Wake Me Up If Something Happens
There is a story that circulates in anthropology departments, LSE's included, about Alfred Gell, one of the discipline's most admired theorists of art, ritual and time, who did his own fieldwork among the Umeda of Papua New Guinea and, later, the Muria Gonds of central India. The story goes that Gell found village life in India so monotonous, after the intensity of his earlier work in New Guinea, that he took to spending large parts of the day lying down, leaving instructions with whoever was nearby to wake him if something interesting happened. He wanted, apparently, to be present for the events worth recording and to sleep quietly through everything else.
I cannot vouch for every detail of this story, and I suspect nobody who now tells it can either. It has the shape of departmental folklore rather than documented biography, repeated at seminars and over drinks because it captures something anthropologists recognize about their own discipline and, not infrequently, about themselves. What makes it worth repeating here is not whether it happened exactly as told, but what the story assumes without ever noticing that it is assuming it. Gell, on this telling, treated his own sleep as empty time, a gap between events, to be endured until the ethnographically interesting resumed. Nobody in any version of the story I have heard ever asks the obvious next question. Where, exactly, was he sleeping? Was it really a hammock, or has the hammock simply become the detail every retelling supplies because a hammock is what a bored anthropologist in the tropics is supposed to sleep in? What hours did this happen, and for how long at a stretch? Was he alone, or within sight of the household whose life he had come to study? Could the people around him actually have woken him if they had wanted to, or had he simply become, for the length of an afternoon, unavailable on his own authority? And what did the village make of a resident foreigner who spent so much of the day visibly, publicly asleep, in the middle of everyone else's working afternoon?
Nobody thought to ask, as far as I know, because the discipline that trained Gell, and that trained me, has never quite believed that sleep is where the interesting material is. The exciting events he wanted to be woken for were, by definition, the ones that could be observed, spoken about, and written up. His own sleeping body, lying somewhere in that village for hours at a stretch, in full view or otherwise, was treated as the one thing in the entire field site not worth a note. That is, more or less exactly, the blind spot this collection is built to close. If even one of the more self-aware anthropologists of his generation could sleep through a large part of his own fieldwork without it occurring to him, or to anyone retelling the story since, that his own hammock might have been the most revealing site in the village, then the case for taking sleep seriously does not need to be argued from first principles. Gell's hammock, whatever it actually was, has already made it.
Making the bed
Across many years of fieldwork in Kolkata and the rural districts around it, sleep kept turning up inside almost everything else I was there to study. But I never saw its importance. It showed up as a line about a nap tucked into the middle of a conversation about stomach trouble, a complaint about noise folded into an account of a religious festival, a remark about a bed that turned out to be describing an entire household's geography. Looked at together, these scenes show sleep behaving as a shared, argued-over, materially arranged and morally coded activity rather than a private biological fact that culture merely decorates afterwards. A bed turns out to be furniture, display, medicine and social rank all at once. A nap turns out to be laziness, ethnic character, digestive error and ordinary sociability, depending entirely on who is describing it. A festival turns out to treat other people's sleep as a resource to be deliberately spent.
Before Five A.M.
For a number of weeks when I was living in Lake Town, a Northeastern suburb of Kolkata, my sleep was all over the place. One morning I woke at 4am, after just a few hours in bed. For some reason I decided I’d go on a walk. I left the apartment, expecting empty streets, and for a while I got them. Then, in the last minutes before sunrise, the pavements filled almost all at once. Hundreds of people appeared for their morning walk: some jogging, most striding briskly in loose pairs and threes, overwhelmingly middle-class by dress and manner, mostly middle-aged or older, more than one of them dressed entirely in white.
What has stayed with me longest is a single detail I very nearly walked past. Those who slept out in the open, on pavements, traffic islands and the raised verandas of shuttered shops, were still visibly asleep at the exact moment the walkers appeared around them. Two populations occupied the same stretch of pavement at the same hour and belonged, in every way that mattered, to two different mornings. For the walkers, five o'clock had already been converted into a small act of virtue, a piece of visible self-discipline performed in clean white cotton before the working day began. For the sleepers on the pavement, the same hour was still simply night, an imposition that had not yet arrived and would not arrive on their own timetable at all.
On other early mornings after that first one, the experience was the same, and the proportions never changed. The walkers arrived at almost exactly the same moment each time, as though the sunrise itself were the organizing signal for an entire social class rather than for the sky alone, while the same handful of pavement sleepers, or others very like them, remained undisturbed by the sudden crowd of moving bodies around them. Nobody among the walkers seemed to register the sleepers at all, and the sleepers, for their part, seemed entirely unbothered by being stepped around by several hundred brisk, cheerful strangers.
What I carried away from all those years is realizing how unevenly the right to define one's own morning was distributed. Respectability was not simply a matter of income or caste position stated abstractly. It showed up, very concretely, in who got to decide when their own night ended, and who had that decision made for them by the arrival of daylight, a passing patrol, or the opening of a nearby shop.
On, Under, and Beside the Bed
Some of the clearest evidence I ever collected of how completely a single bed can organize a household came from a slum settlement in Howrah, the city across the river, where I used to shadow an NGO called German Doctors. Each dwelling unit measured around ten square metres: a small cooking area just inside the door, and behind it one combined room for everything else. The dominant piece of furniture, by a wide margin, was a large, elevated bed, raised on a platform high enough that the space beneath it became usable in its own right. People slept on the bed, under the bed, and beside the bed, roughly five to a unit.
There was no version of a “bedroom” in these units, because there was no version of a living room to distinguish it from. The same bed that hosted the household's daytime conversations and held its more valuable possessions also had to sleep five people a night, some on its surface, some tucked beneath its frame, some laid out on mats along its side. Height did the work that additional rooms do elsewhere. A single piece of furniture organized domestic space not only from side to side but from top to bottom, turning one platform into three or four distinct sleeping positions, each with its own comfort, visibility and implied standing within the household. Whoever slept on top of the bed slept differently, in every sense, from whoever slept beneath it, and the difference was not incidental. It was the household's own quiet way of allocating comfort and status without ever needing to say so out loud.
During the day the same bed also served as the natural place to receive a guest, myself included. More than once I sat on the edge of that same elevated frame while tea was prepared, watching household members move around and beneath it as though the furniture beneath my own feet were simply part of the floor plan rather than an object anyone had chosen. It took me a while to realize that the low, half hidden space under the platform, which I had assumed held stored household goods, was, once the household settled down for the night, somebody's bed.
I had spent a good deal of my own fieldwork thinking about separate children's rooms and the private conjugal bedroom as an assumed, if distant, endpoint of domestic development, the shape a household would take on if only it had enough money and enough space. Suchitra's compound was the plainest possible refutation of that assumption I ever came across. Nobody there was waiting to arrive eventually at a bedroom of their own, as though the bed-under-bed-beside-bed arrangement were simply an unfinished version of something else. The bed had already solved, on its own, a problem that in other households an entire floor plan exists to solve, and it had solved it well enough that the question of a separate room for sleeping never seriously arose.
I found the same lesson again more than once elsewhere, in less crowded and less obviously poor households too. A bed's usefulness for sitting, receiving, eating and talking did not switch off simply because a family could, in principle, have afforded a separate room for each of those activities. Seen from Suchitra's ten square metres, a bed that holds a household's daytime sociability alongside its night-time sleep looks like an arrangement necessity had settled on. Seen against other households further up the income scale, the same arrangement started to look instead like a preference that necessity merely made unavoidable in her case and optional everywhere else.
The Bengali Afternoon
Nowhere did sleep carry more social weight for me, or attract more contradictory opinions, than the ordinary afternoon nap. At a gathering of Marwari businessmen, a young man explained Bengali-owned enterprise's relative economic decline, set against Marwari enterprise, in terms that had nothing to do with capital, credit or connections and everything to do with sleep. Bengalis, he told me, were lazy and unenterprising: content to sing songs in the afternoon and then take a nap after lunch while more industrious communities kept working through the same hours. A Bengali man I knew well, Anil, gave me a much more precise version of the same claim and turned it in an unexpected direction. Bengali neighbourhoods had fewer restaurants, he said, because Bengali households went home for a proper midday meal and then slept through the early afternoon, closing their shops until four or five, whereas Muslim shopkeepers, in his account, ate quickly at a stall and kept their shutters open throughout.
I do not repeat this as reliable ethnographic fact about either community's actual habits. Anil's account was an openly ethnic generalization, offered with a certain relish, drawing a moral line between communities using sleep timing as the marker. What I found valuable in it was precisely its unreliability as description. It showed sleep being used as a live idiom for talking about ethnic character, economic virtue and civilizational rank, in a register that had almost nothing to do with sleep physiology and everything to do with who a speaker felt could be trusted to run a shop.
I heard the same practice contested again from an entirely different direction inside Bengali health discourse itself. Another respondent told me flatly that the after-lunch nap was bad for the health because it interfered with digestion, a claim that slotted neatly into a much wider set of local theories linking food, timing and bodily balance that I was separately trying to document. Between the businessman's contempt, Anil's ethnic accounting, and the digestive objection, the apparently simple act of lying down after lunch turned out to have been argued over from at least three incompatible directions at once, as laziness, as ethnic character, and as a medical error waiting to happen. Nobody I spoke to ever simply told me that Bengalis value the afternoon nap as a settled cultural fact. People nap, get mocked for napping, moralize about who naps and why, and warn one another that it will ruin the digestion, all within the same conversations and sometimes within the same household.
I noticed an obvious connection back to the sleepers on the Lake Town pavements. The afternoon nap being argued over at the Marwari gathering, and defended or attacked in household health talk, was in every case a nap taken behind a door, on a real bed, by people with somewhere private to have it. Nobody in these conversations spent any time discussing the sleep of the men and women who had no shop to close and no bed of their own to retreat to for those same afternoon hours. The moral argument over napping, in other words, was really an argument conducted entirely among people who could all afford to nap in the first place, while an equally real afternoon sleep going on elsewhere in the same city passed without any comment at all.
Napping with a Friend
For a long time, I spent my Saturdays at the home of my friend Anjan in Deshapriya Park. I arrived around 10am. We would sit on the family bed and discuss, usually something to do with poststructuralist theory. We would have lunch around 1:30pm at the dining table, then go back to the family bed and take a nap, maybe for an hour, then when we were both awake again, the housemaid brought us tea and biscuits. More French philosophy, and then I’d head back to my place.
The nap was a vital part of the day, because obviously we’d both be very tired after eating lunch. Two male friends spending an afternoon eating, talking, and then falling asleep on the same bed, before waking, eating, and carrying on the conversation, asked for no justification whatsoever.
Placed next to Suchitra's compound, the scene stopped looking to me like an arrangement forced on people by lack of space. Anjan's middle-class household was not living five to a bed for want of an alternative. The nap on the shared bed was a choice made inside a house with other rooms available. Napping together, on this evidence, was not simply what people did when they had no other option. If this seems strange to Westerners, then what is weird is their way of treating sleep, not the Bengalis’.
Locked In, Locked Out
If the afternoon nap showed me how sleep could become a matter of collective judgement, my own flat showed me how difficult ordinary solitude was to arrange in the first place. For a long stretch of my fieldwork I fell asleep close to midnight and woke close to seven, not because my body demanded exactly that rhythm but because the day outside made almost any other pattern difficult to sustain. I woke around seven because the cleaning woman, the man who delivered the newspaper, and whoever else had business with the household would shortly start ringing the bell. My household was reopened to the outside world every single morning by a small procession of service relationships that had nothing to do with my own preferred waking rhythm and everything to do with theirs.
Solitude, when it came at all, had to be deliberately manufactured rather than simply assumed as a background condition of living alone. I marked one day as the first of that year I had spent in something like complete solitude by Bengali standards, having sent the cleaning woman away well in advance. Even so, the doorbell rang at least four times before evening. During one of that week's Saraswati Puja collections, a group of boys came round asking for contributions, and when I gave them what they clearly felt was too little, they responded by locking me inside my own flat from the outside, a piece of theatre that was more mischievous than threatening but that captured something real about how thin the walls of privacy actually were in that setting.
Even the deliveries and errands that structured an ordinary day turned out, I came to think, to be a kind of scheduled interruption of rest that nobody involved would have described in those terms. The newspaper had to arrive before the household woke properly, which meant someone, somewhere, had already been awake and moving for some time before my own morning even began. Complete solitude by Bengali standards, it turned out, meant a handful of interruptions rather than none at all, and even that modest total had required a deliberate, slightly awkward act of sending someone away in advance. Nothing about withdrawal, sleep very much included, simply arrived on its own in that household. It had to be built, defended and periodically renegotiated, doorbell ring by doorbell ring, and the boys locking me in from the outside was, in its own comic register, a small demonstration of exactly how easily that fragile arrangement could be undone by anyone who felt like undoing it.
Waking the Goddess
Nowhere was the deliberate invasion of other people's sleep more spectacular to me, or more openly acknowledged as deliberate, than during the great autumn festivals. One night, during Durga Puja, I had been asleep for barely an hour when the goddess's image was arriving on the back of a truck, accompanied by drums, cymbals and trumpets loud enough to startle me out of bed. Every night for a week I was woken several times a night by immense drumming, and learned that the local puja organizers actually aim at maximum noise and maximum sleep disturbance. A few people made complaints to the Pollution Control Board, which also governs “sound pollution,” but those complaints go nowhere.
What struck me most was a piece of reasoning I heard offered more than once as though it were self-evident. Taking the goddess's image out at the deepest part of the night was considered more powerful precisely because that was the hour when everyone in the vicinity could be assumed to be asleep. The ritual's impact seemed to depend, quite explicitly, on there being a whole sleeping population available at that exact moment to be woken by it. A year later, during the same festival cycle, I woke again after only two hours of sleep, drumming starting up somewhere close by at first light.
None of this fits into a picture of Durga Puja as one long, undifferentiated night of celebration that simply happens to be loud. What I saw looked closer to a calculated, almost surgical use of a valuable and otherwise private resource. The organizers were not merely making noise as an incidental feature of festivity. They were timing that noise to the moment it would land on the greatest possible number of sleeping bodies, and treating the resulting disruption as a rough measure of the ritual's own success. Waking a neighbourhood, on this logic, was not an unfortunate side effect of honouring the goddess properly. It was very close to being the point of doing it at that hour at all.
I asked, on more than one occasion, whether anyone had ever complained about being woken specifically, rather than about the noise level in some more general sense, and got little more than a shrug in reply each time. Being woken by the goddess's own procession did not seem to register, for most people I spoke with, as a legitimate grievance in the way that a private neighbour's loud party clearly did elsewhere. Sleep interrupted for a deity's sake appeared to belong to a different moral category altogether from sleep interrupted for anyone else's convenience, even though the physical experience of losing an hour or two of rest was identical either way.
Keeping the Village Awake
Kali Puja gave me a much subtler picture of the same underlying problem, because here the management of sleep and waking was explicit, collaborative, and at moments quite funny rather than adversarial. The ritual I attended in a village setting ran from around midnight until close to dawn. A drummer, visibly exhausted well before the end, played through most of it. Children ran about lighting firecrackers between the more solemn parts of the ceremony. At one point an uncle in the gathering took it upon himself to tell a long string of jokes with the openly stated purpose of keeping people awake through the rest of the night, as though collective wakefulness were an achievement that needed active, ongoing maintenance rather than something that could simply be assumed to continue on its own once the ritual had begun.
After the ceremony ended and everyone had eaten, auntie Mangla-di found me a bed at first light, in the living room. I slept until 9am into the morning. Contrary to any simple story of an unbroken all-night vigil, sleep was unevenly distributed. Some family members did not sleep at all. Others had managed short naps during the quieter stretches of the ritual itself, dozing upright or leaning against a wall for twenty minutes before being drawn back into the proceedings when the drumming or the chanting picked up again.
Sleep, jokes, firecrackers, food and eventual rest were woven together across the whole night rather than sorted neatly into one block of ritual wakefulness followed by one block of collective sleep afterwards. Kali Puja was a long night during which different people, using different means and choosing different moments, negotiated exactly how much of it they personally could afford to sleep through, and I have come to think of that uncle's supply of jokes as one of the small working technologies that kept the whole loosely coordinated arrangement running until dawn.
The Ashram Clock
One of the more thorough attempts I encountered in my years of fieldwork to rebuild an entire day around sleep, rather than around any other single activity, was a stay at an ashram, which was built on an unforgiving early schedule. The day began at half past four with bathing, followed by meditation timed to the sunrise, then service work, meals, yoga and a tightly structured sequence of further activity running through until evening. On my first two days there I still needed an afternoon nap and took one without hesitation. By the third day I no longer needed it at all. Nearly everyone in the ashram was asleep by half past nine or ten, and I had not gone to bed that early in twenty years. With yoga nidra, sleep came easily.
The ashram had not persuaded me to sleep earlier through willpower, through an explicit instruction about bedtime, or through any argument I found especially convincing on its own terms. It had rebuilt the entire surrounding day, hour by hour, until early sleep, which had felt close to impossible in my own flat only weeks before, became the easiest and most natural thing in the world to do.
None of the other residents seemed to find the schedule remarkable in the way I did. For them, evidently, half past four was simply the hour the day began, requiring no more comment than seven o'clock required of me back in my own flat. On another occasion, after an early rise and a session of hatha yoga followed by yoga nidra, I slipped back afterwards into a light, dreamy sleep rather than returning cleanly to the day. Practised properly, the exercise seemed to require staying poised right on the edge of sleep without ever quite crossing over into it, close enough that my own body kept mistaking the invitation for permission to go the rest of the way. My repeated small failures to stay on the correct side of that edge were, in their own modest way, some of the clearest evidence I found that the edge itself was real, and that holding a position on it was a skill rather than a default state anyone could simply assume.
Glowing in the Dark
One of the most physically extreme places I ever spent a week in was a small patch of land belonging to a family living deep inside the Sundarbans mangrove forest, where I was staying as a guest during a visit to my friend and fellow anthropologist Annu Jalais, who was doing her own fieldwork there. The particular stretch of forest this family occupied was not supposed to be inhabited at all. It fell within a protected reserve, and their presence there was, strictly speaking, illegal, tolerated rather than sanctioned. They had no land anywhere else, and the reserve, hostile as it was, still offered the one thing an unclaimed patch of forest could: somewhere to grow something and somewhere to put a house. Of everywhere I stayed during all my time in Bengal, this was easily the most hostile environment, a strip of land carved out between the tigers and the crocodiles that used the same water and the same paths in and out of the settlement.
The house itself stood right beside one of the innumerable narrow canals cutting through the forest: a single raised, open terrace under a roof, walled at the back but open to the field in front, with no doors or partitions to speak of. The whole family slept out on that terrace at night, each on a mat, each under a mosquito net, arranged in a row along the platform. There was exactly one actual bed in the household, and it belonged to the head of the family, the pater familias of the compound. As a guest of honour, I was not given a mat of my own alongside everyone else. I was invited to share that single bed with him.
It was only afterwards, from Annu, that I learned what my night in that bed had apparently become, locally, a small piece of folklore. Talking to others nearby about white Europeans, most of whom had never actually seen one in real life, my host had apparently offered me as evidence: You haven’t seen white if you haven’t seen this boy. I was so pale, he told them, that I glowed in the dark. Everyone he said this to seemed to find it enormously funny, and by the time the story reached me it had clearly already done several rounds of the neighbourhood.
I have no way of knowing how much of the joke was really about my complexion and how much was about the sheer improbability of a stranger like me turning up at all, sharing the one bed in a house on an illegally settled strip of tiger country. But it has stayed with me as one of the clearest illustrations in all these notes of how little control a sleeping guest has over what happens to their own image once they have shut their eyes. I had been given the single most valuable thing that household had to offer a visitor, its one real bed, shared with its most senior member, and in exchange I had apparently spent the nights quietly turning into the funniest thing anyone in the area had heard about in some time. Being asleep, it turns out, does not stop other people from talking about you. In a place with as few passing strangers as that one, it may be closer to when the talking really starts.
Giving Up the Bed
Not everyone I met wanted an easier, softer or more comfortable sleep. One friend, Sudeshna, described her own sister as having been spiritually exceptional in a way that had been visible since childhood, despite the family's considerable wealth. The sister, she told me, had only ever slept on the bare floor and had never cared for money or possessions in the ordinary way. Sleeping on a bare floor when a comfortable bed stood available in the very same house was not, in this telling, a sign of deprivation or misfortune. It was closely bound up with a claim to spiritual distinction, a discipline actively chosen rather than a hardship passively suffered.
The same conversation touched on meditation, which Sudeshna said left her sister blissful, sleepy, and prone to what she called nice dreams, folding a demanding spiritual discipline and a soft, contented drowsiness into a single account without treating the two as any kind of contradiction requiring further explanation. This sits awkwardly beside the Marwari businessman's complaint about lazy Bengali napping, and just as awkwardly beside the tantrik consultations I describe below, where too much sleep, sleep at the wrong hour, or an inverted daily rhythm were treated as symptoms of something having gone seriously wrong. Here, giving up a comfortable bed in favour of the bare floor marked something close to the opposite of failure. Sleep, or its deliberate refusal, could be read as laziness, as illness, or as sanctity, depending entirely on who was doing the reading and what they already believed about the person doing the sleeping.
Tulsi, Neem, and the Pillow
Alongside these larger arguments about who ought to sleep, and how much, I kept coming across a much smaller and more practical set of everyday technologies for helping sleep along when it would not come easily on its own. A pharmacist I spoke with described tulsi boiled in milk as a household remedy for getting good sleep during a fever, a small piece of kitchen medicine that treated sleep itself as the direct therapeutic goal rather than merely a welcome side effect of feeling generally better. He also recalled, more uncertainly, his own mother's habit of placing neem leaves under a child's pillow as a protection against malaria, and admitted he had never been entirely sure whether the leaves were meant to physically repel mosquitoes or to act on the body in some other, less mechanical fashion that he could not fully articulate.
Nobody I spoke to seemed to feel that this uncertainty mattered very much in practice. What mattered was that the pillow, that most unremarkable piece of sleeping furniture, was doing at least three jobs at once, providing physical comfort, carrying a specific medicinal substance close to the sleeper's head, and standing guard, however imperfectly understood, against a genuine nocturnal danger. A single object tucked under a sleeping child's head managed to be comfort, medicine and protection all at the same time, without anyone in the household feeling any particular need to sort those functions into separate categories or decide which one was the real reason for the leaves.
Cold Winter Nights, Impossible Afternoons
Sleep, in my own experience there, tracked the temperature with almost embarrassing precision. On the hottest day of a particularly bad stretch, the dining room reached close to thirty five degrees and the study climbed even higher. I slept for perhaps an hour in the afternoon and then needed a further three quarters of an hour simply to summon the will to get out of bed afterwards, my own body flatly refusing to cooperate with the rest of the day's plans. At the opposite extreme, one winter, I went out and bought a blanket against what felt, after months of unrelieved heat, close to arctic, though I now recognise it was entirely mild by most standards.
Neither case was really about temperature as such. Both were about how completely my body's tolerance for its own sleeping conditions had been reshaped by whatever the season had most recently done to it. The same body that had spent months barely tolerating the heat of an ordinary afternoon found a genuinely mild winter night worth buying a blanket for. Sleep, more than almost anything else I experienced there, seemed to reset its own baseline continuously against whatever the climate had most recently inflicted on the body attempting to have it, with very little memory of how things had felt only a season before.
When Sleep Goes Wrong
Across the medically plural landscape I was in West Bengal to study in the first place, disturbed sleep turned out to be one of the most reliable diagnostic signs I came across, cutting cleanly across every system of healing I encountered. A bhodrolok family brought their daughter to a tantrik because she had stopped studying, lost weight, begun sleeping through the afternoon and stayed awake for most of the night, an inversion of the ordinary sleep cycle that was treated, in itself and quite apart from her other symptoms, as evidence that something serious had gone wrong with her. In another household a Muslim mother described a baby who was vomiting, refusing food and getting, in her own words, no proper sleep at all. The baba she consulted diagnosed the trouble as the work of a bad spirit acting on the child.
Elsewhere, Anil explained to me that the so-called bad eye could make a person lazy and sleepy, and drew an explicit contrast between a doctor's tablets and his own mother's preferred ritual treatment for exactly the same complaint. On another occasion he described how violating a prescription associated with a family funeral had been followed, in fairly short order, by laziness, loss of appetite, sleepiness and a general withdrawal from ordinary life, resolved only once a baba had been brought in to identify and correct the underlying problem. None of these were cases where sleep was merely mentioned in passing, as one symptom among many of equal weight. In each one, a change in sleep, too much of it, too little of it, or sleep arriving at the wrong hour of the day, was doing real diagnostic work on its own, standing in for a much wider judgement about whether a person's entire way of being in the world had come apart at the seams.
Taken together, these consultations suggested to me that sleep was rarely treated, by any of the healers or families involved, as a self-contained bodily function to be assessed on its own terms. It functioned instead as a kind of summary reading of a person's whole condition, easier to observe and describe than appetite, mood or study habits taken individually, and for that reason repeatedly pressed into service as the single clearest sign that something else, spiritual, social or medical, had gone wrong somewhere upstream of the sleep itself.
A homeopathic practitioner I came to know well, Dr Deb, put the underlying logic more explicitly to me than anyone else I spoke with in all those years. Only the senior doctors, she told me, the ones she called the big doctors, bothered to ask their patients at length about their dreams and their sleep. The local doctors she encountered in daily practice mostly confined their questions to food and eating habits, treating sleep as marginal at best. Whether or not a practitioner thought to ask about sleep and dreaming had become, in her account, a rough but reliable marker of how far their training and their seriousness as a healer actually went. A dream and sleep history was not, on her view, a minor supplementary detail tacked onto a consultation. It was one of the things that separated a superficial encounter from a properly thorough one, and patients seemed to have absorbed the same distinction without needing it spelled out for them.
A Nightcap in Secret
A young Muslim man I met at the Sufi shrine near Tollygunge Metro Station told me that he drank alcohol every night, but always in the last few minutes before going to bed, because if he drank any earlier in the evening the smell would still be on his breath while other members of his household were awake and likely to notice it. Bedtime, in this one small confession, was doing double duty as both the end of the working day and the last safe window before everyone else in the house withdrew into their own separate unavailability for the night.
A practice that would have been immediately noticed an hour earlier became effectively invisible the moment it was timed to coincide with the household's collective retreat into sleep. He had, in effect, located the one narrow gap in the day when social monitoring was reliably about to switch itself off, and had built a small, entirely private ritual to fit exactly inside it, night after night, without ever needing to negotiate the matter openly with anyone he lived with.
Sleeping for an Answer
Durga Puja and Kali Puja both treat sleep as something to be managed around a ritual that has other purposes. At Tarakeswar, a Shiva pilgrimage town in Hooghly district that draws pilgrims from across India, sleep is not managed around the ritual. It is the ritual. The temple's own founding legend already turns on a dream, in which a search for a hidden object of worship in the forest outside town is resolved when a dream reveals the buried lingam beneath the ground, the discovery that gave the deity, and the town, its name. From a temple founded on a dream, it is a short step to a practice built on deliberately seeking one.
Pilgrims arriving with an unresolved problem, an illness without a diagnosis, a decision without a clear direction, a family difficulty without an obvious remedy, could go to the temple in order to sleep there and wait for Shiva to answer them in a dream. Sleep, on this understanding, is not simply where the pilgrim happens to be when the god chooses to speak. It is the technology the pilgrim has travelled to use, arranged, entered into and waited for with exactly the deliberateness that elsewhere in my own fieldwork went into locking a door, buying a blanket, or timing a nightcap to a household's retreat into sleep.
What makes Tarakeswar worth setting beside everything else I saw is the directness of the transaction it assumes. Elsewhere in my own experience, sleep got people into trouble with neighbours, doctors, employers and each other. At Tarakeswar, sleep is described as a reliable way to secure an audience with the god. A pilgrim who cannot get a hearing anywhere else, from a doctor, from a family elder, from their own uncertain judgement, can still fall asleep, and falling asleep in the right place, having made the right journey, is treated as sufficient grounds to expect an answer. Of everything I gathered over the years, it is the clearest case of sleep being entered into on purpose, as a means rather than an interruption, with a specific, named outcome in view.
Does the Goddess Sleep?
Among the questions I did extensive interviews on was: why is Ma Kali treated like a human being, fed, dressed and put to sleep? If a goddess can be described, entirely naturally and without any sense of strain, as needing to be fed and put to bed, in exactly the vocabulary one would use for top royalty, then the ordinary daily labour of caring for a deity, offering food at set times, waking the image in the morning, settling it for the night, belongs to the same broad category of care as looking after a queen or king. Whether or not Kali is actually imagined, theologically, as requiring sleep in any strict sense, the mere fact that the question could be asked at all, and asked as though it were an ordinary empirical matter rather than a category mistake, is one of the more unexpected doors my fieldwork opened for me.
Learning to Nap
In my early months in Kolkata I treated a nap the way any visitor treats something unfamiliar, marking early rising as remarkable, marvelling at other people's naps, puzzling over why I could not sleep straight through my own building's morning routine, trying yoga nidra and mostly failing at it in one direction or the other. By the later years of my fieldwork, lunch and a nap, an afternoon nap, or a nap before an interview had become almost punctuation to me, small units of time slotted in between interviews, temple visits and everything else, the way a comma or a full stop organizes a sentence without anyone noticing it as a decision. Naps of ninety minutes, two hours, three hours and occasionally four turned up embedded in otherwise ordinary working days, and I no longer felt any need to justify them to anyone, least of all to myself.
After one return from a spell away, I slept solidly until nine the next morning, was woken by the cleaning woman, slept again in the afternoon, and then had a poor night's sleep that I put down, quite precisely and with a straight face, to the combined effect of the nap and an evening cup of black tea. I had gone to Bengal to study how people managed illness, medicine and the body. I came back having quietly acquired an entirely different relationship to my own sleep.
Sleep is not only something an ethnographer observes happening to other people from a comfortable distance. Left in a new place long enough, taking its afternoons and its festivals and its heat on their own terms, it quietly rearranges the ethnographer too.
Grown Up and Still Sharing a Room
One further observation belongs here, from later than the rest and from an ongoing friendship rather than fieldwork proper. A Bengali family I have stayed close to over many years continues to share a single sleeping room as a matter of course, including their daughter, now in her mid-twenties, whenever the family travels together. Nobody in this family is short of space or money. The arrangement is not a holdover from anything like Suchitra's compound or from any deprivation comparable to it. It is simply how this particular family prefers to sleep when they are together, and nobody involved treats it as needing an explanation, an apology, or a plan to eventually grow out of it.
I mention this last because it closes off an escape route that the Howrah material, on its own, still leaves open. Five people fitting on, under and beside one bed in ten square metres can always be read, however unfairly, as a story about what poverty forces people to accept rather than what anyone would freely choose. An adult daughter continuing to share a room with her parents, in a household with no such constraint, cannot be read that way. Between Anjan's shared afternoon nap and this family's shared room, the same pattern turns up twice in my own experience, at two very different points on the economic scale, which suggests to me it was never really about space at all. Sleeping apart, not sleeping together, may be the arrangement that actually needs explaining.
Afterword
Read end to end, this collection turns out to be making a stronger claim than any single scene lets on. Sleep, here, behaves the way food, kinship or money behave in other ethnographies: not a private biological leftover that culture merely decorates, but one of the media through which an entire social order gets negotiated. Nothing in these notes treats sleep as neutral. The Lake Town walkers and the pavement sleepers occupy exactly the same hour and belong to two different mornings, one converting five o'clock into a small daily virtue, the other having the hour imposed on it without any say in the matter. That unevenness, over who gets to decide when their own night ends, is close to the political centre of everything else collected here.
The bed does comparable work at the level of the household. In Howrah it is not furniture in any familiar Western sense, but storage, seating, guest reception, a map of status, and a sleeping platform, all at once, organizing a room vertically as well as horizontally in a way no separate bedroom ever needs to. Once an unconstrained family turns out to keep sharing a single sleeping room by choice, the household economics stop explaining anything. Sleeping apart, not sleeping together, becomes the arrangement that actually needs an account.
The afternoon nap works the same way at the level of argument rather than habit. Nobody in these pages simply says that Bengalis value an afternoon sleep. People nap, get mocked for napping, moralize about who naps and why, and warn each other that it will ruin the digestion, sometimes within the same conversation, and the argument itself, tellingly, is conducted almost entirely among people who can afford a private nap in the first place. The three festivals do a version of the same thing to a whole night at once: Durga Puja treats other people's sleep as something to be deliberately spent, Kali Puja treats collective wakefulness as something that has to be actively maintained with jokes and firecrackers, and Tarakeswar reverses the logic outright, since there sleep is not an interruption of the ritual but the ritual's own technology. The medical material closes the circle by showing sleep doing diagnostic work on its own, standing in for a whole judgement about whether a person's life has come apart, well before any of the local systems it moves through agree on what has actually gone wrong.
None of this would mean much if the ethnographer's own sleep stayed outside the argument, so it does not. The doorbell, the heat, the ashram's four thirty mornings and the diary's slow surrender to the nap are in here as data, not as colour, because a claim about how sleep gets negotiated has no business exempting the person doing the negotiating.
It is worth being honest about what this collection is not. "Bengali" does a good deal of work across these pages even where the material itself keeps contradicting whatever that word is being asked to mean, and the pavement sleepers who open the whole account remain, throughout, an image rather than a voice, seen and interpreted but never actually consulted. Nothing here amounts to a survey, and nothing in it should be read as though it were one. What it offers instead is a set of well-chosen scenes, a handful of contradictions left standing rather than resolved, and a small store of details that keep nagging long after the visit that produced them. If that is a limitation, it is also, I think, the only honest method available for a subject that mostly happens with the observer's own eyes closed.
What sleep turned out to explain, in the end, was not itself, but one of the clearest ways to watch a household, a neighbourhood and a fieldworker all reveal how they distribute trust, hierarchy, care and suspicion among themselves, simply by settling, every single day, the same small and apparently private question of who gets to sleep, when, next to whom, and on whose terms.