The first time I saw a patient die in the hospital was during my third year of medical school, on the pulmonary ward. New to clinical rotations, I sometimes struggled to follow the plot on morning rounds. This particular man had been admitted with breathing problems related to chronic lung disease, and by appearances he did not seem much different than the other patients on the floor: old, tired, sucking hard against an oxygen mask. Our attending physician did not rouse him, though. She listened to his lungs and said to the senior resident, “Does his brother know? Call the brother.” Then, looking at the patient: “He’s dying.”
I checked on him throughout the day, watching his vitals dwindle on the paper chart outside the room where he lay unconscious among white cotton blankets, using his whole sweaty body to breathe. Nurses and technicians came and went, applying cold compresses and pulse oximeters. They had unplugged most of the monitors surrounding his bed, but the television was still on, tuned to something called the C.A.R.E. Channel, which cycled through wide panning shots of lakes and mountains, close-ups of wildflowers and farm animals, all set to a soundtrack of saccharine instrumentals.
This is the hospital death we’ve all been led to fear: a frail and lonely body breaking down among strangers and machines. Such scenes have helped propel the palliative care movement since Dame Cecily Saunders founded the first modern hospice in South London, in 1967. Interwoven among the various justifications for hospice — which may be framed in clinical, moral, or economic terms — is an aesthetic argument that the standard physical templates of healthcare are bleak, sterile, and dehumanizing. Efforts to facilitate dying outside the hospital are driven as often by a desire to avoid futile interventions as by a desire to escape the spaces in which they’re delivered. Design features that we accept when they are part of a patient’s recovery become abhorrent when reconfigured as the setting for death. 1
Hospice — as both space and practice — is thus reactionary by nature. It explicitly negates the institutional paradigm. As Stephen Verderber and Ben J. Refuerzo write, “Palliative architecture rejects the machine for healing, and rejects pure rationalism.” 2 Beyond that rejection, however, the design agenda is rather more variable. Early models of hospice, rooted in a traditional understanding of home, adopted domestic ideals of comfort and familiarity. More recently, we’ve seen spaces that employ dramatic design elements to facilitate a more symbolic engagement with the end of life. For example, the Hospice de Ark (1999), designed by Stan Neuhof in Roermond, The Netherlands, features an elliptical arrangement of inpatient rooms that’s meant to echo the shape of a womb; and in New South Wales, Australia, the Bear Cottage children’s hospice (2001), designed by McConnel Smith & Johnson, plays on childhood motifs with a treehouse-inspired “quiet room.” 3
The uptake of ‘evidence-based design’ suggests that the social project of staging death will eventually be subject to ‘best practices.’
This trend toward more formally ambitious hospice design is portrayed as progress toward a higher architectural mode. Notwithstanding Annmarie Adams’s observation that we know very little “about what constitutes an ideal environment for end-of-life care,” the narrative of formal progress reinforces the sense that such an ideal exists. 4 The uptake of “evidence-based design” suggests that the social project of staging death will eventually be subject to “best practices.” 5 In their survey of modern hospice design, Verderber and Refuerzo caution that there are “no magic formulas,” but they don’t hesitate to offer surprisingly specific guidelines for color (“Above all, avoid yellow and dull, bland hues in the hospice setting”), dimensions (“Private bedrooms should be 20-25 percent larger in size than the typical hospital room. … Provide interesting ceilings with recesses, barrel vaults, and indirect lighting”), and material (“Wood is of the earth, a tree grows with time, and its age rings are visible, symbolizing the change of season. … Wood therefore possesses therapeutic value in the palliative care experience”). 6
Herein lies a paradox. Hospice advocates have rooted their movement in the need to overturn institutional norms, and architects oblige by designing spaces defined by an antagonistic configuration with the status quo. But as this rhetoric gathers strength, it gestures toward a discrete formal endpoint, which puts hospice at risk of becoming similarly institutionalized. It should go without saying that there are no universally advantageous circumstances for death. The rooms that soothe one person may alienate another. As such, I submit that hospice design and practice are necessarily and perpetually unstable. Recognizing the variability in palliative architecture may ease some of the angst associated with the contemporary theatrics of “dying well.”
Death in the Hospital
That pulmonary ward where I watched a man die was in Detroit’s Henry Ford Hospital — in retrospect, a tidy distillation of the design stereotypes associated with 20th century hospitals. Behind the original 1918 pavilion building loomed a 1955 brick tower, its facade punctuated by a grid of small, unadorned windows, and capped with a large sign bearing Ford’s name in vintage script. I remember the interior being defined by narrow hallways and muted tiled floors. The hospital’s namesake, of course, was a pioneer of industrial efficiency and the innovator of an operational paradigm that reshaped American manufacturing and much of civic life. Albert Kahn, the architect of Ford’s Highland Park plant and River Rouge Complex, also designed the University Hospital (1925) in nearby Ann Arbor, where the logic of assembly lines was applied to the spatial organization of patient care. 7 Diagnosis and treatment were defined in modular, process-oriented terms, corresponding to the functional demarcation of space that Adams has shown to be an essential feature of North American hospitals at the time. 8
The archetypal mega-hospital was rational, technological, and highly visible in the urban landscape.
Ford’s hospitals were variations of the archetypal mega-hospital that proliferated in Europe after World War I: rational, technological, and highly visible in the urban landscape. Among the most famous is Alvar Aalto’s Paimio Sanatorium (1929), with its machine-like exterior and avoidance of unnecessary ornament. In healthcare architecture of this era, the new understanding of medicine as a science aligned with a modernist emphasis on clean lines, hygienic surfaces, and overt mechanical accoutrements. 9 As large hospital systems expanded over time, this linear modernism was overlaid with a labyrinthine quality, compounding an atmospheric cold with the risk of getting lost.
These hospitals still dominate the landscape in many European and American cities, despite what some critics see as a tendency to leave visitors depleted and unmoored. Ken Worpole recalls the “memories and mythologies” associated with an inpatient stay: “long corridors smelling of disinfectant; harsh neon lighting; wards and individual rooms which are over-heated and under-ventilated.” 10 Verderber and Refuerzo describe mid-century hospitals as “machines for occupation until death,” expressive of an authoritarian “culture of denial.” 11 Edwin Heathcote laments the emotionally empty architecture that diminishes the sanctity of life’s thresholds: “at the exact moments we are most in need of meaning and spiritual uplift, we find ourselves surrounded by the bleak expression of hygiene and efficiency. … Our existential gateways are manifested as service entrances.” 12
Before engaging with design alternatives, however, we should remember that this aversion is not shared by everyone. To the very ill who do not yet regard their diagnoses as terminal, a modernist, rational design can signify hope amidst uncertainty. Architecture that seems extreme can echo the promise of extreme therapy — of suffering validated by a radical cure — and many patients welcome the aggressive medical interventions associated with this setting. Palliative care advocates may argue their case on economic grounds, noting that 25 percent of Medicare dollars are spent in the last year of life. 13 But of course the last year of life is recognizable only in hindsight; rather more difficult to calculate are the Medicare dollars that rescued individuals from what might otherwise have been their last year.
Patients are often fighting for life right up until the moment they are readying themselves for death. For all its physical unpleasantness, then, the choice to die in the hospital can be regarded as intellectually and spiritually consistent. There are plausible (even admirable) explanations for this attitude. Surgeon Myrick C. Shinall, Jr., for example, has found that religious affiliation correlates with aggressive end-of-life interventions, and separately he theorizes that an ethos of Christian martyrdom might drive some patients to ignore physical discomfort in pursuit of miracles. 14 The modern hospital’s straight lines and stark surfaces aspire to longevity. To understand why some people choose to die in hospitals, we need only recognize that the exercise of making peace with death might be deliberately deferred.
Still, most people, given the option, say they would prefer to die at home. 15 Palliative architecture plays a limited role here, as the “home hospice” paradigm rejects the creation of a new space in favor of subtle modifications to a familiar one. Nurses visit the home to tweak medication regimens; maybe an adjustable bed is moved to the ground floor. Of course, the success of this model depends on the availability, competence, and composure of caregivers who can meet the patient’s changing needs. “At the end of life, things can fall apart quickly,” wrote Karen Brown, in a personal essay about her father’s painful death, published in The New York Times. “We were told a palliative expert would be at my father’s bedside if he needed it. We were not told this would be conditional on staffing levels.” 16
When dying at home isn’t possible, patients often turn to palliative spaces that are configured as home-like. Verderber and Refuerzo trace this motif to mid-20th century hospice prototypes adapted from private residences, and they show how the domestic inflection of these spaces was reinforced by the rhetoric of early advocates like Saunders, who focused on patients’ ability to resume the patterns of independent daily life. To these advocates, domestic hospice spaces were the perfect opposite of the mega-hospital: small, easily navigated, and familiar. 17
Annmarie Adams argues that the interpretation of ‘home’ in hospice centers often represents a stagnant, utopian, middle-class ideal.
And yet this familiarity may also prove constraining. Adams critiques the domestic template of palliative care spaces as a disguise that encloses death in a conventional envelope, allowing society to avoid a direct confrontation with mortality. She argues further that the interpretation of “home” in these spaces represents a stagnant, utopian, middle-class ideal that is inherently normative; it excludes “alternative interpretations of the home as sites of family violence, intense loneliness, repression, and financial hardship.” Patients whose lives have followed variant trajectories may find spaces designed in this style disconcerting, belittling, or even absurd. 18
Still, the trappings of domesticity remain a point of pride for many hospice operators. The Zen Hospice Project in San Francisco has converted a Victorian apartment building on a tree-lined street into a six-bed guesthouse. Promotional materials highlight the building’s large bay windows and high ceilings, the idiosyncrasies of its various rooms, and the cozy furnishings of its common spaces. The organization’s founder, physician B. J. Miller, gave a compelling TED Talk celebrating sensory delight as the highest priority at the end of life. At the guesthouse, cigarette smoking is permitted on the rear deck, and the menu can be readily customized to residents’ appetites. 19 But the architectural backdrop for this sensory exaltation is a willfully quaint space that neatly recedes into the surrounding neighborhood.
Domestic design templates also tend to force a compromise between the depth of feeling a space inspires and the breadth of its accessibility. The Zen Hospice Project, for example, follows a loosely defined ethic of mindfulness. Although the wall art is vaguely Eastern, and there’s a statue of Buddha in the garden, the organization itself is carefully non-denominational. As such, these objects lose their historical weight and become placeholders for platitudes about harmony or contemplation. A spiritual talisman shared among strangers can be only as powerful as it is inoffensive. In a house designed for guests, the arrangements are at best a safe approximation of what most people find pleasing, rather than truly resonant.
Death at the Vanguard
Increasingly, we see hybrid forms that, out of necessity or inventiveness, deviate from these two distinct templates: the mid-century mega-hospital and the middle-class home. St. Christopher’s Hospice (1967) in London, often regarded as the first modern hospice, featured an unexpectedly modernist façade that was likely a consequence of rigid National Health Service regulations. Yet in spite of its spare, institutional appearance, St. Christopher’s modeled a hospice practice that could be adapted to various settings. More recently, design elements from retail and entertainment spaces have been imported into traditional hospitals to blunt the industrial feeling of standardized healthcare. This new, consumer-friendly strain of hospital architecture deviates from the institutional stereotype and enables a different sort of inpatient death. 20
Increasingly, we see hybrid forms that deviate from these two distinct templates: the mid-century mega-hospital and the middle-class home.
Looking toward the future of palliative design, many advocates hail the model of Maggie’s Centres, established in the mid-1990s by writer Maggie Keswick Jencks, who confronted an eventually fatal diagnosis of breast cancer, and her husband, the architectural critic Charles Jencks. There are 22 centers so far, with more on the way, predominantly in the United Kingdom but also in Japan and Hong Kong. At each location, a prominent designer or firm is commissioned to tackle the perceived problem of architecture’s unsatisfactory engagement with terminal illness, and cancer in particular. Importantly, these are non-residential buildings. They are not hospices, but rather meeting places for cancer patients to gather knowledge and foster resilience. Despite this functional distinction, Maggie’s Centres are often held up as models for the entire field of palliative architecture, given their comparatively bold thematic interest in the experience of mortality.
Despite their relatively small physical scale, the buildings qualify as “high visibility, look-at-me architecture,” in Adams’s phrasing, by virtue of their famous designers, formal experimentation, and vivid metaphors. 21 The Centre at Dundee (2003), designed by Frank Gehry, is capped by an asymmetrically undulating metal roof and sited on a hill overlooking a river. In the foreground is a shallow “labyrinth” of grass and stone that Jencks describes as “an allegory for life … there are no dead ends — but you have to trust that you will find a route through.” The Centre at Fife (2006), designed by Zaha Hadid, is a dark, sculptural assembly of triangular planes, enclosing an unexpectedly bright and comforting interior, while the Centre at Swansea (2011), designed by Kisho Kurokawa, is designed as two interlocking crescents, sloping upward at their shared center, intended variously to invoke (again, per Jencks) a hurricane, a spiral galaxy, and yin yang. 22 The buildings in this series diverge markedly in individual form, but they converge with respect to the breadth of the architectural palette and the value placed on symbolism.
Collectively, Maggie’s Centres function as a cultural argument, that art and architecture should play a central role in mediating society’s engagement with mortality.
In The Architecture of Hope, Jencks outlines the mission of Maggie’s Centres in language both expansive and equivocal. The collective identity of these buildings is seen as transitional and genre-defying. According to Jencks, “the typical Maggie’s Centre can be seen as a kind of Non-Type … it is like a house which is not a home, a collective hospital which is not an institution, a church which is not religious, and an art gallery which is not a museum.” 23 Here again we see the tendency of advocates to describe palliative space primarily in terms of what it is not. This language is an expedient but ambiguous way to establish a positive affective valence while not committing to a distinct spatial definition. These self-conscious hybrids have a protean quality, which can be a beacon to disaffected patients of many kinds.
Cleverly, the sheer number of Maggie’s Centres diminishes the extent to which any one design speaks to the overall agenda. Jencks refers to this heterogeneity as “purposeful variety,” which makes possible dramatic and unconventional design choices. If a patient does not respond to a particular building, there may be another nearby that resonates. Consider the Highlands Centre in Inverness (2005), designed by David Page. The building is structured as two intersecting ellipses and its entrance flanked by two elliptical mounds, meant to signal, in Jencks’ words, “the metaphor of dividing cells, since they are crucial in the story of cancer as a form of rogue life.” 24 Inside, an exhibition of drawings states this metaphor more plainly. To find the Inverness design charming, as I admittedly do, requires a certain comfort with and receptivity to biomedical imagery. Yet for some patients, clearly, this symbolism could be more upsetting than nourishing.
One feature that unifies all the Maggie’s Centres is a commitment to progressive aesthetic discourse. Collectively, they function as a cultural argument, that art and architecture should play a central role in mediating society’s engagement with mortality. Design decisions are steeped in theory. Thus, the centrally located kitchen table is not just a table, but a deliberate manifestation of “kitchenism,” a neologism that Jencks defines as “the notion that one architectural meaning essential for self-help is a friendly, convivial atmosphere — with food and drink offered as you come in the door.” 25 This sort of ideological invention advances the argument that palliative architecture is a novel, necessary, and progressive social paradigm, in parallel with the palliative care movement as a whole.
Design decisions are steeped in theory. Thus, the centrally located kitchen table is not just a table, but a deliberate manifestation of ‘kitchenism.’
Much of the praise for Maggie’s Centres comes from stakeholders in this aesthetic discourse, designers and critics for whom symbolic architectural language is both interesting and legible. Yet design literacy is not universal. Jencks’s lament about “anodyne hospital art” and Heathcote’s objection to “kitsch plastic curtains” speak to individual predilections, bounded by class, education, and personal history. 26 If inclusiveness is truly a priority in palliative architecture, we should recognize that insisting on original artistic expression can marginalize those who do not share this background, just as others are unsettled by utopian domestic models.
Banality, too, is in the eye of the beholder. In How We Die, physician Sherwin Nuland puts the “death with dignity” assisted suicide movement on an equal plane with the aggressive end-of-life care model it rejects, in that both are desperate efforts to overcome the problem of mortality. Death with dignity, he argues, is an “expression of the universal yearning to achieve a graceful triumph over the stark and often repugnant finality of life’s last sputterings.” 27 But the presumption that sickness and death must involve spiritual uplift, let alone one that is deliberately mediated by art or nature, may be as much a cliché as cheap hospital curtains and mass-market flowers. Recognizing the commonality of our desires to stage our last acts properly can make any particular vision of those final hours seem not all that special.
The Flickering Hearth
I’d like to think I won’t care very much about the place where I die. I believe in the merits of palliative care but favor an unbraiding of its clinical, economic, and aesthetic justifications. As far as aesthetics go, I’m with Nuland — I’ve seen enough vulnerability at the end of life to know that a bit of grisliness is inescapable. Granted, I’m in my early thirties, and otherwise healthy, and in the lucky position of being able to regard my mortality in theoretical terms. I’ve also spent years on one big hospital campus or another, with the understanding that these spaces have a way of becoming progressively less offensive. Lately the whole hospice conversation seems burdened with enough nervous expectation that there’s something almost refreshing about the idea of sterile white walls, a blank enough canvas on which to sketch out the end.
Were palliative design ever to become truly standardized, its ubiquity would be every bit as depressing as the mid-century mega-hospital.
In a recent interview, physician and philanthropist Shoshana Ungerleider made light of the palliative zeitgeist: “Oh yeah. Death is hot right now. … People outside of medicine hear that I’m interested in end-of-life issues, and they’ll say, ‘Listen. Death is hot.’ Fantastic!” Hospice is now enthusiastically accepted in conversations about the future of biomedicine. At last year’s Exponential Medicine conference, sponsored by a Silicon Valley think tank, Ungerleider’s talk on embracing mortality was presented under the rubric “Redesigning Dying,” which might also have applied to any of the other lectures that day on circumventing mortality. 28 The contradiction barely registers; the anti-institutional stance has become its own sort of convention.
Palliative architecture will persist, of course, but to my mind, its power derives primarily from its unexpectedness. Every Maggie’s Centre is sited near an existing hospital, orbiting it like a satellite, as if to signal that one building type existentially depends on the other. 29 Logically so: hospice aesthetics are incoherent in the absence of a hospital with which to contend. Were palliative design ever to become truly standardized, as popular rhetoric seems so often to will it, its ubiquity would be every bit as depressing as the mid-century mega-hospital. In those final, needful hours, the most comforting hearths are those that feel serendipitously constructed, their warmth actively reclaimed.