle, On Dreams, 458 a 3–9. Ibid., 458 a 17–20. Ibid., 458 a 20–5. Ibid., 459 a 24. Ibid., 459 a 28–34. Ibid., 461 a 1–3, Ibid., 459 b 24–460 a 23. Ibid., 460 a 24–33. Ibid., 460 b 3–16. Aristotle, On Divination in Sleep, trans. J.I. Beare in Barnes, ed., The Complete Works of Aristotle, 463 a 4ff. Mark A. Holowchak, ‘Aristotle on Dreaming: What Goes on in Sleep When the “Big Fire” Goes Out,’ Ancient Philosophy 16 (1996): 405–23. Lloyd, Magic, Reason and Experience, 49ff. Ibid., 52. Aristotle, On Divination in Sleep, 462 b 14–5; 462 b 16. Aristotle, On Dreams, 462 a 8–13. See Lehoux, ‘Tropes, Facts, and Empiricism,’ for a similar example involving a ‘fantastic’ fact rather than a class of experience. Aristotle, On Dreams, 462 a 32–462 b 11. Aristotle, On Divination in Sleep, 462 b 17–18. Ibid., 462 b 24–5. Ibid., 27–30. Ibid., 463 a 21–31. Ibid., 464 a 27–29. Ibid., 463 b 12. 462 b 20–1, 464 a 20. 463 b 15–20.

438

Notes to pages 37–40

74 Ibid., 464 a 1–28. See also Van Lieshout, Greeks on Dreams, 103. 75 Aristotle, On Divination in Sleep, 464 a 21. 76 Steven M. Oberhelman, The Oneirocriticon of Achmet: A Medieval Greek and Arabic Treatise on the Interpretation of Dreams (Lubbock: Texas Tech University Press, 1991), 30–1; Steven F. Kruger, Dreaming in the Middle Ages (Cambridge: Cambridge University Press, 1992), 18–21; Miller, Dreams in Late Antiquity, 42–3. 77 Miller, Dreams in Late Antiquity, 42–4. 78 Kruger, Dreaming in the Middle Ages, 21–34. 79 Jacques Le Goff’s paraphrase of Charles Behr’s position is more elegant: ‘the Ancients were more interested in the results of dreams than in their origins.’ See Jacques Le Goff, ‘Le christianisme et les rêves (IIe-VIIe siècles),’ in Tullio Gregory, ed., I sogni nel medioevo: Seminario Internazionale, Roma, 2–4 ottobre 1983 (Rome: Edizioni dell’Ateneo, 1985), 171–218, 177. All translations are my own unless otherwise noted. 80 See Pierre Bourdieu, The Logic of Practice, trans. by Richard Nice (Stanford, CA: Stanford University Press, 1990), 52–65. 81 Nutton estimates that Galen’s writings in Greek account for approximately 10 per cent of all surviving Greek literature before 350 A.D. (Nutton, Ancient Medicine, 216, 390). 82 Luis García-Ballester, ‘Galen’s Medical Works in the Context of His Biography,’ trans. by Philip Banks, in Luis García-Ballester, Galen and Galenism: Theory and Medical Practice from Antiquity to the European Renaissance (Aldershot, UK: Ashgate, 2002), chapter 1, 12. 83 Ibid. 84 Galen’s use of dreams is discussed in Steven M. Oberhelman, ‘Galen, On Diagnosis from Dreams,’ Journal of the History of Medicine and the Allied Sciences 38 (1983): 36–47. 85 Galen, Galen on the Usefulness of the Parts of the Body, trans. Margaret T. May, 2 vols. (Ithaca, NY: Cornell University Press, 1968), 2: 490–1. 86 As quoted in Miller, Dreams in Late Antiquity, 42. 87 The fragment can be found in Galen, Claudii Galeni opera omnia, ed. C.G. Kühn, 22 vols. (Leipzig: C. Cnobloch, 1821–33, rpt. Hildesheim, 1965), 6: 832–5, and is translated and analysed in Oberhelman, ‘Galen, On Diagnosis from Dreams.’ Whether it is a fragment from the lost treatise Galen mentions in Commentary on Epidemics I, or a later compilation of passages to Galen, is discussed in Oberhelman, ‘Dreams in Graeco-Roman Medicine,’ 139–41. 88 Nutton, Ancient Medicine, 278–84. 89 Oberhelman, ‘Galen, On Diagnosis from Dreams,’ 41–2. 90 Ibid., 41.

Notes to pages 41–5

439

91 Ibid., 44. 92 Artemidorus, The Interpretation of Dreams: Oneirocritica, trans. Robert J. White (Park Ridge, NJ: Noyes Classical Studies, 1975), 31. 93 Ibid., 31. 94 Ibid., 22–3, 184–5. 95 Ibid., 26–7. 96 Luis García-Ballester, ‘On the Origin of the “Six Non-Natural Things” in Galen,’ in idem, Galen and Galenism, chapter 4, 110. 97 On the ‘non-naturals,’ see L.J. Rather, ‘The “Six Things Non-Natural,”’ Clio medica 3 (1968): 333–47; Jerome J. Bylebyl, ‘Galen on the Non-Natural Causes of Variation in the Pulse,’ Bulletin of the History of Medicine 45 (1971): 482–5; and García-Ballester, ‘On the Origin of the “Six Non-Natural Things.”’ 98 Plato, The Republic, ed. by G.R.F. Ferrari, trans. Tom Griffith (Cambridge: Cambridge University Press, 2000), 571b-d. On dreams in Plato, see Lieshout, Greeks on Dreams, 103–36; E. Vegleris, ‘Platon et le rêve de la nuit,’ Ktema 7 (1982): 53–65; and Oberhelman, ‘Dreams in Graeco-Roman Medicine,’ 125–6. 99 Plato, The Republic, 571d–572b, 574b. 100 Nutton, Ancient Medicine, 117; see also Christopher Gill, ‘Did Galen Understand Platonic and Stoic Thinking on Emotions?’ in Juha Sihvola and Troels Engberg-Pedersen, eds., The Emotions in Hellenistic Philosophy (Dordrecht: Kluwer Academic Publishers, 1998), 113–48. 101 See Luis García-Ballester, ‘Soul and Body, Disease of the Soul and Disease of the Body in Galen’s Medical Thought,’ in idem, Galen and Galenism, chapter 3. 102 Nutton, Ancient Medicine, 292–309. 103 Vivian Nutton, ‘Murders and Miracles: Lay Attitudes to Medicine in Classical Antiquity,’ in R. Porter, ed., Patients and Practitioners: Lay Perceptions of Medicine in Pre-Industrial Society (Cambridge: Cambridge University Press, 1985), 23–54. 104 Nutton, Ancient Medicine, 304–5. 105 Martine Dulaey, Le rêve dans la vie det la pensée de Saint Augustin (Paris: Études Augustiniennes, 1973), 181–200; see also Le Goff, ‘Le christianisme et les rêves,’ 185–6. 106 Dulaey, Le rêve, 71–88; Gareth B. Matthews, Thought’s Ego in Augustine and Descartes (Ithaca, NY, and London: Cornell University Press, 1992), 74–89. 107 Augustine, The Trinity, trans. Stephen McKenna (Washington, DC: Catholic University of America Press, 1963), XV.12.21. Similar proofs held for knowledge acquired without the aid of the senses, such as mathematical

440

108

109 110 111

112 113 114 115 116

117

118

119

120

121

Notes to pages 45–8 knowledge: see Augustine, Against the Academicians, trans. Peter King (Indianapolis: Hackett Publishing, 1995), III.11.25. Dulaey, Le rêve, 152ff.; see also Miller, Dreams in Late Antiquity, 232–49. Pagan Neo-Platonist philosophers maintained the same tradition well into the fifth century. For an example, see Lucas Siorvanes, Proclus: Neo-Platonic Philosophy and Science (New Haven, CT, and London: Yale University Press, 1996), 4, 22, and 70. Dulaey, Le rêve, 158–65. Augustine, Confessions, trans. Henry Chadwick (Oxford and New York: Oxford University Press, 1992), X.30.41. William E. Mann, ‘Inner-Life Ethics,’ in G.B. Matthews, ed., The Augustinian Tradition (Berkeley and London: University of California Press, 1999), 140– 65. Le Goff, ‘Le christianisme et les rêves,’ 194–7. Ibid., 195. Kruger, Dreaming in the Middle Ages, 57–82. Augustine, Confessions, X.30.42. Rudolph Arbesmann, ‘The Concept of “Christus Medicus” in St. Augustine,’ Traditio 10 (1954): 1–28; and A.S. Pease, ‘Medical Allusions in the Works of St. Jerome,’ Harvard Studies in Classical Philology 25 (1914): 73–86. See, for example, Darrel W. Amundsen, ‘Tatian’s “Rejection” of Medicine in the Second Century,’ in Ph. J. van der Eijk, H.F.J. Horstmanshoff, and P.H. Schrijvers, eds., Ancient Medicine in Its Socio-Cultural Context (Amsterdam and Atlanta: Rodopi, 1995), 377–92. Darrel W. Amundsen, ‘Medicine and Faith in Early Christianity,’ Bulletin of the History of Medicine 56 (1982): 326–50; Gary B. Ferngren, ‘Early Christianity as a Religion of Healing,’ Bulletin of the History of Medicine 66 (1992): 1– 15. Gilbert Dagron, ‘Rêver de Dieu et parler de soi: Le rêve et son interprétation d’après les sources byzantines,’ in Tullio Gregory, ed., I sogni nel medioevo, 37–55. All these examples come from Ambroise Paré, On Monsters and Marvels, originally published 1573, trans. by Janis L. Pallister (Chicago: University of Chicago Press, 1982). On cosmology, see Fernand Hallyn, The Poetic Structure of the World: Copernicus and Kepler, trans. Donald M. Leslie (New York: Zone Books, 1990); on teratology and natural history, see Lorraine Daston and Katherine Park, Wonders and the Order of Nature, 1150–1750 (New York: Zone Books, 1998); on anatomy, see Nancy Sirasi, ‘Vesalius and the Reading of Galen’s Teleology,’ Renaissance Quarterly 50 (1997): 1–37.

Notes to pages 49–52

441

122 On the telescope and the context of its use, see Albert Van Helden, ‘The Telescope in the Seventeenth Century,’ Isis 65 (1974): 38–58; and Richard S. Westfall, ‘Science and Patronage: Galileo and the Telescope,’ Isis 76 (1985): 11–30. The full title of Kepler’s Dream was Jos. Keppleri mathematici olim imperatorii Somnium seu opus psthumum de astronomia lunari (The Dream or posthumous work on lunar astronomy of Johannes Kepler late imperial mathematician). 123 Hallyn, The Poetic Structure of the World, 258–9. 124 I borrow these terms from Martin Kemp, ‘Temples of the Body and Temples of the Cosmos: Vision and Visualization in the Vesalian and Copernican Revolutions,’ in B. Baigrie, ed., Picturing Knowledge: Historical and Philosophical Problems concerning the Use of Art in Science (Toronto: University of Toronto Press, 1996), 40–85. 125 The novelty of Kepler’s proposal that astronomy should be grounded in a mathematical analysis of physical forces is evaluated in Rhonda Martens, Kepler’s Philosophy and the New Astronomy (Princeton, NJ, and Oxford: Princeton University Press, 2000). Martens argues that Kepler’s interests in astrology and aesthetics are linked to his astronomy through his notion of ‘archetypes,’ which expressed the mind of God and were faithfully mirrored both in symbolism and in experience (48–9). Although Martens does not explicitly discuss Kepler’s Dream, this text seems to exemplify the ideal of harmony between idea, experience, and symbol. 126 John Lear, Kepler’s Dream, with the Full Text and Notes of Somnium, Sive Astronomia Lunaris, Joannis Kepleri, trans. Patricia Frueh Kirckwood (Berkeley and Los Angeles: University of California Press, 1965), 106. 127 Ibid., 107. 128 Ibid., 108–9. 129 Ibid., nn. 62, 66, 67, 74–8, 202, and 205. 130 Siorvanes, Proclus, 4. 131 On dream-genre and Descartes’s philosophy, see Hacking, Historical Ontology, 227–54. For a similar emphasis on genre over content in literary analysis, see Nancy Armstrong and Leonard Tennenhouse, ‘The Interior Difference: A Brief Genealogy of Dreams, 1650–1717,’ Eighteenth-Century Studies 23 (1990): 458–78. On Descartes’s dream scepticism, see G.B. Matthews, Thought’s Ego. Baillet’s text is reproduced and its impact on Descartes’s life and philosophy is analysed in John R. Cole, The Olympian Dreams and Youthful Rebellion of René Descartes (Urbana and Chicago: University of Illinois Press, 1992). 132 Hallyn has suggested that the Copernicanism espoused by Kepler’s demon also ‘prefigured’ the ‘evil genius’ of Descartes, but I am unaware of any

442

133 134 135 136

137 138 139

140

141 142 143

144 145 146 147 148

Notes to pages 52–8 direct evidence that Descartes actually read Kepler’s Dream (see Hallyn, Poetic Structure of the World, 160n.30). Hacking, Historical Ontology, 253. René Descartes, Philosophical Writings, selected, translated, and edited by Elizabeth Anscombe and Peter Thomas Geach (London: Nelson, 1954), 62. Descartes, Philosophical Writings, 127. Thomas Hobbes, Leviathan; or the Matter, Forme and Power of a Commonwealth Ecclesiasticall and Civil, ed. Michale Oakeshott (Oxford: Basil Blackwell, 1946) I.2, 10–11. Ibid., I.2, 18. Ibid., I.2, 13. Burton himself cites a French alchemist, Bernard Georges Penot, on this point. See Robert Burton, The Anatomy of Melancholy, 4 vols., edited by T.C. Faulkner, N.K. Kiessling, and R.L. Blair (Oxford: Clarendon Press, 1989), 1: 251. The classic guide to the religious and political debates of seventeenthcentury England is still Christopher Hill, Puritanism and Revolution: Studies in Interpretation of the English Revolution of the 17th Century (London: Secker and Warburg, 1958), especially chapter 9, ‘Thomas Hobbes and the Revolution in Political Thought.’ See also Blair Worden, ‘Providence and Politics in Cromwellian England,’ Past and Present 109 (1985): 55–99; and David Underdown, Revel, Riot and Rebellion: Popular Politics in England, 1603–1660 (Oxford and New York: Oxford University Press, 1985). Hobbes, Leviathan III.37, 285. Steven Shapin and Simon Schaffer, Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life (Princeton, NJ: Princeton University Press, 1985). Philip Goodwin, Dreames, Historically Discoursed; or a Treatise; wherein Is Clearly Discovered, the Secret yet Certain Good or Evil, the Inconsidered and yet Assured Truth or Falsity, Virtue or Vanity, Misery or Mercy, of Mens Differing Dreames (London: A.M. for Francis Tyton, 1658). See Goodwin’s entry in the Dictionary of National Biography for biographical information. See also Daniel Hipwell, ‘Philip Goodwin (died 1699), divine,’ Notes and queries, Series 8, vol. 4, alt. no. 95 (1893): 326. Goodwin, Dreames, 6. Ibid., 32–3. Ibid., 37. Ibid., 298. On Tryon’s vegetarianism and its precursors, see Serge Hutin, Les disciples anglais de Jacob Boehme aux XVIIe et XVIIIe siècles (Paris: Denoel, 1960), 71–3, 229–30; Hill, Puritanism and Revolution, 282–9; and Anita Guerrini, ‘A Diet

Notes to pages 59–63

149 150

151

152 153

154

155 156 157 158 159

443

for a Sensitive Soul: Vegetarianism in Eighteenth-Century Britain,’ Eighteenth-century Life 23 (1999): 34–42. See also the editor’s introduction to Thomas Tryon, A Discourse of the Causes, Natures and Cure of Phrensie, Madness or Distraction. From A Treatise of Dreams and Visions (Los Angeles: Augustan Reprint Society, 1973). Biographical information comes from Tryon’s own Some Memoirs of the Life of Mr. Tho. Tryon, Late of London, Merchant (London, 1705). Tryon, Treatise of Dreams and Visions, 14, 144. Ibid., 45–6. On the relationship between Paracelsian philosophy and experimentalism, see Charles Webster, ‘Paracelsus, Paracelsianism, and the Secularization of the worldview,’ Science in Context 15 (2002): 9–27; and idem, From Paracelsus to Newton: Magic and the Making of Modern Science (Cambridge: Cambridge University Press, 1982). For Tryon’s views on slavery, see Philippe Rosenberg, ‘Thomas Tryon and the Seventeenth-Century Dimensions of Antislavery,’ William and Mary Quarterly 61 (2004): 609–42. Tryon, A Treatise of Dreams, 225–39. On eighteenth-century medical mechanism, see Brian J. Gibbons, ‘Mysticism and Mechanism: The Religious Context of George Cheyne’s Representation of the Body and its Ills,’ British Journal for Eighteenth-Century Studies 21 (1998): 1–23; Anita Guerrini, ‘The Tory Newtonians: Gregory, Pitcairne, and Their Circle,’ Journal of British Studies 25 (1986): 288–311; idem, ‘James Keill, George Cheyne, and Newtonian Physiology, 1690–1740,’ Journal of the History of Biology 18 (1985): 247–66; and idem, ‘Isaac Newton, George Cheyne and the “Principia medicinae,”’ in Roger French and Andrew Wear, eds., The Medical Revolution of the Seventeenth Century (Cambridge: Cambridge University Press, 1989), 222–45. Compare the prominence of religious and even mystical sentiment in Gibbons’s (‘Mysticism and mechanism’) and Guerrini’s (‘A Diet for a Sensitive Soul’) analysis of Cheyne to the mechanistic ideology emphasized in Roy Porter, Mind-Forg’d Manacles: A History of Madness in England (Cambridge, MA: Harvard University Press, 1987). Anita Guerrini, Obesity and Depression in the Enlightenment: The Life and Times of George Cheyne (Norman: University of Oklahoma Press, 2000). George Cheyne, An Essay of Health and Long Life, 2nd. ed. (London: George Strahan, 1725), 80–1. Emphasis in original. Ibid., p. 77. Ibid., p. 82. George Cheyne, The English Malady; or, a Treatise of Nervous Diseases of All Kinds (London: George Strahan, 1733), 245.

444

Notes to pages 63–5

160 For examples, see John Burton, A Treatise on the Non-Naturals in which the Great Influence They Have on Human Bodies Is Set forth, and Mechanically Accounted for (York: A. Staples, 1738); William Forster, A Treatise on the Causes of Most Diseases Incident to Human Bodies, and the Cure of Them. First, by a Right Use of the Non-Natural; Chiefly by Diet. And Secondly, by Medicine (Leeds: James Lister, 1745); Francis de Valangin, A Treatise on Diet, or the Management of Human Life; by Physicians called the Six Non-Naturals (London: J. and W. Oliver, 1768), 273–8. 161 Vallangin, Treatise, 276–8, 294. 162 For examples, see William Oliver, A Revelation of a Very Extraordinary Sleeper; at Tinsbury, Near Bath. With a Dissertation on the Doctrine of Sensation (London, 1707); and William Hill, A Full Account of the Life and Visions of Nicholas Hart: Who Has Every Year of His Life Past, on the 5th of August, Fallen into a Deep Sleep, and Cannot Be Awakened until Five Days and Nights are expired (London, 1711). On the history of teratology, see Daston and Park, Wonders and the Order of Nature. 163 Burton, Treatise on the Non-Naturals, 287–8. 164 See Gerrit A. Lindeboom, Boerhaave and Great Britain; Three Lectures on Boerhaave with Particular Reference to his Relations with Great Britain (Leiden: Brill, 1974); for an analysis of the peculiar social conditions under which Boerhaave’s work spread to Edinburgh, see Andrew Cunningham, ‘Medicine to Calm the Mind: Boerhaave’s Medical System, and Why It Was Adopted in Edinburgh,’ in Andrew Cunningham and Roger French, eds., The Medical Enlightenment of the Eighteenth Century (Cambridge: Cambridge University Press, 1990), 40–66. 165 See Burton’s account of Chin-Cough (Whooping Cough) in the appendix to his Treatise on the Non-Naturals. 166 John Locke, An Essay concerning Human Understanding, ed. and intro. by Peter H. Nidditch (Oxford: Clarendon Press, 1979), especially II:1 §11, 13, 14, and 16. 167 See Armstrong and Tennenhouse, ‘The Interior Difference.’ 168 See Richard C. Allen, David Hartley on Human Nature (Albany: State University of New York Press, 1999), for biographical details and an analysis of Hartley’s doctrine of vibrations. On the mid-eighteenth century shift from Boerhaave’s hydraulic model of the physiological economy to the doctrine of vibrations, see Porter, Mind-Forg’d Manacles, 176–84. 169 See Allen, David Hartley, 83–129; for a translation of such notions in modern neuroscientific terms, see C.U.M. Smith, ‘David Hartley’s Newtonian Neuropsychology,’ Journal of the History of the Behavioral Sciences 23 (1987): 123–36.

Notes to pages 65–72

445

170 David Hartley, Observations on Man, His Frame, His Duty, and His Expectations (Gainsville, FL: Scholars’ Facsimilies and Reprints, 1966 [1749]), 45–55. Hartley’s understanding of Harvey’s description of the heart’s motion in De motu cordis (1628) seems to have been somewhat deficient, since Harvey argued that it was the action of the heart that drove the blood through the veins. 171 Ibid., 383. 172 Ibid., 385. 173 Ibid., 389. 174 Allen, David Hartley, 104–14; Smith, ‘David Hartley’s Newtonian Neuropsychology,’ 131–3. 175 S.A.D. Tissot, Advice to People in General, with respect to Their Health, trans. from the fifth edition of Avis au peuple (Dublin: James Potts, 1769), 17 and 20. 176 Ibid., 28ff. 177 Ibid., 30. 178 For a cultural history of this phenomenon, see Ekirch, ‘Sleep We Have Lost.’ 179 Locke, An Essay concerning Human Understanding, II:1 §13. 180 S.A.D. Tissot, L’onanisme: dissertation sur les maladies produites par la masturbation, trans. A. Hume, 3rd ed. (London: W. Wilkinson, 1767), 122. 181 Ibid., 99ff. 182 Ibid., 120–4. 183 Ibid., 17. 184 Paine’s Examination of the Prophecies was posthumously published as part three of The Age of Reason. See Thomas Paine, The Age of Reason, Part the Third: Being an Examination of the Passages in the New Testament quoted from the Old, and called Prophecies concerning Jesus Christ: To which Is Prefixed an Essay on Dream ... (London: R. Carlile, 1818). 185 Ibid., 4. 2. Analogize and Experiment 1 Yannick Ripa, Histoire du rêve: Regards sur l’imaginaire des français au XIXe siècle (Paris: Oliver Orban, 1988), 133–54. 2 John V. Pickstone, Ways of Knowing: A New History of Science, Technology and Medicine (Chicago: University of Chicago Press, 2001). 3 Martin S. Staum, Cabanis: Enlightenment and Medical Philosophy in the French Revolution (Princeton, NJ: Princeton University Press, 1980). 4 Pierre Jean Georges Cabanis, On the Relations between the Physical and Moral

446

5 6 7 8 9 10 11 12 13 14

15

16

17

18 19 20 21 22

23 24 25

Notes to pages 73–83

Aspects of Man, trans. George Mora (Baltimore: Johns Hopkins University Press, 1981), 602. Ibid., 615. Ibid., 615–16. Ibid., 615. Oswei Temkin, ‘The Role of Surgery in the Rise of Modern Medical Thought,’ Bulletin of the History of Medicine 25 (1951): 248–59. Cabanis, On the Relations between the Physical and Moral Aspects of Man, 609. Ripa, Histoire du rêve, 133–54. Ekirch, ‘The Sleep We Have Lost.’ William Shakespeare, Hamlet, introduction by Burton Raffel (New Haven, CT: Yale University Press, 2003), 3.2. My thanks to Daryn Lehoux for clarification on this matter. The conclusions I draw are, however, my own. Bonnie Ellen Blustein, ‘The Brief Career of “Cerebral Hyperaemia”: William A. Hammond and His Insomniac Patients, 1854–90,’ Journal of the History of Medicine and Allied Sciences 41 (1986): 24–51; and idem, Preserve Your Love for Science: Life of William A. Hammond, American Neurologist (Cambridge: Cambridge University Press, 1991). For a comprehensive and cross-cultural overview, see Marijke Gijswijt-Hofstra and Roy Porter, eds., Cultures of Neurasthenia: From Beard to the First World War (Amsterdam and New York: Rodopi, 2001). William A. Hammond, ‘On Sleep and Insomnia. Part 1. Physiology of Sleep,’ New York Medical Journal 1 (1865): 88–101; idem, ‘Physiology of sleep,’ New York Lancet: A Family Medical Journal 1 (1866): 43–5. For a critical examination of the reliability of dream introspection, see Eric Schwitzgebel, ‘Why Did We Think we Dreamed in Black and White?’ Studies in History and Philosophy of Science 33 (2002): 649–60. As quoted in Blustein, Preserve Your Love for Science, 165. Lorraine Daston, ‘British Responses to Psycho-physiology, 1860–1900,’ Isis 69 (1978): 192–208. Marie-Jean Lyon, Marquis Hervey de Saint-Dénis, Dreams and How to Guide Them, trans. N. Fry (London: Duckworth, 1982). Blustein, ‘The Brief Career of “Cerebral Hyperaemia,”’ 31. William Alexander Hammond, Sleep, Sleeplessness and the Derangements of Sleep; or, the Hygiene of the Night, a revised edition of Sleep and Its Derangements (1869) (London: Simpkin, Marshall, 1892), 17. Ibid., 87. Emphasis in original. Ibid., 90. Ibid., 245.

Notes to pages 83–6

447

26 Edward Shorter, From Paralysis to Fatigue: A History of Psychosomatic Illness in the Modern Era (Toronto: Maxwell Macmillan, 1992), 239. 27 Ibid., 34. 28 Hammond’s testimony was even sought after as evidence in a murder trial. See A. Patten and B.M. Patten, ‘William A. Hammond, the Dynamograph, and Bogus Neurologic Testimony in old New York,’ Journal of the History of the Neurosciences 6 (1997): 257–63. 29 Blustein, Preserve Your Love for Science, 118–33. 30 As quoted in ibid., 52. 31 Blustein, ‘The Brief Career of “Cerebral Hyperaemia,”’ 48. 32 For examples of the impact of this turn on the practice and profession of neurology, see Kenton Kroker, ‘The Progress of Introspection in America, 1896–1938,’ Studies in History and Philosophy of Biological and Biomedical Sciences 34 (2003): 77–108; and idem, ‘Epidemic Encephalitis and American Neurology, 1919–1940,’ Bulletin of the History of Medicine 78 (2004): 108–47. 33 Blustein, ‘The Brief Career of “Cerebral Hyperaemia,”’ 51. 34 Charles Richet and Paul Broca, Comptes rendus de la Société de Biologie, 12 February 1897. 35 Angelo Mosso, Die Diagnostik des Pulses in Bezug auf die localen Veränderungen desselben (Leipzig: Veit, 1879). 36 W. Bruce Fye, The Development of American Physiology: Scientific Medicine in the Nineteenth Century (Baltimore: Johns Hopkins University Press, 1987). 37 William Henry Howell, ‘A Contribution to the Physiology of Sleep, Based upon Plethysmographic Experiments,’ Journal of Experimental Medicine 2 (1897): 313–45. 38 E.C. Walden, ‘A Plethysmographic Study of the Vascular Conditions during Hypnotic Sleep,’ American Journal of Physiology 4 (1901): 124–61. 39 On Howell’s textbook and its significance, see Fye, Development of American Physiology, 203–4. 40 William Henry Howell, A Text-Book of Physiology for Medical Students and Physicians, 8th ed. (Philadelphia: W.B. Saunders, 1921). 41 Nathaniel Kleitman, Sleep and Wakefulness as Alternating Phases in the Cycle of Existence (Chicago: University of Chicago Press, 1939), 74. 42 John M. O’Donnell, The Origins of Behaviorism: American Psychology, 1870–1920 (New York: New York University Press, 1985). 43 Roger Smith, Inhibition: History and Meaning in the Sciences of Mind and Brain (Berkeley and Los Angeles: University of California Press, 1992), 66–79. 44 J.B.E. Gélineau, ‘De la narcolepsie,’ Gazette des Hôpitaux 54 (1880): 626–8 and 635–7; Pierre Passouant, ‘La narcolepsie au temps de Gélineau,’ Histoire des sciences médicales 15 (1981): 129–35; Francis Schiller, ‘The Semantics of

448

45

46 47 48 49 50

51 52 53 54

55

Notes to pages 87–90

Sleep,’ Bulletin of the History of Medicine 56 (1982): 377–97. On the status of French naval physicians more generally, see Michael A. Osborne, ‘Identity and Distinctiveness in 19th-Century French Naval Medicine,’ Revue scientifique et technique de la défense 54 (2001): 11–18, C. Westphal, ‘Eigenthümliche mit Einschlafen verbundene Anfälle,’ Archiv für Pszchiatrie und Nervenkrankheiten 7 (1877): 631–5; and Franz Fisher, Jr, ‘Epileptoide Schlafzustände,’ Archiv für Pszchiatrie und Nervenkrankheiten 8 (1878): 200–3. See also Charles Féré, ‘Note sur la narcolepsie épileptique,’ Revue de médecine 18 (1898): 430–40. Passouant, ‘La narcolepsie,’ 129. Schiller, ‘Semantics of Sleep,’ 386–7. G. Ballet, ‘Contribution à l’étude du sommeil pathologique: Quelques cas de narcolepsie,’ Revue de médecine 2 (1882): 945–57. Richard Caton, ‘A Case of Narcolepsy,’ Clinical Society of London. Transactions 22 (1889): 133–7. Peretz Lavie makes this suggestion in his, ‘Nothing New under the Moon: Historical Accounts of Sleep Apnea Syndrome,’ Archives of Internal Medicine 144 (1984): 2025–8. Silas Weir-Mitchell, ‘Some Disorders of Sleep,’ American Journal of the Medical Sciences 100 (1890): 109–27. Henri Piéron, Le problème physiologique du sommeil (Paris: Masson, 1913), 196. Anson Rabinbach, The Human Motor: Energy, Fatigue, and the Origins of Modernity (New York: Basic Books, 1990). On biomedical holism, see Christopher Lawrence and George Weisz, eds., Greater than the Parts: Holism in Biomedicine, 1920–1950 (New York: Oxford University Press, 1998). On the significance of the kymograph to physiological discipline, see Merriley Borell, ‘Instruments and an Independent Physiology: The Harvard Physiological Laboratory, 1871–1906,’ in Gerald L. Geison, ed., Physiology in the American Context, 1850–1940 (Bethesda, MD: American Physiological Society, 1987); and Soraya de Chadarevian, ‘Graphical Method and Discipline: SelfRecording Instruments in Nineteenth-Century Physiology,’ Studies in History and Philosophy of Science 24 (1993): 267–91. For an analysis of how such instruments recreated concepts of observation in the life sciences, see Henning Schmidgen, ‘Pictures, Preparations, and Living Processes: The Production of Immediate Visual Perception (Anschauung) in late-19th-Century Physiology,’ Journal of the History of Biology 37 (2004): 477–513. On clinical applications, see Robert G. Frank, Jr, ‘The Telltale Heart: Physiological Instruments, Graphic Methods, and Clinical Hopes, 1854–1914,’ in William Coleman and Frederic L. Holmes, eds., The Investigative Enterprise: Experimental Physiology in

Notes to pages 92–6

56 57 58 59 60

61

62 63 64 65 66

67 68 69

70 71 72

449

Nineteenth-Century Medicine (Berkeley and Los Angeles: University of California Press, 1988). On the origins of the instrument, see H.E. Hoff and L.A. Geddes, ‘The Technological Background of Physiological Discovery: Ballistics and the Graphic Method,’ Journal of the History of Medicine and Allied Sciences 15 (1960): 345–63; and Robert M. Brain and M. Norton Wise, ‘Muscles and Engines: Indicator Diagrams and Helmholtz’s Graphical Methods,’ in Lorenz Krüger, ed., Universalgenie Helmholtz: Rückblick nach 100 Jahren (Berlin: Academie Verlag, 1994). de Chadarevian, ‘Graphical Method,’ 269–70. Nicholas Jardine, The Scenes of Inquiry: On the Reality of Questions in the Sciences (New York: Oxford University Press, 1991). Brain and Wise, ‘Muscles and Engines.’ de Chadarevian, ‘Graphical Method.’ On the local contexts of using EEG for establishing brain death, see E.F. Wijdicks, ‘Brain Death Worldwide: Accepted Fact but no Global Consensus in Diagnostic Criteria,’ Neurology 58 (2002): 20–5; for a cross-cultural analysis, see Margaret Lock, ‘On Dying Twice: Culture, Technology and the Determination of Death,’ in Margaret Lock, Alan Young, and Alberto Cambrosio, eds., Living and Working with the New Medical Technologies (Cambridge: Cambridge University Press, 2000), 233–62. Marta Braun, Picturing Time: The Work of Étienne-Jules Marey (1830–1904) (Chicago: University of Chicago Press, 1992); François Dagognet, Étienne-Jules Marey: A Passion for the Trace, trans. Robert Galeta with Jeanine Herman (New York: Zone Books, 1992). Edouard Toulouse, ‘Nécrologie: Marey,’ Revue scientifique 22 (1904): 673–5. Ibid., 674. Étienne-Jules Marey, ‘Préface,’ Physiologie expérimentale 1 (1875): ii–ix. William Coleman, ‘The Cognitive Basis of the Discipline: Claude Bernard on Physiology,’ Isis 76 (1985): 49–70. Étienne-Jules Marey, ‘Natural History of Organised Bodies,’ Annual Report of the Board of Regents of the Smithsonian Institution for 1867 (1868): 277–304, 287; Braun, ‘Picturing Time,’ 37–41. Marey, ‘Préface,’ ii. Ibid., ii–iii. Sandra Horvath-Peterson, Victor Duruy and French Education: Liberal Reform in the Second Empire (Baton Rouge: Louisiana State University Press, 1984), 180–1. As quoted in Horvath-Peterson, Victor Duruy, 195. Braun, Picturing Time, 38–9. Ibid., 93–9.

450

Notes to pages 97–106

73 Étienne-Jules Marey, ‘Du moyen d’économiser le travail moteur de l’homme et des animaux,’ Physiologie expérimentale 1 (1875): 1–18. 74 Ibid., 1. 75 Ibid., 2. Emphasis in original. 76 Ibid., 3. 77 Étienne-Jules Marey, Animal Mechanism: A Treatise on Terrestrial and Aërial Locomotion (New York: D. Appleton, 1884), 1. 78 Marey, ‘Du moyen d’économiser le travail,’ 7. 79 This claim was not Marey’s own but that of a Berlin engineer – Fehrmann – whose Pferdeschoner (horse-manager) had been tested in 1874 at Halle. Marey thought Fehrmann’s paper to be so important that he translated it and republished it along with his own (Marey, ‘Du moyen d’économiser le travail’). Fehrmann’s Pferdeschoner and Marey’s labour-saving device were practically identical, but Marey was careful to point out that his own work, with its accounts of how speed and effective work could be measured and graphically displayed, was much more precise than that of Fehrmann. 80 Marey, ‘Du moyen d’économiser le travail,’ 13. 81 Marey, Animal Mechanism, 78–101. 82 Ibid., 93. 83 Marey, Animal Locomotion, 94–5. Emphasis in original. 84 Ibid., 100. 85 Étienne-Jules Marey, ‘L’économie de travail et l’élasticité,’ La revue des idées 1 (1904): 161–76. The topic was taken up by Marey’s long-standing collaborator, Auguste Chauveau (1827–1917), who assumed the directorship of the Institut Marey. 86 Angelo Mosso, Fatigue, trans. Margaret Drummond and W.B. Drummond (London: Swan Sonnenschein, 1906), 81. For an equally charming image of Mosso shaking hands with Marey and Kronecker, see Phillipp Felsch, ‘Marey’s Flip Book’ at www.vlp.mpiwg-berlin.mpg.de/essays/data/art31, accessed 25 May 2006. 87 Mosso, Fatigue, 83. 88 Ibid., 7–13. 89 Ibid., 17. 90 On the rise and decline of the practice of identifying subjects of psychological experiment, see Kurt Danziger, Constructing the Subject: Historical Origins of Psychological Research (Cambridge: Cambridge University Press, 1990). 91 Mosso, Fatigue, 92. 92 Ibid., 98–9. 93 Ibid., 241–2. 94 Mosso, Die Diagnostik des Pulses.

Notes to pages 107–13

451

95 Ibid., 12–14. 96 Roger Smith, History of the Human Sciences (New York: W.W. Norton, 1997), 492–529. 97 William James, Principles of Psychology (New York: Henry Holt, 1890), 97–9. 98 William James, ‘What Is an Emotion?’ Mind 9 (1884): 188–205. Emphasis in original. A similar theory was proposed by a Danish physiologist, Carl Georg Lange, around the same time and soon became known as the ‘JamesLange theory of emotion.’ 99 Ibid., 191–2. 100 Mosso, Fatigue, 106. For a similar analysis, see the work of Mosso’s student Josefa Ioteyko, ‘Les défenses psychiques: 1. La Douleur – 2. La Fatigue,’ Revue philosophique 75 (1913): 113–34, 262–73. 101 Mosso, Fatigue, 176. 102 Alain Borreau, ‘Satan et le dormeur: une construction de l’inconscient au Moyen Age,’ Chimère 14 (1991): 41–61; Ian Hacking, Rewriting the Soul: Multiple Personality and the Sciences of Memory (Princeton, NJ: Princeton University Press, 1995), 147. 103 See, for example, Léon Chertok and Isabelle Stengers, A Critique of Psychoanalytic Reason: Hypnosis as a Scientific Problem from Lavoisier to Lacan (Stanford, CA: Stanford University Press, 1992). 104 Amand Marc Jacques de Chastenet Marquis de Puységur, Un somnambule désordonné? Journal du traitement magnétique du jeune Hébert, ed. by Jean-Pierre Peter (Paris: Institut Synthélabo, 1999); Henri F. Ellenberger, The Discovery of the Unconscious: The History and Evolution of Dynamic Psychiatry (New York: Basic Books, 1970), 71–83. 105 James Braid, Neurhypnology, or the Rationale of Nervous Sleep Considered in relation with Animal Magnetism (London: J. Churchill, 1843). 106 Léon Chertok, ‘Centième anniversaire de l’ouvrage “Du sommeil et des états analogues”: De Liébeault à Freud,’ La presse médicale 74 (1966): 2945–6. 107 Auguste Liébeault, Du sommeil et des états analogues, considérés surtout au point de vue de l’action du moral sur le physique (Masson: Paris, 1866). Ellenberger dismisses this story as apocryphal, noting that Liébeault’s theory of sleep was known even by Russian authors (Discovery of the Unconscious, 107). Chertok, on the other hand, cites a critical review in Annales médico-psychologiques (1867) as evidence that Liébeault’s ideas were everywhere rejected (‘Centième anniversaire,’ 2946). 108 Alan Gauld, A History of Hypnotism (Cambridge: Cambridge University Press, 1992), 322–4. 109 Blustein, Preserve Your Love for Science, 159–61.

452

Notes to pages 113–23

110 Chertok and Stengers, A Critique of Psychoanalytic Reason, 28. 111 See, for example, American neurologists’ approach to the intractable problem of epidemics of encephalitis lethargica during the 1920s in Kroker, ‘Epidemic Encephalitis.’ 112 Gauld, A History of Hypnotism, 544. 113 Bernheim, as cited in ibid., 546. 114 Ibid. 115 For examples, see Laura Otis, Organic Memory: History and the Body in the Late Nineteenth & Early Twentieth Centuries (Lincoln: University of Nebraska Press, 1994), and Hacking, Rewriting the Soul. 116 On organic memory and eugenics, see Kroker, ‘Immunity and Its Other: The Anaphylactic Selves of Charles Richet,’ Studies in History and Philosophy of Biological and Biomedical Sciences 30 (1999): 273–96. 117 Ellenberger, Discovery of the Unconscious, 172. Gauld, A History of Hypnotism, has Delboeuf visiting Liébeault’s clinic in 1888 (321). I thank André Le Blanc for encouraging me to pay attention to Delboeuf’s work. 118 François Duyckaerts, ‘Sigmund Freud: Lecteur de Joseph Delboeuf,’ Frénésie: Revue de la Société internationale d’histoire de la psychiatrie et de la psychanalyse 2 (1989): 71–88. 119 Joseph Delboeuf, Le sommeil et les rêves, considérés principlalement dans leur rapports avec les théories de la certitude et de la mémoire (Paris: Alcan, 1885), 166. 120 Ibid., 166 121 Ellenberger, The Discovery of the Unconscious, 86. 122 Ripa, Histoire du rêve, 138–47. 123 Albert Moll, Hypnotism (London: Walter Scott, 1891), 201. 124 Ibid., 210. Emphasis added. 3. The Ends of Darkness 1 Timothy Lenoir, The Strategy of Life: Teleology and Mechanics in Nineteenth Century German Biology (Boston: Kluwer, 1982); Stuart M. Persell, Neo-Lamarckism and the Evolution Controversy in France, 1870–1920 (Lewiston, NY: The Edwin Mellen Press, 1999). 2 Sigmund Freud, Introductory Lectures on Psycho-Analysis, vol. 15, The Standard Edition of the Complete Psychological Works of Sigmund Freud, trans. under the editorship of James Strachey (London: Hogarth Press, 1953), 83. 3 Mikkel Borch-Jacobsen, Remembering Anna O.: A Century of Mystification (New York: Routledge, 1996). 4 Sigmund Freud and Josef Breuer, Studies on Hysteria, vol. 2, Standard Edition, 45.

Notes to pages 124–30

453

5 Sigmund Freud, The Interpretation of Dreams, trans. Joyce Crick (New York: Oxford University Press, 1999), 85. 6 Sigmund Freud, The Complete Letters of Sigmund Freud to Wilhelm Fleiss, 1887– 1904, trans. and ed. by Jeffrey Moussaieff Masson (Cambridge, MA: Belknap Press, 1985), 417–18. 7 Freud, Interpretation of Dreams, 63 (see also 62–73). 8 Sigmund Freud, Introductory Letters on Psycho-Analysis, Parts I and II, vol. 15, Standard Edition, 84–5. 9 This is a persistent theme of Frank Sulloway, Freud, Biologist of the Mind: Beyond the Psychoanalytic Legend (Basic Books: New York, 1979). 10 Freud, Interpretation of Dreams, 374. Emphasis original. 11 Ibid., 382. 12 Ibid., 380–1. 13 Freud, Interpretation of Dreams, 380. Freud never makes clear this allusion to ‘other instances’ in which physiological processes are rendered pathological. Given their contemporary prominence, in the medical and popular press, adverse reactions to diptheria serotherapy might have provided Freud with an obvious example of a functional system gone awry. See Kenton Kroker, ‘Immunity and Its Other: The Anaphylactic Selves of Charles Richet,’ Studies in History and Philosophy of Biological and Biomedical Sciences 30 (1999): 273–96. 14 On the ‘Project’ and Freud’s biology, see Frank Sulloway, Freud, Biologist of the Mind; Lucille B. Ritvo, Darwin’s Influence on Freud: A Tale of Two Sciences (New Haven, CT: Yale University Press, 1990). 15 On energy as concept and metaphor in the physical sciences, see Crosbie Smith, The Science of Energy: A Cultural History of Energy Physics in Victorian Britain (Chicago: University of Chicago Press, 1998), and M. Norton Wise, ‘Mediating Machines,’ Science in Context 2 (1988): 77–113; for the psychological sciences, see Rabinbach, The Human Motor. 16 Alfred Maury, Le sommeil et les rêves, études psychologiques sur ces phénomènes et les divers états qui s’y rattachent, suivies de recherches sur le developpment de l’instinct et de l’intelligence dans leurs rapports avec le phénomènes du sommeil, 3rd (Paris: Didier, 1865), 481. 17 For examples, see Freud, Interpretation of Dreams, 364 and 408. Strachey’s translation created the term ‘cathexis’ for Freud’s Besetzung, while newer translations use derivations of ‘investment’ or ‘charge.’ 18 Ibid., 350–1. 19 Ibid., 408–9. 20 On Delage, see Persell, Neo-Lamarckism and the Evolution Controversy in France, 59–100. 21 Yves Delage, ‘Essai sur la théorie du rêve,’ Revue scientifique 48 (1891): 40–8.

454

Notes to pages 131–8

22 See, for example, Stephan Jay Gould, Ontogeny and Phylogeny (Cambridge, MA: The Belknap Press of Harvard University Press, 1977). 23 Harvie Ferguson, The Lure of Dreams: Sigmund Freud and the Construction of Modernity (London: Routledge, 1996), 134–7. 24 Freud, as quoted in ibid., 137. 25 S.P. Fullinwider, ‘Sigmund Freud, John Hughlings Jackson, and Speech,’ Journal of the History of Ideas 44 (1983): 151–8. 26 Ibid., 157–8. 27 Ellenberger, Discovery of the Unconscious, 486; Sulloway, Freud, Biologist of the Mind, 123–30. 28 For examples, see Kroker, ‘Immunity and Its Other’; and Otis, Organic Memory. 29 Sulloway, Freud, Biologist of the Mind, 330. 30 Carl Jung, ‘L’Analyse des rêves,’ L’Année psychologique 15 (1909): 160–7. 31 On Bergson’s cultural significance, see Mark Antliff, Inventing Bergson: Cultural Politics and the Parisian Avant-Garde (Princeton, NJ: Princeton University Press, 1993); on the scientific and religious debates, see R.C. Grogin, The Bergsonian Controversy in France, 1900–1914 (Calgary: University of Calgary Press, 1988). 32 Henri Bergson, ‘Le rêve,’ Revue scientifique 23 (1901): 705–13. 33 The extent of Bergson’s revisions can be seen his Mélanges, ed. by André Robinet in collaboration with Marie-Rose Mossé-Bastide, Martine Robinet, and Michel Gautier (Paris: Presses Universitaires de France, 1972), 443–63. The 1919 collection in which the revised essay appeared was translated as Mind-Energy: Lectures and Essays, trans. H. Wildon Carr (New York: Holt, 1920). 34 For a philosophical analysis of the links between psychoanalysis and Bergsonism, see Marie Cariou, Lectures bergsoniennes (Paris: Presses Universitaires de France, 1990). 35 Grogin, The Bergsonian Controversy, 22. 36 Henri Bergson, Time and Free Will: An Essay on the Immediate Data of Consciousness, trans. F.L. Pogson (New York: The Macmillan, 1916), 64. 37 Ibid., 181ff. 38 I here repeat Bergson’s own example, but I encourage the reader to think of refrigerators rather than windows in this case. See ibid., 160ff. 39 Ibid., 8–9, 126–37. 40 Henri Bergson, Matter and Memory, trans. N.M. Paul and W.S. Palmer (New York: Zone Books, 1988), 133–77. 41 For an assessment of popular Victorian science, see Bernard Lightman, ed., Victorian Science in Context (Chicago: University of Chicago Press, 1997).

Notes to pages 138–42

455

42 See Geraldine Jonçich, The Sane Positivist: A Biography of Edward L. Thorndike (Middletown, CT: Wesleyan University Press, 1968), 334. For contemporary press coverage, see ‘Columbia Plans Bookstore,’ New York Times (2 February 1913), X8; and ‘Bergson Stands by Will,’ New York Times (5 February 1913), 8. 43 Bergson, Matter and Memory, 155. Emphasis in original. 44 Ibid., 173. 45 For a concise review of Cajal’s theories, see J. Allan Hobson, The Dreaming Brain: How the Brain Creates Both the Sense and the Nonsense of Dreams (New York: Basic Books, 1988), 90–9. 46 Bergson, Matter and Memory, 174–5. 47 Cariou, Lectures bergsoniennes. 48 Patricia Kitcher, Freud’s Dream: A Complete Interdisciplinary Science of Mind (Cambridge, MA: MIT Press, 1992), 186. 49 Pete A.Y. Gunter, ‘Bergson and Jung,’ Journal of the History of Ideas 43 (1982): 635–2. 50 Grogin, The Bergsonian Controvery, 51. 51 John J. Cerullo, The Secularization of the Soul: Psychical Research in Modern Britain (Philadelphia: Institute for the Study of Human Issues, 1982). 52 Grogin, The Bergsonian Controversy, 51. 53 Ibid., 51; Bergson, Mélanges, 509–510. 54 Ibid., 673. Eusapia was famous for her ability to move objects at a distance (telekinesis), but she was also a self-confessed cheat, taking every opportunity to hoodwink her interlocutors – hence the ‘controllers’ who held her on each side. See Harry Price, Fifty Years of Psychical Research (New York: Arno Press, 1975), 74–6. 55 The essay first appeared in the Bulletin de l’Institut générale psychologique in May 1901, was summarized in Revue de philosophie, and was reproduced in Richet’s popular Revue Scientifique all in the same year. See Bergson, ‘Le rêve.’ 56 Bergson, Mind-Energy, 113. 57 Ibid., 122. 58 For an example, see George Trumbull Ladd, ‘Contribution to the Psychology of Visual Dreams,’ Mind 3 (1892): 299–304. 59 Bergson, Mind-Energy, 123–4. 60 For an idiosyncratic description of Bergson, see Edwin Slosson, Major Prophets of To-day (Freeport, NY: Books for Libraries Press, 1968 [1914]). 61 On the use of interrogation in experimental psychology, see Martin Kusch, ‘Recluse, Interlocutor, Interrogator: Natural and Social Order in Turn-ofthe-Century Psychological Research Schools,’ Isis 86 (1995): 419–39. 62 Bergson, Mind-Energy, 125.

456

Notes to pages 142–9

63 Ibid., 126. 64 Bergson, ‘Le rêve,’ 713. His 1919 revision and its 1920 translation eliminated this claim that dreaming of the day’s concerns caused fatigue. 65 The Independent, (23 and 24 October, 1913). Slosson, who dubbed Bergson ‘a modern prophet,’ published the lecture as a monograph the following year as Henri Bergson, Dreams, trans. Edwin E. Slosson (New York: Huebsch, 1914). 66 ‘Pope Denounces Bergson,’ New York Times, 28 August 1913. 67 See, for example, ‘The Philosophy of Henri Bergson and Syndicalism,’ New York Times, 26 January 1911, 4; and editorial, New York Times, 29 August 1913, 8. 68 Slosson, introduction to Bergson, Dreams, 6. 69 Ibid., 7. 70 Ibid. 71 Charles Sherrington to Alexander Forbes, 14th September 1912, Alexandeer Forbes Papers (hereafter AFP), Harvard Medical Library, 15.2.726. 72 Edward Burnett Tylor, Primitive Culture, 2 vols. (New York: Harper and Row, 1958), 1: 429. 73 Edwin G. Boring, A History of Experimental Psychology, 2nd ed. (New York: Appleton-Century-Crofts, 1957), 242–4. For a similar comment by Bergson, see ‘Bergson Here to Lecture,’ New York Times, 3 February 1913, 5. 74 For examples, see John M. O’Donnell, The Origins of Behaviorism: American Psychology, 1870–1920 (New York: New York University Press, 1985); Roger Smith, History of the Human Sciences (New York: W.W. Norton, 1987); and Danziger, Constructing the Subject. 75 See Robert A. Nye, Crime, Madness, and Politics in Modern France: The Medical Concept of National Decline (Princeton, NJ: Princeton University Press, 1984). 76 Théodule Ribot, Les maladies de la volonté (Paris: Alcan, 1884). On Ribot’s place in the history of French psychology, see Jacqueline Carroy and Régine Plas, ‘The Origins of French Experimental Psychology: Experiment and Experimentalism,’ History of the Human Sciences 9 (1996): 73–84; and D.P. Faber, ‘Théodule Ribot and the Reception of Evolutionary Ideas in France,’ History of Psychiatry 8 (1997): 445–58. 77 Edouard Claparède, ‘Esquisse d’une théorie biologique du sommeil,’ Archives de psychologie 4 (1905): 245–9. 78 Walter B. Pillsbury, ‘Edouard Claparède, 1873–1940,’ Psychological Review 48 (1941): 271–8. 79 Edouard Claparède, ‘Edouard Claparède,’ in Carl Murchison, ed., A History of Psychology in Autobiography, 7 vols., 2nd ed. (New York: Russel and Russell, 1961), 1:68ff.

Notes to pages 149–54

457

80 William James to Theodore Flournoy, 31 May 1891, in William James, The Correspondence of William James, 7 vols., edited by John J. McDermott, Ignas K. Skrupskelis, Elizabeth M. Berkeley, and Frederick H. Burkhardt (Charlottesville and London: University Press of Virginia, 1999), 7: 164–5. 81 Letter, James to Flournoy, 12 July 1892, ibid., 7: 304. 82 Pillsbury, ‘Edouard Claparède,’ 272. 83 An influential account of Titchener’s work in the context of Wundtian introspection can be found in Kurt Danziger, ‘The History of Introspection Reconsidered,’ Journal of the History of the Behavioral Sciences 16 (1980): 241– 62; for a description focused on introspection as a practice, see Kroker, ‘Progress of Introspection in America.’ 84 John I. Brooks, The Eclectic Legacy: Academic Philosophy and the Human Sciences in Nineteenth-Century France (Newark: University of Delaware Press, 1998). See also Carroy and Plas, ‘Origins of French Experimental Psychology.’ 85 Smith, History of the Human Sciences, 493. 86 For Richet’s programmatic statement, see his ‘Les procédés de défense de l’organisme,’ Revue scientifique 52 (1893): 801–7. 87 Alfred Binet, review of Claparède, L’année psychologique 12 (1906): 663–4. 88 For a helpful chronology, see Maurice Agulhon, The French Republic, 1879– 1992, trans. Antonia Nevill (Cambridge and Oxford: Basil Blackwell, 1993), 474–80. 89 Henri Piéron, ‘Henri Piéron,’ in Murchison, ed., A History of Psychology in Autobiography. 90 On Binet, see Theta H. Wolf, Alfred Binet (Chicago: University of Chicago Press, 1973); and John Carson, ‘Minding Matter/Mattering Mind: Knowledge and the Subject in Nineteenth-Century Psychology,’ Studies in the History and Philosophy of Biological and Biomedical Science 30 (1999): 345–76. 91 Alfred Binet and Victor Henri, ‘La Psychologie individuelle,’ L’année psychologique 2 (1896): 411. 92 Alfred Binet and Nicholas Vaschide, ‘La psychologie à l’ école primaire,’ L’année psychologique 4 (1898): 1–14. 93 Alfred Binet, ‘Mosso (A.) – La fatigue intellectuelle et physique,’ L’année psychologique 1 (1895): 450–2; Alfred Binet and Nicholas Vaschide, ‘Examen critique de l’ergographe de Mosso,’ L’année psychologique 4 (1898): 253–66. 94 See Binet and Vaschide, ‘La psychologie à l’école primaire,’ idem, ‘Expériences de force musculaire et de fond chez les jeunes garçons,’ L’année psychologique 4 (1898): 15–63; and idem, ‘Un nouvel ergographe, dit ergographe à ressort,’ l’année psychologique 4 (1898): 303–15. 95 Binet and Vaschide, ‘La psychologie à l’école primaire,’ 3. 96 Ibid., 5–6. Emphasis in original.

458

Notes to pages 155–60

97 Wolf, Alfred Binet, 160–75. 98 Alfred Binet, ‘L’évolution de l’enseignement philosophique,’ L’année psychologique 14 (1908): 152–231; see also Grogin, Bergsonian Controversy. 99 Binet, ‘L’évolution de l’enseignement philosophique,’ 227. My clumsy translation loses Binet’s play on the meanings of ‘État’ as ‘being’ and ‘state.’ 100 Wolf, Alfred Binet, 18, 97–8. 101 Carroy and Plas, ‘Origins of French Experimental Psychology.’ 102 Geneviève Paicheler, L’invention de la psychologie moderne (Paris: L’Harmattan, 1992), 15. 103 William Schneider, ‘The Scientific Study of Labour in Interwar France,’ French Historical Studies 17 (1991): 410–46. 104 Georges Ribeill, ‘Les débuts de l’ergonomie en France à la veille de la Première Guerre mondiale,’ Le mouvement sociale 113 (1980): 3–36. 105 Schneider, ‘Scientific Study of Labour,’ 416. 106 Ibid., 426. 107 There is somewhat of a conflict between Wolf’s description of Piéron (Alfred Binet, 20–1) and that of Carroy and Plas (‘Origins of French Experimental Psychology’). While the latter emphasize Piéron’s role in making reactiontime experiments the paradigm of experimental psychology in France, the former suggests that Piéron found these same tests ‘extremely frustrating’ and were perhaps his reason for abandoning Binet’s laboratory. Wolf personally interviewed Piéron in 1960. 108 Nicholas Vaschide and Henri Piéron, La psychologie du réve au point de vue médical (Paris: Librairie J.-B. Baillière et Fils, 1902), 94. 109 Ibid., 95. 110 Eugène Osty, Supernormal Faculties in Man, trans. Stanley de Brath (London: Methuen, 1923), 60. Osty was a long-time supporter of Richet’s métapsychique. 111 Nicholas Vaschide, Le sommeil et les rêves (Paris: Ernest Flammarion, 1911), 175–96. 112 Ibid., 285–94. 113 Ibid., 294–7. 114 Ibid., 22–3. 115 Ibid., 20–1. 116 Ibid., 24. For an alternative (and, I believe, erroneous) take on Vaschide, see J. Allan Hobson, The Dreaming Brain, 69–73. Hobson’s historical introduction to his own neuropsychological theory of dreaming does not mention Claparède but instead credits Vaschide with several of Claparède’s conceits regarding sleep’s positive function as an instinct regulated by the

Notes to pages 161–5

117 118 119

120

121

122 123 124 125

126

127

128 129 130 131

459

brain. He concludes that ‘these ideas of Vaschide’s reflect the first impact of Darwin’s thinking upon theories of sleep,’ but this seems highly unlikely, given his derision of functional explanations in general. Edouard Claparède, ‘La question du sommeil,’ L’année psychologique 18 (1912): 424. Ibid., 426. Emphasis in original. Nathaniel Kleitman, ‘Studies in the Physiology of Sleep, I: The Effects of Prolonged Sleeplessness on Man,’ American Journal of Physiology 66 (1923): 68. Schneider reports that Ribot, Janet, and Lapique suggested Piéron for the position at Toulouse’s laboratory (‘The Scientific Study of Labour,’ 419– 23), but Piéron himself does not mention this in his autobiographical paper (‘Henri Piéron,’ in Murchison, ed., A History of Psychology in Autobiography, vol. 1). Further biographical detail can be found in A. Fessard, ‘Henri Piéron,’ L’année psychologique 50 (1951): vii–xvi, and M. Reuchlin, ‘The Historical Background for National Trends in Psychology: France,’ Journal of the History of the Behavioral Sciences 1 (1965): 115–23. Piéron, ‘Henri Piéron,’ 263; see also Paul Mengal, ‘Henri Piéron (1881– 1964) et les néo-Lamarckiens français,’ in C. Debru, Jean Gayon, and J.-F. Picard, eds., Les sciences biologiques et médicales en France, 1920–1950 (Paris: CNRS, 1994). Persell, Neo-Lamarckism, 101–2. Anne Diara, ‘Le transformisme de Félix le Dantec,’ Revue de synthèse 100 (1979): 407–22. Marc Viré, ‘La création de la chaire d’“évolution des êtres organisés” à la Sorbonne en 1888,’ Revue de synthèse 100 (1979): 377–91. Piéron, ‘Henri Piéron,’ 263. See also Daniela S. Barberis, ‘Moral Education for the Elite of Democracy: The Classe de Philosophie between Sociology and Philosophy,’ Journal of History of the Behavioral Sciences 38 (2002): 355–69. Edouard Toulouse, Nicholas Vaschide, and Henri Piéron, ‘Classification of Psychical Phenomena for Experimental Research,’ Mind 11 (1902): 535–46; idem, Technique de psychologie expérimentale, 2nd ed., revised and expanded by Toulouse and H. Piéron (Paris: Doin, 1911). Edouard Claparède to Henri Piéron, 10 April 1904; postcard, Claparède to Piéron, 24 July 1904; Claparède to Piéron, 25 July 1905, 520 AP 5, Fonds Piéron, National Archives of France (henceforth AP). Claparède to Piéron, 25 July 1905, AP. Edouard Claparède to Henri Piéron, 7 February 1908, AP. Henri Piéron, L’évolution de la mémoire (Paris: Ernest Flammarion, 1910). Hacking, Rewriting the Soul.

460

Notes to pages 165–74

132 Otis, Organic Memory. 133 See William H. Schneider, Quality and Quantity: The Quest for Biological Regeneration in Twentieth-Century France (Cambridge: Cambridge University Press, 1990); and Kroker ‘Immunity and Its Other.’ 134 Piéron, L’évolution de la mémoire, 10. 135 Henri Piéron, ‘Des phénomènes d’adaptation biologique par anticipation rythmique,’ Comptes rendus de l’Académie des sciences 144 (1907): 338–41, 340. 136 Ibid., 340. 137 Piéron, L’évolution de la mémoire, 2–3. 138 Ibid., 3–4. 139 Ibid., 23. Emphasis in original. 140 John B. Watson, ‘Psychology as the Behaviorist Views It,’ Psychological Review 20 (1913): 158–77. 141 Piéron, L’évolution de la mémoire, 346. 142 Charles Richet, La sélection humaine (Paris: Philippe Renouard, 1917), 14. 143 Piéron, L’évolution de la mémoire, 350. 144 Ibid., 352. 145 Piéron, ‘Des phenomènes d’adaptation.’ 146 The most comprehensive historical-philosophical analysis of this practice remains Georges Canguilhem’s The Normal and the Pathological, trans. Carolyn R. Fawcett in collaboration with Robert S. Cohen (New York: Zone Books, 1989). 147 Claparède to Piéron, 25 July 1905, AP. 148 Carroy and Plas, ‘Origins of French Experimental Psychology.’ 149 Henri Piéron, ‘L’étude expérimentale du sommeil normal. La méthode,’ Comptes rendus hebdomadaires des séances de la Société de Biologie de Paris 62 (1907): 307. 150 Marie de Manacéïne, ‘Quelques observations expérimentales sur l’influence de l’insomnie absolue,’ Archives italiennes de biologie 21 (1894): 322–5. 151 Piéron, Le problème physiologique du sommeil, 3–6. 152 Piéron, Le problème physiologique du sommeil, 3. 153 Edouard Toulouse and Henri Piéron, ‘Le méchanisme de l’inversion, chez l’homme, du rhythme nycthéméral de la température,’ Journal de physiologie et de pathologie générale 3 (1907): 425–40. 154 Schneider, ‘The Scientific Study of Labour.’ 155 This is by no means to say that Piéron hated or had complete disregard for animals. Indeed, he once stated that ‘I have always loved animals and only with great reluctance have I performed vivisections’ (Piéron, ‘Piéron,’ 264). But his experiments allowed him to adopt a utilitarian view of their existence in a way not possible with human subjects. His work did attract the

Notes to pages 174–82

156 157 158 159 160 161 162 163 164

461

notice of anti-vivisection movements in Europe and the United States, and a French journalist named Gustave Téry apparently started a press campaign against him. See the notes on experiments sent by Legendre to Piéron from 1906 to 1911 in 520 AP 24. Mosso, Fatigue, 119. On Weichardt, see Reichenbach, The Human Motor, 142–5. Piéron, Le problème physiologique du sommeil, 302. Réné Legendre, ‘The Physiology of Sleep,’ Smithsonian Institution Annual Report, 1911 (New York, 1912), 587–602. Yves Delage, review of Piéron, Le problème physiologique du sommeil, L’année biologique 17 (1912): 579–88, 582. Piéron, Le problème physiologique du sommeil, 228–35. Ibid., 367. Ibid., 385–6. Ibid., 442–4.

4. Inhibition and Disease 1 On Pavlov’s research before the theory of conditional reflexes, see Daniel P. Todes, ‘Pavlov’s Physiology Factory,’ Isis 88 (1997): 165–82; and, for a more comprehensive analysis, see idem, Pavlov’s Physiology Factory: Experiment, Interpretation, Laboratory Enterprise (Baltimore: Johns Hopkins University Press, 2002). 2 Todes, Pavlov’s Physiology Factory, 217–54. 3 Roger Smith, Inhibition: History and Meaning in the Sciences of Mind and Brain (Berkeley and Los Angeles: University of California Press, 1992). 4 Ibid., 59–65. 5 H.S. Curtis, ‘Inhibition,’ Pedagogical Seminary 6 (1898): 65–113, 65. 6 Ibid., 65. 7 Ibid., 76. 8 Smith, Inhibition, 179–90. 9 Jeffrey A. Gray, Pavlov (Brighton, UK: Harvester Press, 1979), 90–113. 10 Shorter, From Paralysis to Fatigue, 41–2. 11 Charles-Édouard Brown-Séquard, ‘Le sommeil normal, comme le sommeil hypnotique, est le résultat d’une inhibition de l’acivité intellectuelle,’ Archives de physiologie normale et pathologique 1 (1889): 333–5; Jacques Gasser, Aux origines du cerveau moderne: Localisations, langage et mémoire dans l’oeuvre de Charcot (Paris: Fayard, 1995), 82ff. 12 For examples, see Elizabeth Sherwood and Martin Sherwood, Ivan Pavlov

462

13 14 15 16 17 18 19

20

21 22 23 24 25 26

27 28 29 30 31

Notes to pages 182–8

(Geneva: Heron Books, 1970); and Gray, Pavlov. On Pavlov’s archive and the ‘scattered’ collection of his research notebooks, see Daniel P. Todes, ‘From Lone Investigator to Laboratory Chief: Ivan Pavlov’s Research Notebooks as a Reflection of His Managerial and Interpretive Style,’ in Frederic L. Holmes, Jürgen Renn, and Hans-Jörg Rheinberger, eds., Reworking the Bench: Research Notebooks in the History of Science (Dordrecht, Boston, and London: Kluwer Academic Publishers, 2003), 203–20. George Windholz, ‘Hypnosis and Inhibition as Viewed by Heidenhain and Pavlov,’ Integrative physiological and behavioral science 31 (1996): 338–49. Ivan Pavlov, Lectures on Conditioned Reflexes, 2 vols., trans. W. Horsley Gantt (New York: International Publishers, 1928 and 1941), 2: 158. Todes, ‘Ivan Pavlov’s Research Notebooks,’ 210–13. Pavlov, Lectures on Conditioned Reflexes, 1: 250. Todes, Pavlov’s Physiology Factory. Sherwood and Sherwood, Ivan Pavlov, 206–10; Todes, ‘Ivan Pavlov’s Research Notebooks,’ 210. W. Horsley Gantt to Walter B. Cannon, 31 May 1923, Walter B. Cannon Archives, Countway Library of Medicine, Harvard Medical Schools (hereafter WBCA), 37.462. Ivan Petrovich Pavlov, ‘The Identity of Inhibition with Sleep and Hypnosis,’ Scientific Monthly 17 (1923): 603–8; idem, ‘New Researches on Conditioned Reflexes,’ Science 58 (1923): 359–61. Ivan Petrovich Pavlov, ‘The Problem of Sleep,’ in Experimental Psychology and Other Essays (New York: Philosophical Library, 1957). Smith, Inhibition, 201. George Dumas, ed., Nouveau traité de psychologie, 7 vols. (Paris: Alcan, 1930– 49), 2: 19–39, 4: 455–522. Pavlov, ‘Identity of Inhibition,’ ‘New Researchers,’ and Lectures on Conditioned Reflexes, 250–4. Dumas, ed., Nouveau traité de psychologie, 2: 31. Ivan Pavlov, Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex, trans. and ed. by G.V. Anrep (New York: Dover Publications, 1960), 152–233. Pavlov, Lectures on Conditioned Reflexes, 1: 156. Dumas, ed., Nouveau traité de psychologie, 2: 39. Ibid., 4: 509. Edward Shorter, A History of Psychiatry: From the Era of the Asylum to the Age of Prozac (Toronto: John Wiley and Sons, 1997), 200–7. George Windholz and L.H. Witherspoon, ‘Sleep as a Cure for Schizophrenia: A Historical Episode,’ History of Psychiatry 4 (1993): 83–93.

Notes to pages 188–92

463

32 The literature on the dramatic increase in asylum populations from the Victorian period is a vast and divisive one among historians of psychiatry. Compare, for example, Gerald Grob, Mental Illness and American Society, 1875– 1940 (Princeton, NJ: Princeton University Press, 1983); Andrew Scull, Social Order/Mental Disorder: Anglo-American Psychiatry in Historical Perspective (Berkeley: University of California Press, 1989); and Shorter, A History of Psychiatry. 33 Gray, Pavlov, 121; Joseph Wortis, Soviet Psychiatry (Baltimore: Williams and Wilkins, 1950), 162. 34 See Theresa C. Smith, No Asylum: State Psychiatric Repression in the Former USSR (Basingstoke, UK: Macmillan, 1996). 35 Gray, Pavlov, 130ff. 36 Pavlov, Conditioned Reflexes, 263–4. 37 R.M. Yerkes and S. Morgulis, ‘The Method of Pawlow in Animal Psychology,’ Psychological Bulletin 7 (1909): 257–73. 38 Randall D. Wight, ‘The Pavlov-Yerkes Connection: What Was Its origin?’ Psychological Record 43 (1993): 351–60; John M. O’Donnell, The Origins of Behaviorism: American Psychology, 1870–1920 (New York: New York University Press, 1985). 39 James B. Watson, ‘Psychology as the Behaviorist Views it’; idem, ‘The Place of the Conditioned Reflex in Psychology,’ Psychological Review 23 (1916): 89– 116. 40 On this point, see Jean-Pierre Peter, ‘Sommeil, rêve, anesthésie, somnambulisme: Le problème de la conscience dans les représentations de l’homme en sommeil,’ Revue d’histoire moderne et contemporaine 43 (1996): 578–92. 41 Much of the following draws upon Kroker, ‘Epidemic Encephalitis.’ 42 See, for example, Oliver Sacks and Joel A. Vilensky, ‘Waking to a New Flu Threat,’ New York Times, 16 November 2005; and Kirsty Duncan, ‘Is Sleeping Sickness Linked to Spanish Flu of 1918?’ Toronto Star, 3 January 2006, A15. 43 Oliver W. Sacks, Awakenings (London: Duckworth, 1973); idem, ‘Postencephalic syndromes,’ in Gerald M. Stern, ed., Parkinson’s Disease (London: Chapman and Hall Medical, 1990); Christopher D. Ward, ‘Encephalitis Lethargica and the Development of Neuropsychiatry,’ Psychiatric Clinics of North America 9 (1986): 215–24; J. Biéder, C. Paresys-Nourry, P. Paresys, and J.D. Even, ‘Documents pour servir à l’histoire de la psychiatrie: L’encéphalite épidémique à la société clinique de médecine mentale,’ Annales médicopsychologiques 147 (1989): 862–6. 44 Ward, ‘Encephalitis Lethargica,’ 223. 45 On the twentieth-century history of virology, see Angela N.H. Creager, The Life of a Virus: Tobacco Mosaic Virus as an Experimental Model, 1930–1965 (Chi-

464

46

47

48

49 50 51 52 53 54 55

Notes to pages 193–6

cago: University of Chicago Press, 2002); and Ton van Helvoort, ‘A Bacteriological Paradigm in Influenza Research in the First Half of the Twentieth Century,’ History and Philosophy of the Life Sciences 15 (1993): 3–21; on the interaction of virus research and epidemic encephalitis, see Kenton Kroker, ‘Creatures of Reason? Viruses at the Pasteur Institute during the 1920s,’ in K. Kroker, P.M.H. Mazumdar, and J. Keelan, eds., Crafting Immunity (London: Ashgate, forthcoming). For biographical detail (with little historical context), see Karoline Freifrau von Economo and Julius Wagner-Jauregg, Baron Constantin von Economo: His Life and Work, trans. R. Spillman (Burlington, VT: Free Press Interstate Publishing, 1937); and Ludo van Bogaert and Jean Théodoridès, Constantin von Economo (1876–1931): The Man and the Scientist (Vienna: Österreichischen Akademie der Wissenshaften, 1979). The former is, in part, the transcript of a celebratory radio address given by Wagner-Jauregg shortly after his student’s untimely death of angina at age fifty-five. The latter includes a translated collection of several of Economo’s papers. Constantin von Economo, ‘Encephalitis lethargica,’ Wiener klinische Wochenshrift 30 (1917): 581–5. A translation of the paper (minus six of the case descriptions) can be found in Bogaert and Théodoridès, Constantin von Economo, 79–84. Maryinez Lyons, The Colonial Disease: A Social History of Sleeping Sickness in Northern Zaire, 1900–1940 (New York: Cambridge University Press, 1992); for the social and intellectual context of experimentation, see Wolfgang U. Eckart, ‘The Colony as Laboratory: German Sleeping Sickness Campaigns in German East Africa and in Togo, 1900–1914,’ History and Philosophy of the Life Sciences 24 (2002): 69–89; and Christoph Gradmann, ‘“It Seemed about Time to Try One of Those Modern Medicines”: Animal and Human Experimentation in the Chemotherapy of Sleeping Sickness 1905–1908,’ in V. Roelke and G. Maio, eds., Twentieth Century Ethics of Human Subjects Research (Stuttgart: Franz Steiner Verlag, 2004), 83–98. ‘A Strange Bargain Is Offered,’ New York Times, 5 January 1923. Eckart, ‘The Colony as Laboratory.’ This is not to imply that all European responses to sleeping sickness were identical – see Lyons, The Colonial Disease. Bogaert and Théodoridès, Constantin von Economo, 39–41, 85–96; and Schiller, ‘Semantics of Sleep.’ L. Mauthner, ‘Pathologie und Physiologie des Schlafes,’ Wiener klinische Wochenshrift 3 (1890): 445–6; Schiller, ‘Semantics of Sleep.’ Bogaert and Théodoridès, Constantin von Economo, 83. Ibid., 85–96. Matheson Commission, Epidemic Encephalitis: Etiology, Epidemiology, Treatment.

Notes to pages 197–202

56

57 58 59

60

61 62

63

64 65

66 67 68 69

465

Report of a Survey by the Matheson Commission (New York: Columbia University Press, 1929). Constantin Freiherr von Economo, Die Encephalitis lethargica, ihre Nachkrankheiten und ihre Behandlung (Berlin: Urban and Schwarzenberg, 1929), trans. as Encephalitis Lethargica; Its Sequelae and Treatment (London: Oxford University Press, 1931). Economo’s 1929 lecture was ‘Sleep as a Problem of Localization,’ Journal of Nervous and Mental Disease 71 (1930): 249–59. Economo, ‘Sleep,’ 249–50. Ibid., 253. J. Victor Haberman, ‘Sleep (Normal and Abnormal) and the Mechanism of Sleep,’ Medical Record (New York) 101 (1922): 265–72. On the viability of the diagnosis of encephalitis lethargica by post-mortem, see Kroker, ‘Epidemic Encephalitis,’ 119. Economo, ‘Sleep,’ 256; for a similar misreading of Pavlov, see Ernst Trömner, ‘Schlafunktion und Schlaforgan,’ Deutsche Zeitschrift für Nervenheilkunde 105 (1928): 191–204. To the best of my knowledge, Pavlov first explicitly spoke of a ‘sleep centre’ in the brain in 1935. See Pavlov, ‘The Problem of Sleep.’ Economo,’ Sleep,’ 259. For an analysis of the journal that replaced Jelliffe’s, see Nissim Mizrachi, ‘From Causation to Correlation: The Story of ‘Psychosomatic medicine’ 1939–1979,’ Culture, Medicine and Psychiatry 25 (2001): 317–43; see also Theodore Brown, ‘Cartesian Dualism and Psychosomatics,’ Psychosomatics 30 (1989): 322–31; and Dorothy Levenson, Mind, Body and Medicine: A History of the American Psychosomatic Society (Baltimore: Williams and Wilkins, 1994). See, for example, Smith Ely Jelliffe, Postencephalic Respiratory Disorders: A Review of Syndromy, Case Reports, Physiopathology, Psychopathology and Therapy (New York: Nervous and Mental Disease Publishing, 1927); and idem, Psychopathology of Forced Movements and the Oculogyric Crises of Lethargic Encephalitis (New York: Nervous and Mental Disease Publishing, 1932). Mizrachi, ‘From Causation to Correlation.’ For example, Isador Abrahamson, ‘Epidemic Encephalitis Lethargica,’ Medical Record (New York) 98 (1920): 969; and idem, Lethargic Encephalitis (New York, 1935). Otto Fenichel, The Psychoanalytic Theory of Neurosis (New York: W.W. Norton, 1945), 258–9. See Kroker, ‘Epidemic Encephalitis.’ ‘Mt. Sinai Hospital Nears Completion,’ New York Times, 1 May 1921, 37. The Times ran at least eight stories on the topic, including ‘Mrs. J.P. Morgan Has Sleeping Sickness,’ 19 June 1925, 1; and ‘Mrs. J.P. Morgan Dies at Glen

466

70 71 72 73 74 75

76 77

Notes to pages 202–9

Cove Home,’ idem, 20 August 1925, 1. On Morgan’s donation, see ‘Sleeping Sickness Fund from Morgan,’ idem, 22 March 1927, 29. Kroker, ‘Epidemic Encephalitis,’ 131–9. ‘Child Slept for a Month,’ New York Times, 16 April 1919, 8 ‘Breaks a 3-Year Sleep,’ New York Times, 29 March 1921, 5. ‘Saved by Talking Machine: Woman Roused from 80-Day Sleep When Other Means Fail,’ New York Times, 24 December 1924, 8. ‘Woman Breaks Two Years’ Sleep,’ New York Times, 21 August 1920, 18. For example, Sylvester R. Leahy and Irving J. Sands, ‘Mental Disorders in Children following epidemic Encephalitis,’ Journal of the American Medical Association 76 (1921): 373–7; G.A. Auden, ‘Behaviour Changes Supervening upon Encephalitis in Children,’ Lancet, 28 October 1922, 901–4; F.G. Edbaugh, ‘Neuropsychiatric Sequelae of Acute Encephalitis in Children,’ American Journal of Diseases of Children 25 (1923): 89–97. My thanks to David Pantalony for directing my attention to these sources. ‘“Badness” Responds to Mild Methods,’ New York Times, 15 November 1931, E7. ‘Boy Pickpocket Says Mania Causes Thefts,’ New York Times, 13 July 1925, 19.

5. Performing Sleep 1 Nathaniel Kleitman, ‘Studies in the Physiology of Sleep: I. The Effects of Prolonged Sleeplessness on Man,’ American Journal of Physiology 66 (1923): 67– 92; idem, ‘Basic Rest-Activity Cycle – 22 Years later,’ Sleep 5 (1982): 311–17. 2 Gérard Lemaine et al., Stratégies et choix dans la recherche: À propos des travaux sur le sommeil (Paris: Mouton, 1977). 3 For an superb example of the problematic nature of this assumption, see Tom Rivers, Tom Rivers: Reflections on a Life in Medicine and Science, an Oral History Memoir Prepared by Saul Benison (Cambridge, MA: MIT Press, 1967). 4 Ian Hacking, Representing and Intervening: Introductory Topics in the Philosophy of Natural Science (Cambridge: Cambridge University Press, 1983); Daston, ed., Biographies of Scientific Objects. 5 On research schools, see Gerald L. Geison and Frederic L. Holmes, eds., ‘Research Schools: Historical Reappraisals,’ Osiris 8 (1993): 1–248; and Kusch, ‘Recluse, Interlocutor, Interrogator.’ 6 Lemaine et al., Stratégies et choix dans la recherche; Schiller, ‘Semantics of Sleep.’ 7 Peretz Lavie, The Enchanted World of Sleep, trans. Anthony Berris (New Haven, CT: Yale University Press, 1996), 18. 8 Moshe Ussoskin, Struggle for Survival: A History of Jewish Credit Co-operatives in

Notes to pages 209–17

9 10 11

12

13

14

15 16 17 18 19 20 21 22 23 24 25 26

27

467

Bessarabia, Old-Rumania, Bukovina and Transylvania (Jerusalem: Jerusalem Academic Press, 1975). Lavie, The Enchanted World of Sleep, 18. Fye, The Development of American Physiology. Dwight J. Ingle, ‘Anton J. Carlson: A Biographical Sketch,’ Perspectives in Biology and Medicine 22 suppl. (1979): S114–S136; see also A.C. Ivy, ‘Anton Julius Carlson,’ Physiologist 2 (1959): 33–9; and L.R. Dragstedt, ‘Anton Julius Carlson, January 29, 1875–September 2, 1956,’ Biographical Memoirs of the National Academy of Science 35 (1961): 1–32. See Gerald L. Geison, ed., Physiology in the American Context, 1850–1940 (Bethesda, MD: American Physiological Society, 1987), particularly the articles by Merrily Borell, Adele E. Clarke, and Louise H. Marshall. On Taylorism in the United States, see Rabinbach, The Human Motor. For a survey of the impact of private philanthropy on American science and biomedicine, see Robert E. Kohler, Partners in Science: Foundations and Natural Scientists, 1900–1945 (Chicago: University of Chicago Press, 1991); and Joel D. Howell, Technology in the Hospital: Transforming Patient Care in the Early Twentieth Century (Baltimore: Johns Hopkins University Press, 1995). Gerald L. Geison, ‘International Relations and Domestic Elites in American Physiology, 1900–1940,’ in Geison, ed., Physiology in the American Context; see also Ingle, ‘Anton J. Carlson.’ Ivy, ‘Anton Julius Carlson.’ Geison, ‘International Relations and Domestic Elites,’ 148. See Tara Abraham, ‘Nicolas Rashevsky’s Mathematical Biophysics,’ Journal of the History of Biology 34 (2004): 333–85, 356–7. Ingle, ‘Anton Julius Carlson’; Michael Bliss, The Discovery of Insulin (Toronto: McClelland and Stewart, 1982). Anton Julius Carlson, The Control of Hunger in Health and Disease (Chicago: University of Chicago Press, 1916), 161. Carlson, The Control of Hunger 160. Ibid., 153–4. Kleitman, ‘Studies in the Physiology of Sleep: I,’ 67. Ibid., 67. Ibid., 67. Ibid., 83. K.H. Collins and A.L. Tatum, ‘A Conditioned Salivary Reflex Established by Chronic Morphine Poisoning,’ American Journal of Physiology 74 (1925): 14–15. Nathaniel Kleitman and George Crisler, ‘A Quantitative Study of a Salivary Conditioned Reflex,’ American Journal of Physiology 79 (1927): 571–614, 571.

468

Notes to pages 218–22

28 See, for example, Maurice Merleau-Ponty, The Structure of Behavior, trans. Alden L. Fisher (Boston: Beacon Press, 1963). 29 For an analysis of the tension between the conceptualization of Pavlov’s dogs as technologies and the view of them as organisms, see Todes, ‘Pavlov’s Physiology Factory.’ The resulting ambivalence regarding holism in Pavlov’s work might explain why there is no mention of him in a recent work on biomedical holism (Lawrence and Weisz, eds., Greater than the Parts). 30 Kurt Goldstein, The Organism: A Holistic Approach to Biology Derived from Pathological Data in Man (New York: Zone Books, 1995). 31 Lancelot Hogben, as cited in W.K. Hancock, Smuts: The Fields of Force, 1919– 1950 (Cambridge: Cambridge University Press, 1968), 191. 32 Jan Christiaan Smuts, Holism and Evolution (London: Macmillan, 1926), 205. 33 Nathaniel Kleitman, Sleep and Wakefulness as Alternating Phases in the Cycle of Existence (Chicago: University of Chicago Press, 1939), 471; see also Dumas, Nouveau traité de psychologie, 4: 455–522. 34 Lawrence and Weisz, Greater than the Parts, 2. 35 Bonnie Ellen Blustein, ‘Medicine as Biology: Neuropsychiatry at the University of Chicago, 1928–1939,’ Perspectives on Science 1 (1993): 416–44. 36 See the articles in a special edition of Perspective on Science 1 (1993), including Gregg Mitman, Jane Maienschein, and Adele E. Clarke, ‘Introduction’ (359– 66); and Sharon E. Kingsland, ‘A Humanistic Science: Charles Judson Herrick and the Struggle for Psychobiology at the University of Chicago’ (445– 77). See also Jack Pressman, ‘Human Understanding: Psychosomatic Medicine and the Mission of the Rockefeller Foundation,’ in Lawrence and Weisz, eds., Greater than the Parts. 37 Franklin McLean, as cited in Blustein, ‘Medicine as Biology,’ 417. 38 ‘Part II of a Document Entitled “The Biological Sciences and Psycho-Neurology” as Prepared for the Consideration of the General Education Board and Dated April 10, 1929,’ Rockefeller Archives Center (hereafter RAC), 1.1, 216A, 6, 73. 39 Blustein, ‘Medicine as Biology,’ 421. 40 Office of the President of the University of Chicago to Max Mason, RAC, 1.1, 216A, 6, 73. 41 Diary entry, Alan Gregg, 8 May 1931, RAC, 1.1, 216A, 6, 73. 42 Franklin C. McLean, ‘A University Department of Psychiatry,’ RAC, 1.1, 216A, 6, 73. 43 Franklin C. McLean, ‘Minutes of the First Meeting of the Committee Appointed by the President to Advise as to the Development of Psychiatry in the University, June 18th, 1931,’ RAC, 1.1, 216A, 6, 73. 44 Lapique’s research program had little impact outside France. See Jean-

Notes to pages 222–31

45 46

47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67

68 69

469

Claude Dupont, ‘Autour d’une controverse sur l’excitabilité: Louis Lapicque et l’École de Cambridge,’ in Claude Debru, Jean Gayon, and Jean-François Picard, eds., Les sciences biologiques et médicales en France, 1920–1950 (Paris: CNRS, 1994). Interview, Louise H. Marshall, 26 July 2002. Marshall was herself a student of Carlson’s during the 1930s. See Kleitman’s letter (18 February 1936) to Dr Robert A. Lambert of the Rockefeller Foundation, outlining his funding sources, RAC, 1.1, 216A, 7, 88. Ingle, ‘Anton Julius Carlson,’ S126. Blustein, ‘Medicine as Biology.’ Carlson to Gregg, 12 March 1934; Gregg to Carlson, 15 March 1934, RAC 1.1, 216A, 7, 88. Kleitman to Gregg, 21 March 1934. Kleitman, ‘Studies in the Physiology of Sleep: I.’ Ibid., 73. Ibid., 90. Kleitman to Gregg, 21 March 1934. Ibid. Alan Gregg, diary entry, 28 March 1934, RAC, 1.1, 216A, 7, 88. Carlson to Gregg, 8 May 1934, RAC, 1.1, 216A, 7, 88. Kleitman to Gregg, 10 September 1934, RAC, 1.1, 216A, 7, 88. Carlson to Gregg, 8 January 1935, RAC, 1.1, 216A, 7, 88. Alan Gregg, diary entry, 20 January 1935, RAC, 1.1, 216A, 7, 88. ‘University of Chicago – Physiology of Sleep. Appraisal, June, 1939,’ RAC, 1.1, 216A, 7, 89. See ‘Kleitman, Nathaniel,’ Current Biography 1957, 308; and coverage in New York Times, 12 November 1939, 2: 7; 7 August 1949, 45; 17 May 1953, 4: 11. ‘University of Chicago – Physiology of Sleep.’ Roy Grinker to Alan Gregg, 23 October 1935, RAC, 1.1, 216A, 6, 75; see also Blustein, ‘Medicine as Biology,’ 425. Roy Grinker to Alan Gregg, 23 October 1935. Kleitman to Gregg, 20 January 1936, RAC, 1.1, 216A, 7, 88. Grinker to Woodward, 27 March 1935, RAC, 1.1, 216A, 6, 76; Grinker to George F. Dick (chair, Medical Department, University of Chicago), 27 January 1936, RAC, 1.1, 216A, 6, 76. Blustein, ‘Medicine as Biology.’ See Tara H. Abraham, ‘(Physio)logical Circuits: The intellectual Origins of the McCulloch–Pitts Neural Networks,’ Journal of the History of the Behavioral Sciences 38 (2002): 3–25.

470

Notes to pages 232–8

70 ‘University of Chicago – Sub-Department of Psychiatry. Appraisal, June, 1938,’ RAC, 1.1, 216A, 6, 78. On Slight’s research interests, see Arthur C. Bachmeyer to Gregg, 22 June 1937, RAC, 1.1, 216A, 6, 76. 71 ‘University of Chicago – Department of Physiology. RF 35026,’ RAC, 1.1, 216A, 7, 88. 72 Alan Gregg, diary entry, 14 January 1936, RAC, 1.1, 216A, 7, 88. 73 Kleitman to Gregg, 14 January 1938, RAC, 1.1, 216A, 7, 89. 74 New York Times, 4 Jul 1938, 10, and 12 November 1939, II: 7. 75 Gregg to Kleitman, 20 January 1938, RAC, 1.1, 216A, 7, 89. 76 ‘University of Chicago – Physiology of Sleep. Appraisal, June, 1939.’ 77 Ibid. 78 Kleitman to Gregg, 19 January 1940; and Gregg to Kleitman, 12 January 1940, in RAC, 1.1, 216A, 7, 89. 79 Nathaniel Kleitman, ‘A Scientific Solution of the Multiple Shift Problem,’ Mining Congress Journal 29 (1943): 15–6; idem, ‘Biological Rhythms and Cycles,’ Physiological Reviews 29 (1949): 1–30; idem, ‘Mental Hygiene of Sleep in Children,’ The Nervous Child: Quarterly Journal of Psycho-Pathology, Psychotherapy, Mental Hygiene, and Guidance of the Child 8 (1949): 63–6; Nathaniel Kleitman and D.P. Jackson, ‘Body Temperature and Performance under Different Routines,’ Journal of Applied Physiology 3 (1950): 309–28; Nathaniel Kleitman and T.G. Engelmann, ‘Sleep Characteristics of Infants,’ Journal of Applied Physiology 6 (1953): 269–82. 80 ‘Problem of Sleep,’ University of Chicago Public Lecture, 8 August 1928, Nathaniel Kleitman Papers, box 28, University of Chicago Library, Special Collectors Research Center. 81 See, for example, Dement, Some Must Watch While Some Must Sleep; Lemaine et al., Stratégies et choix dans la recherché; Schiller, ‘Semantics of Sleep’; and Hobson, The Dreaming Brain. For perspectives from the history of psychology, see Ryan D. Tweney, ‘Programmatic Research in Experimental Psychology: E.B. Titchener’s Laboratory Investigations, 1891–1927,’ in Mitchell G. Ash and William R. Woodward, eds., Psychology in Twentieth-Century Thought and Society (Cambridge: Cambridge University Press, 1987); Arnold H. Gessel, ‘Edmund Jacobson, M.D., Ph.D.: The Founder of Scientific Relaxation,’ International Journal of Psychosomatics 36 (1989): 5–14; and Smith, Inhibition. 82 Kroker, ‘Progress of Introspection.’ 83 See Gessel, ‘Edmund Jacobson.’ Other sources for biographical material include interviews and communication with Jacobson’s former students, Arnold Gessel and Helene Morcos; his laboratory technician, Richard Lange; and his son Edmund Jr.

Notes to pages 239–45

471

84 See Danziger, ‘The History of Introspection Reconsidered,’ 241–62; Tweney, ‘Programmatic Research’; and Kusch, ‘Recluse, Interlocutor, Interrogator.’ 85 Edmund Jacobson, ‘On Meaning and Understanding,’ American Journal of Psychology 22 (1911): 553–77. 86 See A.S. Winston, ‘“As the name indicates”: R.S. Woodworth’s Letters of Reference and Employment for Jewish Psychologists in the 1930s,’ Journal for the History of the Behavioral Sciences 32 (1996): 30–43; and idem, ‘“The defects of his race”: E.G. Boring and Anti-Semitism in American Psychology,’ History of Psychology 1 (1998): 27–51. For a general history, see Leonard Dinnerstein, Antisemitism in America (New York: Oxford University Press, 1994). 87 Interview, Helene Morcos, 11 December 1998. 88 Many women worked in Titchener’s laboratory, but only the men seem to have been exposed to the nasty odours of India rubber or resins – the women sniffed beeswax. See Edmund Jacobson, ‘Experiments on the Inhibition of Sensations,’ Psychological Review 18 (1911): 24–53; idem, ‘Further Experiments on the Inhibition of Sensations,’ American Journal of Psychology 23 (1912): 345–69. 89 Jacobson, ‘Further Experiments on the Inhibition of Sensations,’ 345. 90 The origins of progressive relaxation thus appear to share at least some of the ‘looping effects of human kind’ described by Ian Hacking as integral elements of phenomena such as multiple-personality disorder and child abuse (see Hacking, Social Contruction of What?). This similarity might go some way to explaining why, during interviews, some of my interviewees spontaneously incorporated demonstrations of Jacobson’s techniques into the interview itself. Their actions implied both that the practice was an integral part of their identity and that my understanding of this practice would be transformed if I made an effort to learn it. 91 Edmund Jacobson, ‘The Technic of Progressive Relaxation,’ Journal of Nervous and Mental Disease 60 (1924): 568–78. 92 Ibid., 570; emphasis in original. Jacobson would likely have denied any similarity between his techniques and those of psychoanalysis, however. 93 Edmund Jacobson, Progressive Relaxation: A Physiological and Clinical Investigation of Muscular States and Their Significance in Psychology and Medical Practice (Chicago: University of Chicago Press, 1929), 111. 94 Ibid., 31. 95 Grinker to Gregg, 23 October 1935, RAC, 1.1, 216A, 6, 75. 96 See Kroker, ‘Progress of Introspection,’ 95–6. 97 Flexner to Jacobson, 16 December 1930; Flexner to Jacobson, 7 May 1931; Jacobson to Flexner, 3 September 1931. Simon Flerna Papers (hereafter SFP), B:F 365).

472

Notes to pages 245–52

98 Interview, Helene Morcos, 11 December 1998; interview, Richard Lange, 18 December 1998. 99 Jacobson to Forbes, 14 September 1926, AFP, 10.443. Emphasis in original. 100 Forbes to Jacobson, 23 September 1926, AFP, 10.443. 101 Edmund Jacobson, ‘Action Currents from Muscular Contractions during Conscious Processes,’ Science 66 (1927): 403–4. 102 Ibid., 404. 103 Jacobson published his results in a series of seven articles in the American Journal of Physiology between 1930 and 1931. The following year, he published a summary of his results which was later singled out in the centennial issue of the American Journal of Psychology (1987) as one of the most important contributions to the study of ‘electrophysiological measures’ of the last century. Edmund Jacobson, ‘Electrophysiology of Mental Activities,’ American Journal of Psychology 44 (1932): 677–94. 104 Ibid. 105 Edmund Jacobson, ‘Electrical Measurements of Neuromuscular States during Mental Activities. I. Imagination of Movement involving Skeletal Muscle,’ American Journal of Physiology 91 (1930): 567–608. 106 Otniel E. Dror, ‘The Affect of Experiment: The Turn to Emotions in AngloAmerican Physiology, 1900–1940,’ Isis 90 (1999): 205–37. 107 Jacobson, ‘Electrical Measurements of Neuromuscular States during Mental Activities. I,’ 579. 108 Margaret Floy Washburn, Movement and Mental Imagery: Outline of a Motor Theory of the Complexer Mental Processes (Boston: Houghton, 1916). 109 Walter B. Cannon, ‘The James-Lange Theory of Emotions,’ American Journal of Psychology 39 (1927): 115–24. 110 For a comparative analysis, see Shigehisa Kuriyama, The Expressiveness of the Body and the Divergence of Greek and Chinese Medicine (New York: Zone Books, 1999), 111–52. 111 Jacobson certainly had other sources of revenue, for he had apparently made several million dollars on the stock market, and managed to get most of it out before the crash of 1929 (interview, Helene Morcos, 11 December 1998). 112 Ibid. 113 Edmund Jacobson, You Can Sleep Well: The ABC’s of Restful Sleep for the Average Person (New York: McGraw-Hill, 1938), 82ff. 114 Ibid., ix–x, emphasis in original. 115 Ibid., 204–29. 116 Ibid., 199. 117 Ibid., 144–5. 118 Ibid., 156.

Notes to pages 257–62

473

6. Sleep Finds a Groove 1 Schiller, ‘The Semantics of Sleep.’ 2 Nicolas Adelon, ed., Dictionaire des sciences médicales, 60 vols. (Paris, 1812–22), as cited in Peter, ‘Sommeil, rêve, anesthésie, somnambulisme,’ 579. 3 Herbert H. Jasper, ‘Electrical Signs of Cortical Activity,’ Psychological Bulletin 34 (1937): 411–81. 4 Herbert H. Jasper, ‘Philosophy or Physics – Mind or Molecules,’ in Frederic G. Worden, Judith P. Swazey, and George Adelman, eds., The Neurosciences: Paths of Discovery (Cambridge, MA: MIT Press, 1975). 5 Jasper, ‘Electrical Signs of Cortical Activity,’ 411. 6 Ibid., 470. 7 Lily Kay, The Molecular Vision of Life: Caltech, the Rockefeller Foundation, and the Rise of the New Biology (Oxford: Oxford University Press, 1993). 8 Kay, The Molecular Vision of Life, 26. 9 Sean Dennis Cashman, America Ascendant: From Theodore Roosevelt to FDR in the Century of American Power, 1901–1945 (New York: New York University Press, 1998), 280–318. 10 Kohler, Partners in Science, 265–302. 11 Ibid., 358–91. 12 Ibid., 364–71. 13 Howell, Technology in the Hospital. 14 Ibid., 103–32. 15 On the ECG, see Frank, ‘The Telltale Heart.’ 16 Howell, Technology in the Hospital, 122–5. 17 For examples, see Elliot S. Valenstein, Great and Desperate Cures: The Rise and Decline of Psychosurgery and Other Radical Treatments for Mental Illness (New York: Basic Books, 1986); Scull, Social Order/Mental Disorder; The Mad among Us ; Shorter, A History of Psychiatry; and David Healy, ‘Some Continuities and Discontinuities in the Pharmacotherapy of Nervous Conditions before and after Chlorpromazine and Imipramine,’ History of Psychiatry 11 (2000): 393–412. 18 Prominent examples include Herbert H. Jasper, ‘Charting the Sea of Brain Waves, 1948,’ Journal of Clinical Neurophysiology 14 (1997): 464–9; Mary A.B. Brazier, A History of the Electrical Activity of the Brain: The First Half-Century (London: Pitman Medical Publishing, 1961); Pierre Gloor, ‘Hans Berger and the Discovery of the Electroencephalogram,’ in P. Gloor, ed., ‘Hans Berger on the Electroencephalogram: The Fourteen Original Reports on the Human Electroencephalogram,’ Electroencephalography and Clinical Neurophysiology 28, suppl. (1969); Albert M. Grass, ‘The Electroencephalographic Her-

474

19 20

21

22

23

24

25 26

27 28

29 30 31

Notes to pages 263–7

itage until 1960,’ American Journal of EEG Technology 24 (1960): 133–73; and Jacob Empson, Human Brainwaves: The Psychological Significance of the Electroencephalogram (London: Stockton Press, 1986). The definitive historical study of epilepsy remains Owsei Temkin, The Falling Sickness: A History of epilepsy from the Greeks to the Beginning of Modern Neurology, 2nd rev. ed. (Baltimore: Johns Hopkins University Press, 1971). On the work at Harvard, see RAC, 1.1, 200A, 86 and 87; on Illinois, see RAC, 1.2, 200A, 153; on the Burden Neurological Institute, see RAC, 1.1, 401A, 15. See Kenton Kroker, ‘Washouts: Electroencephalography, Epilepsy and Emotions in the Selection of American Aviators during the Second World War,’ in S. Walton, ed., Instrumental in War: Science, Research, and Instruments between Knowledge and the World (Leiden and Boston: Brill, 2005), 301–38. Cornelius Borck, ‘Electricity as a Medium of Psychic Life: Electrotechnical Adventures into Psychodiagnosis in Weimar Germany,’ Science in Context 14 (2001): 565–90; and Rhodri Hayward, ‘The Tortoise and the Love-Machine: Grey Walter and the Politics of Electroencphalography,’ Science in Context 14 (2001): 615–41. Lemaine et al., for example, relegate EEG to an appendix chronicling advances in the technology since Berger’s initial discoveries (Stratégies et choix dans la recherché). Joseph H. Spear, ‘Cumulative Change in Scientific Production: Research Technologies and the Structuring of New Knowledge,’ Perspectives on Science 12 (2004): 55–85. Brazier, A History of the Electrical Activity of the Brain; Susan Leigh Star, Regions of the Mind: Brain Research and the Quest for Scientific Certainty (Stanford, CA: Stanford University Press, 1989). Richard Caton, ‘The Electric Currents of the Brain,’ British Medical Journal 1 (1875): 278. Hans Berger, ‘On the Electroencephalogram of Man. Third Report,’ Archiv für Psychiatrie und Nervenkrankheiten 94 (1931): 16–60, as translated and reproduced in Gloor, ed., ‘Hans Berger on the Electroencephalogram.’ Gloor, ‘Hans Berger and the Discovery of the Electroencephalogram,’ 4ff. Hans Berger, ‘On the Electroencephalogram of Man. First Report,’ Archiv für Psychiatrie und Nervenkrankheiten 87 (1929): 527–70, in Gloor, ed., ‘Hans Berger on the Electroencephalogram,’ 45. Emphasis in original. Gloor, ed., ‘Hans Berger and the Discovery of the Electroencephalogram,’ 8. See Frank, ‘The Telltale Heart’; and Howell, Technology in the Hospital. Berger, ‘On the Electroencephalogram of Man. First Report,’ 70. Emphasis in original.

Notes to pages 267–73

475

32 Hans Berger, ‘On the Electroencephalogram of Man. Second Report,’ Journal für Psychologie und Neurologie 40 (1930): 160–79, in Gloor, ed., ‘Hans Berger on the Electroencephalogram.’ 33 Berger, ‘On the Electroencephalogram of Man. First Report,’ 81. 34 Ibid., 88–93. 35 Hans Berger, ‘On the Electroencephalogram of Man. Eighth Report,’ in Gloor, ed., ‘Hans Berger on the Electroencephalogram,’ 209–23. 36 Hans Berger, ‘On the Electroencephalogram of Man. Ninth Report,’ in Gloor, ed., ‘Hans Berger on the Electroencephalogram,’ 225–42. 37 See Cornelius Borck, Hirnströme: Eine Kulturgeschichte der Elektroenzephalographie (Göttingen: Wallstein, 2005). 38 Fye, The Development of American Physiology, 26–7. 39 Viktor von Weizsäcker, as quoted in Shorter, A History of Psychiatry, 156. 40 Gloor, ‘Hans Berger and the Discovery of the Electroencephalogram.’ 41 Josiah Macy, Jr Foundation, Josiah Macy, Jr. Foundation, 1930–1955: A Review of Activities (New York: Josiah Macy, Jr Foundation, 1955), 63–70. 42 Marshall, ‘Instruments, Techniques, and Social Units in American Neurophysiology’; Robert G. Frank, Jr, ‘Instruments, Nerve Action and the All-orNone Principle,’ Osiris 9 (1994): 208–35; John Z. Young, ‘Sources of Discovery in Neuroscience,’ in Worden, Swazey, and Adelman, eds., The Neurosciences. 43 See, for example, letter, Edgar Adrian to Alexander Forbes, 16 May 1914, AFP, 1, 2. 44 Edgar D. Adrian, ‘The All-or-None Principle in Nerve,’ Journal of Physiology 47 (1914): 460–74. 45 Frank, ‘Instruments, Nerve Action and the All-or-None Principle,’ 217; see also Adrian to Forbes, 16 May 1914, AFP, 1, 2. 46 Adrian to Forbes, 21 March 1919, AFP, 1, 2. 47 Charles Sherrington to Alexander Forbes, 18 June 1916, AFA, 15, 730. 48 Edgar Adrian to Alexander Forbes, 21 March 1919, AFP, 1, 2. 49 Edgar D. Adrian, ‘The Activity of the Nerve Fibres: Nobel Lecture, December 12, 1932,’ www.nobelprize.org/medicine/laureates/1932/ adrian-lecture.html. Accessed 19 September 2002. 50 J.K. Bradley and E.M. Tansey, ‘The Coming of the Electronic Age to the Cambridge Physiological Laboratory: E.D. Adrian’s Valve Amplifier in 1921,’ Notes and Records of the Royal Society of London 50 (1996): 217–28. 51 Frank, ‘Instruments, Nerve Action and the All-or-None Principle,’ 220. 52 Alexander Forbes, ‘The Laboratory Reacts,’ Atlantic Monthly 118 (1916): 544–51.

476

Notes to pages 274–9

53 Alexander Forbes and Catharine Thacher, ‘Amplification of Action Currents with the Electron Tube in Recording with the String Galvanometer,’ American Journal of Physiology 52 (1920): 409–71. 54 Gerald F. Tyne, Saga of the Vacuum Tube (Indianapolis, IN: Howard W. Sams, 1977), 307ff. 55 Frank, ‘Instruments, Nerve Action and the All-or-None Principle,’ 224. 56 Edgar Adrian to Alexander Forbes, 11 July 1920; see also Adrian to Forbes, 20 August 1920, AFP, 1, 3. 57 Edgar Adrian to Alexander Forbes, 1 March 1921, AFP, 1, 3. 58 Louise H. Marshall and Horace W. Magoun, Discoveries in the Human Brain: Neuroscience Prehistory, Brain Structure, and Function (Totowa, NJ: Humana Press, 1998); cf. David Millett, ‘Wiring the Brain: From the Excitable Cortex to the EEG, 1870–1940,’ (PhD thesis, University of Chicago, 2001). 59 Edgar D. Adrian, ‘The Discovery of Berger,’ in Antoine Rémond, ed., Handbook of Electroencephalography and Clinical Neurophysiology, 1 (1971): 1A–7. 60 Edgar D. Adrian and B.H.C. Matthews, ‘The Interpretation of Potential Waves in the Cortex,’ Journal of Physiology 81 (1934): 440–71; idem, ‘The Berger Rhythm: Potential Changes from the Occipital Lobes in Man,’ Brain 57 (1934): 355–85. On the history of flicker, see John Geiger, Chapel of Extreme Experience: A Short History of Flicker (Toronto: Gutter Press, 2002). 61 Adrian and Matthews, ‘The Berger Rhythm,’ 383. 62 See Hayward, ‘The Tortoise and the Love-Machine.’ 63 See RAC, 1.1, 401A, 15. 64 Early claims for the importance of Forbes’s work came from his student: Hallowell Davis, ‘The Forbes “School” of Neurophysiology at Harvard,’ Electroencephalography and Clinical Neurophysiology 1 (1949): 139–40. For a historical account from the perspective of systems engineering and computing, see David A. Mindell, Between Human and Machine: Feedback, Control, and Computing before Cybernetics (Baltimore: Johns Hopkins University Press, 2002). 65 Walter B. Cannon, ‘The Body Physiologic and the Body Politic,’ Science 93 (1941): 1–10; Stephan J. Cross and William R. Albury, ‘Walter B. Cannon, L.J. Henderson, and the Organic Analogy,’ Osiris 2 (1987): 165–92; Allan Young, ‘Walter Cannon and the Psychophysiology of Fear,’ in Lawrence and Weisz, eds., Greater than the Parts. 66 Alexander Forbes, ‘The Interpretation of Spinal Reflexes in Terms of Present Knowledge of Nerve Conduction,’ Physiological Reviews 2 (1922): 361–414. 67 Forbes, ‘The Interpretation of Spinal Reflexes,’ 412. 68 Charles Sherrington to Alexander Forbes, 15 December 1921, AFP 15, 733. 69 Alexander Forbes to Charles Sherrington, 2 February 1922, AFA 15, 734.

Notes to pages 279–85

477

70 Cross and Albury, ‘Walter B. Cannon’; see also Richard Gillespie, ‘Industrial Fatigue and the Discipline of Physiology,’ in Geison, ed., Physiology in the American Context. 71 Hallowell Davis, ‘Crossroads on the Pathways to Discovery,’ in Worden, Swazey, and Adelman, eds., The Neurosciences, 312. 72 This followed a considerable process of negotiation between Cannon, Forbes, and Davis. See Adrian to Forbes, 26 March 1922, AFP 1, 7. See also the following letters in Walter Bradford Cannon Papers (hereafter WBCA), Harvard Medical Library, 2, 20: Cannon to Davis, 12 February 1923; Davis to Cannon, 8 March 1923; Davis to Cannon, 25 April 1923; Cannon to Davis, 30 April 1923; Davis to Cannon, 14 May 1923 (telegram); Davis to Cannon, 16 May 1923. 73 See, for example, Alexander Forbes to Hallowell Davis, 8 April 1941, AFP 5, 247. 74 See Davis, ‘Crossroads,’ and Hallowell Davis, ‘Alfred Lee Loomis: American Discoverer of the EEG,’ typewritten manuscript (twenty-one pages) of a paper delivered at ‘Hans Berger Day,’ a conference held at the Medical College of Virginia, Virginia Commonwealth University, 21–22 May 1979, Hallowell Davis Papers (hereafter HDP), 21, 60, 2, Becker Medical Library, Washington University, St Louis. 75 Davis, ‘Crossroads,’ 312. 76 Photographic records are featured in Adrian and Matthews, ‘The Interpretation of Potential Waves in the Cortex,’ 440–71, while ink records predominate in idem, ‘The Berger Rhythm: Potential Changes from the Occipital Lobes in Man. 77 Davis, ‘Crossroads,’ 317. 78 On the EEG and epilepsy, see Kroker, ‘Washouts.’ 79 Hallowell Davis to Walter Cannon, 31 July 1934, WBCA, 8, 102. 80 Cannon to Davis, 27 September 1935, WBCA, 10, 123. 81 Davis to Cannon, 3 October 1935, WBCA, 10, 123. 82 Hallowell Davis to Alexander Forbes, 9 June 1934, AFP 5, 245. The letter is typed; the strikeout and emphasis were added by hand. 83 Davis to Forbes, 25 July 1934, AFP 5, 246. Emphasis in original. 84 Davis, ‘Crossroads,’ 317. 85 F.A. Gibbs, H. Davis, and W.G. Lennox, ‘The Electro-encephalogram in Epilepsy and in Conditions of Impaired Consciousness,’ Archives of Neurology and Psychiatry 34 (1935): 1133–48. 86 Kroker, ‘Washouts.’ 87 Walter Cannon to Hallowell Davis, 22 October 1934, WBCA, 8, 102. 88 EERA to Walter Cannon, 9 December 1934, WBCA, 8, 102. 89 See Kroker, ‘The Progress of Introspection,’ 96.

478

Notes to pages 286–95

90 Susan E. Lederer, ‘The Controversy over Animal Experimentation in America, 1880–1914,’ in Nicolaas A. Rupke, ed. Vivisection in Historical Perspective (London and New York: Routledge, 1990); and Donald A. Dewsbury, ‘Early Interactions between Animal Psychologists and Animal Activists and the Founding of the APA Committee on Precautions in Animal Experimentation,’ American Psychologist 45 (1990): 315–27. 91 William L. Laurence, ‘Electricity in the Brain Records. A Picture of Action of Thought,’ New York Times, 14 April 1935, 1 and 32. 92 Hallowell Davis to Walter Cannon, 18 April 1935, WBCA, 10, 123. 93 Kohler, Partners in Science, 201. 94 Kroker, ‘Washouts,’ 315–16. On Loomis, see Luis W. Alvarez, ‘Alfred Lee Loomis: November 4, 1887–August 11, 1975,’ Biographical Memoirs of the National Academy of Sciences 51 (1980): 309–41. On Loomis’s work in creating RadLab, see the interviews conducted with electrical engineers involved in the project at www.ieee.org/organizations/history_center/ oral_histories/oh_rad_lab_ menu.html, accessed 15 May 2005. 95 Hallowell Davis, ‘Alfred Lee Loomis.’ 96 Gibbs, Davis, and Lennox, ‘The Electroenepahalogram in Epilepsy.’ Cf. Alfred Lee Loomis, E. Newton Harvey, and Garret Hobart, ‘Potential Rhythms of the Cerebral Cortex during Sleep,’ Science 81 (1935): 597–8; idem, ‘Further Observations on the Potential Rhythms of the Cerebral Cortex during Sleep,’ Science 82 (1935): 198–200. 97 Robert E. Kohler, ‘Labscapes: Naturalizing the Lab,’ History of Science 40 (2002): 473–501. 98 Alfred Lee Loomis, E. Newton Harvey, and Garret Hobart, ‘Electrical Potentials of the Human Brain,’ Journal of Experimental Psychology 19 (1936): 249–79. 99 Ibid., 255. The reference to Ravel is taken from idem, ‘Cerebral States during Sleep, as Studied by Human Brain Potentials,’ Journal of Experimental Psychology 21 (1937): 127–44, 129. 100 Program, Conference on ‘The Electrical Potentials of the Brain,’ AFP 11, 527. The final papers of the day are described in the program as ‘Potential pattens from the brains of animals. (Speaker to be announced)’ and ‘Paper to be announced.’ 101 ‘List of guests,’ AFP 11, 527. 102 Davis, ‘Alfred Lee Loomis,’ 7. 103 Loomis, Harvey, and Hobart, ‘Electric Potentials,’ 250. 104 Ibid., 275–6. 105 Loomis, Harvey, and Hobart, ‘Cerebral States during Sleep,’ 128. 106 Ibid., 132.

Notes to pages 295–304

479

107 Ibid., 133–4. I have corrected the persistent error of spelling ‘rhythm’ as ‘rythm.’ 108 H. Blake and R.W. Gerard, ‘Brain Potentials during Sleep,’ American Journal of Physiology 119 (1937): 692–703. 109 H. Blake, R.W. Gerard, and N. Kleitman, ‘Factors Influencing Brain Potentials during Sleep,’ Journal of Neurophysiology 2 (1939): 48–60. 110 Ibid., 58. 111 Ralph Waldo Gerard, ‘Organic Freedom,’ in Ruth Nanda Anshen, ed., Freedom: Its Meaning (New York: Harcourt, Brace and Company, 1940). 112 Ibid., 416. 113 Ibid., 427. 114 Ludwig Kast, president of the Macy Foundation (1936), as quoted in Katherine Pandora, Rebels within the Ranks: Psychologists’ Critique of Scientific Authority and Democratic Realities in New Deal America (Cambridge: Cambridge University Press, 1997), 74. 115 Kast, as cited in ibid., 74–5. 116 Hallowell Davis to Dr Frank Fremont-Smith (cc’d to Walter Cannon), 17 January 1938, WBCA, 15, 179. 117 Hallowell Davis, P. Davis, A.L. Loomis, E.N. Harvey, and G. Hobart, ‘Human Brain Potentials during the Onset of Sleep,’ Journal of Neurophysiology 1 (1938): 24–38. See also Davis, ‘Alfred Lee Loomis,’ 8; and Hallowell Davis to Cannon, 8 September 1938, WBCA 15, 179. 118 See Davis to Cannon, 3 May 1935, WBCA, 10, 123. 119 Ibid. 120 Davis to Cannon, 21 December 1937, WBCA, 13, 158. Cannon’s original letter is not in the archive, and Davis never cites the comments directly in his reply. 121 Ibid. 122 Davis to Fremont-Smith, 17 January 1938. 123 Davis was one of the approximately thirty self-proclaimed ‘axonologists,’ a group that included almost all the leading neurophysiological researchers of the day. See Louise H. Marshall, ‘The Fecundity of Aggregates: The Axonologists at Washington University, 1922–1942,’ Perspectives in Biology and Medicine 26 (1983): 613–36. Gerard even identified Davis as one of the group’s leaders – a ‘super-axonologist.’ See the ‘axonologist’ series of letters of 22 October 1932, 10 March 1933, and 7 March 1935, AFP 19, 948, and 950. 124 Grob, Mental Illness and American Society, 279. 125 Pressman, ‘Human Understanding.’ 126 Davis, ‘Alfred Lee Loomis,’ 9.

480

Notes to pages 305–11

127 Blake, Gerard, and Kleitman, ‘Factors Influencing Brain Potentials,’ 58. 128 Kleitman, ‘Studies in the Physiology of Sleep: I.’ 129 Nathaniel Kleitman, ‘The Sleep-Wakefulness Cycle in Submarine Personnel,’ in Committee on Undersea Warfare, National Research Council, A Survey Report on Human Factors in Undersea Warfare (Baltimore: Waverly Press, 1949), 329–41. See also idem, ‘A Scientific Solution of the Multiple Shift Problem,’ 15–16, and Nathaniel Kleitman and D.P. Jackson, ‘Body Temperature and Performance under Different Routines,’ Journal of Applied Physiology 3 (1950): 309–28. 130 See David Meister, The History of Human Factors and Ergonomics (Mahwah, NJ: Lawrence Erlbaum Associates, 1999); and Kroker, ‘Washouts,’ 302–4. 131 Alberto Cambrosio and Peter Keating, ‘The Disciplinary Stake: The Case of Chronobiology,’ Social Studies of Science 13 (1983): 323–53. 132 Nathaniel Kleitman, ‘Biological Rhythms and Cycles,’ Physiological Reviews 29 (1949): 1–30. 133 Ibid., 1. 134 ‘Study of Sleeping Babies Offers Hope to Parents,’ New York Times, 7 August 1949, 45. 135 Arnold Gesell and Frances L. Ilg, Infant and Child in the Culture of Today: The Guidance of Development in Home and Nursery School (New York: Harper, 1943); Benjamin Spock, The Common Sense Book of Baby and Child Care (New York: Duell, Sloan and Pearce, 1946). On the changing culture of childrearing during this period, see William Graebner, ‘The Unstable World of Benjamin Spock: Social Engineering in a Democratic Culture, 1917–1950,’ Journal of American History 67 (1980): 612–29. 136 Nathaniel Kleitman, ‘Mental Hygiene of Sleep in Children,’ The Nervous Child: Quarterly Journal of Psycho-Pathology, Psychotherapy, Mental Hygiene, and Guidance of the Child 8 (1949): 63–6; idem and T.G. Engelmann, ‘Sleep Characteristics of Infants.’ 137 Lemaine et al., Stratégies et choix dans la recherche. 138 Eugene Aserinsky, ‘The Discovery of REM Sleep,’ Journal of the History of the Neurosciences 5 (1996): 213–27. 139 The article in question was, however, published under the title ‘Memories of Famous Neuropsychologists.’ Louise H. Marshall, personal interview, 26 July 2002. Marshall studied under Kleitman in the 1930s and had been one of the driving forces behind American neuroscience history since the early 1980s. 140 Aserinsky, ‘The Discovery of REM Sleep.’ 141 Ibid., 216. 142 Eugene Aserinsky, personal communication, 19 July 1995. Of course, it turned out that what Jacobson (and later, Aserinsky) were recording were

Notes to pages 313–17

143

144

145 146

147 148

149

150

151 152 153

154 155 156

157

481

not ‘eye muscle potentials’ (action currents) at all but the change in direction of the electrical field created by the small electrical potential between the cornea and the retina. My thanks to Jacobson’s laboratory assistant, Richard Lange, for pointing this out in an interview on 18 December 1998. On ‘boundary work,’ see Thomas Gieryn, Cultural Boundaries of Science: Credibility on the Line (Chicago: University of Chicago Press, 1999). The ability of the EEG to generate novelty was equally useful for the cultural field. See Borck, Hirnströme. J.-P. Dupuy, The Mechanization of the Mind: On the Origins of Cognitive Science (Princeton, NJ, and Oxford: Princeton University Press, 2000); see also Marshall and Magoun, Discoveries in the Human Brain. Lemaine et al., Stratégies et choix, 66–8. Frédéric Bremer, ‘Cerveau “isolé” et physiologique du sommeil,’ Compte rendus de la Société de Biologie (Paris) 118 (1935): 1235–41; and idem, ‘Nouvelles recherches sur le méchanisme du sommeil,’ Compte rendus de la Société de Biologie (Paris) 122 (1936): 460–67. Kleitman, Sleep and Wakefulness, 48. G. Moruzzi, and H.W. Magoun, ‘Brain Stem Reticular Formation and Activation of the EEG,’ Electroencephalography and Clinical Neurophysiology 1 (1949): 455–73. Stanley Cobb, ‘Human Nature and the Understanding of Disease,’ in N.W. Faxon, ed., The Hospital in Contemporary Life (Cambridge, MA: Harvard University Press, 1949). H.W. Magoun, ‘Evolutionary Concepts of Brain Function following Darwin and Spencer,’ in Sol Tax, ed., Evolution after Darwin: The University of Chicago Centennial, volume 2, The Evolution of Man: Man, Culture and Society (Chicago: University of Chicago Press, 1960). See, for example, Marshall and Magoun, Discoveries in the Human Brain. Aserinsky, ‘The Discovery of REM Sleep,’ 221. Calvin S. Hall, ‘What People Dream about,’ Scientific American 184 (1951): 60–3; idem, The Meaning of Dreams (New York: Harper, 1953); Grace A. Andrews and Mary Whiton Calkins, ‘Minor Studies from the Psychological Laboratory of Wellesley College: Studies of the Dream Consciousness,’ American Journal of Psychology 12 (1900): 131–4. Aserinsky, ‘The Discovery of REM Sleep,’ 219. Ibid., 222. Nathaniel Kleitman and Esther Kleitman, ‘Effect of Non-Twenty-Four-Hour Routines of Living on Oral Temperature and Heart Rate,’ Journal of Applied Physiology 6 (1953): 283–91. Eugene Aserinsky and Nathaniel Kleitman, ‘Eye Movements during Sleep,’

482

158

159

160 161 162

163

164

165 166

167

168

169

Notes to pages 317–23 Federation of American Societies for Experimental Biology. Federation Proceedings 12 (1953): 6. Eugene Aserinsky and Nathaniel Kleitman, ‘Regularly Occurring Periods of Eye Motility, and Concomitant Phenomena, during Sleep,’ Science 118 (1953): 273–4. The earliest mention I have found is in an article by the German psychiatrist Wilhelm Griesinger: ‘Physio-psychologische Selbstbeobachtungen,’ Archiv für Psychiatrie und Nervenkrankheiten 1 (1868): 200–4. Aserinsky, ‘The Discovery of REM Sleep,’ 223. ‘Notes on Science – Dreams,’ New York Times, 17 May 1953, E11. ‘His [Dement’s] first sleep paper (together with Kleitman) was basically a replicate of the work I had done on normal adults. He used Schizophrenic subjects and although he found nothing startlingly different from the normal adults, he must have satisfied Kleitman that my results were not a fluke.’ (Eugene Aserinsky to Roger Broughton, 7 May 1988, Association for the Psychophysiological Study of Sleep Records (hereafter APSSR), 4, 7. See also Aserinsky, ‘Discovery of REM Sleep,’ 223–4. William Dement, ‘Dream Recall and Eye Movements during Sleep in Schizophrenics and Normals,’ Journal of Nervous and Mental Disease 122 (1955): 263–9, 265. William Dement and Nathaniel Kleitman, ‘Cyclic Variations in EEG during Sleep,’ Electroencephalography and Clinical Neurophysiology 9 (1957): 673–90; William Dement and Edward A. Wolpert, ‘The Relation of Eye Movements, Body Motility, and External Stimuli to Dream Content,’ Journal of Experimental Psychology 55 (1958): 543–53. Norman Malcolm, Dreaming (London: Routledge and Kegan Paul, 1959). For a critique and analysis of Malcolm’s position, see Hilary Putnam, ‘Dreaming and Depth Grammar,’ in R. Butler, ed., Analytical Philosophy, Series 1 (Oxford: Blackwell, 1962); Hywel David Lewis, Dreaming and Experience (London: Athlone, 1968); Ian Hacking, Why Does Language Matter to Philosophy? (Cambridge: Cambridge University Press, 1975). See David Foulkes, ‘Dream Reports from Different Stages of Sleep,’ Journal of Abnormal and Social Psychology 65 (1962): 14–25; idem, The Psychology of Sleep (New York: Scribner’s, 1966); idem, ‘Dream Research, 1953–1993,’ Sleep 19 (1996): 609–24. Calvin S. Hall, Studies of Dreams Reported in the Laboratory and at Home (Felton, CA: Big Trees Press, 1962); idem, ‘Caveat Lector!’ Psychoanalytical Review 54 (1967): 655–61. Dement and Kleitman, ‘Cyclic Variations in EEG during Sleep,’ 346.

Notes to pages 323–33

483

170 William Dement, ‘The Effect of Dream Deprivation,’ Science 131 (1960): 1705–7. 171 Ibid., 1707. 7. Begin the Begin 1 Wilse B. Webb to David Foulkes, n.d., APSSR, 1, 3. The letter was attached to a letter from Foulkes to Webb dated 31 October 1990. 2 Aserinsky, ‘Discovery of REM Sleep,’ 224–5. 3 Canguilhem, The Normal and the Pathological, 207–9. 4 For examples, see Kay, The Molecular Vision of Life ; and Alberto Cambrosio and Peter Keating, Exquisite Specificity: The Monoclonal Antibody Revolution (New York: Oxford University Press, 1995). 5 Albert F. Ax, ‘Editorial,’ Psychophysiology 1 (1964): 1–3. See also idem, ‘Goals and Methods of Psychophysiology,’ Psychophysiology 1 (1964): 8–25; and John A. Stern, ‘Toward a Definition of Psychophysiology,’ Psychophysiology 1 (1964): 90–1. 6 Chester W. Darrow, ‘Psychophysiology, Yesterday, Today, and Tomorrow,’ Psychophysiology 1 (1964): 4–7, 6. 7 Howard Roffwarg, who attended the first meeting, called it a gathering of people ‘interested in studying sleep’ in the Associated Professional Sleep Societies Newsletter (30 March 1986), 3, APSSR 4, 4. Cf. the official agenda, APSSR 1, 5. William Dement also confirmed that dreaming, not sleep, was the primary interest of the people at this first meeting (personal interview, 6 September 2002). 8 Letter, Allan Rechtschaffen to David Foulkes, 23 February 1961, APSSR, 1, 5. 9 Circular letter, Rechtschaffen to APSS participants, 14 December 1961, APSSR, 1, 6. 10 Arthur Shapiro, ‘Welcoming Address,’ APSSR, 1, 7. 11 Dement, interview, 6 September 2002. See also William Dement and Christopher Vaughan, The Promise of Sleep (New York: Dell Books, 1999), 42. 12 Abstract, Charles Fisher and William Dement, ‘Observations on the DreamSleep Cycle during the Course of an Acute Paranoid Psychosis,’ APSSR, 1, 4. 13 Walter W. Hamburger, ‘Summary of Remarks Presented to Conference on Dream Research,’ APSSR, 1, 5. 14 Rechtshaffen, interview, 18 January 1999. 15 Hacking, Historical Ontology, 227–54. 16 Agenda, 1962 APSS meeting, APSSR, 1, 6.

484

Notes to pages 333–41

17 See 1962 APSS meeting, APSSR, 1, 6. 18 For an early example, see S. Woldring and M.N.J. Dirken, ‘Spontaneous Unit-Activity in the Superficial Cortical Layers,’ Acta physiologica et pharmacologica Neérlandica 1 (1950): 369–79. A typical example of using microelectrodes to study evoked potentials can be found in B. Renshaw, A. Forbes, and B.R. Morison, ‘Activity of Isocortex and Hippocampus: Electrical Studies with Microelectrodes,’ Journal of Neurophysiology 3 (1940): 74–105. 19 Michel Jouvet, interview, 13 May 2002. 20 Susan Y. Crawford, ‘Informal Communication among Scientists in Sleep and Dream Research,’ PhD thesis, University of Chicago, 1970. 21 Harry Marks, The Progress of Experiment: Science and Therapeutic Reform in the United States, 1900–1990 (Cambridge: Cambridge University Press, 1997). 22 On the differences between Jouvet and what he dubs the ‘Chicago School,’ see Michel Jouvet, The Paradox of Sleep: The Story of Dreaming, trans. Laurence Garey (Cambridge, MA: MIT Press, 1999), 1–25. See also Michel Jouvet, ‘Paradoxical Sleep: A Study of Its Nature and Its Mechanisms,’ Progress in Brain Research 18 (1965): 20–57. 23 L. Loewenfeld, ‘Über Narkolepsie,’ Münchener medizinische Wochenschrift 49 (1902): 1041–5. 24 Emil Redlich, ‘Zur Narkolepsiefrage,’ Monatschrift für Pszchiatrie und Neurologie 37 (1915): 85–94; and idem, ‘Epilegomena zur Narkolepsiefrage,’ Zeitschrift für das gesamte Neurologie und Pszchiatrie 136 (1931): 128–73. 25 W.J. Adie, ‘Idiopathic Narcolepsy: A Disease Sui Generis; with Remarks on the Mechanism of Sleep,’ Brain 49 (1926): 257–306. 26 Max Levin, ‘Narcolepsy and the Machine Age: The Recent Increase in the Incidence of Narcolepsy,’ Journal of Neurology and Psychopathology 15 (1934): 60–4. 27 Charles Loomis Dana, ‘On Morbid Drowsiness and Somnolence: A Contribution to the Pathology of Sleep,’ Journal of Nervous and Mental Disease 11 (1884): 153–76, 248–50; Carl D. Camp, ‘Morbid Sleepiness, with a Report of a Case of Narcolepsy and a Review of some Recent Theories of Sleep,’ Journal of Nervous and Mental Disease 2 (1907): 9–21. 28 These clinical accounts have been drawn from Luman E. Daniels, ‘Narcolepsy,’ Medicine (Baltimore) 13 (1934): 1–122. 29 K.K. Chen, ‘Two Pharmacological Traditions: Notes from Experience,’ Annual Review of Pharmacology and Toxicology 21 (1981): 1–6. Ephedrine is currently enjoying renewed interest as a traditional (and unregulated) herbal remedy and a controversial recreational psychoactive drug (‘herbal extasy’ and an important ingredient in methamphetamine). James H. Madison, Eli

Notes to pages 342–6

30

31 32 33

34 35

36 37

38 39

40 41 42

485

Lilly: A Life, 1885–1977 (Indianapolis: Indiana Historical Society, 1989), 65– 6. In the early 1930s, the University of Pennsylvania offered graduate medical degrees, and another such program was being developed through the joint effort of Columbia University and the New York Postgraduate School and Hospital. See Rosemary Stevens, American Medicine and the Public Interest (New Haven: Yale University Press, 1971), 198–217. H.A. Cave, ‘Narcolepsy,’ Archives of Neurology and Psychiatry 26 (1931): 50– 101; Daniels, ‘Narcolepsy.’ Daniels, ‘Narcolepsy,’ 105. Ibid., 27. See Curt Rosenthal, ‘Zur Pathogenese, Ätiologie und versorgungsrechlichen Bedeutung der “genuinen” und posttraumatischen echten Narkolepsie,’ Archiv für Psychiatrie und Nervenkrankheiten 96 (1932): 572–608; and Jean Lhermitte and Yves Dupont, ‘Sur la cataplexie et plus spécialement dur la cataplexie du reveil,’ Encéphale 23 (1928): 424–34. Daniels, ‘Narcolepsy,’ 109–10. For example, see J.B. Dynes and K.H. Finley, ‘The Electroencephalograph as an Aid in the Study of Narcolepsy,’ Archives of Neurology and Psychiatry 46 (1941): 598–612; see also D.D. Daly and R.E. Yoss, ‘Electroencephalogram in Narcolepsy,’ Electroencephalography and Clinical Neurophysiology 9 (1957): 109– 20; and H. Strauss, ‘Verbal Report, in the Proceedings of the American EEG Society June, 1956,’ Electroencephalography and Clinical Neurophysiology 9 (1957): 169. Kroker, ‘Washouts,’ 317–23. O.R. Langworthy and B.J. Betz, ‘Narcolepsy as a Type of Response to Emotional Conflicts,’ Psychosomatic Medicine 6 (1944): 121–38; A. Missriegler and B. Karpman, ‘On the Psychogenesis of Narcolepsy: Report of a Case Cured by Psychoanalysis,’ Journal of Nervous and Mental Disease 93 (1941): 141–62; L.A. Spiegel and D.P. Oberndorf, ‘Narcolepsy as a Psychogenic Symptom,’ Psychosomatic Medicine 8 (1946): 28–35. Gerald Vogel, ‘Studies in Psychophysiology of Dreams. III. The Dream of Narcolepsy,’ Archives of General Psychiatry 3 (1960): 421–8. A. Rechtschaffen, et al., ‘Nocturnal Sleep of Narcoleptics,’ Electroencephalography and Clinical Neurophysiology 15 (1963): 599–609; W. Dement, A. Rechtschaffen, and G. Gulevich, ‘The Nature of the Narcoleptic Sleep Attack,’ Neurology 16 (1966): 18–33. Marks, The Progress of Experiment. See Dement and Vaughan, The Promise of Sleep, 202. Dement, Rechtschaffen, and Gulevich, ‘The Nature of the Narcoleptic Sleep Attack.’

486

Notes to pages 348–56

43 A. Rechtschaffen and A. Kales, A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects (Bethesda, Md.: U.S. Dept. of Health, Education and Welfare, Public Health Service, 1968). 8. Insomnia Returns 1 On the fortunes of neurasthenia, see Gijswijt-Hofstra and Porter, eds., Cultures of Neurasthenia. 2 ‘Aktual’ here means ‘present-oriented,’ not ‘real’ or ‘authentic.’ Freud, Introductory Lectures, vol. 16, Lecture XXIV, ‘The Common Neurotic State,’ 385. 3 Robert Kingman, ‘The Insomniac,’ Medical Journal and Record (New York) 130 (19 June and 3 July 1929), 683–7 and 17–21, 18. 4 On addiction and drug control in the United States, see Caroline Jean Acker, Creating the American Junkie: Addiction Research in the Classic Era of Narcotic Control (Baltimore and London: Johns Hopkins University Press, 2002); David F. Musto, The American Disease: Origins of Narcotic Control, 3rd ed. (New York: Oxford University Press, 1999); David T. Courtwright, Dark Paradise: Opiate Addiction in America before 1940 (Cambridge, MA: Harvard University Press, 1982). For Britain, see Virginia Berridge, Opium and the People: Opiate Use and Drug Control Policy in Nineteenth and Early Twentieth Century England (London: Free Association Books, 1999). 5 Caroline Jean Acker, ‘From all Purpose Anodyne to Marker of Deviance: Physicians’ Attitudes towards Opiates in the U.S. from 1890 to 1940,’ in Roy Porter and Mikulas Teich, eds., Drugs and Narcotics in History (Cambridge: Cambridge University Press, 1995), 114–32. 6 On racial tension and addiction in the Canadian context, see Catharine Carstairs, ‘Deporting “Ah Sin” to Save the White Race: Moral Panic, Racialization, and the Extension of Canadian Drug Laws in the 1920s,’ Canadian Bulletin of Medical History 16 (1999): 65–88. 7 Healy, ‘Some Continuities and Discontinuities.’ 8 Peter Temin, Taking Your Medicine: Drug Regulation in the United States (Cambridge, MA: Harvard University Press, 1980); see also idem, ‘The Origins of Compulsory Drug Prescriptions,’ Journal of Law and Economics 22 (1979): 97–9. 9 Jacobson, You Can Sleep Well, 32. 10 Charles O. Jackson, Food and Drug Legislation in the New Deal (Princeton, NJ: Princeton University Press, 1970). For contrasting views on the 1938 act, see Temin, ‘The Origins of Compulsory Drug Prescriptions,’ and chapter 3 of Marks, The Progress of Experiment, and idem, ‘Revisiting “The Origins of Compulsory Drug Prescriptions,”’ American Journal of Public Health 85 (1995): 109–115.

Notes to pages 356–61

487

11 Marks, ‘Revisiting “The Origins,”’ 110. 12 Temin, Taking Your Medicine, 48–9. 13 Steven R. Belenko, Drugs and Drug Policy in America: A Documentary History (Westport, CT: Greenwood Press, 2000), 260–1. 14 Healy, ‘Some Continuities and Discontinuities.’ The following discussion of the early history of the tranquillizers draws upon David Healy, Creation of Psychopharmacology (Cambridge, MA: Harvard University Press, 2002). See also Susan L. Speaker, ‘From “Happiness Pills” to “National Nightmare”: Changing Cultural Assessment of Minor Tranquilizers in America, 1955– 1980,’ Journal of the History of Medicine and Allied Sciences 52 (1997): 338–76; and Mickey C. Smith, Small Comfort: A History of the Minor Tranquilizers (New York: Praeger Publishers, 1985). 15 See chapters three and four of David A. Rochefort, From Poorhouses to Homelessness: Policy Analysis and Mental Health Care, 2nd ed. (Westport, CT: Auburn House, 1997). 16 Daniel Burston, The Crucible of Experience: R.D. Laing and the Crisis of Psychotherapy (Cambridge, MA: Harvard University Press, 2000); Nick Crossley, ‘R.D. Laing and the British Anti-Psychiatry Movement: A Socio-Historical Analysis,’ Social Science and Medicine 47 (1998): 877–89; Norman Dain, ‘Critics and Dissenters: Reflections on “Anti-Psychiatry” in the United States,’ Journal of the History of the Behavioral Sciences 25 (1989): 3–25. 17 Ann Goldberg, ‘The Mellage Trial and the Politics of Insane Asylums in Wilhelmine Germany,’ Journal of Modern History 74 (2002): 1–32. 18 For examples of such advertising during the 1950s and 1960s, see Smith, Small Comfort; for examples from the 1970s, see U.S. Congress, Senate, Committee on Labor and Human Resources, Subcommittee on Health and Scientific Research, Use and Misuse of Benzodiazepines: Hearings before the Committee on Labor and Human Resources, Subcommittee on Health and Scientific Research, 96th Cong., 1st sess., 19 September 1979. 19 Ernest Hartmann, The Sleeping Pill (New Haven, CT, and London: Yale University Press, 1979), 20. 20 Smith, Small Comfort, 29–36; see also Speaker, ‘From “Happiness Pills” to “National Nightmare.”’ 21 U.S. Congress, Use and Misuse of Benzodiazepines, 10. These hearings were led by Senator Edward Kennedy and Senator Gaylord Nelson, both of whom had launched similar hearings regarding anti-competitive behaviour of drug companies during the late 1960s and throughout the 1970s. 22 Andrea Tone, ‘Listening to the Past: History, Psychiatry, and Anxiety,’ Canadian Journal of Psychiatry 50 (2005); 373–80. 23 Marks, ‘Revisiting “The Origins.”’

488

Notes to pages 361–3

24 Marks, The Progress of Experiment, 129; for a biographical account of Kefauver’s Congressional investigations of ‘administered pricing’ in the drug industry, which began in 1959 and culminated with the 1962 amendments, see Joseph Bruce Gorman, Kefauver: A Political Biography (New York: Oxford University Press, 1971), 303–13, 351–9. 25 Federal Register 35 (8 May 1970), 7250–3, as cited in Marks, The Progress of Experiment, 129. 26 Belenko, Drus and Drug Policy in America, 260–2. 27 World Health Organization, Expert Committee on Addiction-producing Drugs, Seventh Report, 1957 (www.undcp.org/bulletin/bulletin_1957-0101_1_page007.html) [accessed 4 December 2005]. 28 Speaker, ‘From “Happiness Pills” to “National Nightmare.”’ 29 ‘Sleep Disorder,’ British Medical Journal 2 (1968): 450. 30 E. Tylden and C. Saville, ‘Dangers of barbiturates,’ British Medical Journal 2 (1970): 49; ‘Sound Sleep,’ British Medical Journal 2 (1970): 492. 31 Crawford, ‘Informal Communication among Scientists in Sleep and Dream Research.’ Crawford claimed that there were a total of 618 people active in the field between January, 1964 and April, 1967. 32 I. Oswald and R.G. Priest, ‘Five Weeks to Escape the Sleeping Pill Habit,’ British Medical Journal 2 (1965): 1093–9. 33 I. Oswald, ‘Drugs and Sleep,’ Pharmacological Review 20 (1968): 272–303. 34 See, for example, F.R. Freemon, H.W. Agnew, and R.L. Williams, ‘An Electroencephalographic Study of the Effects of Meprobamate on Human Sleep,’ Clinical Pharmacology and Therapeutics 6 (1965): 172–6; R.L. Williams and H.W. Agnew, ‘The Effects of Drugs on the EEG Sleep Patterns of Normal Humans,’ Experimental Medicine and Surgery 27 (1969): 53–64; A. Kales and J.D. Kales, ‘Sleep Laboratory and Evaluation of Psychoactive Drugs,’ Pharmacology for Physicians 4 (1970): 1–6. 35 R. Held, B.-A. Schwartz, and H. Fischgold, ‘Fausse insomnie,’ Presse médicale 67 (1959): 141–3; B.-A. Schwartz, G. Guilbaud, and H. Fischgold, ‘Études électroencéphalographiques sur le sommeil de nuit. I. – L’“insomnie” chronique,’ Presse médicale 71 (1963): 1474–6. 36 L.J. Monroe, ‘Psychological and Physiological Differences between Good and Poor Sleepers,’ Journal of Abnormal Psychology 72 (1967): 255–64. 37 See A. Rechtschaffen, ‘Polygraphic Aspects of Insomnia,’ in H. Gastaut, ed., The Abnormalities of Sleep in Man (Bologna: Gaggi, 1969), 109–21; A. Rechtshaffen and L.J. Monroe, ‘Laboratory Studies of Insomnia,’ in A. Kales, ed., Sleep: Physiology and Pathology (Philadelphia and Toronto: J.B. Lippincott, 1969), 158–69. 38 Rechtschaffen and Monroe, ‘Laboratory Studies of Insomnia.’

Notes to pages 364–6

489

39 Anthony Kales, ‘Psychophysiological Studies of Insomnia,’ Annals of Internal Medicine 71 (1969): 624–9. 40 Anthony Kales, ‘Preface,’ in Kales, ed., Sleep, 9–10. UCLA would have been a particularly fertile ground for such laboratory-based psychiatric research. The UCLA Neuropsychiatric Institute had initially aspired to create a research laboratory during the late 1950s but had failed to maintain control over what ultimately emerged as the Brain Research Institute (led by Horace Magoun), which ultimately had only tenuous connections with psychiatric practice. See Norman Q. Brill, UCLA Neuropsychiatric Institute and Hospital: Oral History Transcript, 1988: Norman Q. Brill/Interviewed by Michael S. Balter (Los Angeles: Oral History Program, University of California, 1990). 41 For coverage in the popular press, see ‘Effectiveness of Mood Drugs Sold over Counters Is Studied,’ New York Times, 22 July 1971, 30; and ‘F.D.A. to Review All Drugs Sold over the Counter,’ New York Times, 5 January 1972, 1. 42 ‘U.S. Seizes Cases of Excederin P.M.: Asks Test of Tablet,’ New York Times, 23 January 1970, 49; and ‘U.S. Will Destroy $140,000 in Drugs,’ New York Times, 3 February 1970, 88. 43 Speaker, ‘From “Happiness Pills” to “National Nightmare,”’ 370. See also U.S. Congress, Senate, Subcommittee on Monopoly of the Select Committee on Small Business, Competitive Problems in the Drug Industry. Hearings before the Subcommittee on Monopoly of the Select Committee on Small Business, 90th Congress, 1st sess. to 96th Congress, 2nd sess., 1967–79. 44 See ‘Advertising: Grolier’s Marketing Officer,’ New York Times, 1 June 1970, 49; ‘Drug Spots on TV to be Toned down,’ New York Times, 18 September 1970, 87; ‘Ads Are Blamed for Pill Overuse,’ New York Times, 24 November 1970, 24; ‘Studies Link Drug Use to Parents’ Drug Habits,’ New York Times, 23 July 1971, 20; ‘Drug Ad Inquiry Hears 3 on F.C.C.,’ New York Times, 23 September 1971, 111; ‘“Mood” Pills: People, It Seems, Will Swallow Anything,’ New York Times, 26 September 1971, vi, 10. 45 Speaker, ‘From “Happiness Pills” to “National Nightmare.”’ See also idem, ‘Too Many Pills: Patients, Physicians, and the Myth of Overmedication in America, 1955–1980,’ PhD thesis, University of Pennsylvania, 1992. 46 Healy, The Creation of Psychopharmacology, 129–77. 47 I. Karacan and R.L. Williams, ‘Insomnia: Old Wine in a New Bottle,’ Psychiatric Quarterly 45 (1971): 274–88; see also I. Karacan, et al., ‘New Approaches to the Evaluation and Treatment of Insomnia (preliminary results),’ Psychosomatics 12 (1971): 81–8. Post-REM psychoanalytic studies of sleep disorders had changed little from their earlier incarnations – see ‘Symposium on Neurotic Disturbances of Sleep,’ International Journal of Psycho-Analysis 23 (1942): 49–68.

490

Notes to pages 366–72

48 Karacan and Williams, ‘Insomnia,’ 284; see also Kales and Kales, ‘Sleep Laboratory Evaluation of Psychoactive Drugs.’ 49 Karacan and Williams, ‘Insomnia,’ 285. 50 Quentin R. Regestein and Peter Reich, ‘A Sleep Clinic within a General Hospital Psychiatry Service,’ General Hospital Psychiatry 2 (1980): 112–17. 51 ‘Interview of Doctor Nathaniel Kleitman by Louise Marshall,’ draft transcript, 2–5, Neuroscience History Archives (hereafter NHA). 52 Webb Haymaker, The Founders of Neurology: One Hundred and Thirty-Three Biographical Sketches Prepared for the Fourth International Neurological Congress in Paris (Springfield, Ill.: C.C. Thomas, 1953); the second edition appeared as idem. and Francis Schiller, The Founders of Neurology: One Hundred and FortySix Biographical Sketches by Eighty-Eight Authors (Springfield, Ill.: Thomas, 1970). 53 See the three letters (7 June, 15 June and 7 July 1978) between Haymaker and Magoun in Horace Winchell Magoun Papers (hereafter HWMP), NHA, 8, 16. 54 E.D. Adrian and J.F. Delafresnaye, eds., Brain Mechanisms and Consciousness: A Symposium (Oxford: Blackwell Scientific Publishers, 1954). 55 H.H. Jasper et al., eds., Reticular Formation of the Brain (Boston: Little, Brown, 1958); H.H. Jasper and G.D. Smirnov, eds., ‘The Moscow Colloquium on Electroencephalography of Higher Nervous Activity,’ Electroencephalography and Clinical Neurophysiology, supp. 13 (1960): 1–420. 56 Anne Marie Moulin, ‘The Immune System: A Key Concept for the History of Immunology,’ History and Philosophy of the Life Sciences 11 (1989): 221–36. 57 Letter, Francis Walshe to Horace Magoun, 24 February 1955, HWMP, NHA, 10, 1; see also the letters dated 1 and 15 March 1955 in the same folder. 58 John D. French, Donald B. Lindsley, and H.W. Magoun, An American Contribution to Neuroscience: The Brain Research Institute, UCLA, 1959–1984 (Los Angeles: Brain Research Institute, University of California, 1984). This celebratory book is based exclusively on personal anecdotes – a thoroughgoing historical analysis of the BRI remains to be written. 59 See the eleven-page document entitled ‘BRI Highlights, 1966–1972’ in the folder entitled ‘BRI – Steering Committee for Reorganization, 1971–1976,’ UCLA-BRI, box 10. In the same record series, see also box 47, folder ‘Funding, Institute. 1968–1975,’ in ‘Subject Files, 1944–.’ 60 From the early 1950s until 1961, the BRI operated out of laboratories at federally run Veterans Affairs Hospitals at Long Beach, Brentwood, and Wadsworth, California. The clinicians’ perspective on the battle between the NI and the BRI over the new facility can be found in the ‘Neuropsychiatric Institute and Hospital’ oral histories held in Special Collections at the Charles E.

Notes to pages 372–6

61

62

63 64

65 66 67 68 69

70 71

72

73

491

Young Research Library at UCLA. In particular, see the interviews with Norman Q. Brill, George Tarjan, and Augustus S. Rose. French et al., An American Contribution to Neuroscience, 109–20. See also the BRI Annual Reports in box 1 of University of California at Los Angeles Archives, RG Brain Research Institute, Series 195 – Reports and Publications, 1962–86. See the excerpts from the interview between Francis Brewer and Paul Crandall in ‘Annotated Excerpts from Oral Histories: Program Research and Early BRI,’ HWMP, BRI, 20, 7. French et al., An American Contribution to Neuroscience, 232–3. A copy of Kales’s MSc thesis can be found in the ‘Subject file’ folder entitled ‘Sleep and Wakefulness’ of HWMP, BRI. See also A. Kales, et al., ‘Dream Deprivation: An Experimental Reappraisal,’ Nature 204 (1964): 1337–8. See, for example, Lawrence S. Kubie, ‘The Concept of Dream Deprivation: A Critical Analysis,’ Psychosomatic Medicine 24 (1962): 62–5. The earliest complete statement of this position is David Foulkes, ‘Nonrapid Eye Movement Mentation,’ Experimental Neurology suppl. 4 (1967): 28–38. Hall, Studies of Dreams. Allan Rechtshaffen, interview, 18 January 1999. A. Kales and A. Jacobson, ‘Mental Activity during Sleep: Recall Studies, Somnambulism and Effects of Rapid Eye Movement Deprivation and Drugs,’ 81– 91, in C.D. Clemente, ed., ‘Physiological Correlates of Dreaming,’ Experimental Neurology supp. 4 (1967). The proceedings were published as Kales, ed. Sleep: Physiology and Pathology. See box 47, folder ‘Funding, Institute. 1968–1975,’ RG BRI, Director’s Office, record series no. 371, Administrative Files, 1944–1984, Subject Files, 1944-. Anthony Kales, Edward O. Bixler, and Joyce D. Kales, ‘Role of the Sleep Research and Treatment Facility: Diagnosis, Treatment and Education,’ Advances in Sleep Research 1 (1974): 391–415. For the Kales’s work on the efficacy of sedatives, see, for example, A. Kales and J.D. Kales, ‘Sleep Laboratory Evaluation of Psychoactive Drugs,’ Pharmacology for Physicians 4 (1970): 1–6; A. Kales et al., ‘Hypnotics and Altered Sleep and Dream Patterns: II. AllNight EEG Studies of Chloral Hydrate, Flurazepam, and Methaqualone,’ Archives of General Psychiatry 23 (1970): 219–25; J. Kales et al., ‘Are Over-theCounter Sleep Medications Effective? All-Night EEG Studies,’ Current Therapeutic Research, Clinical and Experimental 13 (1971): 143–151; and A. Kales and J.D. Kales, ‘Evaluation, Diagnosis and Treatment of Clinical Conditions Related to Sleep,’ JAMA 123 (1971): 229–35. Kales, Bixler, and Kales, ‘Role of the Sleep Research Treatment Facility,’ 397.

492

Notes to pages 376–81

74 Rechtschaffen and Kales, eds., ‘A Manual of Standardized Terminology.’ 75 Anthony Kales and Joyce D. Kales, ‘Shortcomings in the Evaluation and Promotion of Hypnotic Drugs. Editorial,’ New England Journal of Medicine 293 (1975): 826–7. 76 Regarding the lack of correlation between REM deprivation and the onset of psychosis, see G.W. Vogel, ‘REM Deprivation. III. Dreaming and Psychoses,’ Archives of General Psychiatry 18 (1968): 312–29. 77 Louis Lasagna, ‘Congress, the FDA, and New Drug Development: Before and after 1962,’ Perspectives in Biology and Medicine 32 (1989): 322–43. 78 Institute of Medicine, Sleeping Pills, Insomnia, and Medical Practice (Washington, DC: National Academy of Sciences, 1979). 79 For a sampling of how the lay press moulded the story of sedative use in the 1970s around the growing criticism of tranquilizers, see the following articles, all taken from the New York Times: ‘U.S. Seizes Cases of Excedrin P.M.; Asks Test of Tablet’ (23 January 1970, 49); ‘U.S. Will Destroy $140,000 in Drugs’ (3 February 1970, 88); ‘Drug Spots on TV to be Toned down’ (18 September 1970, 87); ‘Ads Are Blamed for Pill Overuse’ (24 November 1970, 24); ‘Effectiveness of Mood Drugs Sold over Counters Is Studied (22 July 1970, 30); ‘Studies Link Drug Use to Parents’ Drug Habits’ (23 July 1970, 20); ‘“Mood” Pills: People, It Seems, Will Swallow Anything’ (26 September 1971, iv, 10); ‘F.D.A. to Review All Drugs Sold over the Counter’ (5 January 1972, 1); ‘F.D.A. Official Says Cautious Policy on Licensing New Drugs Has Prevented Medical Disasters in U.S.’ (6 February 1973, 16); ‘Senators Hear Sleep-Aid Drugs Are Ineffective, Possibly a Peril’ (30 October 1975, 1); ‘F.T.C. Plans Curbs on Sleeping Aid Ads’ (31 October 1975, 36); ‘Daytime Sedatives and Aids to Sleep Are Held Ineffective’ (5 December 1975, 1); ‘Surgeon Acquitted of Dispensing Drugs Against Health Law’ (12 May 1976, 81); ‘F.D.A. Cracking Down on Some Sedatives’ (22 June 1977, 24); ‘Federal Study Links Sleeping Pills to 5,000 Deaths Yearly in U.S.’ (28 November 1977, 18); ‘F.D.A. Plans Curb on Some Sedatives Sold over Counter’ (13 June 1978, 1). 80 Institute of Medicine, Sleeping Pills, 65. According to data taken from the IMS National Prescription Audit, barbiturate prescriptions and refills fell from a peak of about twenty million in 1968 to under five million in 1977. Prescriptions for flurazepam, on the other hand, had risen steadily since its introduction in 1970, hitting about thirteen million in 1977 (ibid., 51). 81 Institute of Medicine, Sleeping Pills, 28. 82 Hartmann, The Sleeping Pill. For an analysis that later became a touchstone for a generation of American medical historians, see Charles E. Rosenberg, ‘The Therapeutic Revolution: Medicine, Meaning, and Social Change

Notes to pages 381–7

83 84 85 86 87 88 89 90

91 92 93 94 95 96

97 98 99 100 101

493

in 19th-century America,’ Perspectives in Biology and Medicine 20 (1977): 485–506. Hartmann, The Sleeping Pill, 2 and 132–42. See, for example, Reiser, Medicine and the Reign of Technology. Institute of Medicine, Sleeping Pills, 148. ‘U.S. Study of Sleep Drugs Finds Risks and Overuse,’ New York Times, 5 April 1979, 1. John Elliott, ‘Medical News: Physician Prescribing Practices Criticized; Solutions in Question,’ JAMA 241 (1979): 2353–4, 2359–60. R. Jeffrey Smith, ‘Study Finds Sleeping Pills Overprescribed,’ Science 204 (1979): 287–8. Institute of Medicine, Sleeping Pills, 146. See the exchange of letters between William G. Anlyan, Christian Gillin, and Frederic Solomon (representing the Institute of Medicine) and Bruce H. Medd (from Roche Laboratories), as well as the response by Smith, the author of the original Science article (Science 205 [1979]: 6–8). Institute of Medicine, Sleeping Pills, 149. Guiseppe Moruzzi, ‘Sleep and Instinctive Behavior,’ Archives italiennes de biologie 107 (1969): 175–216. Editorial, Sleep 1 (1978): 115–6. Regestein and Reich, ‘Current Problems.’ Institute of Medicine, Sleeping Pills, 18. On the dopamine receptor theory of schizophrenia as a forerunner of the current technique of using radio-labelling and receptor binding assay systems to develop psychoactive drugs, create diagnoses, and treat mental disorders and thereby create ‘a common language for psychiatry and the pharmaceutical industry’, see Healy, The Creation of Psychopharmacology, 207– 19. I borrow the term ‘platform,’ however, from Alberto Cambrosio and Peter Keating’s study of the transformations of postwar immunology: Biomedical Platforms: Realigning the Normal and the Pathological in Late-TwentiethCentury Medicine (Cambridge, MA: MIT Press, 2003). Daniel F. Kripke, letter, Science 205 (1979): 8. Broughton, personal interview, 5 February 1999. See also Roger Broughton, ‘In Memorium: Henri Gastaut (1915–1995),’ Sleep 18 (1995): 806–7. Roger Broughton, ‘Sleep Disorders: Disorders of Arousal?’ Science 159 (1968): 1070–8. Roger Broughton, ‘A Proposed Classification of Sleep Disorders,’ Sleep Research 1 (1972): 146. Some notable examples include Howell, Technology in the Hospital; Keating and Cambrosio, Biomedical Platforms; and Monica J. Casper and Adele E.

494

102 103 104 105 106 107 108

109 110

111 112

Notes to pages 388–91 Clarke, ‘Making the Pap Smear into the “Right Tool” for the Job: Cervical Cancer Screening in the USA, circa 1940–95,’ Social Studies of Science 28 (1998): 255–90. Diagnostic Classification of Sleep and Arousal Disorders (Rochester, MN: Association of Professional Sleep Societies, 1979), rpt. 1987, 7 (hereafter DCSAD). DCSAD, 11. Ibid., 25. Ibid., 24. Emphasis in original. Ibid., 26. Daniel F. Kripke and Nancy Fisher, ‘Acceptance of the Clinical Sleep Laboratory,’ Sleep Research 4 (1975): 229. All figures in USD. The figure cited here was calculated using the average U.S. Consumer Price Index (charting the changes in prices of goods and services purchased for consumption by urban households) for July 2006. See http://www.bls.gov/cpi/home.htm. Accessed 29 August 2006. ‘An Examination of the National Ambulatory Medical Care Survey,’ Sleep Research 11 (1982): 138. A. Kales et al., ‘Incidence of Insomnia in the Los Angeles Metropolitan Area,’ Sleep Research 3 (1974): 139; C.R. Soldatos, A. Kales, and J.D. Kales, ‘Management of Insomnia,’ Annual Review of Medicine 30 (1979): 301–12. Registein and Reich, ‘A Sleep clinic within a General Hospital Psychiatry Service.’ On the debate over PSG, see Deborah Sewitch, ‘EEG in Insomnia,’ American Journal of Psychiatry 141 (1984): 804; William C. Orr, ‘Utilization of Polysomnography in the Assesment of Sleep Disorders,’ Medical Clinics of North America 69 (1985): 1153–67; Elizabeth A. Jacobs et al., ‘The Role of Polysomnography in the Differential Diagnosis of Chronic Insomnia,’ American Journal of Psychiatry 145 (1988): 346–9; and J.D. Edinger et al., ‘Polysomnographic Assessment of DIMS: Empirical Evaluation of its Diagnostic Value,’ Sleep 12 (1989): 315–22. Critical response to the latter two articles, both of which endorsed the use of PSG in the diagnosis of insomnia, was lively. See Quentin Registein’s critique of Jacobs et al., and the authors’ response in American Journal of Psychiatry 145 (1988): 1483–4; see also exchange between Edinger and Hoelscher’s critics and their response in Sleep 13 (1990): 188– 92. See also Peter J. Hauri, ‘A Cluster Analysis of Insomnia,’ Sleep 6 (1983): 326–28; and R. Matthys, J. de Roeck, and R. Cluydts, ‘Insomnnia: From a Single Entity to a Complex Concept,’ Acta psychiatrica belgica 85 (1985): 390–405. More recent examples include R.R Rosa and M.H. Bonnet, ‘Reported Chronic Insomnia Is Independent of Poor Sleep as Measured by Electroencephalography,’ Psychosomatic Medcine 62 (2000): 474–82; and

Notes to pages 392–8

113 114

115 116

117 118

495

Charles M. Morin, ‘The Nature of Insomnia and the Need to Refine our Diagnostic Criteria,’ Psychosomatic Medicine 62 (2000): 483–5. B. Liljenberg et al., ‘The Prevalence of Insomnia: The Importance of Operationally Defined Criteria,’ Annals of Clinical Research 20 (1988): 393–8. Diagnostic Steering Committee of the American Sleep Disorders Association, The International Classification of Sleep Disorders Diagnostic and Coding Manual (Rochester, MN: ASCA, 1990), 3 (hereafter ICSD). Ibid., 3. On iatrogenic addiction in psychiatry, see Healy, The Creation of Psychopharmacology. On the decline of basic and psychological research in sleep, see Tore A. Nielsen, ‘One Century of Nocturnal Science: Are We Neglecting the Scientific Study of Mental Processes in Sleep?’ Sleep Research 17 (1988): 1; and J. Catesby Ware, ‘Apparent Decline in Basic Sleep Research,’ Sleep Research 17 (1988): 2. For an exception to this trend, see Evelyn J. Marchini et al., ‘What Do Insomniacs Do, Think, and Feel during the Day? A Preliminary Study,’ Sleep 6 (1983): 147–55. Mark R. Pressman, ‘Whatever Happened to Insomnia (and Insomnia Research)?’ American Journal of Psychiatry 148 (1991): 419–20. Charles F. Reynolds III et al., ‘Subtyping DSM-III-R Primary Insomnia: A Literature Review by the DSM-IV Work Group on Sleep Disorders,’ American Journal of Psychiatry 148 (1991): 432–8.

9. Breathe 1 Aspects of the two disorders have recently begun to merge, as insomnia has come to be regarded as a symptom of sleep apnea. See Nancy A. Collop, ‘Can’t Sleep? You May Have Sleep Apnea,’ Chest 120 (2001): 1768–9. 2 Eliot A. Phillipson, ‘Sleep Apnea – A Major Public Health Problem,’ New England Journal of Medicine 328 (1993): 1271–3. 3 Editorial, Sleep 1 (1978): 115. 4 William C. Dement, ‘Foreword’ to Hypersomnia with Periodic Apneas, Advances in Sleep Research, vol. 4, by Elio Lugaresi, Giorgio Coccangna, and Magda Mantovani (New York and London: SP Medical and Scientific Books, 1978). 5 See ‘Symposium on Neurotic Disturbances of Sleep,’ International Journal of Psycho-Analysis 23 (1942): 49–68; G. Moruzzi, ‘Reticular Influences on the EEG,’ Electroencephalography and Clinical Neurophysiology 16 (1964): 2–17; and idem, ‘The Historical Development of the Deafferentiation Hypothesis of Sleep,’ Proceedings of the American Philosophical Society 108 (1964): 19–28. 6 This definition of narcolepsy had been accepted at the First International

496

7

8

9

10

11 12 13

14

Notes to pages 398–400

Symposium on Narcolepsy, held at La Grande Motte in southern France. See Christian Guilleminault, William C. Dement, and Pierre Passouant, eds., Narcolepsy, Advances in Sleep Research, vol. 3 (New York: SP Medical and Scientific Books, 1976). Christian Guilleminault, Frederic L. Eldridge, and William C. Dement, ‘Insomnia with Sleep Apnea: A New Syndrome,’ Science 181 (1973): 856–8.; see also idem, ‘Insomnia, Narcolepsy, and Sleep Apneas,’ Bulletin de physiopathologie repiratoire (Nancy) 8 (1972): 1127–38. I arrived at these figures (used in creating fig. 34) by dividing the number of papers for each category by the total number of papers listed as ‘Current Claims’ regarding ‘Pathology and Disorders Associated with Sleep’ in each year of Sleep Research, and then calculating the mean of these annual percentages over a five-year period. Total percentages do not equal 100, because other sleep disorders, such as sleepwalking, were not included. Roger Broughton and Frederick Snyder, ‘Sleep and Clinical Pathological States,’ in Michael H. Chase, ed., The Sleeping Brain, Perspectives in the Brain Sciences, vol. 1 (Los Angeles: Brain Information Service, 1972), 363–93. For Bremer’s statement about sleep research’s status as a discipline, see his introductory remarks to that same volume. C. Guilleminault and W.C. Dement, ‘Pathologies of Excessive Sleep,’ in Elliot D. Weitzman, ed., Advances in Sleep Research, vol. 1 (New York: SP Medical and Scientific Books, 1974), 345–90. W.C. Dement, ‘Daytime Sleepiness and Sleep “Attacks.”’ in Guilleminault, Dement, and Passouant, eds,, Narcolepsy, 17–42. G. Gunella, ‘Évolution des conceptions pathogeniques sur le syndrome de Pickwick,’ Bulletin de physio-pathologie respiratoire (Nancy) 8 (1972): 981–1003. R.M. Coleman et al., ‘Sleep-wake Disorders Based on a Polysomnographic Diagnosis: A National Cooperative Study,’ Journal of the American Medical Association 247 (1982): 997–1003. Terry Young et al., ‘The Occurrence of Sleep-Disordered Breathing among Middle-Aged Adults,’ New England Journal of Medicine 328 (1993): 1230–5. The Wisconsin study was clearly modelled upon the prospective studies that, since the 1950s, have provided an important epistemological basis to the ‘risk factor’ approach to medical practice. Although several excellent studies of the evolution and consequences of the risk-factor approach have recently appeared, none of them mentions sleep. See Marks, The Progress of Experiment; Robert A. Aronowitz, Making Sense of Illness: Science, Society, and Disease (Cambridge: Cambridge University Press, 1998); and William G. Rothstein, Public Health and the Risk Factor: A History of an Uneven Medical Revolution (Rochester, NY: University of Rochester Press, 2003).

Notes to pages 401–4

497

15 Karl Miller, Doubles: Studies in Literary History (Oxford: Oxford University Press, 1985), 75. James Hogg was the author of Confessions of a Justified Sinner (1824), the story of a man whose aspirations to sainthood are confounded by the actions of a diabolical double. Hogg’s own doubles, one of which seems to have been Macnish, appeared with some regularity in Blackwood’s Magazine. See Miller, Doubles, 16–20. 16 See, for example, Meir H. Kryger, ‘Sleep Apnea: From the Needles of Dionysius to Continuous Positive Airway Pressure,’ Archives of Internal Medicine 143 (1983): 2301–3; and Lavie, ‘Nothing New under the Moon.’ 17 See Charles Dickens, The Posthumous Papers of the Pickwick Club, vol. 2 (London: Chapman and Hall, 1909), ch. 26. 18 Peter Denman, ‘Krook’s Death and Dickens’s Authorities,’ Dickensian 82 (1986): 130–41. 19 Robert Macnish, The Philosophy of Sleep, 2nd ed. (New York: D. Appleton, 1834), 184. 20 Caton, ‘Case of Narcolepsy.’ 21 Minutes from the Clinical Society of London, British Medical Journal 1 (1889): 358 (as cited in Lavie, ‘Nothing New under the Moon’). 22 William Osler, The Principles and Practice of Medicine, 7th ed. (New York and London: D. Appleton, 1909), 439. 23 Ibid., 451. 24 William Osler, The Principles and Practice of Medicine, 4th ed. (New York and London: D. Appleton, 1902), 995–7. 25 W.H. Broadbent, ‘Cheyne-Stokes Respiration in Cerebral Haemorrhage,’ Lancet 1 (1877): 307–9. 26 Weir-Mitchell, ‘Some Disorders of Sleep.’ 27 See, for example, Dana, ‘On Morbid Drowsiness and Somnolence.’ 28 Examples include M.H. Weil, ‘Polycythemia Associated with Obesity,’ JAMA 159 (1955): 1592; H.O. Sieker, E.H. Estes, Jr et al., ‘A Cardiopulmonary Syndrome Associated with Extreme Obesity,’ Journal of Clinical Investigation 34 (1955): 916; J. Howland Auchincloss, Jr, Ellen Cook, and Attilio D. Renzetti, ‘Clinical and Physiological Aspects of a Case of Obesity, Polycythemia and Alveolar Hypoventilation,’ Journal of Clinical Investigation 34 (1955): 1537–45; C. Sidney Burwell et al., ‘Extreme Obesity Associated with Alveolar Hypoventilation – A Pickwickian Syndrome,’ American Journal of Medicine 21 (1956): 811–18. 29 Daniel M. Fox, ‘The Politics of the NIH Extramural Program, 1937–1950,’ Journal of the History of Medicine and Allied Sciences 42 (1987): 447–66; and Victoria A. Harden, Inventing the NIH: Federal Biomedical Research Policy, 1887– 1937 (Baltimore: Johns Hopkins University Press, 1986).

498

Notes to pages 405–11

30 The U.S. Navy sponsored such work in order to determine the minimum amount of oxygen that needed to be carried in high-altitude flight (Louise Marshall, personal interview, 26 July 2002.). See also Seymour L. Chapin, ‘An Active Interface between Medical Science and Aeronautical Technology: The Physiological Investigations for the XC–35,’ History and Philosophy of the Life Sciences 13 (1991): 235–48; and Kroker, ‘Washouts,’ 307–11. 31 C. Sidney Burwell et al., ‘Extreme Obesity.’ This particular work, it should be noted, originated not in the author’s interest in obesity and respiration but in his study of heart disease in pregnancy. See C. Sidney Burwell and James Metcalfe, Heart Disease and Pregnancy: Physiology and Management (Boston: Little, Brown, 1958). 32 ‘Preface’ to K. Akert, C. Bally, and J.P. Schadé, eds., Sleep Mechanisms, Progress in Brain Research, vol. 18 (Amsterdam: Elsevier, 1965), 2. See also G. Moruzzi, ‘Summary Statement,’ in ibid., 241–3, and F. Hoff, ‘Summary and Conclusions from the Internal Medical Aspect,’ in ibid., 244–8. This conference, like so many others in sleep research, was sponsored by the pharmaceutical firm Hoffmann-LaRoche. 33 Richard Jung and Wolfgang Kuhlo, ‘Neurophysiological Studies of Abnormal Night Sleep and the Pickwickian Syndrome,’ in Akert et al., Sleep Mechanisms, 140–59. 34 See R. Bulow and D.H. Ingvar, ‘Respiration and State of Wakefulness in Normals, Studied by Spirography, Capnography and E.E.G.,’ Acta physiologica scandinavica 51 (1961): 230–8. 35 Roger Broughton, personal interview, 5 February 1999. 36 H. Gastaut, C.A. Tassinari, and B. Duron, ‘Étude polygraphique des manifestations épisodiques (hypniques et respiratoires), diurnes et nocturnes, du syndrome de Pickwick,’ Revue neurologique 112 (1965): 568–79. See also idem, ‘Polygraphic Study of the Episodic Diurnal and Nocturnal (Hypnic and Respiratory) Manifestations of the Pickwick Syndrome,’ Brain Research 2 (1966): 167–86. 37 Broughton and Snyder, ‘Sleep and Clinical Pathological States.’ 38 Ibid., 408. 39 Discussion following Ibid., 430. 40 For an example of the neurologists’ opposition to the respirologists’ position, compare Alfred P. Fishman, ‘The Syndrome of Chronic Alveolar Hypoventilation,’ Bulletin de physio-pathologie respiratoire (Nancy) 8 (1972): 971–80, an example of the latter, with C.A. Tassinari et al., ‘Apnoeic Periods and the Respiratory Related Arousal Patterns during Sleep in the Pickwickian Syndrome,’ Bulletin de physio-pathologie respiratoire (Nancy) 8 (1972): 1087– 1102. Some respiratory physiologists had already come around to the neurol-

Notes to pages 411–17

41 42 43

44 45

46 47 48

49 50

51 52 53

54 55

499

ogists’ point of view, however. See also Gunella, ‘Évolution des conceptions pathogéniques.’ Douglas Carroll, ‘Nosology of “Pickwick Syndrome,”’ Bulletin de physio-pathologie respiratoire (Nancy) 8 (1972): 1241–7. Gunella, ‘Évolution des conceptions pathogeniques,’ 985. Dement and Vaughan, Promise of Sleep, 47. Early papers include Guilleminault et al., ‘Insomnia, Narcolepsy, and Sleep Apneas;’ and Guilleminault and Dement, ‘Pathologies of Excessive Sleep.’ E. Lugaresi et al., ‘Effets de la trachéotomie dans les hypersomnies avec respiration periodique,’ Revue neurologique 123 (1970): 267–8. See, for example, Elio Lugaresi, Giogio Coccagna, and Magda Mantovani, Hypersomnia with Periodic Apneas (New York and London: SP Medical and Scientific Books, 1978). Eliot Phillipson, personal interview, 30 April 2002. C.E. Sullivan et al., ‘Reversal of Obstructive Sleep Apnoea by Continuous Positive Airway Pressure Applied through the Nares,’ Lancet 1 (1981): 862–5. The following figures are taken from Ware, ‘Apparent Decline.’ Ware took five key words (‘narcolepsy,’ ‘insomnia,’ ‘dreams,’ ‘cats,’ and ‘apnea’) from paper titles indexed in Sleep Research from 1972 to 1987, and compared their relative frequency. Since cats were by far the most commonly used experimental animal in sleep research during the period, the frequency of this term is a fairly reliable indicator of the fortunes of basic research in the field. Regestein and Reich, ‘A Sleep Clinic within a General Hospital Psychiatry Service.’ U.S. Department of Health and Human Services, Agency for Health Care Policy and Research, Polysomnography and Sleep Disorder Centers, Health Technology Assessment Reports, no. 4 (Washington, DC: Government Printing Office, 1991). As of 1999, the number of accredited centres had risen to 370. See Dement and Vaughan, Promise of Sleep. Colin M. Shapiro et al., ‘Where Are the British Sleep Apnea Patients?’ Lancet 2 (1981): 523. Rosemary Carter et al., ‘Sleep Apnoea,’ Lancet 2 (1981): 873. Aside from Kryger, ‘Sleep Apnea,’ and Lavie, ‘Nothing New under the Moon,’ see also Christian Guilleminault, ‘Obstructive Sleep Apnea: The Clinical Syndrome and Historical Perspective,’ Medical Clinics of North America 69 (1985): 1187–1203; and J.E. Cosnett, ‘Charles Dickens: Observer of Sleep and Its Disorders,’ Sleep 15 (1992): 264–7. U.S. Department of Health and Human Services, Polysomnography and Sleep Disorder Centers, 4 and 15. Ibid., passim.

500

Notes to pages 418–24

56 ‘Sleepwalker Freed in Slaying, Acquitted in Attempted Murder,’ Toronto Star, 15 July 1988, A12; ‘Medical Science Probing Mysteries of Sleep,’ Globe and Mail, 23 September 1989, A1. 57 R. Broughton, R. Billings, et al., ‘Homicidal Somnambulism: A Case Report,’ Sleep 17 (1994): 253–64. The 1999 case of Scott Falater ended rather differently. Falater, a forty-three-year old Arizona man who stabbed his wife fortyfour times with a hunting knife and then hid his blood-stained clothes and bandaged his wounded hand, was found guilty of murder by a Phoenix jury. The sleep researchers, who testified at no charge, were accused by the prosecution of falsifying Falater’s four nights of polysomnographic records, which demonstrated that he fit the profile of a sleepwalker. See Jeff Stryker, ‘Sleepstabbing,’ Salon.com, 8 July 1999, www.salon.com/health/feature/1999/07/ 08/sleepwalking, accessed 28 August 2002. 58 U.S. Department of Health and Human Services, Polysomnography and Sleep Disorder Centers. 59 Ten years later, some respirologists began to argue that polysomnography should be reinstated in the diagnosis of insomnia, because insomnia was in fact a common symptom of sleep apnea. See Collop, ‘Can’t Sleep?’ 60 Steven Epstein, Impure Science: AIDS, Activism, and the Politics of Knowledge (Berkeley: University of California Press, 1996). 61 William Dement, personal interview, 6 September 2002. Michael Chase at the BIS/BRI at UCLA, was also among a number of other sleep researchers involved in these lobbying efforts (Michael Chase, personal interview, 25 July 2002). 62 National Commission on Sleep Disorders Research, Wake up America: A National Sleep Alert, vol. 1, Executive Summary and Executive Report of the NCSDR (Washington, DC, 1993), 17 and 67. 63 Ibid., 24. 64 Ibid., 47. 65 Ibid., 40. The report’s authors used the controversial definition of sleep apnea as a condition in which a patient suffered from at least five apneic episodes per hour of sleep in laboratory conditions. 66 Ibid., 35. 67 Ibid., 47–8. 68 Ibid., 67. 69 See Guilleminault, ‘Obstructive Sleep Apnea.’ 70 J. Wright et al., ‘Health Effects of Obstructive Sleep Apnoea and the Effectiveness of Continuous Positive Airway Pressure: A Systematic Review of the Research Evidence,’ British Medical Journal 314 (1997): 851–60. See also the accompanying editorial, ‘Deep and Shallow Truths.’

Notes to pages 424–6

501

71 Wright et al., ‘Health Effects of Obstructive Sleep Apnoea,’ 851 and 855. 72 See, for example, J.R. Stradling and R.J.O. Davies, ‘Obstructive Sleep Apnoea: Evidence for efficacy of continuous positive airway pressure is compelling,’ British Medical Journal 315 (1997): 368; H.M. Engleman et al., ‘Obstructive Sleep Apnoea: Some Criticisms of Studies Are Unfounded,’ British Medical Journal 315 (1997): 369; and G.J. Gibson and K. Prowe, ‘Obstructive Sleep Apnoea: Review Was Misleading and May Deny Cost Effective Treatment to Patients,’ British Medical Journal 315 (1997): 368–9. 73 Stradling and Davies, ‘Obstructive Sleep Apnoea,’ 368. 74 S.J.G. Semple and D.R. London, ‘Obstructive Sleep Apnoea: Treatment Prevents Road Accidents, Injury, and Death Caused by Daytime Sleepiness,’ British Medical Journal 315 (1997): 368–9; A.I. Pack and T. Young, ‘Superficial Analysis Ignores Evidence on Efficacy of Treatment,’ British Medical Journal 315 (1997): 369. 75 Gibson et al., ‘Obstructive Sleep Apnoea.’ See also ‘Rebuttal from Drs. Davies and Stradling,’ American Journal of Respiratory and Critical Care Medicine 161 (2000): 1778. 76 See John Wright et al., ‘Obstructive Sleep Apnoea: Author’s Reply,’ British Medical Journal 315 (1997): 551; John Wright and Trevor Sheldon, ‘Sleep Apnoea and Its Impact on Public Health,’ Thorax 53 (1998): 410–3; John Wright and Trevor Sheldon, ‘The Efficacy of Nasal Continuous Positive Airway Pressure in the Treatment of Obstructive Sleep Apnea Is Not Proven,’ American Journal of Respiratory and Critical Care Medicine 161 (2000): 1776–8. 77 F. Barbé et al., ‘Treatment with Continuous Positive Airway Pressure Is Not Effective in Patients with Sleep Apnea but No Daytime Sleepiness: A Randomized, Controlled Trial,’ Annals of Internal Medicine 134 (2001): 1015–23. 78 Allan I. Pack and Greg Maislin, ‘Who Should Get Treated for Sleep Apnea?’ Annals of Internal Medicine 134 (2001): 1065–7. 79 P.E. Peppard et al., ‘Prospective Study of the Association between SleepDisordered Breathing and Hypertension,’ New England Journal of Medicine 342 (2000): 1378–84. For a criticism of this study’s tendency to ignore the influence of night-time arousal on blood pressure in favour of the measures of oxygen desaturation that accompany apneic episodes, see R.J. Thomas, ‘Sleep-Disordered Breathing and Hypertension,’ New England Journal of Medicine 343 (2000): 966–7. 80 Yasuyuki Kaneko et al., ‘Cardiovascular Effects of Continuous Positive Airway Pressure in Patients with Heart Failure and Obstructive Sleep Apnea,’ New England Journal of Medicine 348 (2003): 1233–41.

502

Notes to pages 429–31

Epilogue 1 Martin Heidegger, Introduction to Metaphysics, trans. Gregory Fried and Richard Polt (New Haven, CT, and London: Yale University Press, 2000). 2 On Heidegger’s anti-modern stance, see Bruno Latour, We Have Never Been Modern (Cambridge, MA: Harvard University Press, 1993), 65–7. For a concise primer and helpful bibliography on Heidegger’s Nazism, see the ‘Translator’s Introduction’ to Heidegger, Introduction to Metaphysics, xiv–xvii. 3 For a recent example, see Chip Brown, ‘The Man Who Mistook His Wife for a Deer and Other Tales from the Science of Extreme Sleep,’ New York Times Magazine, 2 February 2003, 34–41, 63, 72, 79–80. On the role of the production of novelty in twentieth-century science, see Pickstone, Ways of Knowing, 162–88. 4 F.H.C. Crick and Graeme Mitchison, ‘The Function of Dream Sleep,’ Nature 304, no. 5922 (1983): 111–14. 5 J. Allan Hobson and Robert W. McCarley, ‘The Brain as a Dream-State Generator: An Activation-Synthesis Hypothesis of the Dream Process,’ American Journal of Psychiatry 134 (1977): 1335–68. 6 Jouvet, The Paradox of Sleep, 111–33. 7 Jagdish Hattiangadi, ‘The Mind as an Object of Scientific Study,’ in Christina E. Erneling and David Martel Johnson, eds., The Mind as a Scientific Object: Between Brain and Culture (Oxford: Oxford University Press, 2005), 342–66. 8 Flanagan, ‘Deconstructing Dreams.’

Bibliography

The following is a list of sources that I found helpful in writing this book. I have avoided dividing the works into the conventional categories of ‘primary’ and ‘secondary’ sources for fear that many would fall into both. Or neither. Archives Alan Mason Chesney Medical Archives, Johns Hopkins Medical Institutions, Baltimore, Maryland Adolf Meyer Collection American Philosophical Society Library, Philadelphia, Pennsylvania Simon Flexner Papers [SFP] Becker Medical Library, Washington University School of Medicine, Washington University in St Louis, St Louis, Missouri Hallowell Davis Papers FC022 [HDP] Cornell University Archives, Division of Rare and Manuscript Collections, Ithaca, New York Faculty Biographical Files: E.B. Titchener Harvard Medical Library in the Francis A. Countway Library of Medicine, Harvard Medical School, Cambridge, Massachusetts Alexander Forbes Papers H MS c22 [AFP] Walter Bradford Cannon Papers H MS c40 [WBCA] Rockefeller Archive Center, Sleepy Hollow, NY [RAC] University of California, Los Angeles, California Neuroscience History Archives [NHA] Horace Winchell Magoun Papers [HWMP-NHA]

504

Bibliography

University Archives Record Series no. 371, Director’s Office, Administrative Files, 1944–84, Brain Research Institute [UCLA-BRI] Box 1 UCLA Archives, RG Brain Research Institute, Series 195 – Reports and Publications, 1962–86 Oral Histories, Special Collections, Charles E. Young Research Library University of Chicago Library, Special Collections Research Center, Chicago, Illinois Association for the Psychophysiological Study of Sleep Records, 1961–86 [APSSR] Nathaniel Kleitman Papers [uncatalogued] Personal Communications The following individuals responded to my questions through personal interview, phone conversation, letter, or e-mail. I have included in the endnotes all relevant dates whenever I cite their statements or ideas in the text. Their comments were of invaluable assistance in framing many of the topics discussed in the book and I express my thanks to all for sharing their time and experiences with me. Naturally, I take full responsibility for my interpretation and contextualization of their statements. Eugene Aserinsky Roger Broughton Michael Chase William Dement Arnold Gessel Edmund Jacobson, Jr Michel Jouvet Joe Kamiya Richard Lange Louise Marshalll Peggy Mason Helene Morcos Elliot Phillipson Alan Rechtschaffen Published Sources Abraham, Tara H. ‘Nicolas Rashevsky’s Mathematical Biophysics.’ Journal of the History of Biology 37 (2004): 333–85.

Bibliography 505 – ‘(Physio)logical Circuits: The Intellectual Origins of the McCulloch–Pitts neural Networks.’ Journal of the History of the Behavioral Sciences 38 (2002): 3–25. Abrahamson, Isador. Lethargic Encephalitis. New York: n.p., 1935. Adrian, Edgar D. ‘The All-or-None Principle in Nerve.’ Journal of Physiology 47 (1914): 460–474. – ‘The Discovery of Berger.’ In Handbook of Electroencephalography and Clinical Neurophysiology, vol. 1. Edited by Antoine Rémond. Amsterdam: Elsevier, 1971. Adrian, Edgar D., and B.H.C. Matthews. ‘The Berger Rhythm: Potential Changes from the Occipital Lobes in Man.’ Brain 57 (1934): 355–85. – ‘The Interpretation of Potential Waves in the Cortex.’ Journal of Physiology 81 (1934): 440–71. Agulhon, Maurice. The French Republic, 1879–1992. Translated by Antonia Nevill. Cambridge and Oxford: Basil Blackwell, 1993. Alvarez, Luis W. ‘Alfred Lee Loomis: November 4, 1887 – August 11, 1975.’ Biographical Memoirs of the National Academy of Sciences 51 (1980): 309–41. Antliff, Mark. Inventing Bergson: Cultural Politics and the Parisian Avant-Garde Princeton, NJ: Princeton University Press, 1993. Armstrong, Nancy, and Leonard Tennenhouse. ‘The Interior Difference: A Brief Genealogy of Dreams, 1650–1717.’ Eighteenth-Century Studies 23 (1990): 458–78. Aronowitz, Robert A. Making Sense of Illness: Science, Society, and Disease. Cambridge: Cambridge University Press, 1998. Aserinsky, Eugene. ‘The Discovery of REM Sleep.’ Journal of the History of the Neurosciences 5 (1996): 213–27. Aserinsky, Eugene, and Nathaniel Kleitman. ‘Eye Movements during Sleep.’ Federation of American Societies for Experimental Biology. Federation Proceedings 12 (1953): 6. – ‘Regularly Occurring Periods of Eye Motility, and Concomitant Phenomena, during Sleep.’ Science 118 (1953): 273–4. – ‘Two Types of Ocular Motility Occurring in Sleep.’ Journal of Applied Physiology 8 (1955): 1–10. Association of Professional Sleep Societies. Diagnostic Classification of Sleep and Arousal Disorders. Rochester, MN: Association of Professional Sleep Societies 1987 (repr. of 1979 edition). Attali, Jacques. Histoires du temps. Paris: Fayard, 1982. Avanzini, Guy. Alfred Binet. Paris: P.U.F., 1999. Ax, Albert F. ‘Editorial.’ Psychophysiology 1 (1964): 1–3. – ‘Goals and Methods of Psychophysiology.’ Psychophysiology 1 (1964): 8–25. Barnes, Barry, and Steve Shapin, eds., Natural Order: Historical Studies of Scientific Culture. Beverley Hills, Calif.: Sage Publications, 1979.

506

Bibliography

Bergson, Henri. Dreams. Translated by Edwin. E. Slosson. New York: Huebsch, 1914. – Matter and Memory. Trans. N.M Paul and W.S. Palmer. New York: Zone Books, 1988. – Time and Free Will: An Essay on the Immediate Data of Consciousness. Translated by F.L. Pogson. New York: Macmillan, 1916. Bernard, Claude. An Introduction to the Study of Experimental Medicine. Translated by Henry Copley Green. New York: Collier Books, 1961. Binet, Alfred. ‘Mosso (A.) – La fatigue intellectuelle et physique.’ L’année psychologique 1 (1895): 450–2. – ‘Psychologie individuelle – La Description d’un objet.’ L’année psychologique 3 (1897): 296–332. Blake, H., R.W. Gerard, and N. Kleitman. ‘Factors Influencing Brain Potential during Sleep.’ Journal of Neurophysiology 2 (1939): 48–60. Blustein, Bonnie Ellen. ‘Medicine as Biology: Neuropsychiatry at the University of Chicago, 1928–1939.’ Perspectives on Science 1 (1993): 416–44. – Preserve Your Love for Science: Life of William A. Hammond, American Neurologist. Cambridge: Cambridge University Press, 1991. Borch-Jacobsen, Mikkel. Remembering Anna O.: A Century of Mystification. New York: Routledge, 1996. Borck, Cornelius. ‘Electricity as a Medium of Psychic Life: Electrotechnical Adventures into Psychodiagnosis in Weimar Germany.’ Science in Context 14 (2001): 565–90. Borell, Merriley. ‘Instruments and an Independent Physiology: The Harvard Physiological Laboratory, 1871–1906.’ In Physiology in the American Context, 1850–1940. Edited by Gerald L. Geison. Bethesda, MD: American Physiological Society, 1987. Boring, Edwin G. A History of Experimental Psychology. 2nd ed. New York: Appleton-Century-Crofts, 1957. Bourdieu, Pierre. The Logic of Practice. Translated by Richard Nice. Stanford, CA: Stanford University Press, 1990. – Science of Science and Reflexivity. Translated by Richard Nice. Chicago: University of Chicago Press, 2004. Brain, Robert M. and M., Norton Wise. ‘Muscles and Engines: Indicator Diagrams and Helmholtz’s Graphical Methods.’ In Universalgenie Helmholtz: Rückblick nach 100 Jahren. Edited by Lorenz Krüger. Berlin: Academie Verlag, 1994. Braun, Marta. Picturing Time: The Work of Étienne-Jules Marey (1830–1904). Chicago: University of Chicago Press, 1992. Brooks, John I. The Eclectic Legacy: Academic Philosophy and the Human Sciences in Nineteenth-century France. Newark: University of Delaware Press, 1998.

Bibliography 507 Broughton, Roger. ‘A Proposed Classification of Sleep Disorders.’ Sleep Research 1 (1972): 146. – ‘Sleep Disorders: Disorders of Arousal?’ Science 159 (1968): 1070–8. Broughton, Roger, and Frederick Snyder. ‘Sleep and Clinical Pathological States.’ In The Sleeping Brain: Perspectives in the Brain Sciences. Vol. 1. Edited by Michael H. Chase. 363–431. Los Angeles: Brain Information Service, 1972. Brown, Theodore M. ‘Alan Gregg and the Rockefeller Foundation’s Support of Franz Alexander’s Psychosomatic Research.’ Bulletin of the History of Medicine 61 (1987): 155–82. Bylebyl, Jerome J. ‘Galen on the Non-Natural Causes of Variation in the Pulse.’ Bulletin of the History of Medicine 45 (1971): 482–5. Cabanis, Pierre Jean Georges. On the Relations between the Physical and Moral aspects of Man. Translated by George Mora. Baltimore, MD: Johns Hopkins University Press, 1981. Cambrosio, Alberto, and Peter, Keating. ‘The Disciplinary Stake: The Case of Chronobiology.’ Social Studies of Science 13 (1983): 323–53. Camp, Carl D. ‘Morbid Sleepiness, with a Report of a Case of Narcolepsy and a Review of some Recent Theories of Sleep.’ Journal of Nervous and Mental Disease 2 (1907): 9–21. Canguilhem, Georges. The Normal and the Pathological. Translated by Carolyn R. Fawcett in collaboration with Robert S. Cohen. New York: Zone Books, 1989. Cannon, Walter D. ‘The Body Physiologic and the Body Politic.’ Science 93 (1941): 1–10. – ‘The James-Lange Theory of Emotions: A Critical Examination and an Alternative Theory.’ American Journal of Psychology 39 (1927): 106–24. Cariou, Marie. Lectures bergsoniennes. Paris: P.U.F., 1990. Carlson, Anton Julius. The Control of Hunger in Health and Disease. Chicago: University of Chicago Press, 1916. Carroy, Jacqueline, and Régine Plas. ‘The Origins of French Experimental Psychology: Experiment and Experimentalism.’ History of the Human Sciences 9 (1996): 73–84. Cashman, Sean Dennis. America Ascendant: From Theodore Roosevelt to FDR in the Century of American Power, 1901–1945. New York: New York University Press, 1998. – America in the Age of the Titans: The Progressive Era and World War I. New York and London: New York University Press, 1988. Caton, Richard. ‘A Case of Narcolepsy.’ Clinical Society of London. Transactions 22 (1889): 133–7. – The Electric Currents of the Brain. British Medical Journal 1 (1875): 278. Cave, H.A. ‘Narcolepsy.’ Archives of Neurology and Psychiatry 26 (1931): 50–101.

508

Bibliography

de Chadarevian, Soraya. ‘Graphical Method and Discipline: Self-Recording Instruments in Nineteenth-Century Physiology.’ Studies in History and Philosophy of Science 24 (1993): 267–91. Chertok, Léon. ‘Centième anniversaire de l’ouvrage “Du sommeil et des états analogues”: De Liébeault à Freud,’ La Presse Médicale 74 (1966): 2945–6. Cheyne, George. The English Malady; or, a Treatise of Nervous Diseases of All Kinds. London: George Strahan, 1733. – An Essay of Health and Long Life, 2nd ed. London: George Strahan, 1725. Claparède, Edouard. ‘Esquisse d’une théorie biologique du sommeil.’ Archives de psychologie 4 (1905): 245–9. – ‘La question du sommeil.’ L’année psychologique 18 (1912): 419–59. – ‘The Value of Biological Interpretation for Abnormal Psychology.’ Journal of Abnormal Psychology 1 (1906): 83–92. Coleman, R.M., et al. ‘Sleep-wake Disorders Based on a Polysomnographic Diagnosis: A National Cooperative Study.’ Journal of the American Medical Association 247 (1982): 997–1003. Collop, Nancy A. ‘Can’t Sleep? You May Have Sleep Apnea.’ Chest 120 (2001): 1768–9. Crawford, Susan Y. ‘Informal Communication among Scientists in Sleep and Dream Research.’ PhD thesis, University of Chicago, 1970. Creager, Angela N.H. The Life of a Virus: Tobacco Mosaic Virus as an Experimental Model, 1930–1965. Chicago: University of Chicago Press, 2002. Crick, F.H.C., and Graeme Mitchison. ‘The Function of Dream Sleep.’ Nature 304 (1983): 111–14. Cross, Stephan J., and William R. Albury. ‘Walter B. Cannon, L.J. Henderson, and the Organic Analogy.’ Osiris 2 (1987): 165–92. Dagognet, François. Étienne-Jules Marey: A Passion for the Trace. Translated by Robert Galeta with Jeanine Herman. New York: Zone Books, 1992. – Faces, surfaces, interfaces. Paris: J. Vrin, 1982. Dagron, Gilbert. ‘Rêver de Dieu et parler de soi: Le rêve et son interprétation d’après les sources Byzantines.’ In I sogni nel medioevo: Seminario Internazionale, Roma, 2–4 ottobre 1983, 37–55. Edited by Tullio Gregory. Daniels, Luman E. ‘Narcolepsy.’ Medicine (Baltimore) 13 (1934): 1–122. Le Dantec, Félix. ‘La biologie de M. Bergson.’ Revue du mois 4 (1907): 231. Danziger, Kurt. ‘The History of Introspection Reconsidered.’ Journal of the History of the Behavioral Sciences 16 (1980): 241–62. Darrow, Chester W. ‘Psychophysiology, Yesterday, Today, and Tomorrow.’ Psychophysiology 1 (1964): 4–7. Daston, Lorraine J. ‘British Responses to Psychophysiology, 1860–1900.’ Isis 69 (1978): 192–208.

Bibliography 509 – ‘The Coming into Being of Scientific Objects.’ In Lorraine Daston, ed., Biographies of Scientific Objects. Chicago: University of Chicago Press, 2000. Davis, Hallowell. ‘Crossroads on the Pathways to Discovery.’ In The Neurosciences: Paths of Discovery. Edited by Frederic G. Worden, Judith P. Swazey, and George Adelman. Cambridge, Mass.: MIT Press, 1975. – ‘The Forbes “School” of Neurophysiology at Harvard.’ Electroencephalography and Clinical Neurophysiology 1 (1949): 139–140. Delage, Yves. ‘Essai sur la théorie du rêve.’ Revue scientifique 48 (1891): 40–48. Delboeuf, Joseph. Le sommeil et les rêves, considérés principlalement dans leur rapports avec les théories de la certitude et de la mémoire. Paris: Alcan, 1885. Dement, William. ‘Daytime Sleepiness and Sleep “Attacks.”’ In Narcolepsy. Advances in Sleep Research. Vol. 3. Edited by Christian Guilleminault, William C. Dement, and Pierre Passouant, 17–42. New York: SP Medical and Scientific Books, 1976. – ‘Dream Recall and Eye Movements during Sleep in Schizophrenics and Normals. Journal of Nervous and Mental Disease 122 (1955): 263–9. – ‘The Effect of Dream Deprivation.’ Science 131 (1960): 1705–7. – ‘The History of Narcolepsy and Other Sleep Disorders.’ Journal of the History of the Neurosciences 2 (1993): 121–34. – ‘Foreword.’ In Hypersomnia with Periodic Apneas. Advances in Sleep Research. Vol. 4. Edited by Elio Lugaresi, Giorgio Coccangna, and Magda Mantovani. New York and London: SP Medical and Scientific Books, 1978. – Some Must Watch While Some Must Sleep. San Francisco: W.H. Freeman, 1972. Dement, William, and Nathaniel Kleitman. ‘The Relation of Eye Movements during Sleep to Dream Activity: An Objective Method for the Study of Dreaming.’ Journal of Experimental Psychology 53 (1957): 339–46. Descartes, René. Philosophical Writings. Selected, translated, and edited by Elizabeth Anscombe and Peter Thomas Geach. London: Nelson, 1954. Diagnostic Steering Committee of the American Sleep Disorders Association. The International Classification of Sleep Disorders Diagnostic and Coding Manual. Rochester, MN: ASCA, 1990. Diara, Anne. ‘Le transformisme de Félix le Dantec.’ Revue de synthèse 100 (1979): 407–22. Dowbiggin, Ian. ‘Alfred Maury and the Politics of the Unconscious in Nineteenth-century France.’ History of Psychiatry 1 (1990): 255–87. Dror, Otniel E. ‘The Affect of Experiment: The Turn to Emotions in AngloAmerican Physiology, 1900–1940.’ Isis 90 (1999): 205–37. Dulaey, Martine. Le rêve dans la vie det la pensée de Saint Augustin. Paris: Études Augustiniennes, 1973. von Economo, Constatin. Encephalitis Lethargica: Its Sequelae and Treatment. Translated. London: Oxford University Press, 1931.

510

Bibliography

– ‘Sleep as a Problem of Localization.’ Journal of Nervous and Mental Disease 71 (1930): 249–59. von Economo, Constantin, Karoline Freifrau, and Julius Wagner-Jauregg. Baron Constantin von Economo: His Life and Work. Translated from the second German edition by Ramsay Spillman. Burlington, VT: Free Press Interstate Publishing, 1937. Edelstein, Emma J., and Ludwig Edelstein, Asclepius: A Collection and Interpretation of the Testimonies, 2 vols. Baltimore: Johns Hopkins University Press, 1945. Edelstein, Ludwig. ‘Greek Medicine in Its Relation to Religion and Magic.’ In Ancient Medicine: Selected Papers of Ludwig Edelstein. Edited by Oswei Temkin and C. Lilian Temkin. 205–46. Baltimore: Johns Hopkins University Press, 1967. Ekirch, Roger A. ‘Sleep We Have Lost: Pre-Industrial Slumber in the British Isles.’ American Historical Review 106 (2001): 343–86. Ellenberger, Henri F. The Discovery of the Unconscious: The History and Evolution of Dynamic Psychiatry. New York: Basic Books, 1970. Engleman, H.M., S.E. Martin, I.J. Dreary, and N.J. Douglas. ‘Obstructive Sleep Apnoea: Some Criticisms of Studies Are Unfounded.’ British Medical Journal 315 (1997): 369. Epstein, Steven. Impure Science: AIDS, Activism, and the Politics of Knowledge. Berkeley: University of California Press, 1996. Fearing, Franklin. Reflex Action: A Study in the History of Physiological Psychology. Baltimore: Williams and Wilkins, 1930. Ferguson, Harvie. The Lure of Dreams: Sigmund Freud and the Construction of Modernity. London: Routledge, 1996. Flanagan, Owen. ‘Deconstructing Dreams: The Spandrels of Sleep.’ Journal of Philosophy 92 (1995): 5–27. Fleck, Ludwik. Genesis and Development of a Scientific Fact. Translated by Fred Bradley and Thaddeus J. Trenn. Chicago: University of Chicago Press, 1979. Fleming, Donald. ‘Walter B. Cannon and Homeostasis.’ Social Research 51 (1984): 609–40 Forbes, Alexander. ‘The Interpretation of Spinal Reflexes in Terms of Present Knowledge of Nerve Conduction.’ Physiological Reviews 2 (1922): 361–414. Foucault, Michel. The Birth of the Clinic: An Archaeology of Medical Perception. Translated by A.M. Sheridan Smith. New York: Pantheon Books, 1973. Foulkes, W. David. ‘Dream Reports from Different Stages of Sleep.’ Journal of Abnormal and Social Psychology 65 (1962): 14–25. – ‘Dream Research: 1953–1993.’ Sleep 19 (1996): 609–24. – ‘Nonrapid Eye Movement Mentation.’ Experimental Neurology Suppl. 4 (1967): 28–38.

Bibliography 511 – The Psychology of Sleep. New York: Scribner’s, 1966. Fox, Daniel M. ‘The Politics of the NIH Extramural Program, 1937–1950.’ Journal of the History of Medicine and Allied Sciences 42 (1987): 447–66. Frank, Robert G. Jr. ‘Instruments, Nerve Action and the All-or-None Principle.’ Osiris 9 (1994): 208–35. – ‘The Telltale Heart: Physiological Instruments, Graphic Methods, and Clinical Hopes, 1854–1914.’ In The Investigative Enterprise: Experimental Physiology in Nineteenth-Century Medicine. Edited by William Coleman and Frederic L. Holmes. 211–90. Los Angeles: University of California Press, 1988. Freud, Sigmund. The Interpretation of Dreams. Translated by Joyce Crick. Oxford: Oxford University Press, 1999. – The Standard Edition of the Complete Psychological Works. 24 vols. Translated from the German under the general editorship of James Strachey, in collaboration with Anna Freud, assisted by Alix Strachey and Alan Tyson. London: Hogarth Press, 1953. Fye, W. Bruce. The Development of American Physiology: Scientific Medicine in the Nineteenth Century. Baltimore: Johns Hopkins University Press, 1987. García-Ballester, Luis. Galen and Galenism: Theory and Medical Practice from Antiquity to the European Renaissance. Translated by Philip Banks. Aldershot, UK: Ashgate, 2002. Gasser, Jacques. Aux origines du cerveau moderne: localisations, langage et mémoire dans l’oeuvre de Charcot. Paris: Fayard, 1995. Gastaut, Henri, C.A. Tassinari, and B. Duron. ‘Étude polygraphique des manifestations épisodiques (hypniques et respiratoires), diurnes et nocturnes, du syndrome de Pickwick.’ Revue neurologique 112 (1965): 568–79. – ‘Polygraphic Study of the Episodic Diurnal and Nocturnal (Hypnic and Respiratory) Manifestations of the Pickwick Syndrome.’ Brain Research 2 (1966): 167–86. Gauld, Alan. A History of Hypnotism. Cambridge: Cambridge University Press, 1992. Geison, Gerald L., and Frederic L. Holmes, eds. ‘Research Schools: Historical reappraisals.’ Osiris 8 (1993): 1–248. Gélineau, J.B.E. ‘De la narcolepsie.’ Gazette des hôpitaux 54 (1880): 626–8 and 635–7. Gessel, Arnold H. ‘Edmund Jacobson, M.D., Ph.D.: The Founder of Scientific Relaxation.’ International Journal of Psychosomatics 36 (1989): 5–14. Gibbs, F.A., H. Davis, and W.G. Lennox. ‘The Electroenepahalogram in Epilepsy and in Conditions of Impaired Consciousness.’ Archives of Neurology and Psychiatry 34 (1935): 113–48.

512

Bibliography

Gibson, G.J., and K. Prowe. ‘Obstructive Sleep Apnoea: Review Was Misleading and May Deny Cost Effective Treatment to Patients.’ British Medical Journal 315 (1997): 368–9. Gillespie, Richard. ‘Industrial Fatigue and the Discipline of Physiology.’ In Physiology in the American Context, 1850–1940. Edited by Gerald L. Geison. Bethesda, MD: American Physiological Society, 1987. Gloor, Pierre. ‘Hans Berger and the Discovery of the Electroencephalogram.’ In P. Gloor, eds., ‘Hans Berger on the Electroencephalogram: The Fourteen Original Reports on the Human Electroencephalogram.’ Electroencephalography and Clinical Neurophysiology 28 (supplement) (1969). Grey Walter, W. ‘The Technique and Application of Electro-Encephalography.’ Journal of Neurology and Psychiatry 1 (1938): 359–85. Grob, Gerald N. The Mad among Us: A History of the Care of America’s Mentally Ill. Toronto: Maxwell Macmillan; New York: Free Press, 1994. – Mental Illness and American Society, 1875–1940. Princeton, NJ: Princeton University Press, 1983. Grogin, R.C. The Bergsonian Controversy in France, 1900–1914. Calgary: University of Calgary Press, 1988. Guerrini, Anita. Obesity and Depression in the Enlightenment: The Life and Times of George Cheyne. Norman, Okla.: University of Oklahoma Press, 2000. Guilleminault, Christian. ‘Obstructive Sleep Apnea: The Clinical Syndrome and Historical Perspective.’ Medical Clinics of North America 69 (1985): 1187–1203. Guilleminault, Christian, and W.C. Dement. ‘Pathologies of Excessive Sleep.’ In Advances in Sleep Research. Vol. 1. Edited by Elliot D. Weitzman. 345–90. New York: SP Medical and Scientific Books, 1974. Guilleminault, Christian, Frederic L. Eldridge, and William C. Dement. ‘Insomnia, Narcolepsy, and Sleep Apneas.’ Bulletin de physio-pathologie repiratoire (Nancy) 8 (1972): 1127–38. – ‘Insomnia with Sleep Apnea: A New Syndrome.’ Science 181 (1973): 856–8. Gunella, G. ‘Évolution des conceptions pathogeniques sur le syndrome de Pickwick.’ Bulletin de physio-pathologie repiratoire (Nancy) 8 (1972): 981–1003. Hacking, Ian. Historical Ontology. Cambridge, MA: Harvard University Press, 2002. – Rewriting the Soul: Multiple Personality and the Sciences of Memory. Princeton, NJ: Princeton University Press, 1995. – Why Does Language Matter to Philosophy? Cambridge: Cambridge University Press, 1975. Hall, Calvin S. The Meaning of Dreams. New York: Harper, 1953. – Studies of Dreams Reported in the Laboratory and at Home. Santa Cruz, Calif.: Institute of Dream Research, 1966.

Bibliography 513 – ‘What People Dream about.’ Scientific American 184 (1951): 60–3. Hallyn, Fernand. The Poetic Structure of the World: Copernicus and Kepler. Translated by Donald M. Leslie. New York: Zone Books, 1990. Hammond, William Alexander. Sleep, Sleeplessness and the Derangements of Sleep; or, the Hygiene of the Night. London: Simpkin, Marshall, 1892. Hartmann, Ernest. The Sleeping Pill. New Haven, CT, and London: Yale University Press, 1979. Hattiangadi, Jagdish. ‘The Mind as an Object of Scientific Study.’ In The Mind as a Scientific Object: Between Brain and Culture. Edited by Christina E. Erneling and David Martel Johnson. 342–66. Oxford: Oxford University Press, 2005. Hauri, Peter J. ‘A Cluster Analysis of Insomnia.’ Sleep 6 (1983): 326–28. Hayward, Rhodri. ‘The Tortoise and the Love-Machine: Grey Walter and the Politics of Electroencphalography.’ Science in Context 14 (2001): 615–41. Healy, David. The Creation of Psychopharmacology. Cambridge, MA: Harvard University Press, 2002. – ‘Some Continuities and Discontinuities in the Pharmacotherapy of Nervous Conditions before and after Chlorpromazine and Imipramine.’ History of Psychiatry 11 (2000): 393–412. Held, R., B.-A. Schwartz, and H. Fischgold. ‘Fausse insomnie.’ Presse médicale 67 (1959): 141–3. Hervey de Saint-Denis, Marie-Jean Lyon. Dreams and How to Guide Them. Translated by N. Fry. London: Duckworth, 1982. Hill, Christopher. Puritanism and Revolution: Studies in Interpretation of the English Revolution of the 17th Century. London: Secker and Warburg, 1958. Hobbes, Thomas. Leviathan; or the Matter, Forme and Power of a Commonwealth Ecclesiasticall and Civil. Edited by Michale Oakeshott. Oxford: Basil Blackwell, 1946. Hobson, J. Allan. The Dreaming Brain: How the Brain Creates both the Sense and the Nonsense of Dreams. New York: Basic Books, 1988. Hoff, H.E., and L.A. Geddes. ‘Graphic Registration before Ludwig: The Antecedents of the Kymograph.’ Isis 50 (1959): 5–21. – ‘The Technological Background of Physiological Discovery: Ballistics and the Graphic Method.’ Journal of the History of Medicine and Allied Sciences 15 (1960): 345–63. Holowchak, Mark A. ‘Aristotle on Dreaming: What Goes on in Sleep When the “Big Fire” Goes out.’ Ancient Philosophy 16 (1996): 405–23. Horvath-Peterson, Sandra. Victor Duruy and French Education: Liberal Reform in the Second Empire. Baton Rouge: Louisiana State University Press, 1984. Howell, Joel D. Technology in the Hospital: Transforming Patient Care in the Early

514

Bibliography

Twentieth Century. Baltimore and London: Johns Hopkins University Press, 1995. Ingle, Dwight J. ‘Anton J. Carlson: A Biographical Sketch.’ Perspectives in Biology and Medicine 22 (suppl.) (1979): S114–S136. Institute of Medicine. National Academy of Sciences. Sleeping Pills, Insomnia, and Medical Practice. Washington, DC: National Academy of Sciences, 1979. Jacobs, Elizabeth A., Charles F. Reynolds III, David J. Kupfer, Particia A. Lovin, and Amie B. Ehrenpreis. ‘The Role of Polysomnography in the Differential Diagnosis of Chronic Insomnia.’ American Journal of Psychiatry 145 (1988): 346–9. Jacobson, Edmund.‘Electrophysiology of Mental Activities. American Journal of Psychology 44 (1932): 677–94. – ‘On Meaning and Understanding.’ American Journal of Psychology 22 (1911): 553–77. – ‘Progressive Relaxation.’ American Journal of Psychology 36 (1925): 73–87. – Progressive Relaxation: A Physiological and Clinical Investigation of Muscular States and Their Significance in Psychology and Medical Practice. Chicago: University of Chicago Press, 1929. – ‘The Technic of Progressive Relaxation.’ Journal of Nervous and Mental Disease 60 (1924): 568–78. – You Can Sleep Well: The ABC’s of Restful Sleep for the Average Person. New York: Whittlesey House, 1938. James, William. Principles of Psychology. New York: Henry Holt, 1890. – ‘What Is an Emotion?’ Mind 9 (1884): 188–205. Jardine, Nicholas. The Scenes of Inquiry: On the Reality of Questions in the Sciences. New York: Oxford University Press, 1991. Jasper, Herbert H. ‘Charting the Sea of Brain Waves, 1948.’ Journal of Clinical Neurophysiology 14 (1997): 464–9. – ‘Electrical Signs of Cortical Activity.’ Psychological Bulletin 34 (1937): 411–81. Jasper, Herbert H., L.D. Proctor, R.S. Knighton, W.C. Noshay, and R.T. Costello, eds., Reticular Formation of the Brain. Boston: Little Brown, 1958. Jasper, Herbert H. and G.D. Smirnov, eds. ‘The Moscow Colloquium on Electroencephalography of Higher Nervous Activity.’ Electroencephalography and Clinical Neurophysiology, supp. 13 (1960): 1–420. Jeannerod, Marc. The Brain-Machine: The Development of Neurophysiological Thought. Translated by David Urion. Cambridge, MA: Harvard University Press, 1985. Jouvet, Michel. The Paradox of Sleep: The Story of Dreaming. Translated by Laurence Garey. Cambridge, MA, and London: MIT Press, 1999. – ‘Paradoxical Sleep: A Study of Its Nature and Its Mechanisms.’ Progress in Brain Research 18 (1965): 20–57.

Bibliography 515 Jung, Carl J. ‘L’analyse des rêves.’ L’année psychologique 15 (1909): 160–7. Jung, Richard, and Wolfgang Kuhlo. ‘Neurophysiological Studies of Abnormal Night Sleep and the Pickwickian Syndrome.’ In Sleep Mechanisms, Progress in Brain Research. Vol. 18. Edited by K. Akert, C. Bally, and J.P. Schadé. 140–59. Amsterdam: Elsevier, 1965. Kales, Anthony. ‘Psychophysiological Studies of Insomnia.’ Annals of Internal Medicine 71 (1969): 624–9. Kales, Anthony, Edward O. Bixler, and Joyce D. Kales. ‘Role of the Sleep Research and Treatment Facility: Diagnosis, Treatment and Education.’ Advances in Sleep Research 1 (1974): 391–415. Kales, Anthony, and J.D. Kales. ‘Shortcomings in the Evaluation and Promotion of Hypnotic Drugs. Editorial.’ New England Journal of Medicine 293 (1975): 826–7. – ‘Sleep Laboratory and Evaluation of Psychoactive Drugs.’ Pharmacology for Physicians 4 (1970): 1–6. Kales, Joyce D., T.L. Tan, C. Swearingen, and Anthony Kales. ‘Are Over-theCounter Sleep Medications Effective? All-night EEG Studies.’ Current Therapeutic Research, Clinical and Experimental 13 (1971): 143–51. Karacan, I., and R.L. Williams. ‘Insomnia: Old Wine in a New Bottle.’ Psychiatric Quarterly 45 (1971): 274–88. Keating, Peter, and Alberto Cambrosio. Biomedical Platforms: Realigning the Normal and the Pathological in Late-Twentieth-Century Medicine. Cambridge, MA: MIT Press, 2003. Kingsland, Sharon E. ‘A Humanistic Science: Charles Judson Herrick and the Struggle for Psychobiology at the University of Chicago.’ Perspectives on Science 1 (1993): 445–77. Kleitman, Nathaniel. ‘Basic Rest-Activity Cycle – 22 Years Later.’ Sleep 5 (1982): 311–17. – ‘Biological Rhythms and Cycles.’ Physiological Reviews 29 (1949): 1–30. – Sleep and Wakefulness as Alternating Phases in the Cycle of Existence. Chicago: University of Chicago Press, 1939. – ‘Studies in the Physiology of Sleep: I. The Effects of Prolonged Sleeplessness on Man.’ American Journal of Physiology 66 (1923): 67–92. Kohler, Robert E. Partners in Science: Foundations and Natural Scientists, 1900– 1945. Chicago and London: University of Chicago Press, 1991. Kripke, Daniel F., and Nancy Fisher. ‘Acceptance of the Clinical Sleep Laboratory.’ Sleep Research 4 (1975): 229. Kruger, Steven F. Dreaming in the Middle Ages. Cambridge: Cambridge University Press, 1992. Kubie, Lawrence S. ‘The Concept of Dream Deprivation: A Critical Analysis.’ Psychosomatic Medicine 24 (1962): 62–5.

516

Bibliography

Kuriyama, Shigehisa. The Expressiveness of the Body and the Divergence of Greek and Chinese Medicine. New York: Zone Books, 1999. Kusch, Martin. ‘Recluse, Interlocutor, Interrogator: Natural and Social Order in Turn-of-the-Century Psychological Research Schools.’ Isis 86 (1995): 419–39. Ladd, George Trumbull. ‘Contribution to the Psychology of Visual Dreams.’ Mind 3 (1892): 299–304. Lasagna, Louis. ‘Congress, the FDA, and New Drug Development: Before and after 1962.’ Perspectives in Biology and Medicine 32 (1989): 322–43. Latour, Bruno. ‘Drawing Things Together.’ In Representation in Scientific Practice. Edited by Michael Lynch and Steve Woolgar. Cambridge, MA: MIT Press, 1988. – We Have Never Been Modern. Translated by Catherine Porter. Cambridge, MA: Harvard University Press, 1989. Lavie, Peretz. ‘Nothing New under the Moon: Historical Accounts of Sleep Apnea Syndrome.’ Archives of Internal Medicine 144 (1984): 2025–8. Lawrence, Christopher, and George Weisz, eds. Greater than the Parts: Holism in Biomedicine, 1920–1950. New York: Oxford University Press, 1998. Lear, John, ed. Kepler’s Dream, with the Full Text and Notes of Somnium, Sive Astronomia Lunaris, Joannis Kepleri. Translated by Patricia Frueh Kirckwood. Berkeley and Los Angeles: University of California Press, 1965, Legendre, Réné. ‘The Physiology of Sleep.’ Smithsonian Institution Annual Report. 1911, 587–602. Le Goff, Jacques. ‘Le christianisme et les rêves (IIe–VIIe siècles).’ In I sogni nel medioevo: Seminario Internazionale, Roma, 2–4 ottobre 1983. Edited by Tullio Gregory. 171–218. Rome: Edizioni dell’Ateneo, 1985. Lemaine, G., M. Clémençon, A. Gomis, B. Pollin, and B. Salvo. Stratégies et choix dans la recherche: À propos des travaux sur le sommeil. Paris: Mouton, 1977. Lenoir, Timothy. The Strategy of Life: Teleology and Mechanics in Nineteenth Century German Biology. Boston: Kluwer, 1982. Lewis, Hywel David. Dreaming and Experience. London: Athlone, 1968. LiDonnici, Lynn R. The Epidaurian Miracle Inscriptions: Text, Translation and Commentary. Atlanta: Scholars’ Press, 1995. Liébeault, Auguste. Du sommeil et des états analogues, considérés surtout au point de vue de l’action du moral sur le physique. Paris: Masson, 1866. Lloyd, G.E.R. Aristotle: The Growth and Structure of His Thought. London: Cambridge University Press, 1968. – Magic, Reason and Experience: Studies in the Origin and Development of Greek Science. Cambridge: Cambridge University Press, 1979. Loomis, Alfred Lee, E. Newton Harvey, and Garret Hobart. ‘Cerebral States dur-

Bibliography 517 ing Sleep, as Studied by Human Brain Potentials.’ Journal of Experimental Psychology 21 (1937): 127–144. – ‘Electrical Potentials of the Human Brain.’ Journal of Experimental Psychology 19 (1936): 249–79. – ‘Further Observations on the Potential Rhythms of the Cerebral Cortex during Sleep.’ Science 82 (1935): 198–200. –‘Potential Rhythms of the Cerebral Cortex during Sleep.’ Science 81 (1935): 597–8. Lugaresi, Elio, et al. ‘Effets de la trachéotomie dans les hypersomnies avec respiration periodique.’ Revue neurologique 123 (1970): 267–8. Lugaresi, Elio, Giogio Coccagna, and Magda Mantovani. Hypersomnia with Periodic Apneas. New York and London: SP Medical and Scientific Books, 1978. Lutz, Tom. American Nervousness, 1903: An Anecdotal History. Ithaca, NY: Cornell University Press, 1991. Macnish, Robert. The Philosophy of Sleep. 2nd ed. (1834). Repr. in Daniel N. Robinson, ed., Significant Contributions to the History of Psychology 1750–1920. Washington, DC: University Publications of America, 1977. Malcolm, Norman. Dreaming. London: Routledge and Kegan Paul, 1959. Marchini, Evelyn J., Thomas J. Coates, John G. Magistad, and Shirley J. Waldum. ‘What Do Insomniacs Do, Think, and Feel during the Day? A Preliminary Study.’ Sleep 6 (1983): 147–55. Marey, Étienne-Jules. Animal Mechanism: A treatise on Terrestrial and Aërial Locomotion. New York: D. Appleton, 1884 – ‘Du moyen d’économiser le travail moteur de l’homme et des animaux.’ Physiologie Expérimentale 1 (1875): 1–18. – La méthode graphique dans les sciences expérimentales et particulièrement en physiologie et en médicine. Paris: Masson, 1878. – ‘Natural History of Organised Bodies.’ Annual Report of the Board of Regents of the Smithsonian Institution for 1867. 277–304. – ‘Préface.’ Physiologie expérimentale 1 (1875): ii–ix. Marks, Harry. The Progress of Experiment: Science and Therapeutic Reform in the United States, 1900–1990. Cambridge: Cambridge University Press, 1997. – ‘Revisiting “The Origins of Compulsory Drug Prescriptions.”’ American Journal of Public Health 85 (1995): 109–15. Marshall, Louise H. ‘The Fecundity of Aggregates: The Axonologists at Washington University, 1922–1942.’ Perspectives in Biology and Medicine 26 (1983): 613–36. – ‘Instruments, Techniques, and Social Units in American Neurophysiology, 1870–1950.’ In Physiology in the American Context, 1850–1940. Edited by Gerald L. Geison. Bethesda, MD: American Physiological Society, 1987.

518

Bibliography

Matheson Commission. Epidemic Encephalitis: Etiology, Epidemiology, Treatment. Report of a Survey by the Matheson Commission. New York: Columbia University Press, 1929, 1932, and 1939. Matthews, Gareth B. Thought’s Ego in Augustine and Descartes. Ithaca, NY, and London: Cornell University Press, 1992. Matthys, R., J. de Roeck, and R. Cluydts. ‘Insomnnia: From a Single Entity to a Complex Concept.’ Acta psychiatrica belgica 85 (1985): 390–405. Maury, Alfred. Le sommeil et les rêves, études psychologiques sur ces phénomènes et les divers états qui s’y rattachent, suivies de recherches sur le developpment de l’instinct et de l’intelligence dans leurs rapports avec le phénomènes du sommeil. 3rd ed. Paris: Didier, 1865. Mengal, Paul. ‘Henri Piéron (1881–1964) et les Néo-Lamarckiens français.’ In C. Debru, Jean Gayon, and J.-F. Picard, eds., Les sciences biologiques et médicales en France, 1920–1950. Paris: CNRS, 1994. Merleau-Ponty, Maurice. Phenomenology of Perception. Translated by Colin Smith. London: Routledge and Kegan Paul, 1962. – The Structure of Behavior. Translated by Alden L. Fisher. Boston: Beacon Press, 1963. Miller, Patricia Cox. Dreams in Late Antiquity: Studies in the Imagination of a Culture. Princeton, NJ: Princeton University Press, 1994. Millett, David. ‘Wiring the Brain: From the Excitable Cortex to the EEG, 1870– 1940.’ PhD thesis, University of Chicago, 2001. Mindell, David A. Between Human and Machine: Feedback, Control, and Computing before Cybernetics. Baltimore: Johns Hopkins University Press, 2002. Monroe, L.J. ‘Psychological and Physiological Differences between Good and Poor Sleepers.’ Journal of Abnormal Psychology 72 (1967): 255–64. Moreira, Tiego. ‘Sleep, Health, and the Dynamics of Biomedicine.’ Social Science and Medicine 83 (2006): 54–63. Morin, Charles M. ‘The Nature of Insomnia and the Need to Refine Our Diagnostic Criteria.’ Psychosomatic Medicine 62 (2000): 483–5. Moruzzi, Guiseppe. ‘The Historical Development of the Deafferentiation Hypothesis of Sleep.’ Proceedings of the American Philosophical Society 108 (1964): 19–28. – ‘Reticular Influences on the EEG.’ Electroencephalography and Clinical Neurophysiology 16 (1964): 2–17. – ‘Sleep and Instinctive Behavior.’ Archives italiennes de biologie 107 (1969): 175– 216. Moruzzi, G., and H.W. Magoun. ‘Brain Stem Reticular Formation and Activation of the EEG.’ Electroencephalography and Clinical Neurophysiology 1 (1949): 455– 73.

Bibliography 519 Mosso, Angelo. Fatigue. Translated by Margaret Drummond and W.B. Drummond. London: Swan Sonnenschein, 1906. – Fear. Translated by E. Lough and A. Kiesow. London: Longman, Green 1896.. Nutton, Vivian. Ancient Medicine. New York: Routledge, 2004. – ‘The Medical Meeting Place.’ In Ancient Medicine in Its Socio-Cultural Context. Edited by Ph.J. van der Eijk, H.F.J. Horstmanschoff, and P.H. Schrijvers. 3–26. Amsterdam and Atlanta: Rodopi, 1995. Oberhelman, Steven M. ‘The Diagnostic Dream in Ancient Medical Theory and Practice.’ Bulletin of the History of Medicine 61 (1987): 47–60. – The Oneirocriticon of Achmet: A Medieval Greek and Arabic Treatise on the Interpretation of Dreams. Lubbock: Texas Tech University Press, 1991. – ‘Dreams in Graeco-Roman Medicine.’ In Aufstieg und Niedergang der Römischen Welt: Geschichte und Kultur Roms im Spiegel der Meueren Forschung. Vol. 37. Edited by W. Haase and H. Temporini. 21–156. Berlin and New York: Walter de Gruyter, 1993. O’Donnell, John M. The Origins of Behaviorism: American Psychology, 1870–1920. New York: New York University Press, 1985. Orr, William C. ‘Utilization of Polysomnography in the Assesment of Sleep Disorders.’ Medical Clinics of North America 69 (1985): 1153–67. Oswald, Ian. ‘Drugs and Sleep.’ Pharmacological Review 20 (1968): 272–303. Oswald, Ian, and R.G. Priest. ‘Five Weeks to Escape the Sleeping Pill Habit.’ British Medical Journal 2 (1965): 1093–9. Otis, Laura. Organic Memory: History and the Body in the Late Nineteenth and Early Twentieth Centuries. Lincoln: University of Nebraska Press, 1994. Paicheler, Geneviève. L’invention de la psychologie moderne. Paris: L’Harmattan, 1992. Pandora, Katherine. Rebels within the Ranks: Psychologists’ Critique of Scientific Authority and Democratic Realities in New Deal America. Cambridge: Cambridge University Press, 1997. Passouant, Pierre. ‘Doctor Gelineau (1828–1906): Narcolepsy Centennial.’ Sleep 4 (1981): 241–6. Pavlov, Ivan Petrovich. Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex. Translated and edited by G.V. Anrep. New York: Dover Publications, 1960. – ‘The Identity of Inhibition with Sleep and Hypnosis.’ Scientific Monthly 17 (1923): 603–8. – Lectures on Conditioned Reflexes. 2 vols. Translated by W. Horsley Gantt. New York: International Publishers, 1928 and 1941. – ‘New Researches on Conditioned Reflexes.’ Science 58 (1923): 359–61. – ‘The Problem of Sleep.’ In Experimental Psychology and Other Essays. New York: Philosophical Library, 1957.

520

Bibliography

Persell, Stuart M. Neo-Lamarckism and the Evolution Controversy in France, 1870– 1920. Lewiston, NY: Edwin Mellen Press, 1999. Peter, Jean-Pierre. ‘Sommeil, rêve, anesthésie, somnambulisme: le problème de la conscience dans les représentations de l’homme en sommeil.’ Revue d’histoire moderne et contemporaine 43 (1996): 578–92. Phillipson, Eliot A. ‘Sleep Apnea – A Major Public Health Problem.’ New England Journal of Medicine 328 (1993): 1271–3. Pickstone, John V. Ways of Knowing: A New History of Science, Technology and Medicine. Chicago: University of Chicago Press, 2001. Piéron, Henri. ‘Des phénomènes d’adaptation biologique par anticipation rythmique.’ Comptes rendus de l’Académie des sciences 144 (1907): 338–41. – ‘L’étude expérimentale du sommeil normal. La méthode.’ Comptes rendus hebdomadaires des séances de la Société de Biologie de Paris 62 (1907): 307. – L’Évolution de la mémoire. Paris: Ernest Flammarion, 1910. – Le problème physiologique du sommeil. Paris: Masson, 1913. Pressman, Jack. ‘Human Understanding: Psychosomatic Medicine and the Mission of the Rockefeller Foundation.’ In Greater Than the Parts: Holism in Biomedicine, 1920–1950. Edited by Christopher Lawrence and George Weisz. New York: Oxford University Press, 1998. Pressman, Mark R. ‘Whatever Happened to Insomnia (and Insomnia Research)?’ American Journal of Psychiatry 148 (1991): 419–20. Rabinbach, Anson. The Human Motor: Energy, Fatigue, and the Origins of Modernity. New York: Basic Books, 1990. Rather, L.J. ‘The “Six Things Non-Natural.”’ Clio medica 3 (1968): 333–47. Rechtschaffen, Alan. ‘Polygraphic Aspects of Insomnia.’ In The Abnormalities of Sleep in Man. Edited by Henri Gastaut. 109–121. Bologna: Gaggi, 1969. Rechtschaffen, A., and A. Kales. A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Bethesda, MD: U.S. Dept. of Health, Education and Welfare, Public Health Service, 1968. Rechtschaffen, A., E. Wolpert, W. Dement, S. Mitchell, and C. Fisher. ‘Nocturnal Sleep of Narcoleptics.’ Electroencephalography and Clinical Neurophysiology 15 (1963): 599–609. Rechtshaffen, A., and L.J. Monroe. ‘Laboratory Studies of Insomnia.’ In Sleep: Physiology and Pathology. Edited by A. Kales. 158–69. Philadelphia and Toronto: J.B. Lippincott, 1969. Regestein, Quentin R., and Peter Reich. ‘A Sleep Clinic within a General Hospital Psychiatry Service.’ General Hospital Psychiatry 2 (1980): 112–17. Reuchlin. M. ‘The Historical Background for National Trends in Psychology: France.’ Journal of the History of the Behavioral Sciences 1 (1965): 115–23. Reynolds, Charles F. III, David J. Kupfer, Daniel J. Buysse, Patricia A. Coble, and

Bibliography 521 Amy Yeager. ‘Subtyping DSM-III-R Primary Insomnia: A Literature Review by the DSM-IV Work Group on Sleep Disorders.’ American Journal of Psychiatry 148 (1991): 432–8. Richet, Charles. Essai de psychologie générale. Paris: Félix Alcan, 1887. – ‘Les procédés de défense de l’organisme.’ Revue scientifique 52 (1893): 801–7. Ripa, Yannick. Histoire du rêve: Regards sur l’imaginaire des français au XIXe siècle. Paris: Oliver Orban, 1988. Ritvo, Lucille B. Darwin’s Influence on Freud: A Tale of Two Sciences. New Haven, CT: Yale University Press, 1990. Rothstein, William G. Public Health and the Risk Factor: A History of an Uneven Medical Revolution. Rochester, NY: University of Rochester Press, 2003. Sacks, Oliver W. ‘Postencephalic Syndromes.’ In Gerald M. Stern, ed. Parkinson’s Disease. London: Chapman & Hall Medical, 1990. Schiller, Francis. ‘The Semantics of Sleep.’ Bulletin of the History of Medicine 56 (1982): 377–97. Schmidgen, Henning. ‘Pictures, Preparations, and Living Processes: The Production of Immediate Visual Perception (Anschauung) in Late-19th-Century Physiology.’ Journal of the History of Biology 37 (2004): 477–513. Schneider, William. ‘The Scientific Study of Labour in Interwar France.’ French Historical Studies 17 (1991): 410–46. Schwartz, B.-A., G. Guilbaud, and H. Fischgold. ‘Études électroencéphalographiques sur le sommeil de nuit. I. – L’“insomnie” chronique.’ Presse médicale 71 (1963): 1474–6. Schwitzgebel, Eric. ‘Why Do We Think We Dreamed in Black and White?’ Studies in History and Philosophy of Science 33 (2002): 649–60. Scull, Andrew. Social Order/Mental Disorder: Anglo-American Psychiatry in Historical Perspective. Berkeley: University of California Press, 1989. Sewitch, Deborah. ‘EEG in Insomnia.’ American Journal of Psychiatry 141 (1984): 804. Shapin, Steven, and Simon Schaffer. Leviathan and the Air-Pump: Hobbes, Boyle, and the Experimental Life. Princeton, NJ: Princeton University Press, 1985. Shapiro, Colin M., James R. Catterall, Ian Oswald, and David C. Flenley. ‘Where Are the British Sleep Apnea Patients?’ Lancet 2 (1981): 523. Shorter, Edward. From Paralysis to Fatigue: A History of Psychosomatic Illness in the Modern Era. Toronto: Maxwell Macmillan, 1992. – A History of Psychiatry: From the Era of the Asylum to the Age of Prozac. Toronto: John Wiley and Sons, 1997. Smith, Roger. History of the Human Sciences. New York: W.W. Norton, 1997. – Inhibition: History and Meaning in the Sciences of Mind and Brain. Berkeley and Los Angeles: University of California Press, 1992.

522

Bibliography

Speaker, Susan L. ‘From “Happiness Pills” to “National Nightmare”: Changing Cultural Assessment of Minor Tranquilizers in America, 1955–1980.’ Journal of the History of Medicine and Allied Sciences 52 (1997): 338–76. Spear, Joseph H. ‘Cumulative Change in Scientific Production: Research Technologies and the Structuring of New Knowledge.’ Perspectives on Science 12 (2004): 55–85. Staum, Martin S. Cabanis: Enlightenment and Medical Philosophy in the French Revolution. Princeton, NJ: Princeton University Press, 1980. Stern, John A. ‘Toward a Definition of Psychophysiology.’ Psychophysiology 1 (1964): 90–1. Stevens, Rosemary. American Medicine and the Public Interest. New Haven, CT: Yale University Press, 1971. Sullivan, C.E., F.G. Issa, M. Berthon-Jones, and L. Eves. ‘Reversal of Obstructive Sleep Apnoea by Continuous Positive Airway Pressure Applied through the Nares.’ Lancet 1 (1981): 862–5. Sulloway, Frank. Freud, Biologist of the Mind: Beyond the Psychoanalytic Legend. New York: Basic Books, 1979. Tassinari, C.A., et al. ‘Apnoeic Periods and the Respiratory Related Arousal Patterns during Sleep in the Pickwickian Syndrome.’ Bulletin de physio-pathologie respiratoire (Nancy) 8 (1972): 1087–1102. Temin, Peter. Taking Your Medicine: Drug Regulation in the United States. Cambridge, MA: Harvard University Press, 1980. Temkin, Owsei. The Falling Sickness: A History of Epilepsy from the Greeks to the Beginning of Modern Neurology. 2nd ed., revised. Baltimore: Johns Hopkins University Press, 1971. – Hippocrates in a World of Pagans and Christians. Baltimore: Johns Hopkins University Press, 1991. Tissot, S.A.D. Advice to the People in General, with Regard to Their Health ... Translated by J. Kirkpatrick. London: T. Becket and P.A. deHordt 1765. – Onanism. Originally published 1766. Translated by D.T. Bienville: New York: Garland 1985. Titchener, Edward Bradford. ‘Description vs. Statement of Meaning.’ American Journal of Psychology 23 (1912): 165–82. Todes, Daniel P. Pavlov’s Physiology Factory: Experiment, Interpretation, Laboratory Enterprise. Baltimore: Johns Hopkins University Press, 2002. Toulouse, Edouard. ‘Nécrologie: Marey.’ Revue scientifique 22 (1904): 673–5. Toulouse, Edouard, Nicholas Vaschide, and Henri Piéron. ‘Classification of psychical phenomena for experimental research.’ Mind 11 (1902): 535–46. – Technique de psychologie expérimentale. 2nd ed. Revised and expanded by Toulouse et H. Piéron. Paris: Doin, 1911.

Bibliography 523 United States. Congress. Senate. Committee on Labor and Human Resources. Subcommittee on Health and Scientific Research. Use and Misuse of Benzodiazepines: Hearings before the Committee on Labor and Human Resources. Subcommittee on Health and Scientific Research. 96th Congress, 1st session. 19 September 1979. – Congress. Senate. Subcommittee on Monopoly of the Select Committee on Small Business. Competitive Problems in the Drug Industry. Hearings before the Subcommittee on Monopoly of the Select Committee on Small Business. 90th Congress, 1st session. to 96th Congress, 2nd session., 1967–79. – Department of Health and Human Services, Agency for Health Care Policy and Research. Polysomnography and Sleep Disorder Centers, Health Technology Assessment Reports, no. 4. Washington, DC: GPO, 1991. – National Commission on Sleep Disorders Research. Wake up America: A National Sleep Alert, volume 1, Executive Summary and Executive Report of the NCSDR. Washington, DC, 1993. Van Lieshout, R.G.A. Greeks on Dreams. Utrecht: HES Publishers, 1980. Vaschide, Nicholas. Le sommeil et les rêves. Paris: Ernest Flammarion, 1911. Vaschide, Nicholas, and Henri Piéron. La psychologie du réve au point de vue médical. Paris: Librairie J.-B. Baillière et Fils, 1902. Vogel, G.W. ‘REM Deprivation. III. Dreaming and Psychoses.’ Archives of General Psychiatry 18 (1968): 312–29. – ‘Studies in Psychophysiology of Dreams, III: The Dream of Narcolepsy.’ Archives of General Psychiatry 3 (1960): 421–8. Ward, Christopher D. ‘Encephalitis Lethargica and the Development of Neuropsychiatry.’ Psychiatric Clinics of North America 9 (1986): 215–24. Ware, J. Catesby. ‘Apparent Decline in Basic Sleep Research.’ Sleep Research 17 (1988): 2. Washburn, Margaret Floy. Movement and Mental Imagery: Outline of a Motor Theory of the Complexer Mental Processes. Boston: Houghton, 1916. Watson, James B. ‘The Place of the Conditioned Reflex in Psychology.’ Psychological Review 23 (1916): 89–116. – ‘Psychology as the Behaviorist Views It.’ Psychological Review 20 (1913): 158–77. Webb, Wilse B. ‘Sleep as a Biological Rhythm: a Historical Review.’ Sleep 17 (1994): 188–94. Weir-Mitchell, Silas. ‘Some Disorders of Sleep.’ American Journal of the Medical Sciences 100 (1890):109–27. Wellman, Kathleen. ‘Félix le Dantec et le néo-Lamarckiens français.’ Revue de synthèse 100 (1979): 423–41. Wight, Randall D. ‘The Pavlov-Yerkes Connection: What Was Its Origin? Psychological Record 43 (1993): 351–60.

524

Bibliography

Williams, Simon J. ‘Sleep and Health: Sociological Reflections on the Dormant Society.’ Health 6 (2002): 173–200. – Sleep and Society: Sociological Ventures into the (Un)Known. London: Routledge, 2005. Windholz, George. ‘Hypnosis and Inhibition as Viewed by Heidenhain and Pavlov.’ Integrative Physiological and Behavioral Science 31 (1996): 338–49. – ‘Pavlov and the Pavlovians in the Laboratory.’ Journal of the History of the Behavioral Sciences 26 (1990): 64–74. Windholz, George, and L.H. Witherspoon. ‘Sleep as a Cure for Schizophrenia: A Historical Episode.’ History of Psychiatry 4 (1993): 83–93. Wright, John, and Trevor Sheldon. ‘Sleep Apnoea and Its Impact on Public Health.’ Thorax 53 (1998): 410–13. – ‘The Efficacy of Nasal Continuous Positive Airway Pressure in the Treatment of Obstructive Sleep Apnea Is Not Proven.’ American Journal of Respiratory and Critical Care Medicine 161 (2000): 1776–8. Young, Allan. ‘Walter Cannon and the Psychophysiology of Fear.’ In Greater than the Parts: Holism in Biomedicine, 1920–1950. Edited by C. Lawrence and G. Weisz. New York: Oxford University Press, 1998. Young, John Z. ‘Sources of Discovery in Neuroscience.’ In The Neurosciences: Paths of Discovery. Edited by Frederic G. Worden, Judith P. Swazey, and George Adelman. Cambridge, MA: MIT Press, 1975. Young, Terry, et al. ‘The Occurrence of Sleep-Disordered Breathing among Middle-Aged Adults.’ New England Journal of Medicine 328 (1993): 1230–5.

Index

Academy of Medicine (NYC), 196, 201–2 Achmet, 47 addiction, 14, 350–1; criminalization of, 353–6 and dependency, 380–3; as a medical problem, 357–9, 360– 2, 364–8, 394 Adie, William John, 339 Adrian, Edgar, 11, 270–7 AIDS, 419–21 Albert Merritt Billings Hospital, 223, 226, 232, 237, 330 Alexander, Franz, 220–1, 299, 331 all-or-none hypothesis, 271–2 amplification, 271–5 animals, 5, 10, 15, 29, 31, 80, 89, 96, 101, 105; dreams of, 32, 37; as experimental subjects, 333–4, 337; preparations of, for experiment, 271–2, 313–14. See also cats; dogs animism, 146 Anna O. See Pappenheim, Bertha anti-Semitism, 208–9, 240 Apter, Nathaniel, 314, 321 Aristotle, 29–38 Artemidorus, 41–2 Artistides, Aelius, 23

Asclepian cult, 8, 21–2, 27, 34–5, 45; and abaton, 23–5; rituals of, 22–4 Asclepius, 22, 24–5, 27, 36 Aserinsky, Eugene, 12, 308–12, 316– 21, 324, 327 Associated Professional Sleep Societies (APSS), 14, 15, 326 Association for the Psychophysiological Study of Sleep (APSS), 13, 15, 326–35, 348, 362, 373, 384–7, 391, 396, 399, 419–20 Augustine, 45–8, 69 Ballet, Gilbert, 87 barbiturates, 188–9, 350, 354, 356, 361–2, 379–80. See also drugs; tranquilizers Beard, George, 238 beer, non-alcoholic, 222 behaviour, 10, 167, 241. See also comportement; Watson, James B. Bell, Charles, 109 benzodiazepines, 357–8, 379–80, 383, 390 Berger, Hans, 11, 264–70 Bergson, Henri, 9, 133–45, 151–2, 155, 166

526

Index

Bernard, Claude, 94, 169 Bernheim, Hyppolyte, 111–15, 119 Binet, Alfred, 150–8, 163–4 biogenetic law, 131 blood: circulation of, 71–4, 79–80, 83–4, 98, 266. See also plethysmograph bloodletting, 74 Bluestein, Bonnie Ellen, 83, 85 Bond, Earl D., 203 Borch-Jacobsen, Mikkel, 123 Braid, James, 112 brain localization, 264, 269. See also sleep centre, in brain Brain Research Institute (BRI at UCLA), 363–4, 367–73; funding of, 374–5 Bremer, Frédéric Gaston, 313–14, 399 Breuer, Josef, 122–4 bromides, 87, 189, 343, 355. See also drugs Broughton, Roger, 385–6, 408–9, 418 Brown-Séquard, Charles-Édouard, 83, 130, 181 Burton, John, 63 Burton, Robert, 55 Cabanis, Georges, 72–5, 89, 400–1 Calkins, Mary, 315 Camp, Carl D., 339–40 Canguilhem, Georges, 5, 328 Cannon, Walter B., 186, 212, 248–9, 277, 283–5, 300 Carlson, Anton Julius, 209–15, 222–3, 226–7, 235 cataplexy, 342, 346–7. See also muscular atonia; narcolepsy Catholicism, 151–2 Caton, Richard, 88, 264–5, 402–3 cats: as experimental animals, 334–5

Cave, H.A., 342 Charcot, Jean-Martin, 86–7, 113, 116. See also Salpêtrière Chen, Ko Kuei, 341 Cheyne, George, 61–5 child development: study of, 307–8 children: as experimental subjects, 291, 307–8, 317 Christianity, 38, 45, 47; and dreams, 43–6, 56–7, 69 Cicero, 38, 48 cinematography, 93, 318 Claparède, Edouard, 10, 148–52, 155–6, 160–2, 164–5, 176, 340, 430 Clemente, Carmine, 372 Cobb, Stanley, 293, 314–15 Collège de France, 81, 93, 95, 118, 130, 147, 154 College of Physicians and Surgeons. See Columbia University Columbia University, 138, 197, 202. See also Matheson Commission comportement, 156, 167. See also behaviour; Piéron, Henri Compton, Karl T., 292 consciousness: motor theory of, 246– 8, 252 Copernicanism, 49–51 Cornell University, 238–9, 241 C-PAP (Continuous Positive Airway Pressure), 414–15, 423–6 Crab, Roger, 58–9 Cullen, William, 72 Daniels, Luman E., 342–3, 346 Le Dantec, Félix, 163 Darrow, Chester W., 329–30 Darwin, Charles, 146 Daston, Lorraine, 81 Davis, Hallowell, 12, 280–92, 298–306

Index Davis, Pauline, 280–3, 291, 298–9 degeneration, 147–8, 165–6 Delage, Yves, 130–1, 175 Delboeuf, Joseph, 115–18, 146 Dement, William, 12–13, 320–4, 327– 31, 334, 337, 345, 347–8, 375, 378, 396, 399, 411–13, 419–20 dependency. See addiction; drugs, dependency on depression, 392 Derbyshire, Bill, 282 Descartes, René, 49, 51–5, 327 Dickens, Charles, 401–4 dogs: dreams of, 32, 142; as experimental animals, 170–4, 183, 185, 212, 217, 284. See also animals dreams, 17, 78, 297, 325; and anxiety, 127–8, 131, 159; and autobiography, 23, 39–40, 45, 51–3, 62–3; deprivation of, 12, 323–4, 327–8; and diagnosis, 21, 24–8, 34, 36, 39–41, 44, 47, 57–60, 69, 158–9, 337, 345; and fiction, 49–52; function of, 125–33, 143, 324; incubation of, 23; and madness, 68–70, 72–4, 139, 396; and morality, 43–7, 58–60; natural history of, 36, 37, 41; as private experience, 19–21, 33, 47, 142, 348; and prophecy or divination, 21–2, 35–7, 41–4, 54–60, 68–9; as scientific objects, 6–7, 303–5, 311–24, 326–34, 338, 373–4, 397, 426–7; and self-knowledge, 17, 60, 64; and sensation, 33, 65, 72; and social order, 48, 53–9, 69; theories of, 28, 33, 35, 37, 40–4, 66, 68, 85, 116–18, 121–33, 137–45, 159–60, 251–3; thought in, 33, 64–6, 82, 374; verifiability of, 19–20, 52, 125; wet, 46, 68. See also hypnosis; REM; sleep

527

Dreyfus affair, 147, 151–2 drugs, 30, 76, 86–7, 188–9, 217, 364, 381; advertising of, 359, 365, 376, 382–3; dependency on, 350; and insomnia, 350; overdose of, 14, 361, 379–80, 385; regulation of, 353–7, 361, 365–6, 374–9. See also addiction; suicide drunkenness. See intemperance Dunbar, Helen Flanders, 293 Durham, Arthur E., 81–2 Duruy, Victor, 95–6, 102 École Pratique des Hautes Études, 95, 158. See also Duruy, Victor von Economo, Constantin, 10, 193– 202, 313 EEG, 7, 11, 16, 254–70, 275–8, 282– 308, 310–14, 317–18, 343, 346, 368, 370–1; demonstrations of, 276, 284–5, 290–6; and diagnosis, 363, 366–7, 373; and laboratories, 13 Einthoven, Willem, 267 electrocardiograph (ECG), 261, 267 emotions: study of, 105, 107–11, 132, 159–60, 248–9 encephalitis lethargica, 191–204, 341, 343–4 energy, 129–31, 265 ephedrine sulphate, 341–3 epidemic encephalitis. See encephalitis lethargica epilepsy, 27, 68, 86–8, 262–3, 268, 285, 372–3. See also narcolepsy ergograph, 102–3, 153 Erlanger, Joseph, 292 eugenics, 158, 166–8 evidence based medicine (EBM), 392 evolution, 100–2, 110, 131–4, 143–4, 146, 163, 168, 235

528

Index

Exner, Sigmund, 132 experiment, 72–3, 80, 83, 92–3, 134– 6, 140–1, 160, 164, 315; clinical, 341–2, 348 eye movements, 248, 251–6, 303–4, 309–12, 318. See also REM fatigue, 9, 10, 30, 88–90, 94, 96–111, 139, 142, 148, 174–5 fish: sleep of, 31–2. See also animals Fisher, Charles, 331 Flexner, Simon, 201 Flournoy, Theodore, 149 Food and Drug Administration (FDA), 356, 361, 364–6, 376, 381 Forbes, Alexander, 144, 245–6, 271–81, 288, 302 Foulkes, David, 322, 325, 373 Fremont-Smith, Frank, 299 Freud, Sigmund, 9, 116, 121–34, 139–40, 143–5, 159 Fulton, John F., 292 Galen, 8, 25, 39–44 Galileo, 49–50 galvanometer, 267, 272 Gantt, W. Horsley, 184 Garceau, E. Lovett, 283–4 Gasser, Herbert, 292–3 Gastaut, Henri, 386, 407–8, 411–13 Gélineau, Jean-Baptiste, 86–8, 338 Gerard, Ralph Waldo, 292, 296–8 Giard, Alfred, 163 Gibbs, Erna, 283, 285 Gibbs, Frederick, 283, 285, 311 Gilman, Charlotte Perkins, 243 Goldstein, Kurt, 218 Goodwin, Philip, 57–8, 62, 69 graphical method, 9, 89–97, 152,

160–1, 189–90, 214, 235, 269, 305. See also recording, as practice Gregg, Alan, 11, 221, 223, 225–35, 259, 273, 299 Grinker, Roy, 223, 231–2, 244 Guilleminault, Christian, 375, 378, 399, 411–12 habitus, 16, 17, 75, 426 Hacking, Ian, 52, 165 Haeckel, Ernest, 131 Hall, Calvin S., 315–16, 322–3 Hall, Marshall, 83, 85 Hammond, William Alexander, 8, 79– 85, 112–13, 404 Hartley, David, 65–6 Hartmann, Ernest, 381 Harvard University, 190, 238, 241, 272–3, 280–5, 289 Harvey, E. Newton, 287 Haymaker, Webb, 368 Healey, David, 358 Heidegger, Martin, 429 von Helmholtz, Hermann, 89 Henderson, Lawrence Joseph, 279– 81 Hervey de Saint-Denis, Léon, Marquis de, 81 Hippocratism, 25–8, 34–5 Hobbes, Thomas, 52–8, 68 Hogben, Lancelot, 218 holism, 187, 218–23, 225–6, 269, 279, 298–302, 378 homeostasis, 278 hospitals: transformation of, 260–3 Howell, William Henry, 84 hypertension, 422–6 hypnosis, 8, 9, 111–20, 122–4, 134, 140, 154, 181–4, 294 hypnotoxin, 174–7, 196–9

Index hysteria, 88, 113, 122–4 immunology, 174 Imperial Institute for Experimental Medicine (St Petersburg), 183 influenza, 10, 192, 341; avian, 192 inhibition, 10, 178–87, 191, 198–200, 242, 268, 276, 335, 339 insomnia, 8, 14, 15, 54, 74–80, 243, 249, 374–9, 385; diagnosis of, 392, 418, 427; experimental, 10, 11, 162, 169–75, 216–17, 224–5, 323; psychophysiological, 389–90; and public health, 384, 394, 396–7, 427; status of, 349–51, 360–6, 380–3, 389, 391, 394, 399; as symptom, 351–2; treatment of, 353–5, 359. See also sleep disorders; neurasthenia Institut générale psychologique, 140 Institut psychologique internationale, 140 Institute of Dream Research, 322 Institute of Medicine (U.S.), 351, 378–9, 381–5, 390, 415, 419, 421 instruments, 5, 82, 83, 90–4, 97–100, 107, 168, 226–30, 257, 260–3, 270– 1, 278–9, 283, 289–90; medical, 260–3, 267, 287–8; and surveillance, 13, 329–30, 348. See also EEG; kymograph insulin: discovery of, 211 intemperance, 59, 62–3 intentionalism. See Augustine introspection, 10, 80, 81, 140, 143, 153, 159–62, 212–15, 224–5, 239– 43 Jackson, Hughlings, 131, 225, 246 Jacobson, Edmund, 11–12, 237–54, 256, 305, 311–12, 352, 354–5, 360,

529

367, 430. See also introspection; relaxation James, William, 107–10, 140, 148–9, 238, 248. See also emotions, study of Janet, Pierre, 9, 140, 148–9, 155, 164, 340, 352 Jardine, Nicholas, 92 Jasper, Herbert, 257–8, 292, 370 Jelliffe, Smith Ely, 200–1. See also psychosomatic medicine Johns Hopkins University, 84, 240, 287 Johnson, Samuel, 78–9 Josiah Macy, Jr Foundation, 270, 283, 298–300, 302 Jouvet, Michel, 320, 334–8, 347 Jung, Carl, 131, 133, 139 Jung, Richard, 406–7 juvenile delinquency, 203–4 Kales, Anthony, 362–6, 373–8, 391, 409 Kales, Joyce D., 362, 366, 375–8 Kepler, Johannes, 49–51 Kingman, Robert, 351–4, 360 Kleitman, Nathaniel, 11–15, 84, 162, 205–9, 212, 215–19, 222–37, 255–7, 305–10, 316–20, 323–4, 327, 333, 336–8, 430 Kosinski, Jerzy, 4 Kripke, Daniel F., 385 Kronecker, Hugo, 103, 105–6 Kuhlo, Wolfgang, 406–7 Külpe, Oswald, 239 kymograph, 90–2, 104, 161, 213–14 laboratories, 5, 15, 20, 88, 89, 185, 227, 250–1; bacteriological, 3; and diagnosis, 3, 375–8, 388–91, 405, 409–11, 415; dream, 13, 322, 331–

530

Index

2, 373–4; and education, 95–6, 102–3, 183, 375–7; psychological, 150–4, 158, 161, 239; Rad Lab (MIT), 293; sleep, 286–91, 310, 322, 347–8, 361–7, 374, 382–4, 387, 390–1, 393, 405, 410–13, 416–19 labour, science of, 9, 97, 100–2, 105, 110, 158, 215–16, 236, 306–7 Lapicque, Louis, 222, 257 Lasagna, Louis, 376, 378 Lennox, William, 283 Levin, Max, 339 Lhermitte, Jean, 342 Liébeault, Ambroise August, 111–14 Locke, John, 64–5, 67 Loomis, Alfred Lee, 11–12, 280, 287– 97, 302, 306. See also Tuxedo Park (N.Y.) Lucas, Keith, 270–3 Ludwig, Carl, 90–3, 102 Lugaresi, Elio, 408, 411 Lyons, Maryinez, 194 Macnish, Robert (pseud. James Hogg), 75–6, 400–2 Magoun, Horace W., 313–14, 368–72, 398 Malcolm, Norman, 322 Mammoth Cave (Ky.), 232–3 de Manacéïne, Marie, 115, 171 Marey, Étienne-Jules, 9, 89–106, 140– 1, 155 Mason, Max, 221, 259, 292–3 masturbation, 68 Matheson Commission, 196–7, 202 Matthews, Brian, 276 Maury, Alfred, 118 Mauthner, Ludwig, 194, 196 Mayo Clinic (MN), 341–2

McLean, Franklin, 220–1 mechanism, 60–1, 64–6, 69, 95, 100 memory, 82, 115–20, 124–7, 129, 136– 8, 141–2, 162, 165–6, 168–9 Mesmer, Franz Anton, 111–12 Meyer, Adolf, 201, 240–1 Michael Reese Hospital, 241 microelectrodes, 15, 334, 372 military research, 273–4, 302, 306–7 Moll, Albert, 118–19 Morgan, John Pierpont, Jr, 202 Moruzzi, Giuseppe, 313–14, 369, 383, 398 Mosso, Angelo, 9, 84, 89, 102–10, 132, 266 motility recorder, 228–30 Müller, Johannes, 72, 85, 89–90 murder, 417–18 muscular atonia, 335, 337, 347. See also cataplexy narcolepsy, 13–15, 86–8, 338–48, 389, 399, 402–3, 408, 417. See also sleep disorders Neo-Platonism, 38, 45 neurasthenia, 79, 238, 351–3. See also insomnia, as symptom Neurological Institute (N.Y.), 202 neurology, 79, 83, 85, 86, 113, 122, 130, 191, 194, 405–8, 411; and neuroscience, 368–72 neurophysiology, 136, 270–83, 301 neuropsychiatry, 192, 220–3, 226, 229, 231, 234, 244, 320, 372 neuroscience, 231, 329, 332, 367–8; and neurology, 368–72 Newtonism, 60–1, 64–5 nightmares, 131 non-naturals, 8, 42–3, 60, 67, 75. See also sleep, hygiene

Index Nutton, Vivian, 24, 26 objects, scientific, 3–4 Osler, William, 403–4 Oswald, Ian, 362, 416 Paine, Thomas, 68–9 Pappenheim, Bertha, 123 paradoxical sleep, 335, 337. See also REM pathological anatomy, 72–4 patient activism, 420–2 Pavlov, Ivan, 10–11, 178–91, 198, 207– 8, 212, 214–19, 235 Pelagius, 46 phlebotomy. See bloodletting phrenology, 72, 75–6, 402 physicians: status of in antiquity, 26– 8, 41, 42, 44 physiology, 9, 80, 84, 88, 89, 92–4, 183, 209–12, 216. See also experiment Pickwickian syndrome, 401–14. See also sleep apnea Piéron, Henri, 9, 144, 156–8, 161–77, 188–9, 216, 430 placebo, 381 plants. See sleep, of plants Plato, 43–4 plethysmograph, 106–9 Pliny the Elder, 38, 40 polysomnography (PSG), 363, 388, 391–6, 406–9, 416–19, 426 Pravdich-Neminsky, Vladimir, 265, 267 prophecy, 21–2. See also dreams, and prophecy or divination psychiatry, 14, 188–9, 321; diagnosis in, 384–5; reform of, 301, 336–7, 358–9, 392 psychoanalysis, 12, 121–33, 139, 143–

531

5, 220–3, 251–2, 299–304, 312–16, 321–4, 331, 344–7 psychology, 6, 85, 108, 134–5, 156, 158, 163, 167, 190; functionalist, 145–50 psychophysiology, 14, 71, 81, 108, 329–32. See also instruments, and surveillance psychosomatic medicine, 200–1, 299, 302, 406 public health, 7, 14, 15 Puritanism, 55–6, 59 de Puységur, A.M.J. Chastenet, 111 Rabinbach, Anson, 89 Ramón y Cajal, Santiago, 138 Ravel, Maurice, 290 Rechtschaffen, Allan, 329, 332, 345– 6, 363 recording: as practice, 16, 106, 128, 245–8, 289–91, 329, 347–8. See also graphical method Redlich, Emil, 338 reflex, 83–5, 112, 129, 166, 172, 179, 182–6, 191, 217 Regimen 3, 28 Regimen 4, 27–8 relaxation, 224–5, 237–8, 241–53, 305, 354 REM (rapid eye movements), 6, 12, 13, 17, 207, 255–6, 297, 303–4, 309–10, 316–24, 326–8, 335, 345, 409; dissociation of, 347, 408 REM sleep behaviour disorder, 417 republicanism, 151–2, 155, 162–4, 177, 430 respirology, 415, 422–3 rest cure, 243 reticular formation, 369–72. See also sleep centre, in brain

532

Index

rhythms, biological, 6–7, 31, 62, 85, 166, 169, 172–6, 197, 199, 212–15, 222–3, 232–3, 236, 264–6, 276, 306–7 Ribot, Théodule, 9, 147–8, 169 Richardson, Bruce, 232–3 Richet, Charles, 84, 130, 140, 150, 166, 168 risk factors, 346, 393, 412–13, 422–7 Rockefeller Foundation, 11, 220, 226, 235, 244, 258–63, 299 Rosenthal, Curt, 342 Rufus of Ephesus, 40 On the Sacred Disease, 27–8 Salpêtrière (hospital), 87, 150, 152, 158, 162 Saul, Leon J., 299 scanning hypothesis, 321, 327 scepticism, 45, 52–4, 117. See also dreams, verifiability of schizophrenics: as experimental subjects, 321, 327, 337 Schwartz, Betty, 409 science policy, 258–63, 336 Shakespeare, William, 77 Sherrington, Charles, 144, 181, 218, 270–1, 278–81 Shiller, Francis, 368 Shorter, Edward, 83, 85 sleep: and experience, 4–5, 8, 21, 32, 48; as active, 149–50, 160–1, 164, 198–200, 204, 367, 397–8; first and second, 6, 67, 77; function of, 6, 9, 10, 31, 32, 116–17, 145–8; hygiene, 6, 42, 44, 48, 61–4, 67–8, 75–6, 352; movement in, 29–30; and nutrition, 30, 32; observation of, 25, 31, 159, 250–2, 303, 309–12; as passive, 89, 111, 145–6; and perspiration, 62; of

plants, 29, 172; and populations, 6, 79, 253, 413; as a scientific object, 3, 107, 155–6, 165, 169–71, 191–2, 195, 205–7, 215–16, 232–6, 255–7, 289, 303–6, 309, 318, 328, 330, 333– 4, 372, 394, 406, 428–31; sensation in, 29–32; theories of, 10, 25, 28–33, 40, 54, 61, 65, 71–4, 80–5, 112–20, 136–9, 142, 148–50, 160–1, 164–5, 175–7, 187, 196–200, 217, 236, 337– 40; and therapy, 28, 188–9, 243, 359–60. See also experiment; rhythms, biological; skepticism sleep apnea, 14, 15, 88, 395–8, 400, 413–15, 418, 421; diagnosis of, 425–6; epidemiology of, 399–400, 416–17, 424–6. See also Pickwickian syndrome sleep centre: in brain, 195–200, 204, 225–6, 268, 313–14, 333–4, 338–9 sleep clinics, 3, 6–7, 346–8, 416 sleep deprivation, 16. See also insomnia, experimental sleep disorders, 3, 406, 412, 417–18, 423; and advertising, 412; classification of, 385–94, 415; epidemiology of, 399, 419–23; and medical specialization, 376–8, 393, 396–7, 411, 415, 421–2. See also insomnia; narcolepsy; sleep apnea sleep stages, 289–90, 293–7 sleeping sickness. See encephalitis lethargica; trypanosomaisis sleeplessness. See insomnia sleepwalking, 74, 111–12 Slight, David, 232, 293–4 Slosson, Edwin Emery, 143 Smith, Roger, 184 Smuts, Jan Christian, 218–19 somnambulism. See sleepwalking

Index spiritualism, 140–1, 144–6, 159–60, 163 Stanford University, 346–7, 375, 398, 420 stereotack, 313–14 Strümpell, Adolf, 197 suggestion, 111, 113–16, 119 suicide, 14, 379–80 Sullivan, Colin, 414 Tassinari, C.A., 411 teleology, 29, 31, 37, 122, 125–8, 131– 3, 143, 148–50, 162, 164, 176–7 Tissot, Simon-André, 67–8 Titchener, Edward Bradford, 145, 150, 238–40 Toulouse, Edouard, 150, 156, 158, 163–5, 173 tracheostomy, 408, 413–14 tranquilizers, 358–62. See also benzodiazapenes Trosman, Harry, 331, 345 Tryon, Thomas, 58–9, 62, 66, 69 trypanosomiasis, 193–4 Tuxedo Park (N.Y.), 11–12, 286–98 Tylor, Edward Burnett, 146 University of California (Los Angeles), 363, 369–70, 372–5 University of Cambridge, 272–3, 275, 281 University of Chicago, 11, 14, 206–10, 217–23, 226, 229, 231–7, 245, 344. See also neuropsychiatry University College (Liverpool), 264, 272

533

University of Illinois (Neuropsychiatric Institute), 296, 314 University of Jena, 265 University of Ottawa, 385 de Vallangin, Francis, 63 Vaschide, Nicolas, 153–62, 164–5, 207, 430 vegetarianism, 58–60 Villejuif (hospital), 150, 158, 162, 173 visualization, 50, 82, 207, 227 vivisection, 94 Vleck, Fred, 212–13 Vogel, Gerald, 344–7 Wagner-Jauregg, Julus, 193 wakefulness: enforced. See insomnia, experimental Walsche, Francis M.R., 370 Walter, William Grey, 277 Washburn, Margaret Floy, 248 Watson, James B., 10, 167, 241 Weaver, Warren, 259–60, 292–3 Webb, Wilse B., 325–6 Weir-Mitchell, Silas, 83, 88, 243, 339 Wood, Robert W., 287 Wundt, Wilhelm, 145 Würzburg (school), 239 X-rays, 160–2 Yerkes, Robert, 190–1 The Young Poisoner’s Handbook (1995), 19