VOLUNTARY REFUSAL OF FOOD AND FLUIDS:
ETHNOGRAPHIC VISUALIZATION OF A DEMEDICALIZED DEATH
Francesca Maria Nicosia
B.A., DePauw University, 2002
A thesis submitted to the
University of Colorado Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
Francesca Maria Nicosia
has been approved by
Jean N. Scandlyn
Nicosia, Francesca Nicosia (M.A., Anthropology)
Voluntary Refusal of Food and Fluids: Ethnographic Visualization of a
Thesis directed by Assistant Professor Sarah Horton
Despite patient preference for dying at home, death in the contemporary United
States typically takes place in institutional contexts framed by a biomedical
understanding and ordering of the dying process. Since the development of life-
sustaining technology, the technological imperative to maintain biological life has
contributed to the problem of death in America (Kaufman 2005). In response to
the medicalization of death, including medicines historic aversion to accepting death
as a the ultimate consequence of disease and old age, hospice and palliative care
have sought to bring death and dying out of the shadows and embrace the dying
person and their family by attending to issues of pain and suffering. However,
anthropologists have been slow to critically examine hospice and palliative care as
an alternative to the medical model of dying an alternative that nevertheless shares
many of the same philosophical assumptions about Western subjectivity, patient
autonomy and issues surrounding choice and control at the end of life.
In this thesis, I present a case study of an eighty year-old retired surgeon with
terminal cancer who chose to document his process of intentional cessation of food
and fluids at the end of his life. Dr. Michael Miller chose to stop eating and drinking
in order to achieve what he believed would be a peaceful and gentle death. Through
documenting his dying process on film with the intent to educate others about
voluntary refusal of food and fluids, Michael exerted control over his dying in
order to ultimately give up control and allow nature to take its course. Although
Michael believed he was having a natural death, his dying is an example of
controlling the means of death in order to have the time of social and biological
death coincide (Seale 1998). In addition, the agency exhibited by Michael as an
ethnographic subject by visually documenting his dying a process of narrative
creation which also thoroughly mediated his dying experience was an attempt to
create meaning from his dying that allowed him to maintain his sense of personhood
and gave a sense of purpose to his final weeks of life (Seale 1998, Norwood 2009).
This abstract accurately represents the content of the candidates thesis. I recommend
For Michael an extraordinary human being and teacher.
Many thanks to Karen van Vuuren, without whose direction and inspiration the
documentary film Dying Wish would not have been possible. I also wish to thank
Sarah Horton, Steve Koester, Jean Scandlyn for the guidance and freedom they
provided as my thesis committee. In addition, I want to thank Jim Igoe and Lakshmi
Fjord for their valuable insights and guidance at many points along the way. Thanks
to Guy Micco for his continued support and invitation to screen the film at the
Berkeley Life and Death Potluck Group, an experience that challenged me to more
clearly articulate my theoretical assumptions and issues of representation in this
research. To my fellow graduate students in the Department of Anthropology at the
University of Colorado Denver and all the people who have watched Dying Wish at
venues across the country and abroad in medical schools, hospitals, hospices,
academic conferences and classrooms, film festivals and private homes your
feedback and constructive criticism has been invaluable. Most of all, I wish to thank
Michael Miller and the Miller family for their bravery and willingness to expose
their most private and intimate moments in the public eye and the gaze of a fledgling
TABLE OF CONTENTS
Methods and Background...............................3
Narrating the Dying Process: Creating a Meaningful Life and Death. 16
2. MEDICALIZATION OF DEATH...................................21
3. A BRIEF HISTORY OF HOSPICE AND PALLIATIVE CARE............26
Interlude: The Last Supper..........................30
4. STOPPING EATING AND DRINKING AT THE END OF LIFE...........31
Sociocultural Significance of Food and Fluids.......33
Whats in a Name?: Voluntary Refusal of Food and Fluids...........35
5. TOWARDS A GOOD DEATH......................................43
Embodying the Discourse of Dignity and Quality of Life............44
Subjectivities of Pain and Suffering................46
Choice, Control and Patient Autonomy................52
6. CONCLUSION............................................... 58
Michael lay in the center of a large bed. The room is cast in
shadows except for the evening light streaming in through the
windows. I am standing in the doorway, watching a young
cameraman skirt the perimeter of the bed, capturing this intimate
moment. Michael is surrounded by his partner Jan, his brother
Rodney and two sisters, Joyce and Mardell, who tell him that its
okay to let go as they hold his withered hands and stroke his
wispy white hair. The frailty of Michaels body is made apparent
with each cycle of the rise and fall of his chest, the quiet sound of
his breath barely filling the room. Whats that? Joyce asks
Rodney, as his fingers rest on the inside of Michaels right arm,
locating his pulse. Its the brachial artery, responds Rodney.
Usually you cant see it because people have more .fat.
Today is the thirteenth day of Michaels fast that will soon end
with his death (field notes, September 2007).
This is an ethnographic case study of one mans voluntary cessation of food
and fluids at the end of his life. Dr. Michael Miller, an 80-year old retired thoracic
surgeon with end-stage pancreatic cancer, chose to stop eating and drinking in order
to have what he believed would be a peaceful and gentle death. In 2005, Michael had
been diagnosed two years prior with pancreatic carcinoma which had metastasized
throughout his lymphatic system and three vertebrae. After surgery and one year of
curative therapies, Michael researched the probability of his recovery and decided to
forgo additional treatment. During the next year, he continued to live what he
described as his normal life, including being a vigorous hiker, a point in which
he took great prideevidenced by a scrapbook of pictures of him at the Grand
Canyon and a binder full of the local newspapers weekly features of area parks and
Michael said he knew early on that he did not want to die in the hospital;
although he had spent his career in the operating room, he felt that the hospital is a
stranger. This sentiment is reflected in the considerable attention paid to the
medicalization of death, which has created what Kaufman (2005) refers to as the
problem of death in the United States. Her words are worth quoting at length:
There is a widespread demand to have control at the end of life
through having greater choice. This presents a paradox, for medicine
both provides and constrains that choice. It offers an abundance of
treatments. But it also organizes perhaps in the hospital most
dramatically how one can know the problems of the body, what
makes a person alive or dead, what the role of the family should be,
and the relations among patients, families, and health professionals. It
declares what is important to know and what needs to be imagined.
Then, it ignores other, perhaps equally important things. Thus, it
produces particular forms of dying (9).
Michaels death stands in stark contrast to the types of technologically mediated and
problematic hospital deaths discussed by Kaufman;1 indeed, it can be viewed as a
form of resistance to the dominant biomedical model of dying. Michaels voluntary
1 Kaufman (2005) notes that an equal number of hospital deaths she observed were unproblematic (8),
but given the media attention to and national discussion of the problem of death, her ethnography
focuses precisely on these types of deaths, their structural underpinnings and consequences for dying
patients and their families.
refusal of food and fluids was a way for him to align the timing of his social and
biological deaths (Seale 1998, Norwood 2009) while creating a meaningful narrative
of his dying process. Despite the conflation of selfhood with an agentive body,
Michael asserted his agency even as his body increasingly deteriorated. However,
despite the active steps Michael took to have a non-medicalized death, I show that
his notion of a good death nevertheless relies on the same philosophical
assumptions about Western subjectivity, agency and the importance of patient
autonomy, timing and control at the end of life that shape hospital deaths as well.
Methods and Background
In August and September 2007,1 conducted fieldwork and co-produced a
documentary film (van Vuuren and Nicosia 2008) in Michaels home. The film
Dying Wish and subsequent ethnography of Michael came into existence because
Michael had asked his hospice to inquire if there was someone who was interested in
documenting his decision to stop eating and drinking. Michael was adamant about
his wish to die with dignity and minimal suffering. Through his own research,
Michael had become convinced that when he was closer to death, voluntarily
stopping eating and drinking would afford him a gentle and peaceful end. The
issue of what constitutes a natural death is beyond the scope of this paper,
however, Michael continually spoke of his choice to stop eating and drinking as
hastening his natural death.2 3
Like many medical professionals familiar with the inner workings of the
hospital and the technological imperative to sustain life, Michael had seen too many
patients, with tubes down their windpipes in their last days, and was convinced that
stopping eating and drinking would afford him the type of death he envisioned.
Given biomedicines historic aversion toward death, its inadequacy in dealing with
dying patients and the publics misperception of the effects of stopping nutrition and
hydration of a dying body, Michael took an active part in recruiting us to document
his final weeks of life. He wanted others to see in both the visual and cognitive
sense what an alternative to the typical institutional deathbed scene in the United
States might look like.
In January 2005, two and a half years before he stopped eating and drinking,
Michael contacted his extended family by email to let them know of his eventual
2 Seymour (1999,2000) found that families of patients who died in intensive care units in the United
Kingdom believed the patient had a natural death if it had been an expected, versus sudden, death.
3 In contrast, his hospice physician was adamant that Michael was not hastening his death, but rather
was not prolonging his dying process. This semantic difference is perhaps an artifact of the
physician being an official spokesperson on camera for the philosophy and public assurance that
hospice and palliative care neither hasten nor delay death. In fact, it could be argued that by virtue of
being the subject of a documentary film, Michaels natural dying process was in many ways not
only mediated but extended by the filming process, as discussed below.
plan. His nephew Scott, a Lutheran minister, reflected upon receiving that email:
From that time until now, I have just been in awe of [Michaels] decision [to stop
eating and drinking] and its got me thinking about alternative ways of dying,
because we arent given many alternatives in the media and churches and in society
in general. Michael also contacted his hospice well in advance of being enrolled as
a patient, and informed them that he would at some point stop eating and drinking in
order to hasten his natural dying process.4 By August 2007, Michael had reached the
point in his disease where he was ready to begin the process of allowing his body to
shut down. After a conversation with his hospice, their communications director
contacted the local community radio station, where Karen van Vuuren, a former
hospice colleague of mine, was working. Karen then asked me to join her in an initial
visit to Michaels home because of my interest in and experience with end-of-life
care and graduate studies in medical anthropology.
Karen and I first met Michael and his partner Jan on a Thursday evening in
their home. Michael greeted us at the door, in his signature slacks and collared shirt,
and we gathered around the large black lacquered dining room table. The living
4 In the course of my research, I have found that many people are unaware that hospice is not a
single nationwide entity that provides end-of-life care. In 2008, the United States had 4,850
organizations certified to provide hospice and palliative care service. These include both in-patient
and home-based services and range from small (serving less than 50 patients at a time) to large
(serving more than 1,500 patients a day) organizations. Hospice and palliative care organizations
include private for-profit (49.6%), non-profit (46.2%) or government-run (4.2%) programs and can be
ffeestanding/independent (57.5%), part of a hospital system (21.8% a 1% increase from 2007), part
of a home health agency (19.4%) or part of a nursing home (1.4%) (NHPCO 2009:8-9).
room was filled with plants, artwork and books. I had the sense that the Millers had
transplanted the contents of a much grander and more spacious home into this non-
descript residence after Michael had retired. The atmosphere was one of both
excitement and mutual skepticism. While working for a small non-profit hospice,
Karen and I had both cared for a number of individuals who had stopped eating and
drinking at the end of their lives, whether by decision or default. We were curious
and concerned about the intentions of this man who wanted to expose such a private
and intimate part of his life to public scrutiny. In turn, Michael was curious about our
credentials, opinions and approach. Jan was obviously uncomfortable with the
situation not necessarily Michaels decision to stop eating and drinking but rather
having strangers in their home to document it.
We had initially planned on producing a radio documentary. But, largely due
to Michaels excitement, the project took on a momentum of its own. Two days after
meeting Michael, we had assembled a small film crew consisting of two
undergraduate film students using borrowed cameras, microphones and a makeshift
light kit. Filming began the next Saturday and Michael stopped eating and drinking
the following Tuesday. He was ready to begin his fast yesterday, but in the interest
of the films storyline, we asked him to postpone stopping eating and drinking until
we were able to schedule a time to film him in conversation with his hospice
physician and family. In addition, time was needed to obtain an expedited human
subjects review from the Colorado Multiple Institutional Review Board. Michael
easily agreed to this, and said that since he had already been waiting long enough to
begin his fast, a few more days would not matter.
We were present in the Millers home for varying lengths of time for thirteen
of the next sixteen days prior to Michaels death. Although Michael had given us
full access during this period, the reality of being present in his and Jans home
posed a number of challenges. Most important was the need to respect Jans space
and desire not to be filmed. Although she supported Michaels decision to stop
eating and drinking and understood his desire to document the process, she would of
course have preferred not to have him dying, let alone have it filmed. She reluctantly
welcomed us into their home, and finally agreed to be on camera lest Michael be
portrayed inaccurately as someone whose life partner of twenty years was absent
during this important life transition.5 She explained that its ... difficult for me to
talk about and it will continue to do so as time goes on and he becomes weaker its
going to be more difficult for me to discuss this. Once Michael became confined to
his upstairs bedroom and then later became unresponsive, Jan, along with Michaels
5 Both Michael and Jan were previously married and each had children with their former spouses.
Jans first husband also died from cancer, which only added to the emotional intensity of the situation
younger brother Rodney, a retired geriatric physician, became our primary source of
information on how Michael was progressing, or declining, as it were.
During the first two weeks of fieldwork, we conducted formal and informal
interviews and filmed b-roll with Michael and members of his family.6 We also
filmed interviews with Michaels hospice physician, private home-health aides and
outside experts unaffiliated with Michael or his hospice,7 including an ethicist-
physician, hospice chaplain, sociologist and an end-of-life education specialist. All
taped interviews and other video footage were transcribed verbatim. At the end of
each day I typed up my hand-written notes and additional reflections and
observations. As Michaels fasting progressed, interviews increasingly became less
formal in exchange for a more conversational and observational approach. Julia
Lawton (2000) writes that in her research setting in an in-patient palliative care
facility in the United Kingdom,
formal interviews not only seemed to be too obtrusive to many
patients and their families; in a substantial number of instances
6 B-roll refers to supplemental footage that is captured to intercut with primary interview shots.
7 Michaels hospice physician was filmed under the condition that the hospices executive director
was present during the interview, ostensibly to ensure that his statements were congruent with the
hospices philosophy and policies and would not be construed as supporting euthanasia practices. We
were not permitted to film the hospice nurses who came to the home periodically to check on Michael
and bring medication and supplies. Once Michael was bed-bound, a private home health agency was
hired to provide physical care such as turning and bathing (24-hour caregivers are not covered under
the Medicaid home hospice benefit). These caregivers were given permission by their agency and
agreed to be interviewed and filmed.
they simply were not viable. Some patients, for example, were
heavily sedated during their stay in hospice, whilst others
experienced changes in their mental state, such as becoming very
paranoid or confused. It was, of course, impossible to interview a
patient in a coma (30).
Similar to Lawtons observation of patients in comas, once Michael was
unresponsive, we continued to observe and film him but also spent more time
interacting with and interviewing his family, who were at that point speaking both on
his behalf and about their own experiences.
Lawton (2000) also discusses the shrinking of social worlds experienced by
patients in an in-patient palliative care facility. She explains the sense of loss and
tireless efforts by hospice staff to entice patients to utilize a day room with activities
and the potential to socialize and interact with others. However, most patients were
not interested in socializing, to the dismay of staff (and, it appears, to Lawton).
Michael also experienced and articulated spatial shrinkage. During the three
weeks of fieldwork and filming, Michaels world was confined to his house and
backyard. On the fifth day of his fast, Michael contrasted his current state to his
young grandnieces running through the late summer rain in the front yard, exhibiting
the exuberance and vigor of lives still to be lived. He said, My physical world is 8
8 Like much of Lawtons book, her discussion of shrinking frames the issue in a negative light. While
perhaps the observation needs to be reiterated, is it really surprising or cause for concern that peoples
world would shrink as they are dying? Indeed, the fact that patients experienced spatial shrinkage
seems less problematic to the dying person than anyone else.
shrinking. But we still have this cognitive world to enjoy. Emotional, spiritual,
beautiful, pointing toward a different conception of space and an agentive body than
implied by the pejorative connotation of shrinking. I do not mean to propose that
all people who die in their own homes or from demedicalized deaths will have the
same experience of spatial or bodily reality as Michael, particularly given the varied
and potentially stressed domestic and familial contexts that occur as a result of
home-care (see Brown 2004). But I simply want to point out that shrinkage is not
universally experienced as a negative loss.
As Michaels case shows, not all palliative care patients experience the
deterioration of their bodies and sense of self in the same manner. Michaels dying
challenges conceptions of agency that assume an acting body. Desjarlais (2003)
writes that the end of the body experienced by dying people can for some, be
displaced onto words, or in Michaels case, onto film, when There [i]s little left to
do but die (317). While the decline of Michaels bodily agency would inevitably
result from his decision to stop eating and drinking, through his conscious decision
to move closer toward the physicality of his death, the loss of faculties and ability to
act was in fact being reclaimed as his last act of agency. His eventual death and the
loss of all the physicality and sociality that entails, was for him, not a loss of agency
but precisely its enactment.
Michaels identity of self was not tied to the body per se, but rather his
diminishing physical abilities and his bodys shutting down had created another
means for him to maintain his identity as a physician and teacher. At times during
my three weeks in Michaels home, he identified himself with his mental faculties
and capabilities; once these abilities were slipping, he never distressingly lamented
the transition into another state of being, corporeality or consciousness, but rather
embraced the transitions out of curiosity and desire to document them as an
experiment (made possible by our presence) in watching his natural dying process.
In conducting research with people who are dying, further discussion of
methodological and ethical issues is necessary. Protection of research subjects is
important in any ethnographic endeavor, particularly with populations who are
deemed to be vulnerable.9 Lawton (2000) points out some of the practical, moral
and ethical problems involved in conducting research amongst palliative care
populations (30) including issues surrounding consent. Because of fatigue, physical
discomfort and emotional stressors, Lawtons methodology consisted largely of
informal conversation techniques and direct observation, while also acting as a
9 Given Michaels agency and direction in documenting his dying process, 1 never considered him to
hospice volunteer. The approach of volunteering in order to gain access and trust
with hospice patients and staff seems to be common amongst researchers in hospice
settings (see also Russ 2005). Others who have conducted ethnographic research on
dying within hospital settings have tended to take a more traditional participant-
observer role, including having access to medical charts, physician rounds, staff
meetings and family conferences, which perhaps is reflective of the hierarchy and
structure of hospital culture (see Kaufman 2005, Lock 2001). Frances Norwoods
(2009) approach to studying euthanasia in The Netherlands included observations,
interviews and surveys with researchers, physicians, huisarten (literally, physicians
of the house or home) and 650 of their general practice patients. In addition,
Norwood conducted 25 intensive case studies of patients at the end of their lives;
seventeen of the patients she followed died at home, but they also died in assisted
living facilities, nursing homes and acute-care facilities (9-17). The ideal death was
more accessible at home because
In the Dutch home and in the context of euthanasia discussions dying
individuals are encouraged by huisarten to process what it means to
die in the company of their family and their huisarts. Emotions are a
large part of the experience of dying and in euthanasia talk, fear and
uncertainty two emotions frequently expressed by those who are
dying can be socially processed and culturally addressed (166).
Because research took place in Michaels private home and we were clearly
present in the roles of filmmakers and researcher, it was unnecessary to be present
under the auspice of hospice volunteers. However, my own professional experience
as a hospice worker and the insights I gained from that period which included
caring for a number of individuals who also chose to stop eating and drinking in the
days and weeks leading up to their deaths certainly inform the conclusions I draw
from the ethnography presented here. Given my previous work in hospice and the
intimacy of the setting, it was impossible to maintain strict boundaries between my
roles as filmmaker, ethnographer, caregiver, and on the most basic level, as a human
being experiencing a particularly intimate and emotional part of another persons
life. My experience very much echoes Kaufman and Franklins (2005) observation
that anthropological engagement with death and dying is a case where the boundary
between native and stranger is erased; the boundary between work and life is
A few days before he began his fast, we asked Michael why he wanted to
document his stopping eating and drinking. He answered,
I want this to be documented because of the, uh, potential, potential
to, for innumerable people with terminal diseases to hasten the onset
of death, gently... gently. A compassionate way of, uh, easing into
death. The reports indicate that some even feel a bit better as they
stop drinking and eating and, uh, go to death within a few days .. .
several days.... Frankly, I am hoping that this [film] will be the
greatest contribution Fve made to ... how shall I state it. .. yeah,
this will be a more important and a bigger contribution than I made to
my decades of medicine in the past.10
In entering into this visual ethnographic endeavor, we elicited a series of narratives
from Michael, which he had not anticipated would take the form of a life review in
the process of crafting a narrative for the film. He joked with us, saying, I thought
people would just want to watch me die. We said we could not just show a
voyeuristic deathbed scene and needed to put his life and death in context. Thus,
Michaels dying was shaped by the filming process and our interactions in ways that
are somewhat difficult to distinguish.
On the fifth day after he stopped eating and drinking, Michael was mobile
and lucid. However, we discovered that he had on occasion been taking Ritalin in
order to maintain alertness during on-camera interviews. Initially, I was concerned
that he was unnecessarily taking stimulants to counteract the side effects of the high
doses of narcotics he was taking for pain simply because we were filming and he
wanted to put on a good show, so to speak, potentially contaminating the data.
Would I decide to include Michaels Ritalin in later versions of his story? As it
turned out, Michael had a prescription for Ritalin long before we showed up on the
10 Because the clarity of Michaels speech varied over the weeks I spent with him, I have taken the
liberty of editing out some of the longer pauses, urns and repetition of words or phrases for ease of
reading. However, I have tried to accurately portray and retain the general feel of his speech as his
scene. To some, his use of stimulants would negate the notion of his having some
kind of natural dying experience. However, I came to interpret his pharmaceutical
mediation as simply another attempt to exert control within a situation that
threatened the dissolution of his personhood and ultimately his life.
Along the same vein, because we were creating the documentary at
Michaels behest, to what degree would the film, and later my writing, include the
potential messiness of dying (Cooey 1999) and family dynamics? Although
Michaels death was in fact easy and peaceful with little or no family conflict, which
is not always the case, I am wary of idealizing or sanitizing his experience.11 One
morning we called the house to check in regarding what time to arrive, and were told
that today isnt a good day because Michaels daughter had flown in from out of
state and she did not want us to be there. We later discovered that aside from the fact
that her visit was the final time she would see her father, she was in the midst of her
own treatment for cancer and having a film crew present was simply not an option.
11 Numerous factors including socio-economic status, education level, gender and family support (or
lack thereof) will necessarily affect a persons dying experience. Although Michael fits into a highly
privileged demographic, his experience remains valuable as an atypical example of the way most
Americans die. During the discussion following the presentation of Dying Wish at the 2008 National
Hospice and Palliative Care Organization conference in Dallas, Texas, one of the audience members
criticized the film for potentially being perceived as an idealized death of an educated, upper-middle
class white male, which is a critique I share. However, as a case study, the strength of the film is not
in its advocacy for a particular, often unattainable, experience of dying, but rather that it serves as a
catalyst for discussion about the potentials and constraints of choosing the manner in which one dies
in the United States.
Engaging in biographical ethnography brings with it particular challenges of
representation including a heightened sense of responsibility to the person at the
center of the analysis, the above story a case in point. One of Michaels close friends
was an anthropology professor in another state; Michael had audited a number of
courses taught by his friend and suspected his choice to stop eating and drinking was
of inherent anthropological interest. Periodically, Michael would pose the question:
So, what are you going to write about? to which I would jokingly reply, Well,
youre not dead yet, so I cant say exactly. But the weight of being Michaels
ethnographer bears on my conscience and reminds me of Biehls (2001) remarks that
being Catarinas anthropologist provided a way into present human conditions,
ethnographys core object of inquiry (14).
In his ethnography of two Yolmo Buddhist elders and their deaths, Desjairlais
(2003) has shown that
... by attending carefully to how a person or two within a specific social
setting live out and make sense of their lives, anthropologists can
effectively address the ways personal and interpersonal concerns of
individuals relate to social and cultural processes and so develop more
precise and more integrative understandings of what it means to be a
person, to live a life, to relate to others, to have a body, to be conscious
in time (4-5).
In this way, I hope that Michaels life and death can be instructive. His dying can be
situated not only within the anthropological literature on the medicalization of death,
hospice and palliative care, but also points toward larger cultural and structural
determinants that shape the way in which people may experience subjectivity and
(dis)embodiment at the end of life.
Narrating the Dying Process: Creating a Meaningful Life and Death
Norwood (2009) discusses how euthanasia talk in The Netherlands is
primarily a discursive space that rarely results in euthanasia deaths, instead providing
a way for terminally ill patients to create meaning and maintain their sense of
personhood. Similarly, Michaels motivation for filming his dying process was
driven by his desire give purpose to his final weeks of life congruent with his identity
as a physician. Additionally, similar to Rouses (2004) concept of embodiment-by-
proxy,12 the filming itself served to maintain his personhood and social identity
despite his biological decline. Although Michaels intentions in creating the film
were certainly more altruistic than narcissistic, his social identity would be
maintained long after his death, much like effects of increasingly popular virtual
memorials (see Roberts 2004). Michael said that the time spent with family and
friends leading up to and after he stopped eating and drinking was like attending my
own funeral or memorial service.
12 Rouse (2004) gives four definitions of embodiment-by-proxy to explain the ways in which
parents and professionals attempted to change the others dispositions regarding the medically futile
care of Jasperlynn, a critically ill child of poor African American parents whose value and meaning
were questioned by most involved with her care. Rouses use of embodiment-by-proxy most
congruent with Michaels experience is that the film denotes the potential to alter the perceptions of
the target audience as well as the perceptions of the deploying agent (523).
In some ways, Michaels situation is congruent with Robert Murphys
(1987) reflection on his bodily deterioration resulting from a spinal tumor and
subsequent paralysis. Certainly, their socioeconomic and educational similarities are
reflected in their ability to create a reflexive textual or visual document of their
bodily deterioration, ultimately leading to their deaths. However, Michaels
reflection on his life and dying lacked the bitterness that permeates Murphys
writing, perhaps an artifact both of their disease trajectories, age and personalities.
Yet common to both Michael and Murphy was the desire to create meaning in the
face of their mortality.
The importance of narrative as a means of creating meaning out of illness and
disrupted lives is well understood (Becker 1997, Mattingly 1994, Kleinman 1989). In
her discussion of emplotment as a means of creating a plot out of a succession of
actions with inherent social value, Mattingly (1994) writes that ... we are
motivated, as actors, to create stories while in the midst of acting (812). Although
the ultimate ending to Michaels story was known, the narrative time was organized
within a gap, or as Mattingly puts it that movement toward endings
dominates the experience of time. Michaels agentive control of his own dying
process was an attempt at emplotment of meaningfully moving his narrative
towards a specific ending. While Mattingly focuses on how occupational therapists
create a narrative plot which structures the healing processes of their patients,
Michael attempted to assume control of his own narrative trajectory. Yet neither the
timing of his dying nor the emergence of his storys narrative could be
predetermined despite Michaels attempts to control the situation, for life in time is
neither predictable nor highly controllable (813).13
By documenting his dying process on film, Michael engaged in what Clive
Seale (1998) discusses as a form of therapeutic discourse, creating dialogue about
cessation of food and fluids as one alternative to the medicalization of death. In her
ethnography on euthanasia in The Netherlands, Norwood (2009) places the discourse
of euthanasia talk within Seales framework as a way for dying people, their
families and Dutch society in general to have an open dialogue about death and the
social value of those who are dying. Although Michael placed his choice to stop
eating and drinking in opposition to euthanasia, which is illegal in most of the United
States,14 his decision and the discussion about it that filming provoked functioned
13 Mattingly (1994) outlines six stages of narrative time which also overlap with Michaels
documentation of his dying process: the configuration of narrative time; action and motive as
important structuring devices; the organization of narrative time within a gap; narratives show how
things (and people) change over time; the dramatic nature of narrative time; and the uncertainty of
14 Euthanasia, also referred to as physician-assisted suicide (PAS) or physician-assisted dying (PAD),
is illegal in most of the United States. Exceptions include Oregon the Death with Dignity Act
originated in 1994 as a citizens ballot initiative authorizing the self-administration of lethal
medications and was implemented in 1997 (ODHS 2010); Washington the Washington Death with
Dignity Act was passed as a ballot initiative in 2008 (WSDH); and most recently Montana in 2009
the Montana Supreme Court issued a decision in Baxter V. Montana that protects physicians when
prescribing lethal medication to competent terminally ill patients but did not address the
constitutionality of death with dignity found in a lower court (Johnson 2009).
in much of the same way as euthanasia talk in The Netherlands.
Norwood (2009) observed that most people who initiated a discussion with
their physician about euthanasia do not actually proceed with obtaining a lethal
prescription and those who obtain the prescription often do not take it. Instead,
euthanasia talk is a way for dying people to maintain their sense of personhood
and create a meaningful narrative within the context of their family and broader
Dutch society. By creating the documentary film, Michael engaged in a similar
process of giving meaning to his last weeks of life, enlivening his sense of
personhood and creating the space for discussion and meaning-making within his
family and close circle of friends. Through the film, which has a life of its own,
Michaels social identity continues on, creating the space for a broader dialogue
about the nature of death in America and the cultural values and structural forces
which enable and constrain choices at the end of life.
MEDICALIZTION OF DEATH
Beginning with Hertzs (1960 ) writing on social death, anthropology has long
had an interest in death, funerary ritual, grief and bereavement as culturally mediated
phenomena (see also Mauss 1979, Palgi and Abramovitch 1984). In light of the
decline of ritual in daily Western life, Gorer (1965) found a societal trend toward
avoidance of the bereaved in the United Kingdom. Around the same time, ground
breaking research on death and dying in American hospitals was underway. Glaser
and Strauss (1968) Time for Dying detailed how the dying trajectory was shaped
not only by the disease process but also by the structure and form of medical care
and communication in the hospital. In Passing On, David Sudnow (1967) also
discussed how the perceived social value of patients affected the way they were
treated by medical staff, specifically with regards to resuscitation.15
Psychiatrist Elizabeth Kiibler-Ross (1969) seminal book On Death and
Dying followed over two hundred patients who were dying in the hospital. Kiibler-
Ross placed the dying persons psychological experience at the center of analysis
15 See Rouse 2004 on contemporary issues of worthiness and life-sustaining procedures and
technology in marginal populations.
and outlined five stages of grief.16 In addition to pointing out that nearness to death
does not obliterate personal identity, the essential quality of being human (Kaufman
2005:69), Kiibler-Ross urged physicians to assist their dying patients in cultivating
compassion, forgiveness and engaging in a life review in the hopes of having a
positive dying experience. Michael recalled meeting Kiibler-Ross in the 1970s, an
experience he contrasted to his general education about death as a physician:
My awakening my awakening to the subject [of death] was a
lecture by Elisabeth Kiibler-Ross ... either she or Martin Luther
King are the most powerful speakers Ive ever encountered.... this
gives some insight into the unfortunate paucity of insight on the part
of physicians. This was a conference which was nicely publicized
and ... was attended though by only two physicians, another
physician-intern and I, [but it was] attended by scores of nurses,
social workers, clergy. It was interesting because after the lecture we
broke into groups.... We had in our group a priest who asked,
Who do you think I encounter with the most apprehension about
death? Well, to my surprise it was nuns! I dont understand why
theyre in fact not eager to depart to heaven but it seems not to be
true. It seems that most, all, adhere to, or wish to cling to some life,
even though it merely becomes sort of a biological existence.
Kaufman (2005) also points out that Kiibler-Ross theory of dying as an experiential
process versus a singular death event contributed to dying becoming a process
that was capable of being not only observed and monitored but also expressed
16 Kiibler-Ross five stages of grief are: denial, anger, bargaining, depression and acceptance. The five
stages have been criticized as overly determinant. Michael said he skipped the first four stages and
went straight to acceptance.
through supportive hospice care (70).17 In addition, the interior world of emotions,
psychic need, and spirituality came to be an important aspect of optimally organized
dying, mapped and managed by medicine (70).
Since the advent of medical and technological innovations such as
cardiopulmonary resuscitation, mechanical ventilation and artificial nutrition and
hydration, the dynamics of hospital deaths have changed dramatically. The ways we
think and talk about the dying process and the dying person and the ways we
prepare for death, stave it off, and respond to medical treatments for those near death
are all shaped by social institutions and bureaucratic practices (Kaufman 2005:2).
Gone are the days when patients are simply allowed to linger in the hospital until
death arrives (Glaser and Straus 1968). Instead, the structure of hospitals and
insurance companies now require that patients either be discharged or death be
moved along in a timely manner (Kaufman 2005). Whereas death in the West was
once the purview of the clergy and philosophers (see Aires 1974), the dying process
has now become firmly ensconced within the domain of medicine whether
understood as art, science or bureaucratic form, and is currently the most powerful
framework for understanding the body, the person, dying, and most important, what
to do as death approaches (Kaufman 2005:332). Prior to the advances in life-
17 Kaufman (2005) points out how, In the 1970s birth and death were replaced by birthing and
dying newly conceived socio-medical processes in which the subjective, experiencing person
became the new object of medical knowledge and work (68).
sustaining medical technology, death from a prolonged illness frequently occurred
as a result of dehydration or malnutrition rather than from the disease process itself
(Meares 1994:10). In these instances, there was in fact nothing to do but wait.
Kaufman (2005) brings to light the fallacy of choice at the end of life within
the hospital due to the entrenched pathways into which patients and their families are
forced to navigate but most often do not know what to do. There exists a lack of
attention to the hidden knowledge of medical professionals with regards to what
happens at the threshold between life and death. The knowledge they possess of the
hospital structure, insurance company requirements and potential treatment outcomes
- stands in sharp contrast to what patients and families experience and grasp when
they enter the hospital and confront critical illness (9). In contrast to the average
American, Michael had the advantage of being his own medical expert and possessed
the hidden knowledge that enabled him to know exactly what to do avoid the
hospital to allow nature to take its course.
In their review of The Anthropology of the Beginnings and Ends of Life,
Kaufman and Franklin (2005) write, Anthropologists have often used the margins
of life as a site for examining the making and unmaking of persons and relationships,
social and corporeal bodies and life itself (318). Recent attention to issues of death
and dying has shifted the focus from classic structural-functionalist or symbolic
interpretations of death to analytic frameworks of biopolitics, personhood and
subjectivity in order to investigate issues surrounding the medicalization of the end
of life, the technologization of life itself, social death and abandonment, brain
death and organ transplantation (for a review of recent anthropology on the end of
life, see Kaufman and Franklin 2005). Ultimately, these phenomena point towards
shifting meanings of what it means to be human and to be alive or dead (Kaufman
and Franklin 2005:320). Although the stakes are perhaps higher or more dramatic in
situations of highly medicalized deaths, it is surprising that little ethnographic
attention has been paid to demedicalized or low-technology death in the same way
anthropologists have approached the natural childbirth movement (for example, see
Davis-Floyd and Davis 1996, Davis-Floyd 2001). The following section discusses
the history of the hospice movement as an alternative to the over-medicalization of
A BRIEF HISTORY OF HOSPICE AND PALLIATIVE CARE
Like the abstraction of death with dignity, hospice as is
commonly understood has also become and abstraction, the
quintessential, metaphorical place for the reduction of suffering,
fulfilled dying, and ultimate individualism all values articulated
by the educated middle class (Kaufman 2005:71).
The modem hospice movement began in the late 1960s in response to medicalized
and institutionalized death in the United Kingdom and the United States. As
volunteer efforts, hospices provided an anti-establishment model of death care.
Hospices sought to treat the dying and their families with a more compassionate
comfort care approach that addressed the physical, emotional and spiritual aspects
of dying. Hospice emerged in the same period as the acceptance of Kiibler-Ross
(1969) five stages of grief, the development of symptom management techniques
that could be used outside hospital settings, and perhaps most importantly, the desire
for individual control over ones dying process (Kaufman 2005:70) which was fueled
by the growing consumer rights movement in response to paternalistic physician
control over health care decision-making (Starr 1982).
Michaels hospice physician, Dr. Stark, originally trained in urology, was
asked to describe how hospice approaches the dying process differently than the ICU
or hospital. He replied,
I think the main difference, generally speaking ... is that were more
accepting of [death]. A hospital is geared to fix stuff and [hospice] is
counter to the culture of most of what is done in the hospital to let
go. Philosophically, people are going to be of a different mindset -
the actual care that could be delivered could be very similar.
Although initially viewed as a counter-cultural movement in opposition to the
medicalization of death,
Hospice soon became a means for combining modem medical knowledge
and practice with an institutional response to the expression of individual
need and with an imagined tradition of family and community that was
considered an alternative to the growing bureaucratization, patient
alienation, and use of technology found in the hospital (Kaufman 2005:70).
Although palliative care and hospice are often used interchangeably, for claritys
sake, I refer to palliative care as the medical aspect of pain and symptom management
associated with terminal illnesses that can be used simultaneously with curative
treatment, while hospice refers to the holistic model of care for the dying and their
family that includes attention to physical, psychological, spiritual and social issues.
Central to both models is symptom relief and the palliation of pain and suffering, which,
for physical pain, is predicated on access to pharmaceutical opiate-based analgesics.
Aside from physical pain, hospice attends to other forms of suffering such as spiritual
and existential suffering and increasingly to intractable terminal suffering (Quill and
Byock 2000), which I discuss in greater detail below.
The original modem hospice, St. Christophers Hospice, was founded in London in
1967. Because of differences in the cultural and economic histories and structures of
medical practice, hospice and palliative care in the United Kingdom is different in
practice than its counterpart in the United States (see Lawton 2000). Despite its initial
anti-establishment philosophy, palliative care in the United Kingdom became
bureaucratized and routinized in its articulation with the National Health System
(James and Field 1992, Lawton 2000). However, the division of labor of care in the
United Kingdom is currently undergoing a process of (re)domestication and the
provision of home care is increasingly regarded as the gold standard for the
organisation [sic] of care, in institutional and domestic contexts (Exley and Allen 2007,
see also Higginson, Astin and Imperial 1998).
Within the United States, palliative care has recently gained acceptance within
biomedicine as a certified subspecialty (ABMS 2006, ACGME 2006). Increasing
numbers of medical schools are incorporating end-of-life education into their curricula,
although palliative care remains a marginal medical discipline (Krakauer 2007:392).
Within the United States, home-based care is increasing as a result of both patient
preference and the increasing dismantling of the welfare state that necessitates a reliance
on an unpaid and overwhelmingly female domestic labor force to care for terminally ill
patients at home (Brown 2004).18 And despite the small percentage of people who die as
hospice patients in the United States each year,19
Hospice has become the contemporary symbol of the Anglo-American,
middle-class idea of the good death a patient and family-centered
process in which the foci of attention are personal comfort for the patient
and material and emotional support for family and friends, while the patient
knowingly and reflexively declines toward death (Kaufman 2005:70).
Despite the atypical nature of Michaels dying experience, he nevertheless embodies the
characteristics and values assumed by hospice. In addition, filming his dying process is
perhaps an extreme example of the expectation for hospice patients to knowingly and
reflexively decline toward death.
18 Betweenl980 and 1998, the percentage of Americans dying in the hospital decreased from 54-41%
(Flory, et al. 2004), which may appear consistent with the preference for dying at home. However,
since 1980 there has been an increasing proportion of African Americans and Hispanics dying as
inpatients compared to whites (Flory, et al. 2004, Iwashyna and Chang 2002 and Infeld, Gordon and
Harper 1995), and death in other residences besides private home have also increased (see note 19). In
general, non-whites utilize hospice services at much lower rates than white Americans, attributable to
both cultural and structural barriers (Gordon and Harper 1995, Kemp 2001). 80% of hospice patients
are white (NHPCO 2009:7).
19 According to the National Hospice and Palliative Care Organization, of 2.5 million U.S. deaths in
2008,963,000 patients (38.5% of total U.S. deaths) died while receiving hospice services (NHPCO
2009:4). Less than half (40.7%) of hospice patients die in their own home; over 20% each die either in
a nursing home or inpatient hospice facility, approximately 10% in an acute care hospital and 6% in a
residential facility (which many would considered to be their home) (6).
Interlude: The Last Supper
Last night was bittersweet. As a matter of fact, we had chocolates that
were bittersweet. Michael on the first day of his fast
Michaels last meal was somewhat anti-climactic. There was no fanfare or formal
farewell to food. There was not even a proper sit-down meal, just an assortment of
appetizers spread out on the table buffet-style. Michaels family Jan and her children,
his siblings and nephew were shuffling in and out between the kitchen and dining
room, and settling into the sofas in adjoining living room. Michael walked around the
big black dining room table with a small colorful paper plate and picked up a piece of
salmon next to a note card that read For Mike. I am not sure how much of it he
actually ate. Conversation revolved around reminiscing about family members, life on
the family farm and Sunday dinners growing up in Iowa. Jan later remarked that it
to eat in front of [Michael] when he couldnt eat. But that didnt bother
him.... He didnt want to eat and he wanted to share in the conversation
and the gathering and he still has such a great sense of humor. I was
going to the grocery store yesterday and previously Ive always asked
him before I go what he would like for me to bring him and of course
yesterday I didnt ask him because he .. wasnt going to eat any food.
And when I got back from the grocery store he said, You didnt ask me
what I wanted from the grocery store!
STOPPING EATING AND DRINKING AT THE END OF LIFE
The following section on cessation of food and fluids serves two purposes: 1) to
contextualize the social and corporeal aspects of (stopping) eating and drinking at the
end of life, and 2) to provide an overview of the literature on nutrition and hydration
at the end of life.
At our first meeting, Michael said that he had done his research and although
there was a paucity of published literature on stopping eating and drinking at the end
of life, what he found indicated that it could result in a gentle and peaceful
death. A few days later, he handed me a manila folder with a handful of highlighted
articles one research study and three opinion pieces which are included in the
following paragraphs. The folder also included a letter he had written to the
American Cancer Society asking them to consider promoting the potential benefits of
patients stopping eating and drinking to relieve their suffering as they die. He never
received a response. I took the contents of the manila folder as a starting place for
further investigation of the literature on this subject, which I found to be lacking
across a wide range of disciplines, bioethics excluded.
At the center of Michaels decision to document his cessation of food and
fluids was his concern that both the general public and medical professionals were
unaware or misinformed about the bodily reality and processes that might be
expected when someone stops eating and drinking at the end of a terminal disease.
As I will discuss below, the physiological and biochemical effects of
stopping eating and drinking in terminally ill patients, particularly those with cancer,
often relieve physical and emotional suffering and do not result in a painful or death
by starvation or dehydration. The degree to which this is known and accepted by
medical professionals varies, however, the general public and the media often equate
stopping either oral or artificial nutrition and hydration as starvation or
dehydration, with all of the negative connotations of those words.20 This
sentiment is perhaps most evident in the media discourse and public surveys
surrounding the highly politicized case of Terry Schiavo. In 2005, Schaivo, who had
been in a persistent vegetative state for fifteen years, came to national attention
because of the legal battle between her husband and parents, culminating in a special
session of the U.S. Congress over the legality of the removal of her feeding tube.
201 am unaware of any studies that specifically assess medical professionals knowledge of the
physiological effects of stopping nutrition and hydration at the end of a terminal disease (however, see
Ganzini  for a comparison of nurses experiences with patients who chose Voluntary Refusal of
Food and Fluids versus Physician Assisted Suicide and de la Cruz, Nicosia, Johnson and Aagard
[forthcoming] for an evaluation of Dying Wishs effect on medical students communication about
nutrition and hydration at the end-of-life). My personal experience indicates that while most hospice
professionals understand the palliating effects of VRFF, this is not universal. When my film partner
Karens father was actively dying from lung cancer and had given up eating (but not drinking), his
hospice nurse told his wife that if she could only get him to eat some soup, he would get his strength
Ultimately, Florida courts approved the removal of her feeding tubes, but not without
national controversy and media reinforcement of the notion that withdrawal of
nutrition and hydration, whether oral or artificial, will result in a painful death from
starvation and dehydration (see, for example, Koch 2005).21
Sociocultural Significance of Food and Fluids
Food and fluids have obvious and important nutritional and life-sustaining properties
and the provision of nourishment is both a tangible and symbolic act of nurturing.
The sociocultural significance of eating and drinking is well documented (for an
extensive review, see Mints and Du Bois 2002; see also Marot 1982, Axelson 1986,
Harris 1998, Harris and Ross 1987, and Mclnemey 1992). Eating is often a
social activity where importance lies in social values, meanings and beliefs rather
than in dietary requirements and nutritional values (Marot 1982:203). Food is often
21 The most vocal opposition to removing Schiavos artificial nutrition and hydration came from the
religious right and disability communities, for differing reasons. The right-to-life movements logic in
opposition to withdrawal of life support needs little explanation here. However, the disability rights
movement has a long history of opposition to policies and practices of the right to die movement
concerning the definition of what constitutes a meaningful life worth living (see Not Dead Yet
2010). According to Kaufman (2005), The focus of individual autonomy and on reforms to enhance
self-determination also downplays the voices of those without adequate access to medical services
and/or without political and media clout. Many marginalized people, for example, are worried about
the widespread debate surrounding the right to euthanasia. Their anxiety, that the legislation of
euthanasia could lead to the unwanted deaths of persons deemed less worthy by others, gets far less
attention in public discourse than the cry for the right to control the time of ones own dying (26-7).
associated with gender and social roles, particularly for women (Mclnemey 1992),
who are often primary caregivers (Neal, et al. 1997).
The social and symbolic aspects of eating and drinking are especially
significant at the end of life, however, little research has been done in this area (see
Burge 1996, Holden 1991). Particularly for the family of a dying person, the inability
or refusal to eat can symbolize the reality and immediacy of the dying process. This
can be a significant source of distress for both family members and the dying person,
although the loss of appetite and associated weight loss is often more problematic for
family members than it is for the terminally ill patient (Holden 1991). Caspar (1988)
notes the technology that delivers artificial nutrition and hydration to comatose
patients symbolizes the communitarian aspect of the human family and our
commitment to nurture our more vulnerable members (54). However, the meaning
communicated by this symbol... is intended only for the care givers and society
(54) and does not reflect the ultimate reality of an irreversible coma or other terminal
condition. It is also common for people who are dying to continue to eat in order to
please their families or cause them less distress. Michael was intent on avoiding this
situation and refrained his cessation of food and fluids as relieving his suffering
thereby decreasing his familys suffering as well.
Whats in a Name?: Voluntary Refusal of Food and Fluids
An overwhelming proportion of available literature on cessation of nutrition and
hydration at the end of life includes ethical discussions or opinion pieces (Printz
1992, Bowell and Weng 2000, Chatfield-Mohr and Bhatt 1997, Musgrave 1990,
Caspar 1998, White and Hall 1999, Dunlop, et al. 1995, Janssen and Nordberg 1989,
Sandstead 1990, Miles 1985, Cantor 1985, Wurzbach 1990, Kahn, et al. 2003).
However, the literature on the voluntary or intentional cessation of nutrition and
hydration (as opposed to the withdrawal of artificial nutrition and hydration) is
limited. The phenomenon is referred to in numerous ways, including terminal
dehydration (Zerwekh 1983, Oliver 1984, Printz 1992, Meares 1994), negative
nutritional behavior (Feuz and Rapin 1994) and more recently as voluntary refusal
of food and fluids (VRFF) (Ganzini, et al. 2003). Despite the evidence for cessation
of nutrition and hydration having palliative effects, the various terms described
above generally have a negative connotation. Huffman and Dunn (2002) note this
trend and define terminal dehydration as the withholding intravenous fluids or other
forms of forced hydration (835). However, they prefer a more positive description
of this practice [as] patient choice of limiting fluids to voluntary oral intake (835).
Michaels refraining of the term voluntary refusal of food and fluids as stopping
eating and drinking served to demedcialize the process and signified his agency,
versus if artificial nutrition and hydration were being removed or he was unable to
physically eat or drink as a result of a disease process itself. Plus, voluntary refusal
of food and fluids does not exactly easily glide off the tongue in daily conversation.
By the time we met Michael, he told us that eating and drinking had lost its
pleasure months ago. With Michael present, we asked his hospice physician about
the physiological impact stopping eating and drinking would have. Dr. Stark said,
We could hook Michael up to an IV like we would in a bum unit and
we could give him 5000 calories a day of balanced food, so to speak
[through partial or total parenteral nutrition], and Michaels body
would use almost none of it. It doesnt have the ability to anymore -
biochemically it would actually hurt him. It might make his pain
worse, it might make him more delirious, it might make him prone to
infection, it might have adverse effects ... In Michaels case, calories
and fluids are not a bridge [to good health]. Michaels body doesnt
metabolize them normally, it doesnt put them in the normal places ..
. the process going on in Michael is far bigger than what fluids or
what calories he either takes in or what doctors could put in him.
Given biomedicines imperative to sustain life, it is not surprising that there is
a lack of clinical research focusing on voluntary cessation of nutrition and hydration.
Burge (1996) conducted a clinical review of articles pertaining to dehydration in
terminal patients and found little clinical evidence to guide patients, families, or
clinicians in treating with reduced fluid intake during the terminal phase of life
(2383). Meares (1994) review of the literature on terminal dehydration cites an
opinion piece by Cox (1987), who reports that the claim that terminal dehydration
is not painful is based on observations of over 4,000 patients encountered while he
was a practicing geriatric physician and associate medical director of an Arizona
Mercadentes (1996) review of the nutritional needs for cancer patients does
not deal specifically with the dying process, but addresses factors that modify
nutritional status, including cachexia, nausea and vomiting, decreased caloric intake
or oncologic treatments capable of determining malabsorbtion (10). He concluded
that patients' wishes regarding food and fluid should be taken into primary
consideration while the patients need is the basic principle for any decision (20).
The long-term physiological adaptation to starvation is that the body relies on fuel
sources derived from fat metabolism (McCann, et al. 1994:1265). For terminal
patients or those dying from normal old age, the length of time from cessation of
nutrition and hydration to death varies, depending on previous nutritional status,
body fat and disease trajectory.
In a literature review of terminally ill adults, Hughes and Neal (2000)
concluded that there is little published empirical work on social transactions
between dying people and their caregivers in relation to needs and wishes for food
(1101). Much of the available literature on the end of life in general and cessation of
food and fluid in particular is anecdotal or based on case studies (Huffman and Dunn 22
22 Cachexia is the loss of body mass due to an underlying pathological condition that is unable to be
regained with increased caloric intake.
2002, Bruera and MacDonald 2003, Quill, et al. 2000, Stephany 1991). This point
was noted by Michaels hospice physician who said,
Research on patients who are at the end of life is limited, and its
limited for ethical reasons. This is an extremely vulnerable group of
people and it is very un-palliative to be drawing blood samples to
measure serum markers to see body chemistry changes, to be taking
x-rays, to be watching things in the way we might do it on somebody
in different medical circumstances. And so the research thats done in
end of life will always have an anecdotal quality to it.
Two studies target professional caregivers reasoning associated with feeding
terminally ill cancer patients and found that nurses experience ethical dilemmas with
patients who actively refuse food (Jansson and Nordberg 1989, Davidson, et al.
2000). Holdens (1991) qualitative study with terminally ill cancer patients and their
primary caregivers indicated that the anorexic condition resulting from decreased
food intake is more problematic for family than for the dying person. In their study
of 32 terminally ill patients in a comfort care unit in a long-term care facility,
McCann, et al. (1994) found that in nine instances, patients experienced abdominal
discomfort and nausea when they ate to please their families (1265). However,
another study concluded that [c]aregivers believed that patient-nutrition cessation
was naturally occurring and not physically painful (Meares 1997:7). In a
retrospective survey of Oregon nurses, Ganzini, et al. (2003) found that nurses who
attended hospice patients who stopped eating and drinking in order to hasten their
deaths said they experienced "good" and "peaceful" deaths. These findings were
particularly significant given Oregon's 1997 Death with Dignity Act and the small
percentage of terminal patients who request, receive and self-administer prescribed
lethal medications compared to those who voluntarily refuse food and fluids
(Ganzini, et al. 2000, 2003). Cross-cultural comparison shows that stopping eating
and drinking at the end of life in India has significant religious and cultural value
(Justice 1995) and is a socially and medically acceptable choice in The Netherlands
Stopping eating and drinking often lessens the burden of suffering for dying
patients (van der Riet, et al. 2006, 2008; McCann, et al. 1994). The analgesic effects
of stopping eating and drinking are attributable to the mechanisms whereby food
deprivation leads to an increase of P-endorphin levels in the hypothalamus and
plasma, and water deprivation leads to an increase in levels of dynomorphin, a
powerful opioid, in the hypothalamus (Printz 1992:700). In addition, the creation
and breakdown of ketones produced in energy deficient states may produce analgesic
effects (Printz 1992, 1988). Azotemia (increased levels of nitrogen compounds in the
blood) from starvation or dehydration can induce a semi-conscious state where the
patient feels no pain (Abrams 1987).
Dying people, particularly those with cancer, gradually cease intake of food
and fluids (McCann, et al. 1994, Dunlop, et al. 1995). Meares (1994) found that
terminal dehydration in imminently dying people may be an adaptive process
designed to accommodate the multisystem failure which heralds impending death
(11). In addition, dying people often do not complain of thirst or hunger (McCann, et
al. 1994, van der Riet, et al. 2006, Terman 2006). Dry mouth is the most common
side effect after a patient stops drinking (McCann, et al. 2004) and is easily
addressed by simple techniques such as rinsing the mouth with water, sucking on an
ice cube or by swabbing the inside of the mouth or sucking on a glycerin swab (van
der Riet 2006, 2008). Feuz and Rapin (1994) point out that negative nutritional
behavior can be modified with pain management and attention to the patients
dietary preferences. Although they assume that stopping eating and drinking is
negative behavior and not simply a sign and symptom of the dying process, they
found that adequate pain management and the availability of patient choice in food is
a significant factor in a persons appetite and desire for food. Additionally, some
symptoms may indicate and can be successfully relieved with hydration therapy,
particularly agitated delirium, opioid toxicity (Fainsinger and Bruera 1997) and
hypercalacaemia (Dunlop, et al. 1995).
While some dying patients continue to eat and drink in order to lessen the
distress of family members, Michael and his family viewed his cessation of food and 23
23 Hypercalacaemia is elevated levels of calcium in the blood. Effects include constipation and
associated discomfort, bone pain, kidney stones, depression and confusion, fatigue, anorexia, nausea
fluids as lessening the burden of suffering for both him and his family. On the
seventh day of Michaels fast, he got dressed in the morning but soon decided that he
did not have the energy to come downstairs, an event Rodney described as unheard
of in Mikes history. After that day, Michael remained in his bedroom. Usually
covered by a lavender sheet, he wore only white tank top undershirts, enabling us to
clearly see his increasingly frail and bony body. Rodney, Michaels brother remarked
that stopping eating and drinking not only made it easier for him, but [also for]
those of us who are close to him and love him because its more difficult to see him
getting weaker and weaker every day being more helpless than to see him leave
us, knowing that he cant get any better. As Michael lay in bed the night before he
died, Rodney hypothesized that Michaels dehydration might be contributing to the
lack of death rattle, often be a source of distress for family members.24
Desjarlais (2003) sensory biographies of two Yolmo Buddhist elders sheds
light on how to understand Michaels sensorial and bodily experiences, including
pain and suffering, at the end of his life. Desjarlais aims to show the ways in which
sensory engagements articulate with broader social, personal, and political
24 Death rattle is a common term used to describe the gurgling breathing that often precedes death.
Although not painful for the dying person, it can often be disturbing for family members to witness.
Rodneys hypothesis that Michaels lack of death rattle was due to his dehydration is supported in the
literature with reports that when compared to patients who were hospitalized with intravenous
hydration that required suctioning to clear respiratory secretions, hospice patients who died at home
without intravenous therapy did not produce enough respiratory secretions to warrant suctioning
(Dolan 1983, cited in Meares 1994).
dynamics (4). By filming his dying process and visually depicting his lifeworld
(Good 1994) in the weeks preceding his death, Michael reflexively placed himself at
the center of a national discussion about death a discussion that is intensely
personal, inherently social and undoubtedly political.
TOWARDS A GOOD DEATH
The notion of a good death has particular salience in Western culture (see Aires
1974; Kubler-Ross 1969; Walter 1996; Hart, Sainsbury and Short 1998).25 26 My
intention here is not to review the voluminous body of literature on the good death,
but to situate Michaels dying as an example of one persons attempt to achieve a
particular conception of a good death within the contemporary United States.
Michael wanted a dignified death that was free from pain and suffering and
somehow natural, that is, not prolonged by artificial life-support. Kaufman
(2005) writes that
People want death to be made comfortable by the tools of medicine, which
they expect can eliminate both the disturbing visible signs of the bodys
disintegration and the patients experience of suffering. Yet they also hold
vague ideas that death can somehow be natural and by that they usually
mean peaceful and easy, like the sleep of a child and without the overuse of
drugs or machines (4).
25 Hart, Sainsbury and Short (1998) suggest that the ideology of the good death legitimates a new
form of social control within which socially approved dying and death are characterized by proscribed
and normalized behaviours [sic] and choices. This ideology dominates the social management of
dying and death within the hospice movement and increasingly within the broader community, and
powerfully constrains the choices of dying people.
26 The assemblage of technologies that comprise life-support is seldom referred to as a singular
entity by medical professionals in the hospital. Rather, life support is a lay term that encompasses
the use of a mechanical ventilator, artificial nutrition and hydration and cardiopulmonary
resuscitation. Life support refers both to specific medical procedures and to the various
interpretations of those procedures. The term life support per se is never used in daily hospital
practice, and health professionals do not use it at all. Rather, it is a term that circulates in public talk,
especially to refer to what many people say they do not want if it will only prolong dying (Kaufman
Although Michaels dying is contrasted with medicalized and potentially
problematic deaths that take place in hospitals and other institutions, the power of
the themes that shape hospital death remain dignity, choice, control and timing.
Ultimately, the themes presented by Kaufman point toward cultural ideals of the
meaning of personhood and when embodied subjectivity can and/or should be
Embodying the Discourse of Dignity and Quality of Life
I cant remember many smells from Michaels house. There are
homey smells, none of the familiar smells found in the hospital. No
tantalizing aromas offood. No pet odors, but plenty of earthy plant
smells. Like an indoor jungle, but controlled -just like Mike. I
remember the scent of lotion and lemon-scented glycerin mouth
swabs. The portable commode next to the bed was empty, smelling
more of new plastic than of urine. No stale smell of adult diapers left
unchanged for too long. He refused to wear them and said they were
undignified, but his dehydration left little fluids for him to excrete
anyway [field notes, September 2007].
Dignity and quality-of-life have perhaps become the most important framework for
the orchestration of death in America. Kaufman (2005) notes that
In American society, with its strong emphasis on the ideal of individual
rights, the decision-making power of a person facing death is deemed
necessary and central.... A specific rhetoric of suffering, dignity,
and quality of life shapes those judgments and is deployed often in
hospital discussions about what to do for and about the critically ill
Despite hospice and palliative cares beginnings as an alternative to over-
medicalized deaths, these institutions are not exempt from the same cultural and
ideological roots of positivism and individualism that underlie Western biomedicine
(Gordon 1988). Although removed from the hospital environment, Michael still
understood his own disease and life trajectory through these principles, calculating
each step toward his version of a dignified death. For Michael, dignity was such a
powerful conceptual framework was that even the private home health agency he
hired was called Dignity Care.
Kaufman notes that
For its part, talk about dignity is often connected to ideas of suffering,
especially in attempts to assess the role suffering plays in a patients
experience of critical illness and of hospitalization (whether or not the
patient can articulate his or her suffering). Dignity is also invoked in
the deliberations among staff and family about whether there has been
enough suffering. Dignity may refer both to the practical impacts of
medical treatments on the patients body and existence and to the
intrinsic value of life, regardless of its condition (74-5, italics in
The above quote reflects the specific dynamics at play in hospital intensive care
units. Particularly within hospice and palliative care, where the death of the patient is
most often the expected outcome,27 the ideal dignified death is one with a peaceful
transition, achieved by relieving, as much as possible, physical and non-physical
pain and suffering.
Subjectivities of Pain and Suffering
It is safe to say that most humans tend to avoid suffering given the choice, but
avoidance of suffering at the end of life is not universal across time and space (see
Aires 1972, Walter 1994, Micco, Villars and Smith 2009). Only recently has modem
medicine been able to grant us the coveted pharmaceutical mediation that allows the
peaceful death we not only desire but is also framed by the discourse of the right to
freedom from pain (Micco, Villars and Smith 2009). Particularly in the United
States, Europe and Australia, where the vast majority of the worlds opiate
analgesics are consumed (INCB 1995), there is an expectation that physical pain can
and should be relieved. Michael believed that those who choose to suffer ... I
think are not entitled to ... any great admiration or respect. I see no great merit in
Avoiding physical pain was in fact, Michaels primary reason for having
hospice services. End-stage cancer, in particular, can produce extreme physical pain,
27 Palliative care may be used in conjunction with curative therapies. In 2008, 212,000 patients were
considered live discharges due to extended diagnosis, desire to pursue curative treatment or other
unknown reasons (NHPCO 2009:4).
which is effectively managed with opiate-based or synthetic opioid analgesics.
Michaels body was covered with six clear transdermal fentanyl patches that
supplied him with a constant supply of medicine. The patches were barely noticeable
unless you looked closely at the creases of the skin on his torso and upper arms. In
addition, he augmented the patches with oral fentanyl citrate and/or hydrocodone if
he experienced breakthrough pain pain that occasionally breaks through the
background analgesic barrier.
In Pain as Human Experience: An Anthropological Perspective, Kleinman,
Brodwin, Good and Good (1992) write that
Pain itself poses an obdurate resistance to cultural categories. It is an
experience that simply cannot be avoided, an experience that sets
limits to the meanings given it by cultural beliefs, discourses, or
practices. Something is at stake, frequently desperately so, in the lives
of pain patients. Pain can be a massive threat to the legitimacy of the
everyday world. Viewed in these terms, the anthropological study of
chronic pain addresses the sources, varieties, and consequences of
human suffering (pg).
While the above quote addresses chronic pain, the case of pain at the end of life
appears to present additional issues specific to dying people. Investigations into the
subjectivity of pain are necessarily problematic. Scarry (1985) points out the
inexpressibility of physical pain and the difficulties inherent in putting others pain
into written language: Physical pain does not simply resist language but actively
destroys it, bringing about an immediate reversion to a state anterior to language, to
the sounds and cries a human being makes before language is learned (4). When
moving through the human interior, physical pain is a unique state that interrupts
and challenges the external, sharable world because physical pain unlike any other
state of consciousness has no referential content. It is not of or for anything. It is
precisely because it takes no object that it, more than any other phenomenon, resists
objectification in language (5). If, as Scarry suggests, pain and suffering not only
resist objectification in language but are also conceived of as obstacles to ones life
plans and practical actions and resistance to the flow of experience (Kleinman
1992:125), Michaels efforts to document his (relief of) pain and suffering served to
objectify the pain itself by circumventing its inability to be objectified through
language and attempting to give meaning to his experience through a visual narrative
Pain and suffering are not limited to the physical body, a point well
articulated by anthropologists (Kleinman, Das and Lock 1997, Bourdieu 1999).
Within hospice, there is an understanding that pain and suffering encompasses not
only the physical, but that dying persons may also experience emotional, spiritual or
existential suffering. Recently, the term intractable terminal suffering has been
used to describe pain which cannot be mediated through any means, often leading to
the implementation of palliative sedation the use of sedatives to relieve suffering
by reducing the patients level of awareness to the point of unconsciousness if
necessary (Quill and Byock 2000, AAHPM 2006).
In his exploration of palliative care, Krakauer (2007) addresses how
technologies intervene in and serve as analytical framework for our conceptions and
experiences of death, paying particular attention to the case of not using such
technologies. At the center of the relationship between technology and death is the
issue of suffering: For palliative medicine, neither medical technology nor
technological mastery is what matters most, what calls for a response. What calls is
suffering, and attention to suffering reveals that life-sustaining technology is not
always called for (390). However, palliative medicine does not eschew technology
per se, and is predicated upon the use of state-of-the-art technology in order to
alleviate pain and other physical symptoms while at the same time attending to less
audible forms of suffering that arise from feelings of loss. Krakauer (2007)
By attending to suffering, palliative medicine thinks otherwise than
subjectively and technologically. It encounters the sufferer, the other, in
a way that cannot be characterized merely as objectification by a
thinking subject. The may make palliative medicine necessarily remain
a marginal discipline within medicine (392).
Krakauer focuses on the experiences of Mrs. A, an elderly patient who, after
congestive heart failure, chronic renal failure, emphysema, chronic foot pain and
several strokes which likely resulted in her mild dementia, required several types of
life sustaining technologies to remain alive (382) including hemodialysis, a
mechanical ventilator, organ-replacement and permanent artificial nutrition.
Krakauer shows how the assumptions of medical mastery in which the body is
comprised of disposable and replaceable parts and the Cartesian dream of mastering
death shapes a particular form of thinking subjectivity. The juxtaposition of the
brief ethnographic description of Mrs. A and her familys insistence on maintaining
life-sustaining treatment with palliative cares approach to let dying be in a way
that does not approach dying exclusively with the goal of mastery (391) is meant
to show how palliative care strives to approach each sufferer with no particular
agenda at all (391).
Michaels suffering was twofold: the immediacy of physical discomfort such
as diarrhea and nausea, which would be alleviated by stopping the intake of food;
and the potential suffering he felt he would encounter if structural conditions
prevented him from stopping eating and drinking. When Michaels physical pain was
well controlled, his greatest source of suffering was mental the potential threat that
of not being able to control his own death through stopping eating and drinking.
Michaels mantra was that stopping eating and drinking would allow him to avoid a
wretched ending. When asked to explain what wretchedness he was avoiding,
Michael referred to issues of physical pain and bodily disintegration, but more
importantly to the suffering he would experience from not being in control of the
circumstances surrounding his own death.
When relief of suffering is viewed as a right importantly, a right to which
the vast majority of the dying people in the world are unable to access regardless of
how the role of suffering contributes to diverse notions of a good death should
we at least question the assumption that all forms of suffering at the end of life be
routinely eradicated? The process of the medicalization of death and medicines
status as expert over the dying process has produced a secular world of
medicine, in which we are becoming ever more accepting of the notion that pain
and suffering must be banished from the dying experience. Pain, indeed all suffering,
is meaningless; it should be erased (Micco, Villars and Smith 2009:872). While
being careful not to imply that physical or existential pain is redemptive or
edifying, these authors and palliative care practitioners ask whether there might be
someone for whom ... the suffering of dying is a path of self-understanding or
spiritual awakening (878). More importantly, they question whether through the
overuse of pharmaceuticals to treat all forms of suffering physical, spiritual and
psychological hospice providers have lost sight of the original intent of the hospice
movement to provide a supportive presence for dying patients and their families,
with close attention to all forms of suffering (878). American culture places little
value on pain either in childbirth, death or otherwise. Like Micco, Villars and Smith
(2009), I am in no way suggesting that dying patients pain be denied relief or
suffering be valorized as a rule. But in understanding Michaels dying experience in
historical and cultural perspective, I have tried to point toward his structural position
within contemporary American society that enabled him to access (through hospice
and their provision of pain medication) and create (through his ability to stop eating
and drinking in addition to documenting the process) the means to relieve his
Choice, Control and Patient Autonomy
Underlying the commonly held desire not to be a burden to others at the end of life is
the Western concept of individualism that has been enshrined in medical ethics and
end-of-life discourse within the principle of patient autonomy.28 Michael had
made it clear that he did not want to be a burden to his family. More importantly, he
did not want to be a burden to his partner Jan, who had her own health problems and
was unable to provide the physical care lifting, turning that Michael would
inevitably need as he further declined. When asked whether or not his decision to
28 See Beauchamp and Childress (1979) for discussion of the foundational principles of bioethics:
autonomy, beneficence, nonmaleficence and justice (cited in Kaufman 2005:339, n.5). Regarding
these principles, Kaufman writes that Taken together, the four foundational principles emerge from
Western-American traditions of the law, positivism, and individualism.... Many observers note that
autonomy has become the centerpiece, or dominant principle, in contemporary views of how patients
and health professionals should relate to one another (339, n.5). It is worth noting, that the
individualistic concept of patient autonomy, particularly at the end of life, is not a universal value held
by all Americans (Koenig 1997, Frank, et al. 2008) or cross-culturally (Stonington 2009).
stop eating and drinking was influenced by a desire to make his death easier for those
around him, Michael responded,
I like your question because Ive thought about that a good bit and ..
I believe, and I think that my loved one, are ... attuned to this. In fact
they are sympathetic to my wishes .. they dislike intensely seeing
me suffer. Its an ambivalent feeling for them because of our parting
- its ambivalent for me -1 hate to part with these people that... I
love and ... are important, important in my life. And yet it is
inevitable, we must all part.
Tied up with his decision to stop eating and drinking was Michaels desire to remain
in his own home, thereby being able to fully control his decision-making.
Avoiding dying in the hospital or other institution was perhaps the most
decisive choice Michael could have made in carrying out his end-of-life wishes. In
doing so, Michael hoped to avoid some of the dependency associated with relying on
life-sustaining technologies. One of the most significant consequences of life-
sustaining technology and the medicalization of death has been the objectification of
bodies which are neither dead nor alive, particularly in the case of the maintenance
of biological life for organ transplants (see Lock 2004; Kaufman 2000, 2005). 29
29 Although refusal of treatment is a patients right, including the provision of nutrition and hydration,
the ability of patients to voluntarily refuse food and fluids in institutional contexts is considerably less
than it would be in ones own home. One of the reasons Michael was so adamant about remaining at
home in order to control the circumstances surrounding his own death was because of his 104 year-
old mothers experience as she was dying in a nursing home. She had explicitly stated that she did not
want to be spoon fed, but for various reasons the impetus to provide care, not wanting to be seen
as causing her death and the fallibility of advanced directives the nursing staff did not honor her
wishes and continued to feed her (see Savishnskys  ethnography on food, conflict and control
in a nursing home).
Responding to the lack of attention to issues of embodiment within hospice
and palliative care, Lawton (2000) discusses hospice patients bodily deterioration
and the loss of ability to control embodied actions.
For patients close to death (particularly for those with cancer) this
process of change and loss was experienced in an especially sharp and
painful manner; physical and mental deterioration often being very
rapid at this stage (81).
Within institutional and relational contexts of care, Lawton demonstrates how the
dying patient transitions from subject to object in terms of their dependency on
others to complete basic tasks such as feeding, bathing and other aspects of bodily
care. [A]s their bodily strength and mobility deteriorated significantly, many
hospice patients experienced what appeared to be a tangential loss of self, where the
physical and mental were intimately tied together (89).30 Yet Michaels attempt to
assert his own autonomy and selfhood even as his body deteriorated poses a contrast
to Lawtons analysis, illustrating that selfhood does not exist solely in embodied
30 Because Lawtons (2000) study was conducted within an acute inpatient hospice facility, the
atypical nature of her sample resulted in an ethnographic picture of people dying with particularly
distressing and extreme bodily symptoms (33). According to the NHPCO, the 2008 median length
for hospice stays was 21.3 days (NHPCO 2009:5), however, this figure varies by state. For those who
access emergency hospice services enrolling at an inpatient facility within a few days of death
because of particularly distressful symptoms the pace of physical decline can be more pronounced
than for patients who have been receiving palliative care for a longer period of time.
Does the timing make all that much difference? I think not. I think
that. surely I could live a few weeks longer if I postponed the, the
uh, stopping eating and drinking. But, uh, I dont think, I think that
they would not want to see me suffering. I know that they would not
want to see me suffering even though they would want me to be about
a bit longer. Michael, day six of his fast
Michaels original prediction was that he would last about a week before he would
slip into a coma and die peacefully. However, prognostication is fraught with
difficulty (Christakis 1999) and predicting the exact timing of Michaels death was
no exception. Given that Michael did not have much body fat before he stopped
eating and drinking, everyone involved expected him to die much faster than he did.
Rodney speculated that although Michael tried his best to not take in excess fluids
when sucking on ice chips or swallowing his medication, it was possible he was
absorbing minimal amounts of fluid which could have prolonged his death. On the
ninth day, his hospice nurse brought a pump to supply intravenous pain medication
because of Michaels increasing difficulty swallowing. He now spent most of the day
sleeping on top of an egg crate mattress and with pillows underneath his elbows to
prevent bedsores from forming. Jan would sit on the side of the bed, lovingly
stroking his hair. She reported that Michael doesnt turn much ... he spends most
of his time lying on his back, sleeping with his mouth open.
The importance of time as an organizing principle of contemporary American
dying is explained by Kaufman (2005). Unlike in an intensive care unit, where time
is marked by family meetings convened to determine the fate of a body that could be
indefinitely kept alive by artificial means, Michaels body itself became the marker
of time. With each passing day, we could see the visible signs of decline, his body
becoming thinner, his face sinking in, consciousness waning in and out and his
breathing becoming shallower and less frequent. Unlike many of the scenes
described by Kaufman (2005), Michaels decline was expected and gradual, leaving
little surprises or difficult decisions to be made by his family.
Medicines ability to manipulate the time of a persons biological death has
never been greater, enabling a disconnect between the timing of an individuals
social and biological death (Lock 2004, Kaufman 2005, Seale 1998). Clive Seale
(1998) has suggested that in order to maintain social bonds people who choose
euthanasia or assisted dying do so in order to time their social and biological deaths
to coincide more closely (cited in Norwood 2009:8). Although he never articulated
his decision in this way, Michaels cessation of food and fluids was used to legally
and literally relocate death of the body to coincide with the point in time where [his]
sociality ends (8-9).31 By stopping eating and drinking, Michaels greatest gift to his
family was perhaps relieving them of having to make those tough decisions that arise
at the intersection of the end of social personhood and biological life.
31 Michael believed that stopping eating and drinking was qualitatively different from active
euthanasia in that it was a process that could be reversed if one changes their mind (and it is also legal
in the United States although the ability of dying patients to successfully stop eating and drinking is
dependent upon their social and structural context. Catholic hospitals and nursing homes, for example,
explicitly prohibit patients from voluntarily giving up food and fluids). However, the underlying logic
of aligning the timing of social and biological death is similar.
The medicalization of death in the United States has led to the problem of death
(Kaufman 2005) and left many people concerned with the nature of their dying. By
choosing to stop eating and drinking at the end of his life, Michael took active steps
to have a demedicalized and perhaps more meaningful death. In doing so, he asserted
his agency in order to control and align the timing of his social and biological deaths.
(Seale 1998, Norwood 2009). By choosing to stop eating and drinking, Michael
succeeded in having his version of a good death.
The process of filming his dying, while perhaps an extraordinary endeavor,
not only mediated his dying experience but also served as a way for Michael to
maintain his sense of personhood and identity as a physician and teacher through his
death and afterward. Although his death was framed in opposition to the types of
medicalized deaths that occur in the hospital, the underlying themes that shape the
culture of hospital dying dignity, control, autonomy and timing nevertheless
remained the predominant framework for Michaels understanding of his own dying
process. I hope that Michaels dying experience has shown not only the possibilities
and constraints on agency and the maintenance of personhood at the end of life, but
also points towards the ways in which greater attention to issues of suffering,
subjectivity and embodiment for the dying person should be central to the future
anthropology of hospice and palliative care.
Abrams, F. R.
1987 Withholding treatment when death is not imminent. Geriatrics 42(5):77-84.
1974 Western attitudes toward death from the Middle Ages to the present. Patricia
M. Ranum, trans. Baltimore, MD: The Johns Hopkins University Press.
American Board of Medical Specialties
2006 ABMS Establishes New Subspecialty Certificate in Hospice and Palliative
Medicine. http://www.abms.org/News_and_Events/downloads/ NewSubcert
PalliativeMed.pdf, accessed September 27, 2008.
American Council for Graduate Medical Education
2006 ACGME Board adds hospice and palliative care as new Subspecialty, http://
www.acgme.org/acWebsite/ ewsReleases/newsRel_07_20_06.pdf, accessed
September 27, 2008.
1986 The impact of culture of food-related behavior. Annual Review of Nutrition
2001 Vita: Life in a Zone of Social Abandonment. Berkeley: University of
Biehl, Joao, Byron Good and Arthur Kleinman
200X Subjectivity: Ethnographic Investigations. Berkeley: University of California
Bourdieu, Pierre, ed.
1999 The Weight of the World: Social Suffering in Contemporary Society.
Parkhurst Ferguson, trans. Stanford, CA: Stanford University Press.
2004 Between neoliberalism and cultural conservatism: spatial divisions and
multiplications of hospice labor in the United States. Gender, Place and Culture
Bruera, Eduardo and Neil MacDonald
2003 To Hydrate or not to Hydrate? How should it Be? Journal of Clinical
1996 Dehydration and provision of fluids in palliative care. What is the evidence?
Canadian Family Physician 42: 2383-8.
Cantor, N. L.
1985 Non-feeding as withdrawal of medical therapy. Death Studies 173-180.
1998 Food and Water: symbol and reality. Health Prog 69(4): 54-8.
Chatfield-Mohr, S. and C. Bhatt
1997 Dehydration in the terminally illiatrogenic insult or natural process?
Postgraduate Medical Journal 73(862):476-80.
Cooey, Paula M.
1999 The Messiness of Dying. Journal of Feminist Studies in Religion 15(1 ):96-
Davidson, B., R. Vander Laan, M. Hirschfield, A. Nordberg, E. Pitman and L. Jug
1990 Ethical Reasoning associated with the feeding of terminally ill elderly cancer
patients: an international perspective. Cancer Nursing 13: 286-292.
2001 The technocratic, humanistic, and holistic paradigms of childbirth.
International Journal of Gynecology and Obstetrics 75:S5-S23.
Davis-Floyd, Robbie and Elizabeth Davis
1996 Intuition as Authoritative Knowledge in Midwifery and Homebirth. Medical
Anthropology Quarterly 10(2):237-269.
De la Cruz, Scott, Francesca Nicosia, Dan Johnson and Eva Aagard
Forthcoming Dying Wish: Effects Of A Documentary On Medical Student
Knowledge, Attitudes, Self-Confidence, And Skills In End-Of-Life
2003 Sensory Biographies: Lives and Deaths among Nepals Yolmo Buddhists.
Berkeley: University of California Press.
Dunlop, R. J., J. E. Ellershaw, M. J. Baines, N. Sykes and C. M. Saunders
1995 On withholding nutrition and hydration in the terminally ill: has palliative
medicine gone too far? A reply. Journal of Medical Ethics 21:141-143.
Exley, Catherine and Davina Allen
2007 A critical examination of home care: End of life care as an illustrative case.
Social Science and Medicine 65: 2317-2327.
Fainsinger, R. and E. Bruera
1997 When to treat dehydration in a terminally ill patient? Support Cancer Care
Feuz and Rapin
1994 An observational study of the role of pain control and food adaptation of
elderly patients with terminal cancer. Journal of the American Dietetic Association
Flory, James, Yinong Young-Xu, Ipek Gurol, Norman Levinsky, Arlene Ash and
2004 Place of Death: U.S. Trends Since 1980. Health Affairs 23(3): 194-200.
Ganzini, Linda, Heidi D. Nelson, Terri A. Schmidt, Dale F. Kraemer, Molly A
Delorit and Melinda A. Lee
2000 Physicians Experiences with the Oregon Death with Dignity Act. The New
England Journal of Medicine 442: 557-563.
Ganzini, Linda, Elizabeth R. Goy, Lois L. Miller, Theresa A. Harvath, Ann Jackson,
Molly A. Delorit
2003 Nurses Experiences with Hospice Patients Who Refuse Food and Fluids to
Hasten Death. The New England Journal of Medicine 349:359-365.
Glaser, B. G. and Anselm Strauss
1968 Time for Dying. Chicago: Aldine.
1994 Medicine, Rationality, and Experience: an Anthropological Perspective.
Cambridge: Cambridge University Press.
Gordon, Deborah R.
1988 Tenacious Assumptions in Western Medicine. In Biomedicine Examined.
Margaret Lock and Deborah Gordon, eds. Norwell, MA: Kluwer Academic
Publishers. Pp. 19-56.
1965 Death, grief and mourning in contemporary Britain. London: Cresset Press.
1998  Good to Eat: Riddles of Food and Culture. Prospect Heights, IL:
Harris, Marvin and E.B. Ross, eds.
1987 Food and Evolution: Toward a Theory of Human Food Habits. Philadelphia:
Temple University Press.
Hart, Bethne, Peter Sainsbury and Stephanie Short
1998 Whose dying? A sociological critique of the good death. Mortality 3(1): 65-
1960 Death and the Right Hand. Glencoe, IL: Free Press.
Higginson, Irene J., Paul Astin and Susan Dolan Imperial
1998 Where do cancer patients die? Ten-year trends in the place of death of cancer
patients in England. Palliative Medicine 12(5): 353-363.
1991 Anorexia in the Terminally 111 cancer Patients: The Emotional Impact of the
Patient and the Family. The Hospice Journal 7(3): 73-84.
Huffman, Joan and Geoffrey Dunn
2000 The Paradox of Hydration in Advanced Terminal Illness. Journal of the
American College of Surgeons 194(6):835-9.
Hughes, Nic and Richard D. Neal
2000 Adults with terminal illness: a literature review of their needs and wishes for
food. Journal of Advanced Nursing 32(5): 1101-1107.
Infeld, Donna Lind, Audrey K. Gordon and Bernice Catherine Harper, eds.
1995 Hospice Care and Cultural Diversity, Volume 10. Binghamton, NY: Haworth
International Narcotics Control Board
1995 Report of the International Narcotics Control Board for 1995: Availability of
Opiates for Medical Need. Vienna: INCB. http://www.incb.Org/pdf/e/ar/1995/
suppllen.pdf, accessed March 23, 2010.
Iwashyna, Theodore J. and Virginia W. Chang
2002 Racial and ethnic differences in place of death: United States, 1993. Journal
of the American Geriatrics Society 50(6): 1113-1117.
James, N. and D. Field
1992 The routinization of hospice: charisma and bureaucratization. Social Science
and Medicine 34(12): 1363-75.
Jansson, L. and A. Norberg
1989 Ethical Reasoning Concerning the Feeding of Terminally 111 Cancer Patients.
Cancer Nursing 12: 352-358.
2009 Montana Ruling Bolsters Doctor-Assisted Suicide. New York Times,
December 31. http://www.nytimes.corn/2010/01/01/us/01suicide.html, accessed
April 4, 2010.
1995 The Natural Death While Not Eating: A Type of palliative Care in Banares,
India. Journal of Palliative Care 11(1): 38-42.
Kahn, Marc J., et al.
2003 Allowing Patients to Die: Practical, Ethical, and Religious Concerns. Journal
of Clinical Oncology 21(15): 3000-3002.
Kaufman, Sharon R.
2000 In the Shadow of Death with Dignity: Medicine and the Cultural
Quandaries of the Vegetative State. American Anthropologist 102(l):69-83.
Kaufman, Sharon R.
2005 And a Time to Die: How American Hospitals Shape the End of Life. New
Kaufman, Sharon R. and Sarah Franklin
2005 The Anthropology of the Beginnings and Ends of Life. Annual Review of
2001 Culture and the End of Life: Hispanic Cultures (Focus on Mexican
Americans). Journal of Hospice and Palliative Nursing 3(1): 29-33.
1989 The Illness Narratives: suffering, healing, and the human condition. New
York: Basic Books.
1992 Pain and Resistance: The Delegitimation and Relegitimation of Local Moral
Worlds. In Writing at the Margin: Discourse between Anthropology and
Medicine. Berkeley, LA and London: University of California Press. Pp. 120-146.
Kleinman, Arthur, Paul E. Brodwin, Byron Good, Mary-Jo DelVicchio Good
1992 Pain as Human Experience: An Introduction. In Pain as Human Experience:
An Anthropological Persepctive. Mary-Jo DelVicchio Good, Paul E. Brodwin,
Byron J. Good and Arthur Kleinman, eds. Berkeley: University of California
Kleinman, Arthur, Veena Das and Margaret Lock, eds.
1997 Social Suffering. Berkeley: University of California Press.
2005 The challenge of Terri Schiavo: lessons for bioethics. Journal of Medical
Ethics 31: 376-378.
Krakauer, Eric L.
2007 To Be Freed from the Infirmity of (the) Age: Subjectivity, Life- Sustaining
Treatment, and Palliative Medicine. In Subjectivity: Ethnographic Investigations.
Joao Biehl, Byron Good and Arthur Kleinman, eds. Berkeley: University of
California Press. Pp. 381-396.
1969 On Death and Dying. London: Tavistock.
2000 The Dying Person: Patients experiences of palliative care. London:
2001 Twice Dead: Organ Transplants and the Reinvention of Death. Berkeley:
University of California Press.
1994 The concept of therapeutic emplotment. Social Science and Medicine
1979 The Physical Effect on the Individual of the Idea of Death Suggested by the
Collectivity. In Sociology and Psychology: Essays. London: Routledge and
Keegan Paul. Pp. 35-56.
Meares, Candace Jans
1994 Terminal dehydration: A review. American Journal of Hospice and Palliative
Medicine 11(3): 10-14.
Marcus, George E. and Michael M. J. Fischer
1986 Anthropology as Cultural Critique: An Experimental Moment in the Human
Sciences. Chicago: University of Chicago Press.
McCann, Robert M, William J. Hall, Annmarie Groth-Juncker
1994 Comfort Care for Terminally 111 Patients: The Appropriate Use of Nutrition
and Hydration. The Journal of the American Medical Association 272(16): 1263-
1992 Provision of Food and Fluids in Terminal Care: A Sociological Analysis.
Social Science in Medicine 34: 1271-1276.
Miles, M. H.
1985 The Terminally 111 Elderly: Dealing with the Ethics of Feeding. Geriatrics 40:
Mintz, Sidney and Christine Du Bois
2002 The anthropology of food and eating. Annual Review of Anthropology 31:
1987 The Body Silent. London: Phoenix House.
1990 Terminal dehydration. To give or not to give intravenous fluids? Cancer
National Hospice and Palliative Care Organization
2009 NHPCO Facts and Figures: Hospice Care in America.
_figures.pdf, accessed March 9, 2010.
Neal, Margaret B., Berit Ingersoll-Dayton and Marjorie E. Starrels
1997 Gender and Relationship Differences in Caregiving Patterns and
Consequences Among Employed Caregivers. The Gerontologist 37(6):804-816.
2009 The Maintenance of Life: Preventing Social Death through Euthanasia Talk
and End-of-Life Care in The Netherlands. CITY: Carolina Academic Press.
Not Dead Yet
2010 Fifth Anniversary of Terri Schiavos Death A History Lesson. http://notdead
schiavos.html, accessed April 1, 2010.
1984 Terminal dehydration. The Lancet 2(8403): 631.
Oregon Department of Human Services
2010 Death with Dignity Act. http://www.oregon.gov/DHS/ph/pas/, accessed April
Palgi, Phyllis and Henry Abramovitch
1984 Death: A Cross-Cultural Perspective. Annual Review of Anthropology
1988 Is withholding hydration a valid comfort measure in the terminally ill?
Geriatrics 43: 84-8.
1992 Terminal Dehydration, a Compassionate Treatment. Archives of Internal
2000 Sedation and Voluntary Refusal of Food and Fluids. Annals of Internal
Medicine 132: 403-414.
Quill, Timothy and Ira Byock
2000 Responding to Intractable Terminal Suffering: The Role of Terminal
and Voluntary Refusal of Food and Fluids. Annals of Internal Medicine 182(5):
2004 The Living and the Dead: Community in the Virtual Cemetery. Journal of
Death and Dying 49(l):57-70.
2004 If shes a vegetable, well be her garden: Embodiment, transcendence, and
citations of competing cultural metaphors in the case of a dying child. American
Russ, Ann J.
2005 Loves Labor Paid for: Gift and Commodity at the Threshold of Death.
Cultural Anthropology 20(1): 128-155.
Sandstead, Harold H.
1990 A point of view: nutrition and care of terminally ill patients. American
Journal of Clinical Nutrition 52: 767-769.
Savinshky, Joel S.
2003 Bread and Butter Issues: Food, Conflict and Control in a Nursing Home. In
Gray Areas: Ethnographic Encounters with Nursing Home Culture. Philip B.
Stafford, ed. Santa Fe, NM: School of American Research Press.
Seymour, Jane E.
1999 Revisiting medicalization and natural death. Social Science and Medicine
Seymour, Jane E.
2000 Negotiating natural death in intensive care. Social Science and Medicine
1985 The Body in Pain: The Making and Unmaking of the World. Oxford & New
York: Oxford University Press.
1982 The social transformation of American medicine: the rise of a sovereign
profession and the making of a vast industry. New York: Basic Books.
1991 Nutrition for the terminally ill. Home Health Care Nurse 9: 48-9.
2006 Two Weeks. 102 min. Custom Productions: Los Angeles.
1967 Passing On. Englewood Hills, NJ: Prentice Hall.
2006 The Best Way to Die. Carlsbad, CA: Life Transition Publications.
Van der Riet, et al.
2006 Nutrition and hydration at the end of life: pilot study of a palliative care
experience. Journal of Law and Medicine 14(2): 182-98.
Van der Riet, et al.
2008 Palliative care professionals' perceptions of nutrition and hydration at the end
of life. International Journal of Palliative Nursing 14(3): 145-51.
van Vuuren, Karen and Francesca Nicosia
2008 Dying Wish. Boulder, CO: WordWise Productions. 29 minutes.
1994 The Revival of Death. London: Routledge.
Washington State Department of Health
2010 Death with Dignity Act. http://www.doh.wa.gov/dwda/, accessed April 4,
White, K.S. and J.C. Hall
1999 Ethical dilemmas in artificial nutrition and hydration: food for thought.
Nursing Case Management 4(1): 34-6.
1990 The dilemma of withholding or withdrawing nutrition. Image: Journal of
Nursing Scholarship 22: 226-230.
1983 The dehydration question. Nursing 13: 47-51.